Bioasphalt and Biochar
from Pyrolysis of Urban Yard Waste
by
Daniel R. Hill
Submitted in partial fulfillment of the requirements
for the degree of Master of Science
Department of Civil Engineering
Case Western Reserve University
January 2012
CASE WESTERN RESERVE UNIVERSITY
SCHOOL OF GRADUATE STUDIES
We hereby approve the thesis/dissertation of
_Daniel R. Hill______
candidate for the _Master of Science______degree*.
(signed)_Aaron A. Jennings______
(chair of the committee)
_David Zeng______
_Xiong “Bill” Yu______
______
(date) _4 August 2011______
*We also certify that written approval has been obtained for any proprietary
material contained therein.
ii Table of Contents List of Tables……………………………………………………………………………..iii List of Figures……………………………………………………………………………..v Acknowledgements……………………………………………………………………..viii Abstract…………………………………………………………………………………...ix
1. Introduction ...... 2 2.1. Recent Interest in Pyrolysis as a Sustainable Technology ...... 3 2.2. Desirability of Non-Petroleum-Based Asphalt Binders ...... 10 2. Literature Review...... 12 2.1. Yard Waste Generation and Management...... 12 Generation and Current Management……………………………………………….12 Greenhouse Gas Emissions and Carbon Storage……………………………………13 Issues in Yard Waste Composting…………………………………………………..14 2.2. Pyrolysis ...... 16 Slow Pyrolysis………………………………………………………………………17 Fast Pyrolysis………………………………………………………………………..17 Reactor Types……………………………………………………………………….18 2.3. Biochar ...... 23 Properties of Biochar………………………………………………………………..23 Carbon Sequestering Potential of Biochar…………………………………………..25 Added Environmental Benefits of Biochar…………………………………………26 2.4. Bio-Oil ...... 27 Chemical Properties of Bio-Oil……………………………………………………..27 Physical Properties of Bio-Oil………………………………………………………29 Applications of Pyrolysis Liquids…………………………………………………..30 2.5. Asphalt Binder Properties and Testing Procedures ...... 35 Rheological Properties of Binders…………………………………………………..35 Pavement Deformations Related to Binder Rheology………………………………37 Testing and Specifications for Binders……………………………………………...38 Specifications for Asphalt Binders in Ohio…………………………………………43 Testing on Bio-Oil…………………………………………………………………..44 3. Materials and Methods ...... 44 4.1. Sample Collection ...... 44 4.2. Sample Preparation ...... 47 4.3. Testing Methods ...... 49 Drying……………………………………………………………………………….49 Pyrolysis…………………………………………………………………………….51 Chemical Evaluation of Condensate………………………………………………...59 4. Results and Discussion ...... 64 5. Summary and Conclusions ...... 75
iii 6. Appendix ...... 79 6.1. Appendix A: Methods and Results of Growth Experiments ...... 79 Experiment 1 - Soybeans……………………………………………………………79 Experiment 2 – Corn………………………………………………………………...82 Experiment 3 – Soybeans…………………………………………………………...84 6.2. Appendix B: Ohio Asphalt Binder Specifications ...... 88 7. References ...... 90
iv List of Tables
Table 1: Example Organizations Promoting Biochar Use ...... 6
Table 2: Moisture and Pyrolysis Testing Results for All Samples ...... 64
Table 3: Moisture and Pyrolysis Testing Results for Branch Samples ...... 66
Table 4: Moisture and Pyrolysis Testing Results for Leaf Samples ...... 66
Table 5: Evaluation of Pre-Drying Effect on Yields and Byproduct Collection ...... 68
Table 6: Preliminary Chemical Analysis of Off-Gas Condensate ...... 74
Table 7: Yield Results from First Biochar Growth Experiment with Soybeans...... 79
Table 8: Yield Results from Second Biochar Growth Experiment with Corn ...... 82
Table 9: Yield Results from Third Biochar Growth Experiment with Beans...... 85
v List of Figures
Figure 1: Nutrient Poor Oxisol (left) and Terra Preta Oxisol (right) ...... 4
Figure 2: Terra Preta Formations Discovered in Brazil ...... 5
Figure 3: Initial Bench Scale Pyrolysis Reactor Setup ...... 7
Figure 4: Soybean Plants in Biochar Growth Experiments ...... 8
Figure 5: Corn Plants in Biochar Growth Experiments ...... 9
Figure 6: Schematic of Fluidized Bed Reactor ...... 19
Figure 7: A Rolling Thin Film Oven ...... 42
Figure 8: Pressure Aging Vessel ...... 43
Figure 9: Yard Waste in Biodegradable Bags ...... 46
Figure 10: Yard Waste Brush Pile (Oak Branches) ...... 47
Figure 11: Yard Waste Log Pile ...... 47
Figure 12: Rhododendron Leaves Prepared for Pyrolysis ...... 48
Figure 13: Oak Branches Prepared for Pyrolysis ...... 49
Figure 14: Dried Willow Branches ...... 50
Figure 15: Dried Black Locust Leaves ...... 50
Figure 16: Pyrolysis Reactor Connected to Discharge Piping in Furnace ...... 52
Figure 17: Heated Piping Directing Off-Gases into Collection Flask ...... 52
Figure 18: Second Collection Flask and Water-Cooled Condensers ...... 53
Figure 19: Schematic of Pyrolysis Reactor and Collection System ...... 53
Figure 20: Second Collection Flask Filled with Dense Off-Gases ...... 55
Figure 21: Maple Branches Before Pyrolysis ...... 55
Figure 22: Maple Branches After Pyrolysis (Char) ...... 56
Figure 23: Maple Samaras Before Pyrolysis ...... 56
vi Figure 24: Maple Samaras After Pyrolysis (Char) ...... 57
Figure 25: Off-Gases Concentrating in Glass Fittings Above First Collection Flask ...... 58
Figure 26: Setup of Boiling Reflux Flasks Under Condensers for COD Test ...... 61
Figure 27: Comparison of Byproduct Collection and Moisture Content ...... 68
Figure 28: Byproduct Collection Over Project Duration ...... 69
Figure 29: Moisure Contents of Deciduous and Coniferous Samples ...... 70
Figure 30: Organics Contents of Deciduous and Coniferous Samples ...... 71
Figure 31: Bottled Pyrolytic Liquids ...... 73
Figure 32: Dry Soybean Stem Yield versus Char Levels in Soil ...... 81
Figure 33: Dry Bean Yield versus Char Levels in Soil ...... 81
Figure 34: Dry Plant Yield versus Biochar Levels in Soil ...... 83
Figure 35: Dry Plant Yield versus Topsoil Level in Pot ...... 84
Figure 36: Dry Bean Yield versus Biochar ...... 86
Figure 37: Dry Bean Yield versus Topsoil Addition Levels ...... 87
Figure 38: Dry Bean Yield versus Sand Addition Levels ...... 87
vii Acknowledgements
I would like to thank my thesis advisor, Dr. Jennings, for his support, guidance, and patience. This work would have been impossible without his input. I would also like to thank both Dr. Zeng and Dr. Yu for serving on my defense committee and for their questions and comments. Their advice is greatly appreciated.
I would also like to thank my family for their support and encouragement throughout the duration of this project.
viii
Bioasphalt and Biochar
from Pyrolysis of Urban Yard Waste
by
Daniel R. Hill
Abstract
Pyrolysis is proposed as an alternative management method to composting for
urban yard waste recycling. Pyrolysis of yard waste biomass creates both biochar, a
carbon-sequestering soil amendment, and bio-oil, a viscous liquid that could be used as a renewable source for non-petroleum-based asphalt binders. A bench scale pyrolysis reactor was used to test over 50 samples of yard waste for moisture, organics, and char content. The average moisture content of all samples was 31.0%, average organics content was 30.8%, and average char yield was 38.1%. Branches and leaves showed similar char yields and byproduct recovery rates. Higher feedstock moisture did not hinder recovery rates, but dewatering was necessary for a quality bio-oil of sufficient viscosity. Aqueous fractions of bio-oils had high COD levels, and may require additional attention for disposal as wastewater.
ix 1. Introduction
This project investigates the utilization of pyrolysis as a management method for the
recycling of yard waste. Pyrolysis is a process encompassing a variety of methods for the
thermal degradation of organic substrates in the absence of oxygen. Pyrolysis is a process
that has been used for the production of charcoal for millenia. Currently, the most frequent management method for the recycling of yard waste in the United States is composting. With this method, the yard waste recycling only leads to one product: compost. Pyrolysis of all biomass feedstocks (including that of yard waste) can produce multiple products: biochar (a solid product) and bio-oil (a liquid product). Biochar can be used as a carbonaceous soil amendment which improves fertility and serves as a carbon sequestering agent, preventing additional greenhouse gas (GHG) emissions. Bio-oil consists of viscous organics which can have a variety of applications, including the potential use as a non-petroleum-based asphalt pavement binder. Both of these products could play roles in reducing greenhouse gas (GHG) emissions from waste management and reducing the demand of fossil fuels within the asphalt industry.
This project was designed to explore the feasibility of recovering sufficient amounts of bio-oil from yard waste pyrolysis to support the production of bioasphalt. A successful yard waste pyrolysis process could produce a non-petroleum-based asphalt binder that would be of value to the transportation industry. This product results from a carbon negative process that sequesters more CO2 than it releases. The amount of bio-oil that can
be produced from yard waste, the physical and chemical properties of this bio-oil, the
pyrolysis operating conditions under which bio-oil production is optimized, and a host of
other unit process and operation details are all unknown, but the potential advantages of
2 this yard waste management strategy are intriguing. The work described here represents original research that begins to answer basic questions about the potential of yard waste pyrolysis to produce a practical bio-oil product that could be used to produce bioasphalt.
The fundamental approach applied in this research is described in a later section.
Before discussing these project details, it seems prudent to document the context from
which this project emerged. Pyrolyzing yard waste is not an obvious approach for yard
waste management or the production of useful byproducts. It is, however, plausible and
emerged as an option during exploratory research based on a growing interest in
pyrolysis as a sustainable technology. Therefore, the following sections present some of
the background information that led to this project.
1.1. Recent Interest in Pyrolysis as a Sustainable Technology
Processes based on heating an organic substrate in the absence of oxygen (known as
pyrolysis) and in the absence of a bulk liquid phase (known as anhydrous pyrolysis)
predate recorded history. Pyrolysis has been used to produce primary products (e.g. char,
charcoal, coke) and byproducts (e.g. tar, pitch, resin) for millennia. Modern applications
have led to a wide variety of process innovations (e.g. flash and fast pyrolysis), reactor
configurations, and target products. Pyrolysis is used to produce fuels and other liquid
and gas phase organics, as a method to analyze complex organics, and to manage organic
wastes (e.g. for scrap tires). However, until recently, its energy demands and byproduct
discharges would probably not have placed it high on the list of sustainable technologies.
This is changing because of recent interest in pyrolysis conducted to produce soil
amendment charcoal (hereafter referred to as biochar) or bio-oils.
3 Pyrolysis emerged as a technology of renewed interest following the rediscovery of terra preta do indio, or terra preta, which refers to dark, unexpectedly fertile, pre-
Columbian, anthropogenic soils of the Brazilian Amazon basin (see Figures 1–2). The formations were reported in the 1950s but not documented until the publication of Wim
Sombroek’s book Amazon Soils in 1966 (Maris 2006). The formations are noteworthy because they apparently sustained large human populations (ca. 450 BC–950 AD) in areas otherwise characterized by thin and unproductive soils believed to be incapable of sustaining agriculture. Similar formations have since been identified in Ecuador, Peru, southern and western Africa, and in the Far East. It is clear that they are anthropogenic, but it is not known if they were created intentionally or as the byproduct of other activities (the former is considered to be most likely). It is known that their key ingredient was charcoal, which can moderate soil pH, improve moisture retention, enhance mineral availability, and support beneficial microbial activity.
Figure 1: Nutrient Poor Oxisol (left) and Terra Preta Oxisol (right) (Verheijen et al. 2009)
4
Figure 2: Terra Preta Formations Discovered in Brazil (adapted from Lima et al. 2002)
The potential benefits of biochar soil amendments are thought to be substantial. The process sequesters carbon (in soil, charcoal appears to have a half-life of hundreds of years) and may improve crop productivity. This potential has spawned numerous international organizations dedicated to the development of the process (see Table 1).
Groups worldwide are conducting field trials to examine crop yield improvements and soil scientists are working to quantify the mechanisms by which biochar improves soil productivity. Equipment manufacturers are also showing interest. Companies in at least
13 nations (Australia, Brazil, Canada, Denmark, Germany, Hungary, India, Korea, the
Netherlands, New Zealand, Russia, the United Kingdom, and the United States) are producing biochar equipment or running pilot scale reactors (Rasmussen 2009).
Commercial scale plants are operating in Australia (ANZBRN 2009), Canada
5 (Dynamotive Energy Systems 2009), India, Brazil, the Philippines (Australian Biochar
2009), and in the United States (Austin 2009; Gunther 2009).
Table 1: Example Organizations Promoting Biochar Use
Organization Internet Address
Biochar.org http://www.biochar.org/joomla
International Biochar Initiative http://www.biochar-international.org/node/648
Biochar Fund http://biocharfund.org
The UK Biochar Research Center http://www.geos.ed.ac.uk/sccs/biochar
Biokohle.org http://biokohle.org
Australia and New Zealand Biochar Research Network http://www.anzbiochar.org
Canadian Biochar Initiative http://www.biochar.ca
Wiser Earth European Biochar Initiatives http://www.wiserearth.org/organization
Outback Biochar http://www.outbackbiochar.com
Biocharinfo (CarboZero Foundation) http://www.biochar.info
Biochar Europe http://www.biochar-europe.org
Support Biochar http://www.supportbiochar.org
Numerous agricultural and forestry byproducts have been evaluated as feedstocks for pyrolysis, but it appears that no one has examined the potential benefit of using urban yard waste as the feedstock. This is a mistake! The improved agricultural productivity from biochar alone may make the management of yard waste with pryolysis financially viable. The sequestering of CO2 and bio-oil production provide additional economic value and convert an even higher portion of the waste into useful products. Taken together, the merits of yard waste pyrolysis present exciting new opportunities in solid waste management, agriculture, and trasportation materials research.
It should be noted that there is a wide variety in the technical sophistication being applied to pyrolysis research. Many researchers have devised sophisticated bench-top reactors to explore the kinetics and thermodynamics of the process. However, these reactors use very small samples (often only a few grams) that have been processed to small particle sizes by grinding, shredding, or pulverizing. They can produce valuable
6 data, but are not appropriate for “production” pyrolysis. To produce larger quantities,
reactors must accommodate larger volumes of less-processed feedstock. Production reactors range from crude units constructed from steel drums to sophisticated mobile and fixed production plants.
The reactor used for this research is a bench-top unit developed by modifying a muffle furnace to house a 0.95 L pyrolysis chamber. This reactor was used to study operating conditions and product yields. It can operate at temperatures greater than 500
°C, has digital controllers that allow for ramped temperature profiles, and has off-gas condensers for improved byproduct recovery (see Figure 3).
Figure 3: Initial Bench Scale Pyrolysis Reactor Setup
This reactor is adequate for the production of biochar, and to produce enough material to support plant growth studies conducted in biochar-amended soils. CWRU is adjacent
7 to urban areas with high yard waste generation rates, so the examination of yard waste as
a feedstock for pyrolysis began. This led to the evaluation of yard waste components and
to plant growth studies conducted in CWRU’s research greenhouse. The growth studies
were not intended to duplicate the worldwide effort to evaluate the agricultural benefits
of biochar, but to determine if there was anything unusual about biochar derived from
yard waste that would inhibit its use. Results were promising and showed no evidence of growth inhibition (see Figures 4–5). The work described here grew out of an observation made during the effort to manufacture sufficient biochar to support growth studies.
Figure 4: Soybean Plants in Biochar Growth Experiments
8
Figure 5: Corn Plants in Biochar Growth Experiments
The results of the growth studies were mixed. For “good” soils (i.e. soils that would
be characterized as topsoil), no significant improvement in crop yield could be detected
and at the highest amendment levels used (50% biochar by volume), yield decreased.
However, in soils that would be characterized as “poor” (i.e. sand or glacial till), biochar
amendments improved yield by approximately 20%. These data are included in Appendix
A.
During the pyrolysis of sufficient yard waste to support these growth studies, an operational problem was encountered with the pyrolysis reactor. It was discovered that once the off-gas byproducts left the reactor, they cooled rapidly and could condense and plug the discharge piping. Unchecked, this could become a significant operational problem. However, it was soon realized that this could also be an opportunity. The thick, viscous, tar-like product has several potential applications. Based on this observation,
9 efforts started to explore the potential of using pyrolysis to produce raw bio-oils of sufficient quantity and quality to be of use in the transportation industry for use as a non- petroleum-based asphalt binder.
1.2. Desirability of Non-Petroleum-Based Asphalt Binders
The asphalt industry is faced with increasing prices and decreasing availability of the petroleum used in making conventional asphalt pavement. The U.S. Bureau of Labor
Statistics reports that asphalt binder prices have risen by 25% in the last five years and spiked at more than 300% in 2008. Demand for asphalt was expected to reach nearly 40 million tons in 2009 and demand will undoubtedly lead to additional cost increases.
Concerns about the cost, availability, and environmental baggage of using petroleum-
based materials have led to the evaluation of several alternative materials. Research has
examined the use of wood cellulose, wood lignin, bottom and fly ash, waste tires, and
coal mining waste as pavement components. Of these wastes, bio-oil produced from
cellulose or waste tires is thought to represent the most environmentally friendly,
abundant, and cost effective opportunity for reducing the amount of petroleum-based
materials in asphalt binders.
Research has shown that lignin and lignin products have potential for use in the
asphalt industry. Significant among these studies was the work conducted by Gargulak
and Lebo (1999), and Sundstrom et al. (1983). This work explored various lignin uses,
including as asphalt binders, concrete admixtures, well drilling mud, in dust control, in
vanillin production and in dispersants. Williams et al. (2008) also conducted research that
used fast pyrolysis and fractionation to extract lignin and lignin products from organic
feedstocks. Researchers have also investigated the use of lignin as a biological polymer in
10 retarding the aging (or oxidation) of asphalt pavements (Bishara et al. 2005; Dizhbite et al. 2004; McCready and Williams 2007; Ouyang et al. 2006). This function of lignin bio- oil serves to prolong asphalt pavement life by reducing aging-related failures such as thermal and fatigue cracking. In the latest research conducted by McCready and Williams
(2007), lignin was found to have a profound effect on widening the performace grade range of asphalt binders. However, as observed by Allen (1980), the properties of bio-oils produced are influenced by the source of the biomass feedstock, and by the condition under which they are manufactured. This is especially noteworthy given the wide range of variability in the materials that comprise yard waste.
In 2005, USEPA commissioned an assessment of pavements that included non- petroleum-based asphalts (Cambridge Systematics, Inc. 2005). USEPA’s interest stemmed from the potential of these products to reflect more sunlight and reduce heat island effects in urban areas. The Shell Oil Company has experimented with the use of vegetable oil-based asphalts on two Norwegian roads (Anderson et al. 2008). They found lower emissions levels than typical petroleum-based binders. Ecopave Australia has also released a bioasphalt product called GEO320 made from sugar and molasses (Johnson
2005). Although innovations are being pursued overseas, there is little mention of research on non-petroleum-based asphalts on the official websites of the National Asphalt
Pavement Association (NAPA 2009), the Asphalt Institute (AI 2009), or the National
Center on Asphalt Technology at Auburn University (NCAT 2009). In fact, the Asphalt
Institute’s mission states that its chief interest is “to promote the use, benefits, and quality performance of petroleum asphalt.” It appears that research on non-petroleum-based binders would be on the cutting edge of this emerging technology.
11 2. Literature Review
2.1. Yard Waste Generation and Management
Generation and Current Management
USEPA reported that 32.9 million tons of yard waste were generated in 2008. This represented 13.2% of the total municipal waste stream. From the generated yard waste,
21.3 million tons (64.7%) were recovered for either reuse or recycling. This number did
not take into account the extent of source reduction at the household level through
backyard composting, leaving grass clippings to decompose on the lawn, or other
methods (USEPA 2009). As of 2006, 23 states had at least a partial ban on landfill
disposal of yard waste. The types of bans vary. For example, Nebraska bans leaves and
grass disposal from April 1–November 30, but in Maryland, yard waste collected
separately from municipal solid waste (MSW) is banned from landfill disposal year-
round (Arsova et al. 2008).
Composting programs are the most prevalent management method for yard waste
recycling, and 3,505 composting programs were reported in 2007. Most of these
composting programs were in the Midwest region of the U.S. (approximately 1,600). The
Northeast region had the second highest number of programs with approximately 1,100
(USEPA 2008). Another less prevalent management method of yard waste recycling is
mixed waste composting. This is different than typical yard waste composting because
yard waste is not handled on its own, but is combined with food waste, paper, and wood.
In 2007, there were 16 mixed waste composting facilities, handling approximately 1,500
tons per day (USEPA 2008).
12 There are three main types of yard waste composting, which are signified by their type of management process: windrow, aerated static pile, and in-vessel composting.
Windrow composting is most frequent, and consists of leaving yard waste in long piles open to the air for decomposition. The size of the windrows is variable, and depends on the type of windrow turner used. Turning is necessary at regular intervals to maintain proper aeration. Typical sizes of windrows are approximately 15 m long by 5 m wide by
2.5 m high (Brewer and Sullivan 2003). Buffer zones around windrows are suggested to be 60 m for purposes of odor control (Komilis and Ham 2004). Depending on the feedstock, the windrows will mature over a range of 130–170 d (Brewer and Sullivan
2003; Komilis and Ham 2004).
In many areas, yard waste is disposed of in sanitary landfills. However, USEPA strongly discourages this practice and many authorities at the state, county, and municipal levels have mandated other forms of disposal. Often communities must collect and dispose of yard waste separately from municipal solid waste. With ever-increasing personnel demands and costs for equipment and fuel, separate yard waste collection is an increasing financial burden. Furthermore, there are few waste management alternatives for yard waste.
Greenhouse Gas Emissions and Carbon Storage
Estimates of total greenhouse gas (GHG) emissions from yard waste composting are unavailable from USEPA, as the biogenic emissions from yard waste composting are not calculated. Following guidelines from the Intergovernmental Panel on Climate Change,
USEPA focuses on tracking the extent of anthropogenic GHG emissions, as these emission sources are subject to human control. USEPA estimates of GHG emissions from
13 yard waste composting include only the collection and transportation of the waste from its source to the composting facility, and the mechanical turning of the windrows for aeration. These are the only two phases in yard waste management with any anthropogenic GHG emissions. The combined emissions of these two phases amount to
0.01 metric tons carbon equivalent (MTCE) per ton of yard waste compost, but given the estimates of carbon storage using yard waste compost, there is a net carbon flux of −0.05
MTCE per ton of yard waste compost (where a negative sign denotes carbon storage)
(USEPA 2006). This signifies that in terms of anthropogenic GHG emissions, yard waste composting is a healthy technology. But as stated before, this number ignores the biogenic emissions of CO2 and other GHGs. With yard waste pyrolysis, biochar application sequesters greater fractions of carbon.
Issues in Yard Waste Composting
Although composting of yard waste is a method which recycles a significant fraction of its portion of the MSW stream (64.7%), there are negative issues with the resulting product. Research investigating the effect of pesticides such as 2,4-
Dichlorophenoxyacetic acid (2,4-D) in yard waste composts has been conducted. Michel et al. (1995) found that only 47% of added 2,4-D was mineralized (i.e. experienced complete biodegradation). Less than 1% of added carbon was present in water from the mature compost, signifying a small potential of leaching problems from windrows.
Müller et al. (1998) documented the potential for thermophilic microorganisms to degrade most major types of environmental pollutants (alkanes, benzene and toluene, chlorinated hydrocarbons, etc.). Typical yard waste composting will enter this thermophilic phase once the active respiration of mesophiles raises the internal
14 temperature of the pile at the beginning of the composting process (Fogarty and Tuovinen
1991). Despite the presence of thermophilic degradation in yard waste composting, total
elimination of herbicides does not seem to be achieved. The degree to which composting
manages biological and chemical yard waste contamination remains an open question.
Another issue surrounding yard waste composting is the establishment of a standard
evaluation of compost maturity. Yard waste compost maturity is not the same as stability.
Stability measures the resistance of further degradation of the compost, and is determined by oxygen uptake rates or CO2 evolution rates. Maturity describes the effectiveness of a
compost used for a specific function (e.g. mulch for landscaping versus a soil amendment
for vegetable crops). The effectiveness of various tests will vary depending on the
feedstock source and on the end use of the compost. Tests have ranged from C:N ratio,
inorganic N concentrations, and cation exchange capacities (Sullivan and Miller 2001;
Brewer and Sullivan 2003).
Given these potential problems with yard waste composting, the management of this solid waste can pose difficulties for communities. Composting yard waste is not an ideal solution for multiple reasons:
• Composting has large demands on time, space, and energy.
• Yard waste compost is not in high demand as a consumer product.
• Composting yard waste can have aesthetic problems (e.g. odor)
• Composting yard waste can lead to concerns about plant and animal pathogen
distribution which may not be fully deactivated in the composting process.
• Composting yard waste does not fully destroy the residuals of lawn care
chemicals such as insecticides and herbicides.
15 • Composting yard waste releases all of the CO2 potential of the organic waste.
2.2. Pyrolysis
Pyrolysis is a process of thermochemical decomposition of organic material in the
absence of oxygen. Pyrolysis methods yield products in the solid, liquid, and gas phases.
Specific products and amounts in each phase depend on the feedstock and the process
method used. In general, methods with slower heating rates and longer off-gas residence times produce a greater fraction of solid products (e.g. charcoal, char, or biochar), while methods with faster heating rates and shorter off-gas residence times produce a greater fraction of liquid products. Shorter residence times prevent the off-gases or vapors from continuing to react with each other (Mohan et al. 2006). Typical methods of pyrolysis include traditional/carbonization, conventional/“slow,” “fast,” flash-liquid, flash-gas, hydro-pyrolysis, methano-pyrolysis, ultra pyrolysis, and gasification (Bridgwater 2003;
Demirbas 2005). The main product of interest from traditional and conventional/slow pyrolysis is charcoal or char, and yields are typically around 35% by weight. In fast and flash-liquid pyrolysis the product of interest is liquids with yields around 75% by weight.
Flash-gas, ultra pyrolysis, and gasification pyrolysis processes focus on the production of gas, with yields around 85% (Bridgwater 2003; Demirbas 2005). The difference between slow and fast pyrolysis generally based on temperature and heating rates, but there is no explicit distinction between the two (Mohan et al. 2006). Specific characteristics, properties, and applications of the products of pyrolysis are discussed in the following sections.
16 Slow Pyrolysis
Slow pyrolysis has low heating rates (1–20 °C/min) and requires char residence times
ranging from hours to days (Brewer et al. 2009). Slow pyrolysis is similar to the
traditional method of producing charcoal, which has been used for millennia. In
traditional methods of pyrolysis, piles of wood were covered with either mud or bricks in
order to restrict oxygen availability. This setup results in the highest charcoal yield of all
pyrolysis processes. Pyrolysis leaving a pure or near-pure carbon is also referred to as carbonization. Modifying slow pyrolysis process characteristics like heating temperature and heating rate can lead to higher yields for bio-oils (Williams and Besler 1996).
Fast Pyrolysis
Fast pyrolysis is performed at higher heating rates (up to 1000 °C/min) and at carefully controlled temperatures (close to 500 °C) (Brewer et al. 2009; Bridgwater
2003). The development of fast pyrolysis began during the oil crisis in the 1970s with experiments to develop liquid fuels from pyrolysis liquids. The feedstocks of interest were lignocellulosic biomass sources (Czernik and Bridgwater 2004).
Fast pyrolysis processes focus on the production of liquid products, referred to in the
literature as pyrolysis oils, pyrolytic liquids, bio-oil, or biomass oils. The essential
features of fast pyrolysis reactors are (1) high heating and heat transfer rates, (2) a
moderate and carefully controlled reaction temperature around 500 °C for maximization
of liquid products, (3) short vapor (or off-gas) residence times of around 2 s or less, and
(4) rapid cooling and condensation of off-gases in order to produce liquid bio-oil product.
There are many different reactor configurations that have been developed for use in fast
pyrolysis. These include ablative, fluid bed, circulating fluid bed, entrained flow, and
17 rotating cone reactors. Apart from the reactor design, other important design
considerations include pre-drying of the feedstock to minimize the water fraction of the
liquid product, grinding to ensure high heating and heat transfer rates to particles, and
char separation from liquid as its presence accelerates polymerization and increase of
viscosity (Bridgwater and Peacocke 2000).
Reactor Types
• Ablative
Ablative reactors receive ground biomass particles and apply high pressure against a
heated surface at high speeds and elevated temperatures (e.g. 600 °C). There are different
configurations of this reactor type, including ablative mill, tube, and vortex reactors.
Ablative reactors receive biomass from screw feeders which are in turn fed with grinders
or hoppers that preheat the biomass and lower the moisture content of the feedstock.
Biomass particles are then applied with high pressure against heated solids within the
reactor. Steam or nitrogen purging avoids long residence times and excessive char
formation. Off-gases and products are then pushed through cyclones to remove char particles and directed through condensers to capture gases (Bridgwater and Peacocke
2000).
Examples of ablative reactors include the laboratory-scale reactor at the University of
Aston in the UK, the vortex configuration reactor at the National Renewable Energy
Laboratory, and the ablative tube at BBC in Canada. These were the only ablative reactors reviewed by Bridgwater and Peacocke (2000) that were still operational at the time of publication. Other reactors had been either dismantled or abandoned.
18 • Fluidized Bed
Fluidized bed and circulating fluidized bed reactors transfer heat from a source to the biomass particles with a combination of convection and conduction. The reactor requires small particles of usually not more than 3 mm, as larger ones will lower the heat transfer
(Bridgwater et al. 1999). The diagram in Figure 6 shows the schematic of a bench top
fluidized bed reactor constructed at the Center for Sustainable Environmental
Technologies at Iowa State University. A carrier gas transports biomass from a feeder
into a reactor surrounded by radiation heaters and thermocouples for temperature
monitoring. Residence times are around 0.5 s and the gas outlet leads to a cyclone for
char removal and an ice bath for gas condensation and capture (Iowa State Univ. 2010).
Bio-oil yields for fluidized bed reactors typically range from 55–75 wt % (Iowa State
Univ. 2010; Bridgwater and Peacocke 2000).
Figure 6: Schematic of Fluidized Bed Reactor (Iowa State Univ. 2010)
19 Fluidized bed reactors and circulating fluidized bed reactors are the most popular reactor configurations in fast pyrolysis due to their ease of operation and potential for scale-up. The technology is well understood and the heat transfer is very efficient in these reactors (Bridgwater and Peacocke 2000; Bridgwater 2003). Examples of these reactors in operation include those at the Institute for Wood Chemistry at BFH in Germany,
Dynamotive Inc. in Canada, Resource Transforms International in Canada, Union Fenosa in Spain, and at the University of Waterloo in Canada (Bridgwater and Peacocke 2000).
• Circulating Fluidized Bed
Advantages of circulating fluidized bed reactors are similar to fluidized bed reactors in that the technology is well understood and has superior temperature control. An advantage of circulating reactors over regular fluid beds is that the residence of char is the same as that of the off-gases. Char residence times are longer in regular fluid beds.
This is beneficial for greater production of pyrolytic liquids. Circulating fluid beds are suitable for large throughputs, but still need to be tested at larger scales (Bridgwater
2003).
• Entrained Flow
Two entrained flow reactors for fast pyrolysis were reviewed by Bridgwater and
Peacocke (2000). A downflow reactor constructed in 1991 for research in Belgium was utilized and with a mass flow rate of 84.0 kg/h yielded 16.2 wt % char, 39.9 wt % oil, and
14.9 wt % water. The reactor was abandoned in 1993 due to lack of funding. GTRI, a company based in the USA, constructed an entrained flow reactor for fast pyrolysis in
1983. Yields by weight ranged from 48-60% for oil (with water), and 8-20% for char,
20 with reactor temperatures ranging from 499-525 °C. The reactor was abandoned in 1990
(Bridgwater and Peacocke 2000).
Bohn and Benham (1984) researched a tubular entrained flow reactor which could be operated in a continuous mode, but required regular replacement of the costly high alloy tube due to its exposure to the heat source on the outside and to high-temperature steam on the inside.
• Rotating Cone
A rotating cone reactor for fast pyrolysis was developed from 1989-1997 at the
University of Twente in the Netherlands. Development began with an idea similar to ablative reactors in its method of heat transfer, with particles of biomass being slid across a heated metal surface. The feed of biomass particles is introduced onto an impeller mounted at the base of the rotating cone. The particles are then catapulted spirally up and around the interior surface of the cone along with an excess flow of inert particles (e.g. sand). The feed pyrolyzes as it travels along the surface, and the char is ejected from the top of the cone. This concept is compact and does not require a carrier gas which decreases demand for space and costs of secondary oil collection. Cone surface temperatures are similar to that of ablation surfaces previously described, at around 600
°C. Initial experiments set the rotation of the cone at 900 rpm. Further modifications have included the decrease of cone volume to decrease gas residence times, and the introduction of an internal recycling scheme for the sand once the char (that is collected with the sand) is combusted after removal from the reactor volume (Bridgwater and
Peacocke 2000).
21 • Transported Bed
The main feature of a transported bed reactor consists of a chamber in which high
temperature recirculated sand is contacted with biomass particles that are ground to
approximately 6 mm and dried to approximately 10% moisture. The reactor has an
upflow, where products are passed through a series of cyclones for solids removal, and
then off-gases are cooled and condensed in a multiple-stage system that recirculates gas for the heating of sand. The control is such that residence times for off-gases can be set very low, and optimized for chemicals production (on the order of hundreds of milliseconds). With controls set for longer residence times, the transported bed system more completely cracks the lignin fraction of the biomass feedstock and promotes the production of bio-oil (Bridgwater and Peacocke 2000).
• Vacuum Moving Bed
One example of the vacuum moving bed reactor system is the patented Pyrocycling process. This process is performed under reduced pressures, heating rates and temperatures than most fast pyrolysis processes (450 °C, total pressure 15 kPa). After the feedstock is ground and dried, it is introduced in a vacuum to the main part of the reactor where the feedstock is conveyed along two plates. These plates are heated by mixtures of molten salts that are maintained at 530 °C by a heater supplied with non-condensable gases captured from the pyrolysis process. A vacuum pump removes the off-gases and directs them through a series of condensers which separate products into heavy oils, light oils, and an aqueous fraction for recovery. Typical yields from this process with dried fir/spruce bark feedstock are 35 wt % bio-oil, 20 wt % pyrolytic water, and 34 wt % char, with the remainder being non-condensable gases. For a feedstock of spruce wood, yields
22 are 47 wt % oil, 17 wt % pyrolytic water, and 24 wt % char, with the remainder being non-condensable gases (Bridgwater and Peacocke 2000).
2.3. Biochar
In pyrolysis, the majority of the carbon present within the feedstock is fixed in the solid product that results from the process. This is referred to as char. The most typical type of char is charcoal, from wood and/or sawdust pyrolysis, which has been produced with traditional methods for millenia. Another form of char is coke, created from coal pyrolysis. The char of interest to this project is biochar, which is biomass-derived char with a specific end use: application as a soil amendment for both fertility enhancement and carbon sequestering.
Traditionally, chars have been used as combustible fuels or as activated carbons in adsorption applications such as air purification, groundwater remediation, and drinking water filtration. As previously mentioned, use of char as a soil amendment grew out of a rediscovery of terra preta soils in the Amazon basin in the 1960s. Initial interest on biochars grew out of its ability to improve soil fertility by supplying and retaining nutrients. Further interest on biochar grew from its potential to sequester carbon.
Properties of Biochar
Properties of biochar can vary with the process temperature, method of pyrolysis, and reaction time used in its formation. Brewer et al. (2009) compared the properties of chars from slow pyrolysis (at 500 ºC), fast pyrolysis (at 500 ºC), and gasification (at 730 ºC and
760 ºC) with two different feedstocks (switchgrass and corn stover). Analysis of physical properties included measurement of particle density and Brunauer-Emmet-Teller (BET) surface area. Chemical properties tested included measuring the contents of moisture,
23 volatiles, fixed carbon, and ash. Other evaluations included elemental contents, higher
heating values (HHVs), mineral content, and aromaticity. Findings indicated that fast
pyrolysis and gasification led to chars with smaller particle sizes than slow pyrolysis,
believed to be due to the more rapid volatilization leading to increased porosity and
fragmentation. Gasification chars were fine powders, while fast pyrolysis chars were very
fine powders with the lowest BET surface areas (e.g. with corn stover, 7.0 m2/g for fast
pyrolysis vs. 20.9 m2/g in slow pyrolysis). The chars had low BET surface areas overall
in comparison to commercial activated carbons. Particle densities of chars increased as
process temperature increased (Brewer et al. 2009).
Moisture, volatiles, fixed carbon, and ash contents were evaluated on a weight
percentage basis. Higher ash contents were observed in biochar samples with lower fixed carbon content. Both the switchgrass and corn stover chars had high ash contents (32.4–
54.6 wt %), with a significant portion of the ash content being silica. An increased silica content in ash can lead to slagging in fuel applications for char. Both the low BET surface areas compared to commercial activated carbons and the high ash and silica content signify that chars from switchgrass and corn stover might be better utilized as biochar soil amendments than as fuels or activated carbons (Brewer et al. 2009).
The investigations by Brewer et al. (2009) focus on the properties of the individual chars and not on the behaviors of the chars within and overall effect on the soil environment to which they are applied. The pyrolysis design parameters of process temperature and conversion method can also have an effect on the long-term stability of the char within the soil, the specifics of which still seem to be an unanswered question.
For example, biochar produced at 400 ºC had a much greater stability against oxidation
24 by ozone than biochar produced at 1000 ºC. Experimental results have varied as widely
as the choice of feedstocks for biochar. Both quick and slow decompositions of biochar
have been reported. Despite these varied findings, the black carbon that is the main
component of biochars has been found to be one of the oldest fractions of carbon in soils.
This signifies that on the whole, biochars may decompose in terms of the original
biochemical form but will not suffer significant mass losses (Lehmann et al. 2006).
Carbon Sequestering Potential of Biochar
Lehmann et al. (2006) published a review of the use of biochar as a carbon
sequestering agent and its potential for use on a wider scale. Consideration and evaluation
of urban wastes (yard waste in addition to site clearing, pallets, and wood packaging) as a
potential source for biochar production was included. The 3.9 × 1013 g of urban wastes
generated in the United States alone was estimated to have an annual biochar production potential of 1 × 1013 g/yr. Levels of urban wastes at the global level were not available,
but the biochar production potential on this scale was estimated to be 3 × 1013 g/yr. When
also considering other sources of biomass waste materials as feedstocks (forest residues,
mill residues, rice husks, and groundnut shells), total annual biochar production potential
was estimated at 1.62 × 1014 g/yr on the global scale (Lehmann et al. 2006).
Lehmann et al. (2006) also estimated the maximum levels of biochar applications that
soils could tolerate before crop and timber yields were adversely affected. This was
performed in an attempt to optimize soils as a carbon sink. This is a difficult question to
answer definitively, as different crops will respond to biochar-amended soils in different
ways. For example, beans grown with biochar additions of 6.0 × 107 g C per hectare resulted in yields similar to that of beans grown without biochar amendments, while most
25 results demonstrate increasing yields up to 1.40 × 107 g C per hectare. In general,
findings indicate that crops respond positively to biochar amendment levels up to 5.0 ×
107 g C per hectare (Lehmann et al. 2006). The best and most consistent yields with
biochar amendments are typically in soils that are highly degraded (Roberts et al. 2010).
A life cycle assessment (LFA) of biochar systems was conducted which included
estimations of the energetic and climate change potentials with three different types of
feedstocks: an energy crop (switchgrass), an agricultural residue (corn stover), and yard
waste (Roberts et al. 2010). The estimation of the net GHG emissions from 1 metric ton
(1000 kg) of yard waste feedstock (dry) was -885 kg CO2e (where a negative sign
denotes emissions reduction). This value was the largest reduction of all feedstocks
evaluated, mostly due to there being no emissions associated with the production of yard
waste, only transport. This end value results from three components: stable C, reduced
soil N2O emissions, and avoided compost emissions. Results from the assessment indicated that there is more energy produced through the slow pyrolysis/biochar system than consumed for each feedstock. With respect to yard waste specifically, the net energy of 1 metric ton (dry) is +4043 megajoules (MJ), where a positive value represents energy generated. This consists of +424 MJ from avoided composting, +3507 MJ from excess syngas and heat energy production, with the remainder being from avoided fossil fuel use
(Roberts et al. 2010).
Added Environmental Benefits of Biochar
Biochar applications have demonstrated emissions reductions of significant GHGs other than CO2. At additions of levels 20 g/kg, biochar addition to soil nearly eliminated
methane emissions. Biochar decreased nitrous oxide emissions from soybeans by 50%
26 and from grass stands by 80%. A possible explanation for this extended GHG
suppression is through the improved aeration that the biochar amendments provide to the
soil by stabilizing the carbon levels. A possible explanation for the lower levels of nitrous
oxide emissions is the higher C/N ratio, which slows N cycling within soils and therefore
emissions of nitrous oxides (Lehmann et al. 2006).
The benefits of biochar amendments are not limited to the suppression of GHG
emissions from soils. Black carbon present in soils which has a similar composition to
that of biochar can buffer ammonia and also exhibits ammonia adsorption. These actions
limit the volatilization of ammonia from soils and therefore from agricultural crops.
Biochars have also been shown to adsorb compounds such as ammonium, nitrate,
phosphate, and other hydrophobic organic pollutants. Despite these added benefits, this
has not yet resulted in economic applications for land users. At present, these added
environmental benefits of extended GHG emissions reductions and ammonia adsorption
from biochar addition are poorly understood, and warrant more research (Lehmann et al.
2006).
2.4. Bio-Oil
Chemical Properties of Bio-Oil
The chemistry of bio-oils is complex and the composition can contain up to 300
substances (Czernik and Bridgwater 2004). Because of this composition, the difficulty
associated with total chemical characterization of bio-oils is significant. Findings can vary widely based on the feedstock used for pyrolysis. Physical properties of bio-oil can provide more pertinent information indicating its potential performance as an asphalt binder.
27 • Corrosiveness/pH
Pyrolysis liquids have very low pH values, typically 2–3. This is due to the presence
of large amounts of organic acids, mostly formic and acetic. This corrosiveness affects
materials in construction such as aluminum, steel, and some sealing materials, but not
stainless steel. Higher water content and elevated temperatures can also increase the corrosiveness of the bio-oil (Czernik and Bridgwater 2004; Raouf and Williams 2010).
Storage has no effect on the pH value of the bio-oils (Czernik et al. 1994).
• Water Content
The water content of bio-oils varies depending on feedstock used in its formation. It
results from the feedstock’s original moisture content and from the dehydration reactions
occurring during pyrolysis. Typical water contents of bio-oils range from 15–30 wt %. At
these levels, water is typically miscible with oligomeric components derived from lignin
within the bio-oil. This is due to solubilization from polar hydrophilic compounds such as
low molecular weight acids, alcohols, and hydroxyaldehydes, which typically arise from
the decomposition of carbohydrates. Drying feedstocks before pyrolysis can decrease the
water content of the product bio-oil, and is desirable for some applications (Czernik and
Bridgwater 2004; Raouf and Williams 2010). Water content of bio-oils can increase in
storage, as condensation reactions can occur over time (Czernik et al. 1994).
Water content can have both advantages and disadvantages when present in bio-oils,
depending on application or end use. In applications for internal combustion engines, it
generally improves the flow capabilities by reducing oil viscosity, which can aid in the
pumping and atomization of the fuel. It also promotes a more uniform temperature profile
within combustion cylinders, and leads to lower NOx emissions. Its negative effects
28 include the increase of ignition delays, the decrease of combustion rates, and the lowering of heating values (Czernik and Bridgwater 2004; Raouf and Williams 2010).
• Volatility
Boiling temperatures of bio-oils can vary due to their chemical composition. Bio-oils can contain amounts of nonvolatile materials such as sugars and oligomeric phenolics apart from the volatile fractions of water and organics. Slow heating of bio-oils can result in polymerization of the different reactive components. This leads to bio-oils boiling below the level of 100 ºC. Distillation ends in the range of 250–280 ºC, which results in
35–50% of the original material as residue. This signifies that total evaporation before combustion is not possible with bio-oils, which is necessary for some applications
(Czernik and Bridgwater 2004). This property of bio-oils is important, as typical temperatures for handling asphalt fall in the range of 100–165 ºC. Because of this, bio- oils can be considered as alternative asphalt binders to petroleum-based asphalt pavement materials (Raouf and Williams 2010).
Physical Properties of Bio-Oil
As mentioned in previous sections, the analysis of the chemical structure and properties of bio-oil is difficult due to the complexity of the material. Therefore physical properties which are more easily quantified and measured can illustrate the behavior as potential replacements for petroleum-based asphalt binders. Physical and rheological properties of asphalt binders, including viscosity, are related to some significant distresses and fatigues of asphalt pavements such as raveling, cracking, rutting, and
29 stripping (Raouf and Williams 2010). These pavement distresses are discussed in a
following section.
Applications of Pyrolysis Liquids
Czernik and Bridgwater (2004) published a review of the applications of fast
pyrolysis liquids/bio-oils. There have been numerous projects in developing bio-oils for
producing liquid fuels and they have been tested in furnaces, boilers, engines, and
turbines, but few projects have achieved commercial success. Bio-oils have also been
developed for chemicals production.
• Liquid Fuel Combustion
One commercially successful application of bio-oil use in heat generation was
developed at Red Arrow Products in Wisconsin. Different mixtures of bio-oil, char, and
gas are used in a burner that provides space heating requirements for the entire plant. The
bio-oil is fed to the combustor separately from the char and gas, and is atomized with air
before delivery through a stainless steel nozzle. Analysis of emissions in 1994 had CO at
17%, NOx at 1.2%, and formaldehyde at 0.2% of permitted emissions levels (Czernik and
Bridgwater 2004).
Other research with a burner/boiler system has been conducted in Finland. The boiler was supplied by a dual fuel burner that utilitzed different mixtures of bio-oil and fuel oil.
Operation on bio-oil alone was possible with modifications to stabilize combustion.
Analysis showed that emissions were below permitted levels for CO and NOx, but
particulates were at high levels. Different bio-oils were tested and they performed less
ideally with high viscosities, water fractions, or solids fractions. In general, emissions
30 were lower than those of burning typical fuel oil apart from high particulates (Czernik
and Bridgwater 2004).
Projects developing bio-oils for combustion in diesel engines typically required the
addition of additives such as nitrated alcohol (methanol). Ignition delays were typically
longer than that of fuel oils, and problems were encountered with clogging from coke
formation and deposit formations on parts. Smoother operations were achieved with bio-
oils that were hot-filtered (particulates removed) due to lower molecular weights and
water content.
Bio-oils from slow pyrolysis of forest and agricultural residues were tested for
combustion in gas turbines in the late 1980s. These bio-oils had similar properties to fast pyrolysis bio-oils but higher carbon content and viscosity. Combustion efficiency was measured at 95%. Emissions of CO were higher but CH and NOx levels were lower than permitted levels of petroleum-based fuels. Slag buildup was identified as a potential future problem. Tests through the full power range of a 2.5 megawatt turbine found that
NOx and SO2 emissions were lower than that of diesel fuel, but particulates were higher.
Tests conducted on a 75 kilowatt dual fuel system, which at full power produced 73% of power that would be produced by a diesel fuel alone. Approximately 40% of the power came from bio-oil, and 60% came from diesel. In the dual fuel mode, emissions from CO
and CH were higher, and NOx levels were lower (Czernik and Bridgwater 2004).
Combustion in Stirling engines is typically applied for small-scale combined heat and power production. Bio-oil fuels applied to a 25 kW Stirling engine, observing that emissions were below German standards and that there were no apparent deposits or
31 residues. Despite this, the efficiencies of electric and thermal production were not as high as typical fuels (50–60%) (Czernik and Bridgwater 2004).
In general, the applications of bio-oil for use as transport fuels have problems with low heating value, incompatibility with typical transport fuels, high solids content, high viscosity, incomplete volatilization, and chemical instability. These problems can sometimes be solved with simple physical methods, but some need extensive chemical processing. Because of these problems, bio-oil as an addition to diesel fuel seems to be the simplest solution. This method is not economic, as the surfactants needed to emulsify the bio-oil and diesel are expensive and the mixed bio-oil/diesel fuels create higher levels of corrosion or erosion. Two main methods for upgrading bio-oil alone to a transport fuel are hydrotreating and catalytic vapor cracking, both resulting in the full deoxygenation of the bio-oil (Czernik and Bridgwater 2004).
• Chemical Production
While traditionally wood pyrolysis liquids and oils were a major source of chemicals like acetic acid, methanol, turpentine, and tars, the majority of these can now be produced at lower costs as products derived from natural gas, crude oil, or coal. Bio-oil is composed of over 300 identified compounds, but separation of every one of these is not cost-effective, and some amounts are small. Because of this, chemical production from bio-oil has focused either on the liquid as a whole or on the major groups present within the pyrolysis liquids, such as carbonyls, carboxyls, and phenolics (Czernik and
Bridgwater 2004).
Applications using the bio-oil as a whole include Dynamotive Corporation’s
BioLime, which uses the carboxylic acids and phenols present in bio-oil to react with
32 lime to form calcium salts. BioLime contains approximately 50% water and 7–14% calcium and is successful in capturing SOx emissions from coal combustors when
injected as a liquid suspension in the flue gas stream. The organic calcium salts and
compounds in BioLime are approximately four times more efficient in capturing acid
gases than lime acting alone (Czernik and Bridgwater 2004).
Bio-oil has also been used as an ingredient in a slow-release nitrogen fertilizer. The
high concentrations of carbonyl groups in bio-oil are reacted with ammonia, urea, or
other substances containing an NH2 group. When added to an organic matrix, the
resulting fertilizer shows lower leachability than typical mineral fertilizers. Its carbon
content also mimics the biochar fertilizers, where the carbon content is sequestered
(Czernik and Bridgwater 2004).
There are also chemical production applications of bio-oil from the fractionation or separation of bio-oil. The most common form of separating bio-oil is through water
addition, which separates the the water-soluble and water-insoluble fractions (the water-
insoluble fraction is also termed pyrolytic lignin). The water soluble fractions have been
used as meat browning agents and smoky flavors in food flavorings, and have been
commercialized by Red Arrow Products in Wisconsin (previously mentioned for using
bio-oils in a liquid fuel application). These products are in competition with products
mostly produced from slow pyrolysis. Another application of the water-soluble fraction
of bio-oil includes use as an environmentally-friendly de-icer, from its containing carboxylic acids. Although the application of de-icers from bio-oil is feasible in the technical sense, the expense of the process is not economic in the face of the
33 comparatively low cost of typical calcium chloride de-icers (Czernik and Bridgwater
2004).
The water-insoluble fraction of bio-oil, or pyrolytic lignin, can be applied as a replacement of phenol in phenol-formaldehyde resins. Replacements have been shown at
30–50% of phenol levels, and successfully used as adhesives in manufacturing of plywood and particleboard, although no commercial applications have yet been established. Commercial development has been undertaken by companies such as
Louisiana Pacific, Weyerhauser, and A.C.M. Wood Chemicals (Czernik and Bridgwater
2004).
The production of specific chemicals from bio-oil is also possible, but is expensive due to the difficulty of separation of bio-oil at large scales. Main chemicals of interest within bio-oils include glycolaldehyde, levoglucosan, and levoglucosenone.
Glycolaldehyde is typically the most abundant compound in bio-oil apart from water, and is the most active meat-browning agent in liquid smoke. Glycolaldehyde production from pyrolysis is more effective if the feedstock is primarily cellulosic, leaving pyrolysis liquids lignin-free. Levoglucosan has potential for the production of pharmaceuticals, surfactants, and biodegradable polymers, but has a high production price and is difficult to isolate from other pyrolysis liquids. The isolation of levoglucosenone from other pyrolysis liquids is relatively easier than that of levoglucosan, and has similar potential uses, but commercial application has not yet been achieved (Czernik and Bridgwater
2004).
34 • In Bio-Binders
Using fast pyrolysis oils for use as bio-binders has been pursued recently by Williams
and Raouf. Their research parallels the goal of this project, albeit with different
feedstocks and pyrolysis methods. Fast pyrolysis was utilized, and three homogeneous,
individual feedstocks were analyzed for use as bio-binders: corn stover, switchgrass, and oakwood (Raouf and Williams 2010). Results from their project are summarized in the next section.
2.5. Asphalt Binder Properties and Testing Procedures
Rheological Properties of Binders
Rheological properties of asphalt binders that affect field performance include
viscoelasticity, temperature susceptibility, shear susceptibility, and age hardening (or
oxidation). These characteristics are summarized below.
• Viscoelasticity
Viscoelasticity refers to the dual viscous and elastic behavior of a material when
subjected to loading and unloading. Petroleum-based asphalt binders demonstrate viscoelastic behavior because at high temperatures they behave as viscous liquids, while at low temperatures they behave as elastic solids, reverting back to original shapes when
unloaded. This dual behavior aids in the production process (Raouf and Williams 2010).
• Temperature Susceptibility
Temperature susceptibility is the rate of change of the binder consistency with a
change in temperature. Binders with high temperature susceptibilities are undesirable for
multiple reasons. At high temperatures viscosities can be too low and lead to mixing
35 problems during compaction, and at low temperatures viscosities can be too high and lead to low-temperature cracking. Because of this variability, viscosity must be assessed at high, low, and medium temperatures to fully evaluate the performance of the binder.
Petroleum-based asphalt binders are ideal for their job in part because of their specific temperature susceptibilities. At high temperatures, it behaves like a viscous liquid and facilitates mixing. At medium temperatures, it behaves as a viscoelastic solid. At low temperatures it behaves like an elastic solid, acting as a glue which holds the matrix of pavement aggregate together. Temperature susceptibility is quantified with the evaluation of viscosity temperature susceptibility (VTS), a unitless value. VTS calculation of 50 common binders in the United States varied from 3.36 to 3.98 (Raouf and Williams
2010).
• Shear Susceptibility
Shear susceptibility is the rate of change of viscosity with the shear rate. It is also referred to as the shear index. Newtonian fluids do not demonstrate shear susceptibility, as their viscosity is independent of shear rate by definition. Non-Newtonian fluids will demonstrate shear susceptibility, as their viscosities increase with shear rate. In general, a lower gain in shear susceptibility leads to better pavement performance.
• Age Hardening (Oxidation)
Rheological properties of asphalt binders change over the course of production and throughout the service of the pavement. Age hardening (or oxidation) is one process through which those properties can change. Oxidation occurs from the interactions of oxygen with the binder, which lead to an increase in stiffness. This is of particular interest with binders from bio-oils, as they consist of organic compounds which will react
36 with oxygen. Rates of oxidation increase at higher temperatures. Although age hardening is also referred to as oxidation, other overall contributing factors include polymerization and volatilization. Polymerization is the linking of similar molecules to form long chains of larger molecules, which increases hardening rates. Volatilization is the evaporation of lighter fractions of binders, which is a function of temperature. This evaporation does not contribute significantly to long-term aging. The overall combined effects of these three factors contribute an increase in binder viscosity (Raouf and Williams 2010).
Pavement Deformations Related to Binder Rheology
Pavement deformations that are related to the physical performance of the asphalt binder include raveling, cracking, rutting, and stripping. These deformations are not exhaustive, as there are other deformations related to only properties from the asphalt aggregate. Knowledge of these deformations and fatigues will give readers a better understanding of pavement performance as a whole and provide a better background for the purposes of physical testing on binders, discussed later within this report (Raouf and
Williams 2010).
• Raveling
Raveling is a deformation in which solid aggregate particles of the asphalt pavement become separated or dislodged. It occurs within the pavement from the surface downward or from the edges inward, and is an effect of the age hardening of the binder.
The increased viscosity from age hardening increases brittleness and decreases adhesiveness, which is why the solid aggregate particles become separated or dislodged
(Raouf and Williams 2010).
37 • Cracking
There are two different types of cracking related to binder properties: load associated
cracking and low-temperature cracking. Load associated cracking is also known as
fatigue or alligator cracking, and it is related to binder consistency. Low-temperature cracking results from binders exhibiting high stiffness at low temperatures. Other properties that may have an impact on low-temperature cracking are binder consistency and temperature susceptibility. In general, to avoid low-temperature cracking care must be taken to use a binder with sufficiently low stiffness in cold climate applications (Raouf and Williams 2010).
• Rutting
Rutting is a phenomenon where solid aggregate gradually moves under repeated loadings, like the formation of depressions from wheel paths. The shape and texture of the pavement aggregate are most closely related to these distresses, but utilizing pavement binders with higher viscosity or stiffness can aid in minimizing extensive rutting (Raouf and Williams 2010).
• Stripping
Stripping of pavements is the weakening of the bond between the binder and the aggregate, or loss of adhesiveness. It can result from moisture damage or incursion.
Utilizing pavement binders with a sufficiently high viscosity or stiffness can aid in minimizing stripping (Raouf and Williams 2010).
Testing and Specifications for Binders
The current system for testing of asphalt binders came from Superpave (Superior
Performing Asphalt Pavements), a product of the Strategic Highway Research Program
38 that was established in 1987. Among other things, Superpave established performance-
based asphalt binder specifications that directly related laboratory results to field
performance. Superpave specifications differ from previous specifications in overall
format. Physical and rheological property requirements for every performance grade must
be the same, but what differs is that the binder must maintain those same properties over
differing temperature margins (Raouf and Williams 2010). These temperature margins
range from an upper limit based on an average seven-day maximum design temperature
to a lower limit based on minimum design temperatures. Design temperatures are
dependent upon the climate of asphalt applications. Examples of different performance
grades are 58-28 (from -28–58 °C), and 64-22 (from -22–64 °C) (AI 2005). Binder specifications are extremely useful, but results from testing will not fully illustrate the performance of the binder in the field. Other characteristics that affect performance include mixing proportions and aggregate shape (Raouf and Williams 2010).
Physical and rheological testing on binders is typically performed on the original
binder, on residues from treatment with a rolling thin film oven, and on residues from
treatment with a pressure aging vessel. Both rolling thin film oven and pressure aging
vessel residues are designed to simulate the various phases of binder aging. Treatment
with a rolling thin film oven simulates binder aging and hardening from the production
and construction phases. Treatment with a pressure aging vessel simulates binder aging
and hardening from in-situ conditions and long-term service. Testing on these residues in
addition to the original, unmodified binder provides better indications of binder
performance in real world applications. Testing equipment and specifications for each are
described below (Raouf and Williams 2010).
39 • Rotational Viscometer
A rotational viscometer is used to measure the rotational/kinematic viscosity of the original binder at high temperatures (e.g. at 135 °C). The rotational viscometer equipment consists of a cylindrically-shaped spindle that is inserted into a temperature-controlled container filled with the binder sample. Torque applied from a motor rotates the spindle within the viscous sample. The torque necessary to maintain a constant speed for the spindle within the binder sample is measured and is directly related to the viscosity. For
Superpave performance grade asphalts, rotational viscosity must be no more than 3 pascal-seconds at 135 °C (AI 2005, Raouf and Williams 2010).
• Dynamic Shear Rheometer
A dynamic shear rheometer is utilized to measure binder properties at intermediate to high temperatures. Specific binder properties measured by dynamic shear rheometers are the shear modulus (G*) and phase angle (δ). The shear modulus is the ratio of the applied shear stress and the resulting shear strain. The phase angle is indicative of the viscoelastic behavior: a phase angle of 90° indicates fully viscous behavior, while a phase angle of 0°
indicates fully elastic behavior. Binders at 20 °C have typical phase angles of 45°.
Results from the dynamic shear rheometer are expressed as G*/sin(δ). For performance grade asphalts, this value must be 1.00 kilopascals or greater with the original binder. On rolling thin film oven residues, specifications require G*/sin(δ) values of 2.20 kilopascals or greater. On pressure aging vessel residues, Superpave requires G*/sin(δ) values of no more than 5,000 kilopascals (AI 2005).
40 • Bending Beam Rheometer
Bending beam rheometers are employed to demonstrate the stiffness at low (−12 °C or −18 °C) temperatures and cracking potentials of a binder sample. This testing is necessary to determine the low temperature performance grade for a binder, and is performed only on pressure aging vessel residues. Bending beam rheometers measures two values: the creep stiffness, S, and the m-value. For Superpave performance grade asphalts, S values must be 300 megapascals or less. The m-value for a binder must be
0.300 or more. If the m-value for a binder complies but the S value is too high, direct tension testing must be performed on the sample (AI 2005).
• Rolling Thin Film Oven
Samples for rolling thin film oven treatment are placed in cylindrical bottles and placed horizontally within the chamber of the oven (see Figure 7). These bottles are rotated while being subjected to a constant heating temperature and air flow, which gradually expose more thin films that simulate the binder covering the asphalt aggregate
(AI 2005). For Superpave specifications, the heating temperature is set at 163 °C for 85 minutes (Raouf and Williams 2010). The mass change is determined, as heating leads to volatilization of lighter fractions. Also, the residue is subjected to testing in a dynamic shear rheometer to determine if the viscoelastic properties of the original binder have been maintained. Typically, viscosity increases by 2-3 times through this oven treatment
(AI 2005; Raouf and Williams 2010).
41
Figure 7: A Rolling Thin Film Oven (Humboldt Mfg. Co. 2011a)
• Pressure Aging Vessel
Binder treatment with a pressure aging vessel simulates 5-15 years of service aging.
A typical pressure aging vessel is shown in Figure 8. This is done by subjecting a binder sample to high temperature and pressure for an elongated period (e.g. 100 °C at 2.10 MPa for 20 hr). In order to appropriately develop residues to simulate aging, the samples subjected to pressure aging vessel treatment must first be treated by a rolling thin film oven so that the resulting residue will demonstrate simulated aging from construction and service (AI 2005).
42
Figure 8: Pressure Aging Vessel (Humboldt Mfg. Co. 2011b)
Specifications for Asphalt Binders in Ohio
The Asphalt Institute maintains a directory of individual state specifications for performance grade asphalt binders. In Ohio, similar to Superpave specifications, tests performed on the original binder are specific gravity (ASTM D 70), flash point
(AASHTO T 48), rotational viscosity (AASHTO T 316), and dynamic shear (AASHTO
T 315). Tests necessary for the RTFO residue are mass loss (AASHTO T 240) and dynamic shear. Tests necessary on the PAV residue are dynamic shear, creep stiffness
(AASHTO T 313), and direct tension (AASHTO T 314). Necessary equipment for these tests include a rotational viscometer, dynamic shear rheometer, rolling thin film oven,
43 pressure aging vessel, and bending beam rheometer. The entire form with all
specifications is included in Appendix B.
Testing on Bio-Oil
Raouf and Williams (2010) tested physical, chemical, and rheological aspects of
binders from bio-oils. Physical tests included separation (ASTM D 7173) to evaluate
blending from polymer additions and specific gravity for comparison with typical
bitumen. Chemical testing consisted of Fourier Transform Infrared (FTIR) Spectroscopy
and Gas Chromatography/Mass Spectrometry (GC/MS) to quantify oxidative aging and
to identify chemical bonds and functional groups. Rheological testing consisted of
measurement of shear susceptibility, viscosity temperature susceptibility, and viscosity-
shear rate.
3. Materials and Methods
3.1. Sample Collection
Yard waste samples were collected from the Case Western Reserve campus, or
directly from tree lawns in neighborhoods and suburbs surrounding the campus. The
majority of samples originated from the cities of Cleveland Heights and Shaker Heights,
Ohio. Cleveland Heights (population = 46,000, area = 8.1 mi2) collects yard waste on a
weekly basis, but only from tree lawns. Yard waste must either be bagged or bundled for
collection (City of Cleveland Heights 2011). Shaker Heights (population = 28,000, area =
6.3 mi2) collects yard waste on a weekly basis as well. Yard waste must be bagged for
collection. In addition, loose leaf piles are collected directly from tree lawns in April and
from October 15–December 15. Most yard waste is hauled to a commercial composting
44 facility, although the city also provides free wood chips to residents (City of Shaker
Heights 2011). In both of these communities, residential lots are large and the
landscaping is mature. Both communities use a combination of standard collection
vehicles, vacuum trucks, front-end loaders, and street sweepers to manage yard waste.
Compost samples were also collected from public piles in the cities of Shaker Heights
and Beachwood, and from Boyas Excavating, Inc. in Valley View. Samples from 2008
were acquired in late June through mid-July. Samples from 2009 and 2010 were mostly acquired in late summer, from mid-July to mid-August. Different types of yard waste were collected as much as possible in this project because of the inevitable variability in the composition of the yard waste stream, as residents will not always be discarding waste from the same trees, bushes, or plants. Evaluation and testing of a range of samples would help to isolate any potential types or components of yard waste that would be most beneficial for the production of either bio-oils or biochar.
In summer, yard waste was deposited on tree lawns for pickup in three forms: in biodegradable bags, in piles of mixed waste, or in piles of waste from one specific origin
(see examples in Figures 9–11). Some waste was bundled, but most often it was not. The volumes of deposited yard waste varied greatly. Piles could range from only a pair of biodegradable bags to ten or more, or from a few branches from one type of tree to large piles of assorted trimmings stretching 15 feet or more along the curb. In autumn, most yard wastes were from fallen leaves and were not bagged. Piles of a mix of leaves were concentrated on the tree lawn or spilled out into the street slightly.
Samples were typically collected only if there was an appreciable amount of both leaves and branches for testing, and only if the origin of the waste (i.e. the type of plant)
45 could be identified. On some occasions, only selected components such as fruits, seeds, or nuts were collected (e.g. pine cones or buckeyes). Collected samples were preserved as collected in 5 gal plastic buckets or 20 gal lidded plastic bins. Some bins were vented to enable drying.
Figure 9: Yard Waste in Biodegradable Bags
46
Figure 10: Yard Waste Brush Pile (Oak Branches)
Figure 11: Yard Waste Log Pile
3.2. Sample Preparation
In preparation for testing, samples were separated into their components (i.e. leaves/needles, branches and/or stems, and fruit). They were reduced to a manageable size with either garden clippers or by hand. Such separation was not needed with single
47 component samples such as acorns or pine cones. Once prepared, samples were split into two aliquots: one for drying and the other for pyrolysis. Drying measured the moisture content, and pyrolysis testing measured organics content, char yield, and byproduct collection. Preparation of the yard waste samples for both the drying and pyrolysis processes were identical, but drying tests utilized smaller sample sizes. Prepared samples for testing are shown in Figures 12 and 13.
Figure 12: Rhododendron Leaves Prepared for Pyrolysis
48
Figure 13: Oak Branches Prepared for Pyrolysis
3.3. Testing Methods
Drying
To measure the moisture content of each yard waste component, specimens of known mass were left to dry in crucibles (also of known mass) overnight (approximately 18–20 hr) in a drying oven set at 97 °C. After drying overnight, the mass of the dried samples were recorded to determine the mass lost in the process (i.e. the moisture content). The moisture content (MC) of the yard waste sample as a percentage was calculated with Eq.
1.
MC (%) = 100 × [1 − (mD − mC)/(mW − mC)] (Eq. 1)
Where mC is the mass of the crucible, mW is the mass of the crucible with yard waste prior to drying (or “wet”), and mD is the mass of the crucible with dried yard waste. Figures
14–15 show two examples of samples after drying.
49
Figure 14: Dried Willow Branches
Figure 15: Dried Black Locust Leaves
50 Pyrolysis
Pyrolysis testing required a reactor that prevented sample exposure to oxygen as the
yard waste was heated. The principal part of the reactor used was a 0.95 L (1 U.S. quart)
steel vessel, modified with an off-gas discharge port that could be piped out of the muffle
furnace heat source and into an off-gas management system. This reactor was designed with an emergency blow-off lid that would prevent the generation of excessive pressure in the reactor if the discharge piping became plugged.
After it was filled with a sample of yard waste and sealed, the pyrolysis reactor was set in a muffle furnace (see Figure 16). The reactor was attached to copper discharge piping, directing off-gases into a 250 mL collection flask on top of the muffle furnace.
Adjustable electric heating tape was wrapped around the copper discharge piping in order
to maintain off-gas temperatures consistent with the heating setting of the muffle furnace.
The temperature of the pipe was monitored by thermocouples which were installed
underneath the heating tape. The first thermocouple monitored the temperature at the end
of the copper discharge piping, where a second thermocouple (added in later trials)
monitored the beginning of the pipe. The off-gases entered the collection flask through a
glass fitting that was connected with high temperature silicone sealing tape. This fitting
directed the gas flow downward, encouraging the condensation of denser products.
Products staying in gas form continued to flow up through a second fitting and towards a
second collection flask of volume 1,000 mL. The open top of the second glass fitting was
covered with a screw-on test tube cap to prohibit off-gas loss. Figure 17 illustrates the
setup of the copper discharge pipe, heating tape, thermocouples, and first collection flask
on the top of the muffle furnace.
51
Figure 16: Pyrolysis Reactor Connected to Discharge Piping in Furnace
Figure 17: Heated Piping Directing Off-Gases into Collection Flask
The second collection flask (1,000 mL) was attached to the first by plastic tubing. It was also connected to a series of water-cooled condensers. This was configured to allow condensing off-gases to drain back into the collection system. Any gaseous byproducts
52 that did not condense within the system were discharged through the exhaust of a laboratory hood. Figure 18 shows this section of the off-gas collection system. Figure 19 shows a schematic of the entire pyrolysis reactor system.
Figure 18: Second Collection Flask and Water-Cooled Condensers
Figure 19: Schematic of Pyrolysis Reactor and Collection System
53 In early trials of pyrolysis testing (2008 samples), the discharge piping was unheated
and directed the off-gases directly to the 1,000 mL collection flask and the water-cooled
condensers. This resulted in a larger fraction of the denser organics concentrating and
condensing within the discharge piping. This operational problem required constant
clearing and maintenance to achieve adequate gas flows. To avoid this operational
problem and to increase the collection of off-gases, the configuration with the 250 mL
collection flask was used. Furthermore, the addition of heating tape around the discharge
piping was implemented with thermocouples to monitor temperature. The impacts of these system modifications are discussed in the following section.
Pyrolysis testing began once setup was completed and all necessary connections were made between the reactor, discharge piping, flasks, tubes, and condensers. The muffle furnace was set to a temperature of 250 °C for approximately 45 min. After this initial period of heating, the furnace setting would be raised to 350 °C. While some off-gassing occurred during the first 45 min of the process (mostly steam), significant off-gassing
(dense white smoke) would begin to travel through the collection system after the interior temperature of the furnace would reach 350 °C (see Figure 20). The duration of pyrolysis at 350 °C was variable (depending on the rate of off-gas discharge), and heating was turned off once significant off-gassing into the first collection flask concluded. After the furnace had cooled, the reactor was detached from the collection system and the pyrolyzed yard waste sample was removed. Figures 21–24 show two different samples of yard waste both before and after pyrolysis.
54
Figure 20: Second Collection Flask Filled with Dense Off-Gases
Figure 21: Maple Branches Before Pyrolysis
55
Figure 22: Maple Branches After Pyrolysis (Char)
Figure 23: Maple Samaras Before Pyrolysis
56
Figure 24: Maple Samaras After Pyrolysis (Char)
The two collection flasks containing the byproducts were consolidated in a glass bottle after recording the byproduct mass collected in each flask. The byproducts were a mixture of water, organics, and small suspended particles of yard waste char. In early trials, a portion of off-gases escaped from around the cap on the top of the glass fittings.
In all trials, some off-gases would solidify within the glass fittings, within the plastic tubing between the flasks, or in the water-cooled condensers (premature condensation within the glass fittings is visible in Figure 25). After transferring the byproducts to a bottle, a fraction of bio-oil would be left within the collection flasks, as it was too viscous to flow into the bottle. This required an acetone wash of the collection flasks and glass fittings after each pyrolysis trial. The waste acetone from these washes were consolidated and dewatered to improve bio-oils collection.
57
Figure 25: Off-Gases Concentrating in Glass Fittings Above First Collection Flask
• Data Analysis
In order to calculate the organic content of yard waste samples, four parameters were
needed: the mass of the reactor (mR), the mass of the reactor with yard waste sample before pyrolysis (mN), the mass of the reactor with pyrolyzed yard waste (mP), and the
moisture content of the yard waste (calculated with Eq. 1). The organics content (OC) of
the sample by mass percentage was given with Eq. 2.
−1 OC (%) = 100 × [mN − mP − (mB − mR) × (MC/100)] × (mN − mR) (Eq. 2)
Char content, or yield (CY) as a percentage from each component of yard waste was calculated with Eq. 3.
−1 CY (%) = 100 × (mP − mR) × (mN − mR) (Eq. 3)
58 Byproduct collection efficiency was calculated with six parameters: the mass of the
reactor with yard waste sample before pyrolysis (mN), the mass of the reactor with
pyrolyzed yard waste (mP), the masses of the empty collection flasks (mA1 and mB1) and
the masses of the collection flasks with byproducts (mA2 and mB2). The byproduct collection was calculated with Eq. 4.
−1 Collection (%) = 100 × [(mA2 − mA1) + (mB2 − mB1)] × (mN − mP) (Eq. 4)
The results of Eq. 1–4 for all samples tested are included and discussed in the next
section.
Chemical Evaluation of Condensate
To evaluate the collected byproduct as a wastewater, numerous tests were undertaken with a collection of samples. Tests employed included a chemical oxygen demand (COD) test, and evaluation of pH, specific conductance, total solids, and total volatile solids.
• COD Evaluation
This test was performed as a measure of the oxygen equivalent of the organic matter content in a condensate sample. The evaluation of byproduct COD followed the open reflux method (508 A) in the sixteenth edition of Standard Methods for the Examination of Water and Wastewater, in which a sample is refluxed in a strongly acidic solution with an excess of potassium dichromate (K2Cr2O7) (APHA 1985). After refluxation and
cooling, the sample is titrated with ferrous ammonium sulfate to the endpoint of added
ferroin indicator to measure COD. The standard reflux time of 2 hr was reduced to 1 hr.
Reagents of potassium dichromate (0.0417M), and sulfuric acid were needed, as well as ferroin indicator solution, standard ferrous ammonium sulfate (FAS) titrant (at approximately 0.25M), and mercuric sulfate (HgSO4) powder. A series of 3-4 reflux
59 flasks were prepared with a small amount of mercuric sulfate powder and wastewater sample. One reflux flask was designated as a “blank” to compare with an unrefluxed blank flask to standardize the FAS solution with. The other flasks had differing amounts of sample to confirm the COD of the sample could be compared over different concentrations. If a wastewater sample size is too large, there is no residual potassium dichromate remaining after the reaction to titrate. So multiple flasks must be prepared to ensure that those levels are not exceeded. In the case of the condensate samples, 100-to-1 dilutions had to be prepared because all potassium dichromate was consumed in the reaction without dilution.
After combining the mercuric sulfate and condensate sample, 10 mL of potassium dichromate solution were added to each flask, as well as 6 glass beads (to avoid large bubbles developing during boiling/refluxation). The reflux flasks were connected to the series of water-cooled condensers and hot plates, and 30 mL sulfuric acid reagent were added through the top of the water jackets once the water stream was turned on. Hot plates were then set at boiling temperatures and maintained for 1 hr. The apparatus with the array of connected water-cooled condensers and hot plates is shown with three refluxing samples in Figure 26.
60
Figure 26: Setup of Boiling Reflux Flasks Under Condensers for COD Test
After refluxation and digestion and cooling, small amounts of ferroin indicator (only
3 or 4 drops) are added to samples. The potassium dichromate remaining in the sample is titrated with ferrous ammonium sulfate in order to measure the amount of potassium dichromate consumed. The oxidizable organic matter was then calculated in terms of oxygen equivalent. Before calculating the COD level of a sample, the ferrous ammonium sulfate was standardized with Eq. 6. VFAS denotes the volume of FAS titrant required for a
“blank,” unrefluxed sample to pass the endpoint of the ferroin indicator.
MFAS = 6 × 0.0417M × 10.0 mL / VFAS (Eq. 6)
61 The COD of a refluxed sample was then calculated with Eq. 7. The variable a is the
titrated volume of FAS for the blank refluxed sample, the variable b is the titrated volume
of FAS for the refluxed sample with organics, and V is the volume of original
“wastewater” sample included in the reflux flask during the COD test.
–1 COD (mg/L) = 8000 × (a – b) × MFAS × V (Eq. 7)
Results from Eq. 7 for each COD test performed are included in Table 6 in the
Results section.
• pH
The pH of byproduct samples was measured with a pH meter calibrated to either 4.00,
7.00, or 10.00. For all of the samples tested, liquid byproducts were acidic and the meter was calibrated with a buffer solution of 4.00. The pH meter was first calibrated to the same value as a buffer solution (in this case, 4.00, as all samples of pyrolytic liquids were acidic). The meter was then rinsed with distilled water and inserted into a small beaker with bio-oil sample. Once stable, the pH value was recorded. These data are included in
Table 6 in the next section.
• Specific Conductance
This procedure followed the manual provided by Jennings (2008). Measuring the specific conductance of liquid samples first required the calibration of a conductance meter and conductivity cell to standard NaCl solutions. Standards of 10, 25, 50, 100, 200,
400, 500, and 1,000 mg/L NaCl were prepared from a 2 g/L stock solution. Measuring the stock solutions began with the most dilute and proceeded to the highest concentrated solution. Once this was done, calibration of the meter was verified by checking that an
Excel plot of the calibration data was adequately linear.
62 After successful calibration, the specific conductance of each of the pyrolysis liquids were measured and recorded. These data are included in Table 6 in the Results and
Discussion section.
• Total Solids, Total Volatile Solids, Volatile Solids
This procedure followed that of the laboratory manual by Jennings (2008). Collected liquids were ignited in evaporation dishes and mass changes were recorded to determine the fraction that had volatilized. First, evaporation dishes were ignited at 550 degrees C for 1 hr in a muffle furnace to ensure any excess moisture or volatile solids no longer remained in the dishes. The masses of each dish were recorded after cooling. Known volumes of liquid byproduct samples were transferred to the dishes, and then the samples were dried in an oven at 80 degrees C for 24 hrs. After 24 hrs of drying, dishes were extracted from the oven and cooled in a desiccator. After cooling the masses of each dish were measured and recorded. The calculation of total solids (TS) was then calculated with
Eq. 5, where m1 denotes the mass of the clean dish, m2 denotes the mass of the dish after drying, and V denotes the known volume of sample transferred to the dish.
TS = (m2 – m1)/V (Eq. 7)
Total volatile solids (TVS) was determined with one additional procedure step. The dried and cooled dishes were ignited at 550 degrees C for 15 min in a muffle furnace.
After extraction and cooling in a desiccators, the mass of the dishes were again measured and recorded. Total volatile solids were calculated with Eq. 6, where m3 denotes the mass of the ignited dish, m2 denotes the mass of the dish after drying, and V denotes the sample volume.
TVS = (m2 – m3)/V (Eq. 8)
63 The volatile solids of each liquid sample by percentage was calculated with Eq. 7, and expressed as a percentage.
VS (%) = 100 × TVS/TS (Eq. 9)
The results of calculating Eq. 5-7 are included in the next section in Table 6.
4. Results and Discussion
Table 2 shows results for each sample tested throughout the course of the project.
Data include year of collection, yard waste type (tree/plant source), component (tree/plant part), moisture content (calculated from Eq.1), organics content (calculated with Eq. 2), char yield (calculated from Eq. 3), and byproduct collection (or byproduct recovery rate, calculated from Eq. 4).
Table 2: Moisture and Pyrolysis Testing Results for All Samples
Byproduct Year Yard Waste Type Component Moisture (%) Organics (%) Char (%) Collection (%) 2008 Beech Branches 25.98 30.87 43.16
2008 Beech Leaves 6.67 38.22 55.11
2008 Compost Boyas Mix 39.46 3.89 56.65
2008 Compost Mature Mix 57.92 13.87 28.21 2008 Cottonwood Branches 38.18 26.66 35.17 2008 Cottonwood Leaves 50.00 14.48 35.52 2008 Grass Clippings 68.77 18.93 12.30 2008 Leaves Assorted 21.24 25.69 53.07 2008 Maple Branches 19.00 46.50 34.51 2008 Maple Leaves (dead) 11.66 35.50 52.85 2008 Maple Leaves (live) 11.11 32.68 56.21 2008 Mugo Pine Branches 11.70 45.57 42.73 2008 Mugo Pine Leaves 53.10 21.65 25.25 2008 Oak Branches 24.50 37.27 38.23 2008 Oak Leaves 19.36 37.60 43.04 2008 Pear Branches 41.15 27.31 31.54 2008 Pear Leaves 8.29 37.01 54.70 2008 Pine Cones New 68.39 13.63 17.98 2008 Pine Cones New - Brown 40.26 23.11 36.62 2008 Pine Cones New - Green 61.03 19.46 19.51 2008 Pine Cones Old 52.27 11.13 36.60 2008 Fir Branches 45.41 8.82 45.78 2008 Fir Leaves 51.75 17.35 30.90 2009 Acorns Immature 12.92 47.80 39.28 70.9
64 2009 Acorns Mature 9.89 48.58 41.53 68.7 2009 American Elm Branches (dried) 0.00 62.99 37.01 22.2 2009 American Elm Branches 39.12 30.65 30.23 74.1 2009 American Elm Leaves 10.64 38.07 51.29 59.1 2009 Birch Leaves 8.30 40.79 50.91 61.9 2009 Black Locust Branches 31.74 35.47 32.79 85.0 2009 Black Locust Leaves 8.90 37.94 53.16 61.9 2009 Buckeye Fruit 10.37 51.64 37.99 23.5 2009 Buckeye Husks 9.49 45.79 44.72 46.9 2009 Juniper Mature - Leaves/Stems 7.42 43.76 48.82 57.9 2009 Magnolia Branches 45.56 13.26 41.18 17.7 2009 Magnolia Leaves 10.57 36.95 52.48 51.2 2009 Rhododendron Leaves 9.60 34.23 56.17 57.4 2009 Sycamore Branches 8.08 50.88 41.04 54.3 2009 Sycamore Leaves 9.31 37.56 53.13 46.5 2009 Willow Branches 42.14 28.11 29.75 31.2 2010 Willow Leaves 10.59 37.96 51.45 58.3 2010 Blue Spruce Branches 43.27 20.97 35.76 68.3 2010 Blue Spruce Leaves/Needles 52.59 21.91 25.50 81.5 2010 Maple Branches 37.68 35.64 26.68 79.0 2010 Maple Leaves 13.03 37.61 49.36 58.9 2010 Maple Samaras 36.82 31.31 31.87 74.0 2010 Norwegian Pine Branches 43.60 33.27 23.13 82.5 2010 Norwegian Pine Leaves/Needles 54.84 23.31 21.85 77.2 2010 Oak Branches 34.39 37.72 27.89 42.7 2010 Oak Leaves 46.30 14.03 39.67 64.3 2010 Rhododendron Branches 45.83 30.17 24.00 81.0 2010 Rhododendron Leaves 57.36 18.94 23.70 73.2 2010 Unkn. Conifer Branches 46.04 28.45 25.51 84.2 2010 Unkn. Conifer Leaves/Needles 53.68 22.25 24.07 82.6 Average 31.06 30.84 38.10 61.25 Standard Deviation 19.58 12.42 11.72 18.94
One of the goals of the research was to determine if there were significant performance differences between distinct components of yard waste. If there were significant performance differences, this information could help collection efforts concentrate on components of yard waste that would be most beneficial for bio-oil production. The most common types of yard wastes in the samples tested for this project were branches and leaves. The yields of these two components are shown in Tables 3–4.
Although it was not obvious, results indicated that branches have higher average moisture contents than leaves (30.6% versus 21.3%). This result is apparently due to leaves drying more rapidly than bulkier plant tissues after being collected as yard waste. Branches also
65 have higher average organics content than leaves (34.2% versus 31.9%). Leaves have higher char yields (46.8% versus 35.1% in branches) and byproduct collection efficiencies (59.2% versus 54.1% in branches). These numbers illustrate that branches would be a modestly more effective feedstock for the production of bio-oil. However, a collection system that separates branches from leaves is difficult to envision when leaves are nearly as effective in producing bio-oil. Additionally, the economic viability of the overall process will depend on the beneficial uses of the biochar and other off-gas products in addition to the bio-oils. Therefore, the overall optimum operating strategy may require optimization for the production of some other component.
Table 3: Moisture and Pyrolysis Testing Results for Branch Samples
Year Yard Waste Type Component Moisture (%) Organics (%) Char (%) Byproduct Collection (%) 2008 Beech Branches 25.98 30.87 43.16 2008 Cottonwood Branches 38.18 26.66 35.17 2008 Maple Branches 19.00 46.50 34.51 2008 Mugo Pine Branches 11.70 45.57 42.73 2008 Oak Branches 24.50 37.27 38.23 2008 Pear Branches 41.15 27.31 31.54 2008 Fir Branches 45.41 8.82 45.78 2009 American Elm Branches 39.12 30.65 30.23 74.1 2009 Black Locust Branches 31.74 35.47 32.79 85.0 2009 Magnolia Branches 45.56 13.26 41.18 17.7 2009 Sycamore Branches 8.08 50.88 41.04 54.3 2009 Willow Branches 42.14 28.11 29.75 31.2 2010 Maple Branches 37.68 35.64 26.68 79.0 2010 Oak Branches 34.39 37.72 27.89 42.7 2010 Rhododendron Branches 45.83 30.17 24.00 81.0 2009 American Elm Branches (dried) 0 62.99 37.01 22.2 Average 30.65 34.24 35.11 54.15 Standard Deviation 14.40 13.36 6.53 26.70
Table 4: Moisture and Pyrolysis Testing Results for Leaf Samples
Year Yard Waste Type Component Moisture (%) Organics (%) Char (%) Byproduct Collection (%) 2008 Beech Leaves 6.67 38.22 55.11 2008 Cottonwood Leaves 50.00 14.48 35.52 2008 Maple Leaves 11.11 32.68 56.21 2008 Maple - dead Leaves 11.66 35.50 52.85 2008 Mugo Pine Leaves 53.10 21.65 25.25 2008 Oak Leaves 19.36 37.60 43.04 2008 Pear Leaves 8.29 37.01 54.70
66 2008 Fir Leaves 51.75 17.35 30.90 2009 American Elm Leaves 10.64 38.07 51.29 59.1 2009 Birch Leaves 8.30 40.79 50.91 61.9 2009 Black Locust Leaves 8.90 37.94 53.16 61.9 2009 Juniper - Mature Leaves 7.42 43.76 48.82 57.9 2009 Magnolia Leaves 10.57 36.95 52.48 51.2 2009 Rhododendron Leaves 9.60 34.23 56.17 57.4 2009 Sycamore Leaves 9.31 37.56 53.13 46.5 2009 Willow Leaves 10.59 37.96 51.45 58.3 2010 Maple Leaves 13.03 37.61 49.36 58.9 2010 Oak Leaves 46.30 14.03 39.67 64.3 2010 Rhododendron Leaves 57.36 18.94 23.70 73.2 2008 Leaves Assorted 21.24 25.69 53.07 Average 21.26 31.90 46.84 59.17 Standard Deviation 18.00 9.18 10.02 6.51
Another goal of the project was to determine if drying the yard waste feedstocks prior to pyrolysis significantly increased yields of bio-oil. A drying step would not be difficult to implement, but would increase requirements on both area and cost for the pyrolysis process. The hypothesis in this aspect of the project is that if yard waste had less moisture to expel, byproducts might have higher organic concentrations and might favor higher molecular weight products. Comparisons between un-dried and dried feedstocks are included in Table 2 and emphasized in Table 5. In both cases, despite the dried feedstocks having lower moisture content and higher organics content than un-dried feedstocks, the byproducts collection was less efficient with dried feedstocks. This relationship between feedstock moisture and byproduct collection is illustrated further in
Figure 27, and includes all tested components. After applying linear regression analysis, the results of Figure 27 illustrate that although there is a modest correlation that seems to favor higher moisture content, the correlation is not sufficient to conclude that pre-drying is advantageous. If anything, the data indicate that higher moisture content somehow assists in the formation or collection of bio-oils.
67 Table 5: Evaluation of Pre-Drying Effect on Yields and Byproduct Collection
Year Yard Waste Type Component Moisture (%) Organics (%) Char (%) Byproduct Collection (%) 2009 American Elm Branches (dried) 0.00 62.99 37.01 22.24 2009 American Elm Branches 39.12 30.65 30.23 74.14 2009 Rhododendron Leaves 9.60 34.23 56.17 57.39 2010 Rhododendron Leaves 57.36 18.94 23.70 73.21
Figure 27: Comparison of Byproduct Collection and Moisture Content
Another research task was to determine the effectiveness of modifications to the pyrolysis process design. After experimenting with alternative reactor designs, the efforts of this project concentrated on improving the performance of the discharge piping. The idea was that if temperatures in the discharge piping were more carefully maintained, bio- oils could be encouraged to condense at more convenient locations within the collection system than within the discharge piping, causing operational problems. Figure 28
68 illustrates byproduct collection of the system as a function of time as efforts were made to “tune” both production and collection of bio-oils. With exceptions made for a few uncharacteristic results achieved early in the experimental program, the results demonstrate that these efforts were able to make a significant improvement in the consistency of collection, as well as a modest improvement in the magnitude of collection. Linear regression analysis of these data indicate that there is a correlation between later trials and modestly higher recovery rates of byproduct.
Figure 28: Byproduct Collection Over Project Duration
Figures 29 and 30 compare the moisture and organics contents of the major species of deciduous and coniferous tree species tested. One might assume that the conifers would have higher organic content and therefore higher yields of bio-oils, but this was not the
69 case from samples tested. The degree to which these would yield equivalent amounts of bio-oils is unknown.
Figure 29: Moisure Contents of Deciduous and Coniferous Samples
70
Figure 30: Organics Contents of Deciduous and Coniferous Samples
Based on the overall results of Table 2, the average biochar yield was 38.1%. The remainder of the mass (61.9%) was given off as byproducts. Of these byproducts, the average collection efficiency (i.e. the recovery rate) was 61.2%. The remainders of the
byproducts (mostly smoke and water vapor) were discharged into the ventilation system.
To put these numbers in perspective, the annual production of biochar and bio-oils is
estimated below if a yard waste pyrolysis process with these properties were applied to
the whole of yard waste generation in Cuyahoga County, Ohio.
Annual U.S. Yard Waste Production Estimate: 32.9 × 106 tons (USEPA 2009)
Total U.S. Population Estimate: 307,006,550 c (USCB 2009a)
Cuyahoga County Population Estimate: 1,275,709 c (USCB 2009b)
71 Annual Biochar Production:
32.9 × 106 tons × (307,006,550 c)−1 × 1,275,709 c × 0.381 = 52,100 tons
Annual Byproduct Production:
32.9 × 106 tons × (307,006,550 c)−1 × 1,275,709 c × 0.619 = 84,600 tons
In the production of these byproducts, a portion of it would be aqueous. Using the average moisture content measured for all samples of yard waste, this fraction is estimated below.
Aqueous Byproduct:
32.9 × 106 tons × (307,006,550 c)−1 × 1,275,709 c × 0.311 = 42,500 tons
Organic Byproduct:
32.9 × 106 tons × (307,006,550 c)−1 × 1,275,709 c × 0.308 = 42,100 tons
However, it should be noted that not all of the organic fraction was actually collected in liquid phase. Some of this escapes in the gaseous discharge. The overall byproduct collection efficiency of this fraction was 61.2%. Obviously, these estimates assume that all yard waste is collected and successfully pyrolyzed, but the production potential seems adequate to generate a material and revenue stream that would help offset the cost of the pyrolysis operation.
One consideration that has not yet been addressed is that pyrolysis also yields an aqueous byproduct. This may be released as steam, but doing so intentionally would
72 substantially reduce the recovery of the organic fraction and could lead to serious air
pollution control compliance issues. Generally, it is preferable to condense the off-gas into a liquid phase and then separate the organic phase from the aqueous phase. However, as Figure 31 illustrates, there can be a great deal of variability in the nature of the aqueous phase depending on the yard waste feedstock.
Figure 31: Bottled Pyrolytic Liquids
Table 6 presents a preliminary chemical characterization of the collected byproduct.
Note that these solutions are dominated by organic acids (i.e. high COD, low pH, and
high percentages of volatile solids). Additional analysis should be conducted to determine
to what extent valuable byproducts can be extracted from this fraction, and to what extent
the management of this as wastewater will lead to additional costs.
73 Table 6: Preliminary Chemical Analysis of Off-Gas Condensate
COD Specific Total Solids Total Volatile Volatile Sample (mg/L) pH Conductance (mS) (g/L) Solids (g/L) Solids (%)
Blue Spruce Branches 256,000 2.43 1.41 72.26 69.24 95.82
Blue Spruce Leaves/Needles 140,000 3.38 2.18 31.65 30.75 97.15
Conifer Branches 145,000 2.58 1.21 32.97 30.91 93.76
Conifer Leaves/Needles 78,000 3.85 3.85 19.40 18.52 95.42
Norwegian Pine Branches 140,000 2.62 1.29 30.59 30.06 98.25
Norwegian Pine Leaves/Needles 91,000 3.85 3.22 23.56 23.20 98.47
Average 141,667 3.1 2.2 35.1 33.8 96.5
Standard Deviation 62,784 0.7 1.1 19.0 18.1 1.8
Separation of the aqueous and organics fractions was necessary to produce a bio-oil
that would be of sufficient viscosity for use in an asphalt binder.
For pyrolysis to be a feasible process for yard waste management, the off-gas
condensate, or bio-oil, must be of sufficient viscosity for use in an asphalt binder. As the
samples in Figure 31 illustrate, the collected byproduct does not typically have a
sufficiently high organic concentration. Because of this, separation processes of the
aqueous and organic fractions of the off-gas condensate are necessary. For this project, organics fractions were enriched through a combination of both physical separation and from drying. Some collected condensate samples had been in storage for a period of time before enrichment that the more dense organics had settled to the bottom of their containers. In these cases, aqueous fractions were poured out into a separate container,
while organics fractions of many samples were consolidated. This provided an enriched
74 (or developed) bio-oil sample. This enriched bio-oil sample in a beaker was then inserted into a drying oven at 70 °C over a span of 3–5 days, driving additional water off and producing a bio-oil sample of higher organic concentration.
In other cases, a fraction of organic condensate would be retained in containers after physical separation by pouring. With an acetone wash, the remaining organics were consolidated into another beaker, left open in a hood for the acetone to evaporate.
Evaporation of the acetone also resulted in an enriched bio-oil that could be delivered for binder testing. Further research and process innovations in the separation of these aqueous and organics fractions of pyrolytic liquids would be necessary to deliver higher quality bio-oils for more extensive binder testing.
5. Summary and Conclusions
This study was designed to evaluate the fundamental feasibility of producing useful quantities of bio-oil from pyrolysis of yard waste. To accomplish this, the research evaluated the moisture, organics, and char content of over 50 distinct yard waste components, and the research also evaluated the recovery efficiencies of the biochar and byproducts. The overall average byproduct generation rate was 61.9% (31.1% aqueous and 30.8% organic). Of this approximately 61.2% was recovered by the collection system used in this project.
The variability of moisture content, organics content, and byproduct collection were grouped by type of yard waste. This was done to investigate whether any typical yard waste component would be more beneficial for bitumen production. In the comparison between branches and leaves, neither type has a significantly higher organic fraction
(34.2% and 31.9%, respectively). This fact is relevant in that, if collecting bio-oils from
75 yard waste pyrolysis were implemented at a larger scale, extensive feedstock separation would not be necessary.
The benefit of drying feedstock prior to pyrolysis was also investigated. In comparing byproduct collection of un-dried and dried samples, pre-drying decreased the total byproduct collection rates. This was partly due to the increased viscosity of the byproducts after drying. In looking at the same relationship between byproduct collection and moisture content over all samples tested, a sample’s moisture content seemed to have no major impact on byproduct production or collection.
In retrospect, the laboratory apparatus used was more effective at accomplishing the basic pyrolysis steps than in managing the recovery and separation of byproducts.
Improvements were made during the course of the project by installing heating eelements to control gas discharge temperatures in strategic sections of the discharge piping, and these helped stabilize and improve recovery. However, the off-gas management system should be modified for further applications. An improved off-gas management apparatus is desirable in order to increase the organic production of the organic fraction and to help reduce the potential air discharge complications of the process.
The difficulties arising from transferring bitumen from collection flasks to storage containers or bottles also warrants more attention. Although in an aqueous solution, a portion of higher viscosity bitumen in the flasks could not be transferred to separate containers without an acetone wash. Of those bitumen samples that were transferred to separate containers, they were eventually consolidated for the creation of a larger sample.
They were easily transferred as they had a significant aqueous fraction. After consolidation of these aqueous bitumen products, dewatering was necessary to increase
76 the organic concentration. As the dewatered samples in this study had various viscosities, further exploration and standardization of this dewatering process could help deliver a bio-oil product with more uniform properties. Further research should probably make use of a rotary evaporator to help implement the separation and concentration of the organic fraction.
The physical properties of yard waste bio-oils and asphalt binders made from these bio-oils should be explored in detail. This project demonstrated that bio-oils could be generated from yard waste, and that it is likely that the volume would be sufficient to support the manufacture of asphalt binders. However, the properties of the resulting product are unknown. A considerable amount of additional research would be required to determine how to best use this material, and to quantify the properties of the products.
Experience has also demonstrated that some of the byproducts of pyrolysis may represent significant aesthetic or environmental problems. The gas phase discharges can be malodorous and may lead to air pollution compliance obligations. Likewise, the aqueous phase discharges will probably require treatment. The properties of these discharges should be quantified and both liquid and gas treatment systems should be examined to ensure that both discharges can be successfully treated.
Finally it is important to keep in mind that yard waste pyrolysis is desirable for several reasons. It will yield bio-oil, but this is probably not sufficiently valuable in and of itself to justify the yard waste pyrolysis process. The process also sequesters carbon, allows for yard waste collection efficiencies, produces a biochar soil amendment of potential value to agriculture, and may yield other byproducts of commercial value. A
77 successful implementation of yard waste pyrolysis will require the success of many elements of the process that were not evaluated in the work described here.
78 6. Appendix
6.1. Appendix A: Methods and Results of Growth Experiments
Experiment 1 - Soybeans
The first biochar growth experiment consisted of investigating effects of char addition to soil in the growth of soybeans. In this experiment, chars were derived from both
pyrolysis of yard waste and from commercial hardwood charcoal. Before addition into
the soil in each of the sample pots, chars were ground to sizes of approximately 1 cm or
less. Chars were then added to planting pots in percentages by volume to either sand,
topsoil, or till. The sand, topsoil, and till were not mixed together in sample dishes.
Dishes were all set on the same watering schedule. After harvesting, the beans and stems
of each plant were separated and their “wet” masses were recorded. After overnight
drying at 90 degrees C, their dry masses were also recorded. These data are included in
Table 7.
Table 7: Yield Results from First Biochar Growth Experiment with Soybeans
Dish Soil Type Char (vol %) Char Type Dry Stem Yield (g) Dry Bean Yield (g) 1 Sand 0 - 1.32 0.53 2 Sand 0 - 1.46 1.83 3 Sand 20 Yard Waste 1.31 0.58 4 Sand 20 Yard Waste 1.80 0.27 5 Sand 10 Yard Waste 1.67 2.89 6 Sand 10 Yard Waste 0.85 1.19 7 Sand 20 Commercial 1.74 3.33 8 Sand 20 Commercial 2.28 3.23 9 Sand 10 Commercial 2.10 2.52 10 Sand 10 Commercial 1.77 0.97 15 Sand 10 Commercial 0.30 0.03 17 Topsoil 0 - 6.03 26.55 18 Topsoil 0 - 6.47 24.31 23 Topsoil 20 Commercial 6.82 28.23 24 Topsoil 20 Commercial 7.61 27.57
79 25 Topsoil 10 Commercial 6.34 22.53 26 Topsoil 10 Commercial 5.81 24.7 28 Till 0 - 0.98 1.82 29 Till 20 Commercial 2.21 8.86 30 Till 20 Commercial 0.00 0 31 Till 10 Commercial 2.16 7.29 32 Till 10 Commercial 0.31 1.06 33 Till 20 Yard Waste 3.82 1.48 35 Till 10 Yard Waste 0.00 0 37 Sand 50 Commercial 1.21 0.50 38 Sand 50 Commercial 2.46 0.95 39 Topsoil 50 Commercial 5.35 21.85 40 Topsoil 50 Commercial 5.96 27.52 Average 2.86 8.66 Standard Deviation 2.33 10.83
Figures 32 and 33 illustrate the dry stem and bean yields versus the level of char
added in soil. The highest bean yields were observed at levels of 20% char addition. For
the two soil types tested with 50% char addition (topsoil and sand), bean yields dropped,
but not significantly. Stem yields at 50% char addition showed comparable yields to 20% addition levels. Bean yields appear to approach a maximum level with char addition at approximately 20%, but the amount of data is such that a definitive relationship cannot be stated.
80
Figure 32: Dry Soybean Stem Yield versus Char Levels in Soil
Figure 33: Dry Bean Yield versus Char Levels in Soil
81 Experiment 2 – Corn
The second growth experiment measured effects of biochar additions to soils for a corn crop. In this experiment, plant samples were grown in mixtures of topsoil, sand, and biochar. Till was not used as a soil type, and the topsoil and sand were mixed together in different fractions along with biochar. As before, all samples were watered on the same schedule. In harvesting of the corn, the entire plant was removed from the pot and stored in a plastic bag. The mass of the plant was recorded both “wet” and after overnight drying at 80 degrees C. Yield masses were measured by recording the mass of the entire corn plant, not by just the ears of corn themselves. Dry yields are included in Table 8.
Table 8: Yield Results from Second Biochar Growth Experiment with Corn
Sample Topsoil (vol %) Sand (vol %) Biochar (vol %) Dry Plant Yield (g) 50 100 0 0 7.81 51 100 0 0 5.90 52 100 0 0 7.88 53 85 0 15 8.52 54 85 0 15 7.82 55 85 0 15 10.30 56 75 0 25 9.87 57 75 0 25 8.71 58 75 0 25 11.22 59 65 0 35 9.78 60 65 0 35 10.44 61 65 0 35 11.51 62 50 0 50 8.10 63 50 0 50 13.96 64 50 0 50 10.20 65 20 0 0 4.75 66 20 80 0 7.87 67 20 80 0 7.81 68 15 80 15 9.52 69 15 70 15 10.09 70 15 70 15 10.16 71 15 70 25 0.87 72 15 60 25 11.40 73 15 60 25 7.38 74 15 60 35 9.69 75 15 50 35 4.12 76 15 50 35 8.70 77 10 50 50 7.57 78 10 40 50 7.57 79 20 40 50 6.20 80 20 40 0 6.55
82 81 20 80 0 4.70 82 20 80 0 3.20 83 15 80 25 6.12 84 15 25 25 2.52 85 15 25 25 6.00 88 10 50 50 2.21 Average 7.76 Standard Deviation 2.88
Figures 34 and 35 demonstrate the effect of biochar addition to the soil in graphical form. Results from these trials were consistent with those of the first crop in that intermediate levels of biochar addition (20%-35%) demonstrated slightly higher mass yields than the maximum biochar addition levels at 50%. Figure 35 shows that topsoil levels at around 60% yielded the greatest corn plants.
Figure 34: Dry Plant Yield versus Biochar Levels in Soil
83
Figure 35: Dry Plant Yield versus Topsoil Level in Pot
Experiment 3 – Soybeans
In this experiment soybeans were grown in biochar amended soils. Soils were a mixture of differing levels of both topsoil and sand. Each soybean plant was watered on the same schedule and frequency. Data recorded from harvested plants included the number and mass of stems/stalks, the number and mass of beans, and the dry mass of both stems and beans. Drying was performed on the harvested samples overnight. This data is shown in Table 9.
84 Table 9: Yield Results from Third Biochar Growth Experiment with Beans
Sample Topsoil (vol %) Sand (vol %) Biochar (vol %) Beans Dry Stalk Yield (g) Dry Bean Yield (g) 1 0 100 0 50 3.35 25.47 2 0 100 0 25 2.32 12.03 3 0 80 20 57 4.44 26.23 4 0 80 20 43 3.43 17.95 5 0 90 10 52 4.07 25.74 6 0 90 10 44 3.54 22.29 7 0 80 20 60 5.08 24.97 8 0 80 20 48 4.7 27.44 9 0 90 10 32 2.46 15.73 11 0 100 0 32 2.74 15.2 14 0 80 20 33 3.24 14.77 16 0 90 10 32 2.53 14.52 23 80 0 20 65 7.04 27.84 24 80 0 20 35 3.22 17.18 25 90 0 10 80 7.63 31.46 37 0 50 50 29 3.57 13.02 38 0 50 50 22 1.95 11.08 50 100 0 0 29 2.87 14.13 52 100 0 0 24 2.84 8.18 53 85 0 15 63 6.73 36.25 54 85 0 15 65 7.25 30.3 57 75 0 25 59 6.63 31.4 58 75 0 25 65 6.48 38.01 59 65 0 35 78 7.57 35.94 60 65 0 35 64 7.3 35.77 61 65 0 35 60 6.22 28.74 62 50 0 50 57 4.74 25.83 63 50 0 50 63 8.65 30.8 65 20 80 0 57 3.87 25.75 66 20 80 0 55 3.64 25.12 67 20 80 0 49 5.07 24.59 68 15 70 15 57 5.05 24.13 69 15 70 15 47 5.13 22.74 70 15 70 15 52 4.52 25.52 71 15 60 25 62 6.82 36.21 72 15 65 25 55 3.27 21.42 73 15 60 25 55 4.36 24.28 74 15 50 35 54 3.43 22.78 75 15 50 35 51 3.81 21.05 76 15 50 35 75 8.94 38.29 77 10 40 50 49 2.67 19.19 78 10 40 50 52 3.44 22.66 79 10 40 50 57 5.54 24.13 80 20 80 0 49 4.49 20.86 81 20 80 0 55 3.34 25.15 82 20 80 0 69 6.47 30.52 83 15 60 25 70 5.75 36.5 84 15 60 25 52 2.87 24.31 85 15 60 25 58 6.35 27.79 86 10 40 50 39 5.5 25.33 87 10 40 50 37 4.05 23.02 Average 24.50 Standard Deviation 7.34
85 Figures 36 and 37 display the effect of biochar and topsoil levels in soil on the yield
of beans. Figure 36 shows results similar to that of the first two growth experiments, in
that intermediate levels of biochar addition resulted in the greatest bean yield masses. At
higher levels of biochar addition (50%), yield levels dropped slightly. Figure 38 shows
results consistent with that of the second growth experiment, where topsoil levels of
approximately 50-60% led to the greatest yields. Overall, these growth experiments
display that char additions to soil improve dry yields the most at levels of 20-35%, and lower only slightly at higher levels (i.e. 50%). This means that high levels of biochar application in soils are not necessary to achieve the greatest productivity, which would decrease labor demands and costs if implemented at a larger scale.
Figure 36: Dry Bean Yield versus Biochar
86
Figure 37: Dry Bean Yield versus Topsoil Addition Levels
Figure 38: Dry Bean Yield versus Sand Addition Levels
87 6.2. Appendix B: Ohio Asphalt Binder Specifications
88
89 7. References
AI (Asphalt Institute). (2005). Summary of the standard PG testing and criteria.
t_9_7_2005.pdf>. 11 July 2009. AI (Asphalt Institute). (2008). State Binder Specification Database – Ohio. e_Binder_Specs/OH_DOT_Binder_Spec_052108.pdf>. 15 Mar 2009. AI (Asphalt Institute). (2009). AI (Asphalt Institute). (2011). “State Binder Specification Database – Ohio.” e_Binder_Specs/OH_DOT_Binder_Spec_010111.pdf>. Accessed 14 Feb 2011. Allen, B. R. (1980). “Pretreatment Methods for the Degradation of Lignin.” Asphalt Institute. (2003, Asphalt Institute). Superpave Performance Graded Apshalt Binder Specification and Testing, Superpave Series No. 1 (SP-1), Lexington, KY, USA. Anderson E., Vasudevan, J., Stoker, D., Hagen, E., and Bragstad, R. (2007). “Road Trials with Vegetable Oil Based Binders in Norway.” 23rd World Road Congress, Paris, France, Sep 2007. 90 ANZBRN (Australia and New Zealand Biochar Research Network). (2009). “Project.” APHA (American Public Heath Association). (1985). Standard Methods for the Examination of Water and Wastewater, 16th ed. Greenberg, A. E., Trussell, R. R., Clesceri, L. S., and Franson, M. A. H., eds. Washington, DC. Arsova, L., van Haaren, R., Goldstein, N., Kaufman, S. M., and Themelis, N. (2008). “The State of Garbage in America.” BioCycle, 49(12). Austin, A. (2009). “Mantria, CDI Develop Commercial Biochar Plant.” Biomass Magazine Online, commercial-biochar-plant>. 9 July 2009. Australian Biochar. (2009). “Biochars – A New Generation.” Bishara, S. W., Robertson, R. E., and Mohoney, D. (2005). “Lignin as an Antioxidant: a Limited Study on Asphalts Frequently Used on Kansas Roads.” 42nd Annual Peterson Asphalt Conference. Laramie, WY. 91 Bohn, M. S., and Benham, C. B. (1984). “Biomass Pyrolysis with an Entrained Flow Reactor.” Industrial and Engineering Chemistry Process Design and Development, 23, 355-363. Brewer, C. E., Schmidt-Rohr, K., Satrio, J. A., and Brown, R. C. (2009). “Characterization of Biochar from Fast Pyrolysis and Gasification Systems.” Environmental Progress & Sustainable Energy, 28(3), 386-396. Brewer, L. J., and Sullivan, D. M. (2003). “Maturity and Stability Evaluation of Composted Yard Trimmings.” Compost Science & Utilization, 11(2), 96-112. Bridgwater, A. V. (2003). “Renewable fuels and chemicals by thermal processing of biomass.” Chemical Engineering Journal, 91, 87-102. Bridgwater, A. V., Meier, D., and Radlein, D. (1999). “An overview of fast pyrolysis of biomass.” Organic Geochemistry, 30, 1479-1493. Bridgwater, A. V., and Peacocke, G. V. C. (2000). “Fast pyrolysis processes for biomass.” Renewable and Sustainable Energy Reviews, 4, 1-73. Cambridge Systematics, Inc. (2005). “Reducing Heat Islands: Compendium of Strategies – Cool Pavements.” 92 City of Cleveland Heights, The. (2011). “Yard Waste – Leaves, Grass, Brush.” City of Shaker Heights. (2011). “Collection & Removal – Leaves, Grass, Brush & Snow Collection.” Czernik, S., and Bridgwater, A. V. (2004). “Overview of Applications of Biomass Fast Pyrolysis Oil.” Energy & Fuels, 18, 590-598. Czernik, S., Johnson, D. K., and Black, S. (1994). “Stability of Wood Fast Pyrolysis Oil.” Biomass and Bioenergy, 7(1-6), 187-192. Demirbas, A. (2005). “Pyrolysis of ground beech wood in irregular heating rate conditions.” Journal of Analytical and Applied Pyrolysis, 73, 39-43. Demirbas, A. (2007). “Progress and recent trends in biofuels.” Progress in Energy and Combustion Science, 33, 1-18. Dizhbite, T., Telysheva, G., Jurkjane, V., and Viesturs, U. (2004). “Characterization of the radical scavenging activity of lignins–natural antioxidants.” Bioresource Technology, 95(3), 309-317. 93 Dynamotive Energy Systems. (2009). “Dynamotive Provides Update on CQuest Biochar Test Programs and Production Noting that Biochar Initiatives Gain Ground in the US.” cquest%E2%84%A2-biochar-test-programs-and-production-noting-that-biochar- initiatives-gain-ground-in-the-us/>. 20 Sep 2009. Fogarty, A. M., and Tuovinen, O. H. (1991). “Microbiological Degradation of Pesticides in Yard Waste Composting.” Microbiological Reviews, 55(2), 225-233. Gargulak, J. D., and Lebo, S. E. (1999). “Commercial Use of Lignin-Based Materials.” Lignin: Historical, Biological, and Materials Perspectives. Glasser, W. G., Northey, R. A., and Schultz, T. P., eds. American Chemical Society, Washington, D.C. 304-320. Gunther, S. (2009). “Biochar 101.” Mother Nature Network. Humboldt Manufacturing Company. (2011a). “Revolving Shelf Oven for Asphalt Testing.” Humboldt Manufacturing Company. (2011b). “Pressure Aging Vessel for Asphalt Binder Aging Operations.” 94 Iowa State University Center for Sustainable Environmental Technologies. (2010). “Current Research: Bench Top Fluidized Bed Reactor.” Feb 2011. Jennings, A. A. (2008). ECIV 368 Environmental Engineering Laboratory Manual, Spring 2008. Case Western Reserve University. Cleveland, OH. Johnson, D. (2005). “GEO320 Technology Improves Shell Bitumen.” _Bitumen_46855.html>. 9 Aug 2008. Komilis, D. P., and Ham, R. K. (2004). “Life-Cycle Inventory of Municipal Solid Waste and Yard Waste Windrow Composting in the United States.” Journal of Environmental Engineering, 130(11), 1390-1400. Lehmann, J., Gaunt, J., and Rondon, M. (2006). “Bio-Char Sequestration in Terrestrial Ecosystems – A Review.” Mitigation and Adaptation Strategies for Global Change, 11, 403-427. Lima, H. N., Schaefer, C. E. R., Mello, J. W. V., Gilkes, R. J., and Ker, J. C. (2002). “Pedogenesis and pre-Colombian land use of ‘Terra Preta Anthrosols’ (‘Indian black earth’) of Western Amazonia.” Geoderma, 110, 1-17. 95 Maris, E. (2006). “Black Is the New Green.” Nature, 442, 624-626. McCready, N. S., and Williams, R. C. (2007). “The Utilization of Agriculturally Derived Lignin as an Antioxidant in Asphalt Binder.” Proceedings of the 2007 Mid-Continent Transportation Research Symposium, Ames, Iowa. August 2007. Michel, F. C., Reddy, C. A., and Forney, L. J. (1995). “Microbial Degradation and Humification of the Lawn Care Pesticide 2,4-Dichlorophenoxyacetic Acid during the Composting of Yard Trimmings.” Applied and Environmental Microbiology, 61(7), 2566-2571. Mohan, D., Pittman, C. U., and Steele, P. H. (2006). “Pyrolysis of Wood/Biomass for Bio-oil: A Critical Review.” Energy & Fuels, 20, 848-889. Müller, R., Antranikian, G., Maloney, S., and Sharp, R. (1998). “Thermophilic Degradation of Environmental Pollutants.” Advances in Biochemical Engineering/Biotechnology, 61, 155-169. NAPA (National Asphalt Pavement Association). (2009). Jan 2009. 96 NCAT (National Center for Asphalt Technology). (2009). Ouyang, C., Wang, S., Zhang, Y., and Zhang, Y. (2006). “Improving the Aging Resistance of Asphalt by Addition of Zinc Dialkyldithiophosphate.” Fuel, 85, 1060-1066. Raouf, M. A., and Williams, R. C. (2010). “Development of Non-Petroleum Based Binders for Use in Flexible Pavement.” 594_Williams_Non-PetroleumBinder.pdf>. Roberts, K. G., Gloy, B. A., Joseph, S., Scott, N. R., and Lehmann, J. (2010). “Life Cycle Assessment of Biochar Systems: Estimating the Energetic, Economic, and Climate Change Potential.” Environmental Science & Technology, 44(2), 827-833. Sullivan, D. M., and Miller, R. O. (2001). “Compost quality attributes, measurements and variability.” Compost utilization in horticultural cropping systems. Stofella, P. J., and Kahn, B. A., eds., CRC Press. Boca Raton, FL, 95-120. Sundstrom, D. W., Klei, H. E., and Daubenspeck, T. H. (1983). “Use of Byproduct Lignins as Extenders in Apshalt.” Industrial & Engineering Chemistry Product Research and Development, 22(3), 496-500. 97 USCB (U.S. Census Bureau, Population Division). (2009). Annual Estimates of the Population for the United States, Regions, States, and Puerto Rico: April 1, 2000 to July 1, 2009. NST-EST-2009-01. December 2009. USCB (U.S. Census Bureau, Population Division). (2010). Annual Estimates of the Resident Population for Counties of Ohio: April 1, 2000 to July 1, 2009. CO-EST2009- 01-039. March 2010. 39.xls>. 12 Jul 2010. USEPA (U.S. Environmental Protection Agency). (2006). Solid Waste Management and Greenhouse Gases: A Life-Cycle Assessment of Emissions and Sinks (3rd Edition). Washington, D.C. USEPA (U.S. Environmental Protection Agency). (2008). Municipal Solid Waste in the United States: 2007 Facts and Figures. EPA530-R-08-010. Washington, D.C. USEPA (U.S. Environmental Protection Agency). (2009). Municipal Solid Waste Generation, Recycling, and Disposal in the United States: Detailed Tables and Figures for 2008. Washington, D.C. 98 Verheijen, F. G. A., Jeffery, S., Bastos, A. C., van der Velde, M., and Diafas, I. (2009). Biochar Application to Soils - A Critical Scientific Review of Effects on Soil Properties, Processes and Functions. EUR 24099 EN, Office for the Official Publications of the European Communities, Luxembourg, 149pp. Williams, P. T., and Besler, S. (1996). “The Influence of Temperature and Heating Rate on the Slow Pyrolysis of Biomass.” Renewable Energy, 7(3), 233-250. Williams, R. C., Satrio, J., Rover, M., Brown, R. C., and Teng, S. (2008). “Utilization of Fractionated Bio Oil in Asphalt.” Journal of the Transportation Research Board, 09(3187), 1-19. 99