Research Summary & Recommendations For Advanced …



|Berkshire Energy Laboratory |

|Research Summary & Recommendations For Advanced Research |

|Prepared for Ollie Curme and Noa Simons |

|Thomas Horgan |

|10/14/2008 |

Table of Contents

1. Introduction

1. Alternative Energy Perspective

2. Research Goals

3. Energy Economics

2. Basic Background Chemistry

1. Wood/Sludge/Fuels

2. Combustion

3. What makes a good fuel?

3. Primary Research on Wood and Sludge Conversion Routes

1. Gasification

2. Wood to Diesel/Gasoline

3. Sludge to Diesel

4. Conclusions and Recommendations

5. Appendix I: Citing Requirements and Equipment LIsts

Executive Summary

The conversion of biomass (both wood and sludge) to fuels involves chemical processes which break down complex hydrocarbons into simpler molecules that combust with much higher energy efficiency. Fossil fuels start with an advantage over bio-fuels since the earth has already done most of the work. Having realized this in the early 20th century, human beings have built a global infrastructure to support the use of fossil fuels for energy. For bio-fuels to compete, feedstocks must be plentiful and inexpensive, conversion processes must be highly efficient and the biofuels produced must be comparable so that existing infrastructure can be maintained.

Though renewed concerns over the environment and fossil fuel availability have resulted in renewed interest in biomass conversion, much of the research in the area focuses on process development of relatively old technologies. In general the industry (like fuel cells) is focusing on improving efficiencies for processes with little hope of reaching commercial viability. They are supported in their efforts by well intentioned organizations desiring to cultivate a ‘green image’, unwilling however to demand the technological and economic rigor required to accomplish the final goal. To avoid making the same mistake, I’ve employed concepts from the field of energy economics as a standard for comparison of the conversion routes we’re investigating.

Early research of sludge conversion by liquefaction lead to Bio-Petrol and their process for synthesizing diesel fuel. While questions remain, the energy balance and economics are highly favorable compared to all other processes studied. Earlier research from Japan (Itoh process) on a simpler process model is even more economically favorable. Sludge conversion then is recommended as the primary research area for Berkshire Energy Labs with efforts focusing on verification of a liquefaction process on a lab scale and subsequent research into variants and upgrading methods patentable by BEL.

Biomass conversion to liquid fuels is primarily accomplished by either fast pyrolysis or fischer tropsch (FT) synthesis. The fuel quality varies drastically and while pyrolysis is the simpler process the fuel oil produced is extremely low grade. FT synthesis produces diesel fuels of exceptional quality and while currently it’s a multi step process on the industrial scale, opportunities exist for redesigning the reaction sequence to support use at a residential scale. This is recommended as a secondary research area for BEL with efforts focusing on process design and controls to achieve higher diesel yields directly from the FT reactor.

I envision a waste to energy plant synthesizing liquid fuels from municipal sludge, woody biomass and ultimately any carbonaceous feedstock. Initial work would be around development of ‘proven’ or at least demonstrated methods such as liquefaction and fischer tropsch synthesis. Later work could be in advanced process development (catalysis, reactor design etc) funded by revenue from fuel sales.

1. Introduction

Our work is concerned with the conversion of biomass and municipal sludge to fuels of one form or another. It’s helpful to begin by stepping back and briefly characterizing exactly what occurs in the conversion of mass from less to more useful forms. In so doing we will provide a perspective for the more detailed conversion discussions that follow.

The application of heat and pressure in sufficient amounts to any complex substance causes chemical bonds to break and reform resulting in progressively simpler molecules as the process proceeds. Ultimately, only elements remain and even these can be further degraded into subatomic particles if we carry the process to extremes. Different bonds with different energies break and form at different temperatures and pressures, consuming or generating heat in the process. Chemists and chemical engineers work to control the destruction and formation of chemical bonds by controlling heat and pressure specifically to convert mass from less to more useful forms with minimal energy loss to the surroundings. This is done through the design of reaction vessels and control systems as well as the reaction sequence itself. For example, solvents may be used or catalysts introduced to break chemical bonds at lower energies, donating and recovering intermediate elements or compounds at key points in the reaction, increasing the reaction rate without consuming the catalyst.

Of particular interest in the field of energy are the chemical bonds between hydrogen and carbon. Human beings have advanced largely because we’ve realized that these bonds are rather easily broken in a self-catalyzing sequence, producing large amounts of heat with explosive force that can be used to drive engines and generators. Over the history of the earth heat, pressure and natural catalysis have combined to form increasingly complex hydrocarbons culminating in biological life. When this bio-mass dies, the same heat and pressure applied over time convert it back to more simple molecules in the form of fossil fuels such as crude oil, coal and natural gas. Initially, humans relied almost exclusively on the direct combustion of dead bio-mass for our energy needs. Later we realized that with relatively little input energy, the simpler hydrocarbons present in fossil fuels can be refined to burn with much higher energy output than biomass. The earth having already done most of the work, fossil fuels quickly became the fuels of choice to power the industrial revolution and as we know, an entire infrastructure has been built around them. What this means is that all over the world, pipes, pumps, engines, generators and a variety of other equipment have all been designed around a set of fuels that have a fairly narrow range of densities and heating values.

Concern over the continued use of fossil fuels is growing for a number of well known reasons: supplies are thought to be dwindling, over-reliance on unstable foreign sources and increased CO2 loading in the atmosphere believed by many to be contributing to global warming. These have come together to spur renewed interest in alternative/renewable energy sources such as wind, solar, geothermal and biomass conversion – the source we’re interested in at Berkshire Energy Labs. More specifically, we’ve decided to focus our attention on the conversion of wood and treatment plant sludge to fuel.

The preceding narrative points to some of the issues we face in the conversion of biomass and sludge. In comparison to fossil fuels, biomass contains more complex hydrocarbons which do not combust cleanly or efficiently. Conversion of biomass to clean burning fuels requires significant input energy which is an added cost over fossil fuel. To take advantage of existing technology and infrastructure, the fuel grade produced needs to be quite high.

Energy and Economic Comparisons

Through Ollie’s research we had down-selected to the two primary areas already mentioned however it’s beneficial to compare the processes in terms of energy efficiency and financial potential. The comparisons in table 1 below were made using data from commercial and pre-commercial processes available on the internet (Bio-Petrol, Itoh Paper, Enertech, Anerobic Digestion Info Sheet, NH Bio Oil Study, Choren Technology, Community Power Systems). Input energy data were not available for Wood-to-diesel and wood-to-bio-oil and were estimated using heat capacity, operating temperatures and mass flow rates of feedstock.

[pic]

Table 1 – Energy Economics

The Choren Fischer Tropsch process stands out as being highly energy efficient with the Bio-Petrol and Itoh processes second. Note that the energy efficiency calculation is made as follows using Natural Gas as the presumed input fuel for convenience:

Energy Efficiency = 100 x HHV of Fuel Produced/(HHV of Feed + HHV of Natural Gas)

Energy efficiency is a ‘property’ of the conversion process and can only be improved by improving the heating value of the fuel. The FT process is highly energy efficient because of the excellent heating value of the diesel fuel produced compared to wood. The Bio-Petrol and Itoh processes are efficient for similar reasons.

There is much on the internet about Net Energy Gain and Net Energy Return on Investment (NEROI) but no one can agree on what inputs should be included in the calculations. As a result its’ not yet clear to me how anyone can call there process a ‘net energy producer’. In any event, it’s an unnecessary complication. The energy efficiency is a reasonable indicator of the environmental value of a process for conversion of bio-fuels. Processes which call themselves ‘Net Energy Producers’ are claiming only that the energy content of the synthesized fuel exceeds the input energy required to make it. If you subtract the heating value of the feedstock, all processes have a net energy loss.

Included in table 1 is a comparison based on data available from treatment plants in the Saratoga area. From the NYS Department of Environmental Conservation, there are approximately 40 waste water treatment plants (WWTPs) within a 50 mile radius of Saratoga. Of those, 18 are either incinerating or burning their activated sludge (no post treatment such as aneraobic digestion). If we could get access to 75% of that sludge, we could produce up to 3,600 gals of 85% no 2 oil per day (using a Bio-Petrol type process).

Table 2 below is an economic comparison of the processes based on information available on the internet. The calculations in this table ignore capital costs but do include an estimate of transportation costs. The processing cost assumes electricity as the energy source and the maintenance and operating costs are a percentage of the production rate plus transportation.

[pic]

Table 2 – Economic Analysis

Under these assumptions, sludge to diesel by liquefaction appears to be the most economically viable solution by a wide margin. This is largely due to the relative quality of the fuel produced and the added revenue brought in through the sludge removal.

2. Basic Background Chemistry

In our analysis of alternative fuels we require a shared understanding of the basic chemistry of fuels and combustion.

Chemistry of Wood

Wood consists of about 50% cellulose, 25% hemi cellulose and 25% lignin (Figure 1).

[pic]

Figure 1 – Chemical Structure of Wood

Note the number of –OH bonds present in the wood molecule. Also note the number of aromatic carbon rings containing carbon double bonds. Wood burns less efficiently than fossil fuels in part because oxygen and water first have to be driven off before combustion of the remaining coal can take place. Figure 2 below is a representation of the chemical structure of coal. It is mostly carbon in the form of aromatic rings.

[pic]

Figure 2 – Chemical Structure of Coal

Chemistry of Treatment Plant Sludge

Sewage sludge is more difficult to identify given the wide variety of components present. Figure 3 shows a proposed structure for the organic components from the Bio-petrol patent. The general structure is similar to wood in the number of aromatic rings present.

Figure 3 – Chemical Structure of Sewage Sludge (Organic Components)

The organic portion of sludge contains mainly lipids, proteins and carbohydrates. The inorganic portion is mainly silt and clay with lower levels of heavy metals. Organics (volatile solids) comprise on average about 80% of the total solids.

Elementally, crude oil and sewage sludge have similar Hydrogen to carbon ratios which is what motivated the conversion research in the first place.

|Type of organic matter |Elemental composition, %wt daf |

| |C |H |N |S |O |H/C |

|Crude oil |84.0-87.0 |11.0-14.0 |0.1-0.3 |0.5-3.5 |1.0-3.0 |1.5-1.9 |

|Coal |66.0-86.0 |5.7-7.0 |0.5-1.9 |0.4-3.5 |8.0-29.0 |0.9-1.3 |

|Wood |48.0-52.0 |5.8-6.2 |0.1-1.5 |- |40.0-45.0 |1.4-1.5 |

|Cellulose |44.4 |6.2 |- |- |49.4 |1.7 |

|Lignin |63.0 |6.0 |- |- |31.0 |1.1 |

|Fats |76.0-79.0 |11.0-13.0 |- |- |10.0-12.0 |1.7-2.0 |

|Albumines |50.0-55.0 |6.5-7.5 |15.0-18.0 |0.3-2.5 |21.5-23.5 |1.7-1.8 |

|Sewage sludge |23.0-44.0 |4.5-6.0 |2.5-7.5 |0.3-1.8 |16.0-24.0 |1.2-1.7 |

Table 3 – Elemental Components of Sludge & Other Feedstocks

Chemistry of Fossil Fuels

Fossil fuels and their bio-fuel counterparts are a mixture of mostly straight chain (aliphatic) hydrocarbons, with some aromatics and other impurities. The straight chain hydrocarbons mainly determine the combustion properties and various fuels are characterized accordingly. Table 3 gives the carbon number ranges for common fossil fuels. We’ll reference these numbers throughout this document:

[pic]

Table 4 – Carbon Numbers of Common Fuels

Gasoline for example is a mixture of simpler, clean-burning, straight chain hydrocarbons (such as octane), and lesser amounts of iso-octane and aromatics such as benzene and toluene.

[pic]

Figure 4 – Gasoline as mixture of straight chain hydrocarbons and additives

Because diesel fuel undergoes less refining then gasoline, the distribution of straight chain (paraffins) and other types of hydrocarbons (aromatics, napthenes, iso-paraffins) is much broader.

[pic]

Table 5 – Constituents of a Typical Transportation Diesel Blend

It’s instructive to compare the two in terms of engine performance. Gasoline engines are spark engines and some energy is lost in pre-ignition of the vapors. Diesel engines are compression engines which means that the explosion in the cylinder occurs due to compression (not a spark). With no pre-ignition losses, diesel engines apply more useful work through larger compression ratios and longer strokes leading to improved miles per gallon. On the down side, diesel blends contain more particulates and are hotter burning resulting in increase NOx emissions (see below).

Chemistry of Biofuels

The biofuels we’ve researched to date include syngas from wood, pyrolysis oils, fischer-tropsch diesel and methanol (wood alcohol).

Syngas: The products of wood gasification in air are N2, H2 and CO gas mainly with smaller amounts of methane (CH4) and some other trace compounds. If steam is used as the oxidant, higher amounts of methane are produced. Table 6 below is a typical syngas composition by volume in air [5]:

Nitrogen N2 51%,

Carbon monoxide CO 27 %,

Hydrogen H2 14 %,

Carbon dioxide CO2 4.5%,

Methane CH4 3 %,

Oxygen O2 0.5%

Tar CH1.522O0.0288 Trace

Table 6 – Syngas Composition by Volume in Air

Though tar is present in small amounts, it constitutes the biggest problem in reliable operation of gas turbines from syngas. The tar quickly builds up inside the engine cylinders causing sticking, and other problems that reduce engine efficiency.

Pyrolysis Oils: Pyrolysis refers to the process of heating in the absence of air. It occurs as a prelude to gasification and is what drives off the volatiles leaving the char that is ultimately gasified. In the production of pyrolysis oils, the biomass is heated above gasification temperatures in a closed container, breaking enough of the cellulose bonds to liquefy it. The hot liquid is rapidly quenched and remains in liquid form. Pyrolysis oils can be properly described as ‘liquid wood’ since the molecular structure and heating value are similar. The uncontrolled process results in a complex mixture of water and various organics including a number of acids.

Fischer Tropsch Diesel: One of the conversion routes of syngas to diesel involves the fischer tropsch process. Here syngas is converted via a high temperature catalytic process to light gasses, a slurry of longer chain liquid fuels (gasoline & diesel) and even longer chain waxes. The waxes can be cracked downstream to increase the yield of diesel. FT diesel actually has a higher heating value than fossil diesel due to the relative absence of aromatics, sulpher and other contaminants. For use as a transportation fuel it is blended with aromatics to bring the heating value down.

Methanol: In a relatively simple reaction sequence discussed below, methanol can be produced from syngas. The chemical formula for methanol is CH3OH. It is a very clean burning fuel with an excellent heating value but it is also highly toxic and flammable.

Dimethlyether (DME): DME (CH3OCH3) has a number of uses in addition to serving as a substitute for fossil fuels. It can be synthesized from methanol catalytically with the application of heat according to the reversible reaction: 2CH3OH → CH3OCH3 + H2O

Corn Ethanol: The chemical formula for ethanol is CH3-CH2-OH

Biodiesel (from transesterification): Made from vegetable oil, biodiesel molecules are short chain esters synthesized in methanol (usually). The R and R’ in the diagram below are either methyl (CH3-) or ethyl (-CH2CH3) groups. They are clean burning and can be used as a direct substitute in diesel engines.

Basic Combustion Chemistry

For the various fuels described above there are only a handful of atomic bonds that we are concerned with (C-H, C-C, C=C, C-O, C=O, H-H, see figure 5 below). The relative strength of these bonds is specifically what determines the energy with which a particular fuel burns. More specifically, the energy released during combustion is the difference between the energy required to break the bonds and the energy released when the new bonds form.

Figure 5 – Combustion Reaction and Bond Energies

By convention energy releases are negative and energy inputs are positive. As implied from the figure above, in complete combustion of a hydrocarbon in oxygen, there are only two products – carbon dioxide and water. This may be generally written:

CxHyOz + wO2 ( mCO2 + nH2O

The subscripts x, y, z and the coefficients w, m and n depend upon the type of hydrocarbon being combusted. Since all combustion processes we are concerned with occur in air (with impure reactants, uncontrolled temperature, etc) incomplete combustion occurs. This may be generally written:

CxHyOz + wO2 + (79/21)wN2( gCO + mCO2 + nH2O + pC(Coal) + qNOx

There are a number of things worth noting. One is that CO and NOx are toxic. CO is a common product of incomplete combustion reactions and NOx , requiring higher temperatures to form, is a common component of diesel exhaust. Another thing worth noting is that energy spent forming CO, Nox and coal is wasted and should be avoided when designing combustion based processes.

What makes for a good fuel?

We are now in a position to describe on a molecular level what makes for a good combustion fuel. This is most easily done by comparing the higher heating values from table 7 below to the molecular formulas, structures and bond energies listed above.

[pic]

Table 7 – Heating Values and Data for Common Fuel Constituents

Note that the higher heating value (used in all of our energy calculations – lower heating values also include the heat of vaporization of water) compares the energy of breaking various C, O and H bonds to the energy of formation of carbon dioxide and water. So straight chain hydrocarbons such as Octane and Heptane, the major components of gasoline and diesel, have very high heating values due to the large number of low energy H-C bonds. Conversely, aromatics such as benzene and toluene have lower heating values because of the number of higher energy, carbon double bonds that have to be broken before CO2 and H2O can form.

For the development of alternative fuels to gasoline and diesel, a ‘good fuel’ fuel will have a similar composition in terms of aliphatics and aromatics and therefore be similar in density for pumping and storage. We’ll see that it’s possible to make even better burning fuels from wood using fischer tropsch but this is not necessarily ‘good’ for engines. It’s a bit more difficult to make comparable fuels from sewage sludge due to the variety of sludge components but it may well be suitable for home heating with no upgrading required.

3. Primary Research on Wood and Sludge Conversion Routes

1. Gasification

Processes which produce liquid fuels from woody biomass begin with either gasification or pyrolysis (the first chemical step in gasification). We will begin with gasification as it precedes the conversion processes required for the synthesis of liquid fuels. Figure 6 describes various conversion routes some of which will be discussed below

Figure 6 – Biomass Conversion Routes

In gasification biomass is subjected to heat in a low oxygen environment resulting in a ‘syngas’ of mostly hydrogen and carbon dioxide. The syngas can be combusted to provide heat, power or work or processed into liquid fuels. Gasifiers were used extensively during World War II as fuel supplies were impacted by allied bombing and valuable experience was gained. Interest in gasification waned after the war with the discovery of new fossil fuel supplies and improved refining technology.

Chemistry: Gasification occurs in three steps. Initially as heat is applied (often through combustion of a small amount of the biomass) and temperatures approach 100C, water is driven off. As temperature increases above 300C, pyrolysis reactions occur and volatiles (H2, CH4, tars) are driven off leaving the solid char. Finally, gasification of the char occurs above 700C with following reactions dominating. The water may be provided by combustion and volatilization or may be introduced separately:

C + H2O ( H2 + CO 131 kJ/mol

C + CO2 ( 2CO 172 kJ/mol

C + 2H2 ( CH4 -75 kJ/mol

The heat for the first two reactions is supplied by the combustion and pyrolysis reactions.

Current Technology: In small scale applications up and downdraft gasifiers are most commonly used.

[pic]

Figure 7 – Up and Down Draft Gasifier Designs

Community Power Corporation, manufacturers of the BioMax Wood-Gas Generator, employ the downdraft design as does the Econoburn gasification based boiler from Alternative Fuel Boilers. For industrial applications, a fluidized bed is more commonly used (Figure 8).

Figure 8 – Fluidized Bed Gasifier

The bed material is generally either sand or char from the process or both and serves a heat transfer medium.

Research Opportunities: Initially we were considering developing a wood-gas generator for commercial sale, on the scale of a Home Depot, 5kW Diesel Genset. We had not considered gasification on a production scale though it is relevant as the first step in the Fischer Tropsch synthesis. The primary areas requiring development are related to cleaning of tars from the syngas and the ability to control the H2/CO ratio of the final product. The later is particularly important to Fischer Tropsch synthesis as it is a primary factor in the determination of the relative gasoline/diesel fraction in the product. For tar removal and destruction, down-draft fixed bed gasifiers are being studied as a promising development however complete destruction is not achieved and fouling issues remain a concern.

2. Wood to Diesel and Gasoline

As indicated earlier, gasoline and diesel consist mainly of straight chain hydrocarbons of varying lengths with aromatics and other compounds blended in to control combustion temperature. Recalling that syngas is about half inert nitrogen, 25% CO and 15% H2, we can see that a reaction sequence must be designed to break the CO and H2 bonds and initiate carbon chain formation with hydrogenation of the remaining carbon bonds. The current state of the technology consists of two process routes. The fischer tropsch synthesis makes a direct conversion of syngas to hydrocarbons which then require upgrading to separate fuel grades (gasoline, diesel, others). Gasoline is also synthesized from syngas via methanol either directly through the Mobil process or indirectly, through DME synthesis.

Fischer Tropsch Synthesis:

The Fischer Tropsch (FT) synthesis, developed in 1923, is more exhaustively researched than methanol conversions. Though the exact reaction chemistry remains uncertain, FT synthesis has much to recommend it. The quality of fuels produced is extremely high and the distribution of fuel products synthesized is controllable through the reaction temperature, selection of catalysts, pressure, etc. The reaction scheme requires a number of processing steps however and the syngas needs to be very pure to avoid fouling of the catalysts. The reactions are also highly exothermic leading to temperature control difficulties that broaden the range of products.

Process Design: FT synthesis is a multistep process that begins with tar removal from the syngas (Figure 9). The cleaned gas enters a reactor where a wide range of hydrocarbons are synthesized. Secondary cracking and separation processes are necessary to separate the light fractions and break down the waxy deposits.

Figure 9 – Fischer Tropsch Synthesis Process Design

Chemistry: The exact chemistry of the FT Synthesis is an area of research particularly with respect to the mechanisms of chain-growth, however the table below summarizes the key reactions. The desirable reaction is the formation of straight chain paraffin’s and olefins. The reversible water gas shift reaction occurs in the presence of product water. Undesirable alcohols can form with oxygen and the catalyst itself can cause CO2 and water to form or bond directly with carbon.

[pic]

Table 8 - Primary Reactions of Fischer Tropsch Synthesis

The uncertainty of the mechanism has pushed modeling toward use of probabilistic functions to estimate product distributions. None are particularly accurate but some are good enough to give a sense of the effect of various processing parameters. The most popular of these is the Anderson Schultz Flory approach. They start with the assumption that some estimate of the probability of carbon chain growth is available (α). In other words, for a given set operating conditions there is a finite probability that C-C bond will form continuing chain growth, versus the opposite probability that a C-H bond will form and terminate chain growth. The probability estimate is itself the subject of intensive research. One of the better papers relates it to the fraction of H2 and CO in the syngas, and the reaction rates of hydrogen with C1, C5 and C6 bonds (known values).

[pic]

Figure 10 – Hydrocarbon Fraction vs Chain Growth Probability

Figure 10 indicates that the maximum production of a gasoline/diesel fraction corresponds to α = 0.7 to 0.9. This is equivalent to a H2/CO ratio in the syngas of between 2 and 2.5. I caution here that this gives only an indication of direction and that a number of other factors are involved.

The reaction rates of hydrogen with C1, C5 and C6 bonds are used as constant values in the chain growth probability estimates but in reality are functions of temperature and other things. FT synthesis is carried out at either low temperature (200 – 240C) on cobalt catalysts or high temperature (300-350C) on either iron or cobalt catalyst. The lower temperatures favor the longer chain hydrocarbons and the higher temperatures the shorter. There is significant cross over however. In addition to temperature, chain growth is effected by pressure, catalyst type, syngas quality, degree of catalyst fouling and the reactor design itself. There are other considerations as well including undesirable side products (olefins, alpha olefins) leaving the research opportunities wide open.

Reactor Designs: The design of FT reactors is complicated by some features of the reaction sequence and the reaction products themselves. First recall that as the number of carbons in the chain grows, the products go from gases (C1-C3) to liquids (C5 to C20) to waxy solids (C20 +). At any one time the contents of the reactor contain a slurry of liquids and waxes, thicker or thinner depending on conditions, with gasses bubbling up through to the top. As syngas is pumped in it contacts the catalyst, releasing hydrogen and carbon which react with the nearest open carbon site from whatever length hydrocarbon is nearby. If hydrogen reacts, chain growth is terminated and if a carbon reacts chain growth is continued. If no hydrogen or carbon is nearby, a carbon double bond may form in the hydrocarbon creating less desirable olefins, terminating chain growth. Other side reactions can occur such as water gas shift drifting back and forth between water, carbon monoxide, carbon dioxide and hydrogen. The oxygen freed from the catalyzed syngas can also react with the hydrocarbons creating alcohols or to oxidize the catalyst itself. Finally the carbon can bond directly to the catalyst causing coke formation.

So inside an FT reactor is a chaotic mess with ever changing composition. The catalyst which drives the reaction is often prevented from contacting the syngas because it may be gummed up with wax or coke or it may be oxidized. Moreover, each reaction occurring inside the reactor releases a different amount of heat increasing or decreasing the likelihood of other reactions in the mix. As reactor design problems go, this one is not trivial and a number of reactor design have been proposed to keep as much control over the reaction as possible (Figure 11).

[pic]

Figure 11 – Fischer Tropsch Reactor Designs

The key issue in FT synthesis is controlling the temperature at the reaction site. The difficulty of this is evident and the general approach for those interested I using FT synthesis for diesel production is to run the process at lower temperatures, make waxes and crack them in a secondary step.

Economics: Choren, has a pilot plant located in Friedberg, Germany making diesel from woody biomass (most other organizations are using coal or methane based syngas). For a completely self sufficient process, approximately 41lbs of wood (at less than 40% moisture) are required for every gallon of transportation grade diesel produced. At 3 to 8 cents per lb of wood and about $4.00 per gallon of diesel, the revenue is between about $1 and $3 per pound of wood processed.

Corporate Activity: A number of companies are working with fischer tropsch processes using syngas from coal or natural gas. Coal and NG syngas is much cleaner than wood syngas and therefore more suitable for fischer tropsch. The aforementioned Choren has had industrial scale success converting woody biomass to diesel. Their Carbo-V process which produces their Sun Diesel is pictured below. Note the number of process steps required to clean the syngas properly before the FT reactor. Also note that there are multiple upgrading steps following the FT reactor which are not shown.

Syntroleum out of Tulsa, Oklahoma is working on commercialization of a $150M bio-processing plant, scheduled for completion in 2010. They’ve developed an advanced refining capability (Bio-Synfining™ ) that can take as inputs restaurant fats, oils, etc as well as fischer tropsch products including bio-mass.

[pic]

Figure 12 – Choren Fischer Tropsch Process

Other Processes: In general the conversion of woody biomass to straight chain liquid hydrocarbons is accomplished by applying, heat, pressure and catalysis (solvents) in clever ways to maximize the efficiency of the conversion. Fischer Tropsch is the only such process sufficiently advanced for commercialization but it has deficiencies.

• It’s a multistep process that is capital intensive. The basic processes are gasification, gas cleaning, fischer tropsh synthesis and upgrading but each of these could itself be multistep.

• The fischer tropsch process is difficult to control leading to a wide range of hydrocarbon products.

Some other routes are being studied include the following:

• Near/Super Critical Biomass Liquifaction: Biomass can be heated in water to critical temperatures and pressures causing the formation of carbon and hydrogen radicals that condense on cooling to simpler hydrocarbons.







• Catalytic Depolymerization



• Pyrolysis & Liquifaction Upgrading



• Bergius Process



Research Opportunities at BEL : There are two possible applications of fischer tropsch technology we may wish to investigate:

• An FT or FT like technology that could be scaled down in a continuous process for use in combination with a residential scale diesel generator

• A commercial scale FT process that could co-produce diesel fuel for sale along with the sludge-to-diesel liquefaction processes described below.

As previously noted, while the FT process produces extremely high purity aliphatic hydrocarbons, the range of products is wide and difficult to control. To address this, manufacturers lean toward processing at lower temperatures to syntheses waxes to be upgraded in an additional process step. These steps known as cracking, are similar to those used in refining of fossil fuels and involve subjecting the waxes to high temperatures or introducing a catalyst. Catalytic cracking yields higher quality products and fluid catalytic cracking (FCC)is the most popular method in use. FCC is primarily used for gasoline production but also has applications for diesel.

A considerable amount of academic research is in progress to model and better define the specific reaction mechanisms of FT synthesis. This may be in part due to the renewed interest in alternative energy. At BEL, our research would be directed toward process engineering and reactor design to address the following:

• Control and process design methodologies to narrow the range of hydrocarbons produced, eliminating the need for secondary processing steps.

• Reactor designs that do not require separation of catalyst from waxy slurries

3. Sludge to Diesel

Early on we identified municipal treatment plant sludge as a potential feedstock for an alternative energy process largely because a fee paid to have it removed. Sub/Super Critical water treatment or hydrothermal liquefaction also emerged as a promising conversion route out of some studies conducted at Pacific Northwest National Labs among others. The studies at PNNL ultimately resulted in the sludge-to-oil (STORS) technology which has been piloted in Northern California and soon-to-be in China, in combination with a denitrification technology/ammonia recovery process. The Colton, California demonstration occurred between May and November of 2000 however few details are available on the specifics of the process.

Compared to bio-mass in general there is relatively little academic research in the conversion of sludge to fuel (apart from aerobic/anerobic digestion). The more recent studies seem to focus on gasification and pyrolysis both of which produce a fuel of lower quality than liquefaction based processes. Typically, the research approach to sludge is as a waste disposal problem. Sludge liquefaction was studied in the past mainly as a means of degradation that also had the potential to recover some value.

In my early investigation, I came across Bio Petrol – a small group of PhD’s from Israel seeking funding to build a pilot scale sludge-to-diesel, thermochemical liquefaction processes in cooperation with the City of Jerusalem. More recently I’ve been evaluating some research out of Japan that compares quite favorably to the Bio Petrol concept. Sufficient information is available on these two processes to use them as a basis for the discussions that follow. The key difference between the two is that the Bio Petrol process is a two stage process in a solvent, designed to extract all of the oil in liquid form. The Japanese process is straight liquefaction in water, recovering some of the oil in liquid form and burning the remaining oil adsorbed in the solids to fuel the process.

Hydrothermal/Thermochemical Liquefaction:

Heating a substance in water to temperatures of 300/350C and pressures of 12 to 20 MPa is known as hydrothermal liquefaction (also hydrothermal upgrading, sub critical treatment - for reference the critical temperature of water is 373.9C and the critical pressure is 22 MPa). If a solvent or catalyst is added it is referred to as thermochemical liquefaction. The point of either process is to break down the complex hydrocarbons present in sludge into radicals and condense them back into aliphatic hydrocarbons (oils) with improved heating values. The topic does not seem to be exhaustively researched however a fair amount of research was conducted on liquefaction processes in the 1970’s and 80’s, mainly on its application to woody biomass. A handful of studies have been conducted on application of the process to treatment plant sludge, most of which occurred in Japan.

Process Design: Both the Bio-Petrol and Japanese processes begin with dewatered sludge (~20% solids cake), the form most commonly landfilled or incinerated by treatment plants. The Bio-Petrol process pictured below in figure 13 has been tested on the bench and they are in the process of procuring funding for a 1 ton per day pilot plant. The process requires additional dewatering to a solids content of about 90% before the dry sludge is fed into the slurry preparation vessel where solvent is added (the solvent is the re-circulated oil fraction that flashes above 350C). The slurry is fed into the reactor and heated to 350C and the products separated to light oils, gas and solids. The solid slurry enters the second pyrolysis reactor and its vapor products are then separated into more light oil, water and a fraction that undergoes a final distillation step to separate the heavy oil from an even heavier fraction used as solvent. The light and heavy oils are mixed in the final product.

[pic]

Figure 13 – Bio-Petrol Process Diagram

The Japanese process described in the paper by Itoh et al and pictured below in figure 14, has been tested in a 5 ton per day pilot plant. The liquefaction process is hydrothermal (in water, not solvent) though a solvent extraction is conducted separately on the distillate (and the solids for testing purposes). The dewatered sludge is charged to two piston driven injection tanks and forced into the reactor. The phase separator (separator 1) conveys the vapors from the reactor over the feed to pre-heat it, before they are further cooled and separated to distillate and gases. The remaining slurry from separator 1 is sent to a flash tank where the bottoms are drawn off and the gas condensed.

[pic]

Figure 14 – Thermochemical Liquefaction Process

As indicated earlier, it’s possible in the Biopetrol process to extract nearly all of the available oil and sell it. If however the oil is used to dry the sludge to the > 90% solids content required, the final yield is about 3% by weight on a wet sludge basis (of 37.2 MJ/kg oil). The Itoh process yields about 2.5% (of 38.7 MJ/kg oil). On an energy basis Biopetrol yields about 7.8% and Itoh about 6.4%.

What makes the Itoh process more economically viable (table 2) is that it requires much less input energy per kg of oil produced. In the BioPetrol process, the oil is used to dry the sludge and a substantial amount of electricity is required to power the two large reactors. In the Itoh process, the dewatered sludge is charged to the reactor as-is and the oil-soaked solids are burned as-is for the reactor heat. The Itoh process does not specify the energy requirements but applying conservative estimates for the sludge pump and injection tanks, it appears to be less than half of Bio-Petrol. The result is an estimated 30% increase in revenue.

Chemistry: The chemistry of liquefaction in water and solvents is even less well characterized than FT synthesis. Very generally, it proceeds in a similar manner. The carbon-carbon and carbon hydrogen bonds present in the organic matter are broken under the applied heat and pressure creating elemental and compound radicals which re-form as simpler, aliphatic hydrocarbons, alcohols, ketones and a number of other things (over 70 different compounds have been indentified in sludge liquefaction oil). Catalysts such as sodium carbonate have been used as well as solvents including higher boiling fractions of the liquefaction products themselves (Bio-Petrol).

Table 8 below lists various compounds found in oil from sludge liquefaction products. The neutral fraction in this study accounted for over half of the oil and it contains the compounds found in fuel oils (alkanes, alkenes, aromatics) though specific fractions of each are not available in the literature.

[pic]

Table 8 – Composition of Sludge Liquefaction Products

In short the reaction chemistry of hydrothermal and thermochemical liquefaction is not discussed in detail in any academic literature that I have access to.

Reactors: For a bench scale reactor system a number of autoclave-type reactor designs are available. For pilot scale, both Bio-Petrol and Itoh appear to be using a reactor of the type pictured below (US Patent US7329395B2).

[pic]

Figure 15 – Hydrothermal Liquefaction Reactor

The feed enters on the left through a screw mechanism and is heated to liquefaction temperatures in sections 3 and 4. Cooling occurs in section 5. They are relatively long and narrow – the Itoh process reactor is 0.43m in diameter and 7.25m in height.

A second reactor design (figure 16) has been developed by Fessbender out of Batelle. It’s designed to address problems with high temperature and pressure reactors including shaft leakage and corrosion which are accelerated by the process conditions.

[pic]

Figure 16 – Fessbender HTL Reactor

The inner mechanism is the liquefaction reactor and the outer shell contains a non-compressible fluid pressurized to the operating pressure. The seals on the inner reactor are pressure balanced and less subject to wear and leakage. The slurry enters through 115 is heated electrically by the apparatus at 120, The scraper mechanism in the middle scrapes the sides and stirs the slurry for uniform heating and finally the slurry leaves the reactor at 128.

Economics: The economics of sludge-to-oil processes have already been discussed but it bears repeating that they significantly over shadow any biomass conversion process yet studied. There is limited data available for capital costs estimates however the previously mentioned STORS plant in Colton, California was $3 million dollars for a plant processing 20,000m3/day. The Bio-Petrol business plan requests $3.8 million for a plant expected to process 231 m3/day. I’ve mentioned that the Bio-Petrol plant is a two stage process and includes additional hardware for drying however insufficient information is available to properly compare the capital required for the two processes beyond that.

Corporate Activity: The previously mentioned STORS technology was started by Batelle in cooperation with Pacific Northwest National Labs and licensed to Thermoenergy. They designed, built and operated the 5 mgd STORS facility at the Colton, California Wastewater Treatment Plant during 2000. This large plant, also utilized Thermoenergy’s ARP to convert raw, digested and waste-activated sludge into fuel (either oil or coal) and ammonium sulfate crystals (fertilizer).

The status of the Bio-Petrol project is that they have completed a lab-scale demonstration and are seeking funding for a 60 TPD (wet) plant to be followed by a 240 TPD pre-commercial plant. Their concept is to build the conversion facilities at the treatment site.

What little academic literature is available comes mostly from Japan. Some of the studies imply that the technology is being applied there but I can find very little corporate activity or public information. It may be that the sludge-to-oil plants operational in Japan are directed and funded by government entities.

Other processes: The two main routes under study for conversion of municipal sludge to fuels involve either hydrothermal/thermochemical liquefaction or gasification followed by fischer tropsch or other upgrading processes.

• Steam-Hydrogasification (FT Based Process for sludge and other feedstocks):



• Enertech: Sludge to Coal



• Nedo: Sludge Treatment by HTL:



• Castion: Thermal Treatment of Municipal WWTP Sludge for low quality fuel oil



• Conversion of Petrochemical Plant WWT Sludge by HTL:



• Patent for Converting Municipal WWTP Sludge to Oil using CO:



Research Opportunities at BEL : Essentially all of the public information available (corporate and academic) claims that municipal treatment plant sludge can be converted to fuel oils at a profit based either on lab-scale research or pilot studies. The aforementioned STORS technology appears to have been the most successful commercial attempt to date however they ceased operations in 2001 for undisclosed reasons and it’s not clear if they were making oil or coal. The obvious question is if this technology is so good, why aren’t more people doing it? It may well be related to the price of crude oil (less than half of what is now in 2001) as well as the increasing popularity of aerobic/anerobic digestion methods. Economically, oil synthesis looks only marginally better than digestion if the conversion is conducted on site and presumably has much higher capital costs (note that a large portion of the revenue in our business model comes from removal fees). There are also a number of technical issues that are not well addressed in the literature and require research on a lab scale.

• What is the true quality of the oil produced? Can it be sold as heating oil without further upgrading?

• Can the oil be upgraded to transportation grade fuel and what are the yields?

• How well do the liquefaction reactors hold up to continuous use? What are the maintenance costs?

• What possibilities are there to introduce catalysis and does it improve yields?

Fortunately the academic literature provides enough detail that reproducing the lab set up should be straight forward.

4. Conclusions and Recommendations:

This work was initially undertaken with two concepts in mind – the development of a wood gas generator and municipal sludge to oil conversion through liquefaction. The motivation for the wood-gas generator was the desire to have electricity available in emergency situations such as if the grid was down and liquid fuels unavailable. Sludge conversion was motivated by a vast feedstock supply which is itself a source of revenue. A gasification based technology for a wood-gas generator is not new and has been attempted commercially by Community Power Corporation in Denver with very little progress over the last fifteen years. We elected to focus on conversion of biomass to liquid fuels for possible application in both a residential scale unit and on a larger scale for fuel synthesis. With sludge conversion, initially efforts were focused on vetting Bio-Petrol for possible investment. More recently, I’ve been investigating details of developing a sludge liquefaction process of our own.

Through my research and economic calculations I’ve developed the opinion that a waste-to-energy plant producing liquid fuels may be an idea whose time has come. I envision a plant receiving not only municipal sludge, but also saw mill and paper mill waste and ultimately any carbonaceous solid waste stream. The general processes I’ve studied either gasify carbon based material then liquefy it or liquefy it directly. They are not restricted to either woody biomass or municipal sludge for feedstocks and either feedstock works in both (with process modifications). A key to profitability may be economies of scale. The economics of a sludge-to-oil plant are only marginally improved over digestion methods if the plant is built on site. The additional revenue from removal fees from multiple plants makes the process much more promising even with transportation costs. This with added revenue from woody biomass conversion presents a promising scenario.

Whereas there are certainly a number of research opportunities in FT synthesis and thermochemical sludge liquefaction we may be able to pursue them in parallel with scaling up a version of processes already proven or at least demonstrated. The benefit of this approach is that it provides an opportunity for hands on experience with both pilot and production scale conversion processes and will help motivate better ideas on how to improve them. As seen from the website, several papers have been written testing different solvents and catalysts and some of the concepts have been patented. Some look promising but it’s not really possible to evaluate them on a production basis without pilot plant experience.

References

[1] Dinesh Mohan, Charles U. Pittman, Jr., Philip H. Steele, Pyrolysis of Wood/Biomass for Bio-oil: A Critical Review, Energy & Fuels 2006, 20, 848-889

[2]

[3]

[4]

[5] Taylor, Charles Fayette (1985). Internal-Combustion Engine in Theory and Practice - Vol.1. Cambridge: The MIT Press, pp. 46-47. ISBN 0-262-70027-1. 

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

Appendix 1: Citing Requirements & Equipment List

[pic]

Appendix 1: Citing Requirements & Equipment List Continued

[pic]

-----------------------

Lignin

Iso-Octane

Benzene

Toluene

Pyrolyzer (400/500C)

Combustion Chamber & Gasifier (1400C))

Heat Exch Deduster WGS Scrubber FT Reactor Sun Diesel

Hopper

Wood chips

Air or Steam

SynGas

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download