I first met Scott Hoover at the Grassroots Biodiesel Conference in Pittsboro in January of 2005. A bunch of us were on our way to Fort Lauderdale to attend the NBB conference, and our imaginations were focused on an impending train trip.
Scott was on his way to a Masters Degree and what held his imagination was the energy balance of our homemade fuel.
I once dipped my toe into an understanding of energy balance, and I wrote about it from my brother’s kitchen table on Georgian Bay. I did a back of the envelope calculation which I published in Energy Blog.
So when Scott wanted to delve deeper, I was delighted.
Once he began this project, two problems immediately emerged. The first was that Piedmont Biofuels isn’t very well suited to the scientific method. We are not very good at keeping records. We often run on gut feel. Communications is not our strong suit. Scott was able to overcome this by checking his horror and plugging in assumptions where there were holes in our data.
The second problem was that his work went way over my head in about the first fifteen minutes of the project. He overcame this obstacle by doing the work himself.
When he finished his work, he stepped me through his results and I was delighted. Scott’s numbers were way more appealing than the ones I had previously come up with , and they squared with some energy balance calculations done by Tom Leue at Homestead.
In keeping with the open source nature of Piedmont Biofuels, we would like to offer all of this research for free to anyone that can benefit from it. Please remember that attribution is a cornerstone of the open source philosophy, so if you use any of this, make sure to give credit to Scott Hoover.
Other than that, I think this is a spectacular piece of work, and I am grateful to Scott for having undertaken it. Enjoy.
By Scott Hoover
declaration | abstract | acknowledgements | glossary | introduction | conceptual framework | methodology | summary | bibliography
School of Engineering Science
Division of Science and Engineering
The project contained herein is my own account of research into the energy balance of a small-scale grassroots biodiesel production facility and any errors are solely my own.
Scott L. Hoover
Several studies have estimated the fossil energy balance of biodiesel production from straight vegetable oils such as soybean or canola oil. However, these results apply mainly to large processing facilities using only straight vegetable oils as feedstock. Using restaurant waste vegetable oil as feedstock, one would expect the fossil energy balance to be much better because of the much limited number of operations needed to obtain the feedstock. This study estimates the fossil energy balance at Piedmont Biofuels co-op; a small-scale, grassroots biodiesel production facility using waste vegetable oil as feedstock. A computer spreadsheet program was designed to facilitate co-op members and other small scale producers in calculating the energy balance of a batch process. The results show that Piedmont Biofuels has a positive fossil energy balance of 7.8:1.
Energy Balance. An indication of the energy efficiency of a process. Expressed as a ratio – Fuel Product Energy:Input Energy
Fossil Energy Balance An indication of the renewable nature of a fuel. A fossil energy balance greater than 1 indicates the process is renewable. Expressed as – Fuel Product Energy:Fossil Energy Inputs
Homebrew Biodiesel. The production of biodiesel in small batches (1 quart to 150 gallons) for non-commercial use.
Embodied Energy. Fuel energy value plus a surcharge for the production process from raw materials.
Lifecycle Analysis. Consideration of all steps in a production process from extraction of raw materials from the ground through to their final use.
Petroleum Diesel. Diesel fuel refined from crude oil.
Process Energy Balance. An indication of the amount of energy needed in the processing of feedstock energy into its final fuel product form – exclusive of the energy contained in the feedstock itself. Expressed as – Fuel Product Energy:Process Energy
Straight Vegetable Oil (SVO). Oil derived directly from the plant crops. Common crops for producing SVO include soybean and rape (canola).
Transesterification. Term for the chemical reaction between vegetable oil and methanol using potassium hydroxide as a catalyst to produce biodiesel fuel and glycerol.
Waste Vegetable Oil (WVO). The used grease and oils from restaurant deep fryers.
Diesel cars have long enjoyed greater popularity in many countries such as Germany and France; Americans however, have favored gasoline powered cars because of the higher price of petroleum diesel and concerns about pollution and noise. Except for commercial use – transportation fleets, off road vehicles, and a few heavy-duty pick up trucks – there are few diesels to be found. In recent years, diesels have been gaining popularity in the USA. Several manufacturers, such as VW and Mercedes, are targeting the household market with diesel sedans and SUVs. Modern diesels are cleaner, quieter, and are admired for their efficiency and fuel use flexibility. More people are running diesel vehicles on biodiesel or straight vegetable oil. Likewise, homebrew biodiesel, or production of biodiesel fuel in small batches for noncommercial use, has grown in popularity, partially due to the internet which has spread information and techniques among homebrewers. Home power production no longer seems limited to just electricity and domestic hot water. Do-it-yourselfers are now making their own transportation fuel.
Piedmont Biofuels Co-op, in Pittsboro, NC, promotes homebrewing of biodiesel from waste vegetable oil as one of its activities. Co-op members who advocate this type of fuel are invariably asked, “but how much energy does it save? Is it efficient?” The answer does not come easily. Back of the envelope calculations will easily show there is more energy in the fuel than it takes to produce, but by how much? How does the process compare with other fuels and other techniques of producing biodiesel? There are no firm answers.
Many studies have reported on the energy balance of biodiesel production using feedstock such as soybean oil rather than waste vegetable oil. However there is a major difference between the energy balances depending upon the source of the feedstock; using waste vegetable oil (WVO) to produce biodiesel essentially recycles an otherwise waste product. Whereas soybean oil (soyoil) used to make biodiesel starts its life on a farm where soybean crops are grown. It is clear that these processes are not the same and thus the energy balances will not be the same. The energy balance of a straight vegetable oil process does not reflect the energy balance of a process using waste vegetable oil.
This paper will attempt to quantify the fossil energy balance of producing biodiesel using waste vegetable oil. It will serve as both a benchmark for other homebrewers to use for comparison and to use as an educational tool for Piedmont Biofuels education and outreach. Homebrew production is a batch process and therefore energy balance will apply per batch of biodiesel produced. In addition, a computer spreadsheet program was created to be used by Piedmont Biofuels and other homebrewers to calculate the energy balance of a batch process.
To lay the groundwork for the project, the first section of the paper will introduce the work of Piedmont Biofuels, familiarize the reader with the biodiesel production process, explain the concept of energy balances, and present energy balances of various fuels for comparison purposes. The second half of the paper will describe the methodology of determining the energy balance of Piedmont Biofuels: collection of data, analyzing the data, and a discussion of the results.
Piedmont Biofuels is a grassroots co-op near Pittsboro, North Carolina with about 75 members. Members come from all walks of life to use or support the use of biofuels. The objectives for Piedmont Biofuels (http://www.biofuels.coop) and its members and member/workers are wide ranging and include the following:
- Provide a space where worker members can make their own fuel from waste vegetable oil.
- Provide pure commercially produced biodiesel (B100) to the community.
- Have a USDA Research Farm for oilseed crop research.
- Have an elaborate glycerin composting facility.
- Do education and outreach on both biodiesel and engine modifications that enable people to use straight vegetable oil (SVO) as fuel.
- Lobby the North Carolina legislature, as well as national representatives, on behalf of biodiesel and alternative fuels.
- Have an intern program that allows people to live on-site and learn about all facets of operations.
The facility consists of a home which has been converted into the refinery and office space, a fueling station, farm, and housing for interns. A view of the co-op can be seen in figure 1. The co-op also owns a 1600 gallon oil tanker truck which is used to procure and deliver commercial soydiesel to 5 biodiesel fueling stations in the area.
The co-op began with the efforts of Lyle Estill and other community leaders. (Estill, 2004, Spring/Summer) The first batches of biodiesel were made, as is often the case, in a kitchen blender. Working through a fledgling biofuels course at the local community college, batches were upsized to produce larger and larger yields of fuel. Many failed batches of biodiesel were produced as they experimented with different processes and formulas. Eventually, they were able to produce biodiesel on a consistent basis and the reactor size grew into the present 75 gallon yield reactor. As the biodiesel yields increased in size, so did the interest in biodiesel. More and more people were taking the biofuels course and demanding biodiesel. At times, eleven people might show up to help make a ten gallon batch of biofuel. Everyone would drive away with a great sense of accomplishment, but hardly enough fuel to get them home. It was decided to form a co-operative in order to continue the work they were doing and be able to supply the demand for commercial biodiesel. Piedmont biofuels is constantly growing and is in the process of upsizing to a 120 gallon reactor. In addition, there are plans to open a commercial biofuel refinery producing about a million gallons of biodiesel a year from straight vegetable oil.
PIEDMONT BIOFUELS BIODIESEL PRODUCTION PROCESS
Biodiesel is relatively easy to make, inexpensive, and uses simple and cheap equipment as compared to ethanol production. This makes it especially attractive to people who often make, or homebrew their own biodiesel at home in small batches. The simple chemical process of biodiesel production is called transesterification. In the process, vegetable oil is reacted with methanol using potassium hydroxide (KOH) as a catalyst. Methanol replaces the glycerin in the WVO. The products are biodiesel and glycerol. To better understand the energy balance of the process, the production process as done at the refinery is explained in the following paragraphs and figure 2 shows the process diagrammatically. These are presented for conceptual understanding of the process rather than as complete instructions. Portions of this section are taken from Alovert. (2005)
Transportation and Treatment of Waste Vegetable Oil
The co-op produces biodiesel mainly from waste vegetable oil (WVO). This is the used oil left over from restaurant fryers. The co-op collects WVO in 200 to 250 gallon batches in large plastic containers. The used oil is run through a screen to filter out pieces of food, breading, or other large particles. The used oil needs to be dried first to remove as much of the water from the oil as possible. Water in the oil, in the worse case, will prevent the transesterification process and yield a brown gel that is unusable. Dewatering is done by putting 75 gallons (one batch in this case) of the WVO in an insulated vessel with a jacket heat exchanger. Hot water, heated using solar thermal panels, is circulated through the jacket heat exchanger. Previously, gas was used to heat the water, but later the solar hot water system was built from scratch to increase the efficiency of the process. The heat helps the oil and water to separate. After the water has settled on the bottom, the watery oil at the bottom is drained out a spigot leaving the dry oil in the vessel.
While the oil is drying a titration should be done on a sample of the oil to determine how much free fatty acid is in it, and thus how much additional potassium hydroxide (KOH) catalyst to add to the methanol. Free fatty acids (FFAs) are present in WVO and will combine with KOH to produce soap. If extra KOH isn’t added to compensate for this reaction, there would not be enough KOH available as catalyst, and the biodiesel would not be fully converted. Too much KOH and an unusable glop can result. It is important that the titration be done carefully so an accurate amount of KOH can be added. Titration is a simple process and involves keeping track of how much KOH/water mixture it takes to neutralize the FFA in an oil sample. A pH indicator is used to tell when the solution reaches about pH 8.5. The results of titration are used to determine the amount of catalyst to add to the methanol. At piedmont biofuels, .7477oz of KOH is added per gallon of oil plus the extra amount determined by titration. At this point, a liter batch is usually made to test how the reaction will proceed.
Mixing Methanol and Potassium Hydroxide
After the amount of catalyst has been determined, the catalyst needs to be mixed with methanol. Methanol is flammable and dangerous. The safest way to handle methanol is in a closed system which reduces the possibility of contacting the methanol or breathing its fumes. The methanol (22% by volume of oil- about 16.5 gallons for the 75 gallon batch) is piped from its storage tank directly into carboys (sturdy HDPE plastic jugs). The amount of KOH determined in the titration is then measured and added to the methanol in the carboy. The carboy is immediately re-sealed, picked up, and gently swirled to begin dissolving the catalyst. The carboy should set for 15-30 minutes, and then be swirled and let set again. This process should continue until visual inspection shows that all the catalyst has dissolved.
The methanol/KOH solution and WVO are then ready to be reacted. The still hot WVO is pumped from the drying vessel into the reaction vessel. The reaction vessel is a 100 gallon vessel with a circulating pump which circulates the mixture through an external mixer that provides intimate contact between the reactants. The lid on the methanol carboy is replaced with a lid with a hose barb attached. The methanol is piped to the reaction vessel and the circulating pump agitates the mixture for about two hours to insure a complete reaction.
The mixture is allowed to set unheated for 8-36 hours to let the darker byproduct (mainly glycerol) settle to the bottom. The remainder is biodiesel and some contaminants. The glycerol and byproducts are drained off the bottom through a spigot in the reactor vessel. The Piedmont Biofuels batch process will yield about 75 gallons of biodiesel (about the same as the beginning amount of WVO used) and about 16.5 gallons of glycerol (about the same as the beginning amount of methanol used). The glycerol at Piedmont Biofuels is either composted or fed to goats on the farm (apparently they love the stuff and prefer the feed mixed with glycerin over the straight feed).
Washing the Biodiesel
Bubble washing with water will remove most of the impurities from the biodiesel. The biodiesel is pumped into a large bubblewashing tub. This is a large flat-bottomed, open tub with an aquarium air stone in the bottom attached to an aquarium air pump. Water (about 40 gallons for the 75 gallon batch of biodiesel) is gently added to the biodiesel so the two liquids don’t mix. The water will settle to the bottom of the biodiesel. Turn on the aquarium air pump, and let it bubble for a few hours. The rising air bubbles carry with them a layer of water surrounding them. This moisture will take in soap and water soluble contaminants. When the air bubble gets to the surface, it pops, and the water turns into a water droplet and falls back down to the bottom of the tub, picking up more contaminants on the way. Thus, the contaminants will remain in the water in the bottom of the tub. The wash water should be drained and changed to remove the accumulating contaminants. The biodiesel should be washed this way two or three more times for about 6 to 8 hours each bubbling. When the wash water is clear and has the same pH as tap water and the biodiesel is clear, the washing if finished.
The biodiesel should be left in the open tub so the water can settle and evaporate out of the fuel. This can take a few hours to a few weeks depending on the quality of the fuel. The better the fuel is washed, the shorter the drying time.
The final step is to filter the biodiesel to 5 microns to remove any abrasive particulates. The fuel is now ready for use in any diesel engine with no modifications.
Primary Energy Balance An energy balance is simply a ratio of the energy of the fuel product to the energy inputs. For example: .83:1 means for every unit of energy input used, .83 units of energy are available in the fuel product. However, it is very important to identify the type of energy balance being defined. Sheehan, Camobreco, Duffield, Graboski, and Shapouri (1998, Sect. 4) identify three main types of energy balances: Total primary energy balance (the .83:1 example above), process energy balance, and fossil energy balance.
Total Primary Energy Balance
Total primary energy balance considers the cumulative energy content of ALL resources extracted from the environment. There are two subsets of primary energy: feedstock energy (energy value of crude oil in the case above), and process energy (energy used in the operations to extract, transport, and refine the crude oil into a fuel product). Of these two subsets, the type of energy balance based on process energy is the only one of importance. Total primary energy balances are in keeping with laws of thermodynamics. Energy can neither be created nor destroyed – only changed in form.
Process Energy Balance
A second type of energy balance is process energy balance. It involves only the process energy subset of primary energy. Process energy is energy needed in the process of converting feedstock into final fuel form, but does not contribute to the energy of the fuel product itself. The energy of the feedstock is not considered in a process energy balance. Process energy balance is an indication of the efficiency of the process used in producing fuel from feedstock. Petroleum Diesel has a process energy balance of 4.98:1. For each unit of energy that goes into the processing of crude oil, 4.98 units are available in the fuel product. This is not a violation of the laws of thermodynamics, because the feedstock energy inputs are not considered.
Fossil Energy Balance
Biodiesel homebrewers are mainly concerned with the fossil energy balance because they are concerned about the renewable nature of biodiesel. The fossil energy balance tracks all energy inputs (whether from feedstock subset or process subset) that are from fossil sources such as coal, petroleum, and natural gas. The higher the fossil energy balance is greater than 1, the more renewable we consider the fuel. A fossil energy balance of 1:1 is still nonrenewable because no energy is lost in the process of converting the fossil energy to a usable fuel. The fossil energy balance for petroleum diesel is .83:1. This ratio indicates the nonrenewable nature of petroleum diesel: it has a much higher input of fossil fuels than biodiesel because the feedstock itself, petroleum, is a fossil fuel. WVO, the main feedstock at Piedmont Biofuels, is a recycled waste product from renewable sources, and is not counted as a fossil fuel. It should be expected that the fossil energy balance of WVO to be higher than that of petroleum diesel.
Energy balances provide a way to compare different processes and production facilities. As indicated above, it is important to define the type of balance being given. It has happened that the fossil energy balance of one process has been compared to the process energy balance of another and other gross misuses of energy balances. Most often, as in this report, when an energy balance is given, but not specified by type, it is the fossil energy balance. Likewise, fossil energy balance is the focus of this paper.
The energy balance reported here is a study of the life cycle inventory for biodiesel produced from restaurant waste vegetable oil. Life cycle inventories trace feedstock inputs from their first extraction of raw materials from the ground through their processing into their usable form. Petroleum diesel feedstock traces its lifecycle back to crude oil in the ground. Waste vegetable oil starts its life as a waste in a restaurant dumpster. Because WVO is a waste product, the energy balance can exclude the energy involved in crop agriculture, transporting the crop to crushers, extracting the vegetable oil at the crusher, and transport of vegetable oil to the restaurants. Life cycle inventory for this study excludes any end uses of the fuel. However, it includes energy resources consumed in the following steps
- Transportation of WVO to Piedmont Biofuel’s refinery
- Treatment (dewatering) of the WVO
- Conversion of WVO to biodiesel
- Treatment (washing) of the biodiesel
Energy credits are not applied to the production of glycerol (a byproduct) because of its low value, although the glycerol produced at Piedmont Biofuels is utilized on the farm. This report does not attempt to quantify emissions, water usage, waste streams, cost effectiveness, externalities or anything other than energy balance.
Energy balances also have to reflect embodied energy. In some cases, embodied energy and lifecycle analysis are considered to be similar concepts. For this paper, the two are distinguished differently: whereas lifecycle analysis is applied to the feedstock subset of primary energy, embodied energy applies to the process subset of primary energy. Embodied energy is a surcharge added to the value of the energy used to account for its production from fossil fuel raw materials or for materials used to capture the energy. For example, if 30 kWh are used in a process, then a surcharge of 80 kWh is added to account for the losses involved in producing that 30 kWh. Production of electricity is only about 30% efficient when all losses are considered.
REVIEW OF LITERATURE
Biodiesel Made from WVO
There are no reports of energy balances on production of biodiesel from WVO other than “back of the envelope” type calculations. These calculations range from 4.13:1 to 5.26:1. Leue (personal communication, 2004) bases his calculations on published reports of energy balances on production of biodiesel from soybean oil. The conversion process is much the same for production of soydiesel and biodiesel from WVO. However, there are some major differences in the lifecycles which could be expected to lead to different energy balances. Another problem with this approach is one of economies of scale. Homebrew biodiesel is done in small batches whereas the refineries the energy balances are taken from are large scale facilities. This disparity also may be cause for a difference in energy balance. In addition, Piedmont Biofuels uses renewable energy for many operations in its process, while published reports usually assume all inputs are from fossil fuels. Thus energy balance of Piedmont biofuels may be expected to be higher than that of the published reports. A more direct approach would be preferable for determining the energy balance.
Biodiesel Made from Soybean Oil (Soydiesel)
Several published studies report the energy balances of biodiesel produced from soybean oil. The frequently cited fossil energy balance for soyoil biodiesel (soydiesel) is 3.2:1. (Sheehan et al. 1998) That is, for every unit of fossil energy input, 3.2 units of fuel energy are produced. This report by the National Renewable Energy Laboratory (NREL) evaluated the energy balance of petroleum diesel as cited in the Energy Balances section above. This NREL report is also a lifecycle analysis of the energy balance. It doesn’t include credits for the glycerol byproduct (as some reports do) because the study’s focus was on the use of soydiesel fuel and because it didn’t envision a market for the vast amount of glycerol produced. This study also assumes that fossil fuels are used for all inputs. For example, tractors used in agriculture are assumed to run on petroleum diesel whereas they could just as well be run on biodiesel.
A report by Ahmed, Decker, and Morris (1994) evaluated energy balance of soydiesel for three cases: national average- 2.51:1, industry best- 3.24:1 and industry potential- 4.10:1. However, credits for process co-products, soy meal and glycerol, were included in this evaluation whereas they weren’t in the NREL report. (Sheehan et al., 1998) When co-product credits are taken away, the national average energy balance falls to 1.45:1. Energy in transportation is not mentioned in the ILSR report whereas it is specifically mentioned as being included in the NREL report. Energy in transportation is a small proportion of all energy inputs into the process, yet when counted, it would lower the energy balance given in the ILSR report.
Energy balances of biodiesel should increase with improved technology. One such improvement is production using a continuous (rather than a batch) process which recovers methanol and process water. West Central Soy claims an energy balance of 7:1 using such a process. (Biodiesel: A cleaner greener fuel, 2003) This figure, however, is not a lifecycle analysis and it is unclear whether this is a fossil or process energy balance. Without knowledge of the details of the energy balance, the 7:1 ratio seems to be of questionable use.
In the USA, soybeans are the primary crop for refining into biodiesel. Around the world, many different crops, with various oil energy values, are used for refining into biodiesel. In Europe, oilseed rape, a relative of canola with higher oil energy content than soybean, is commonly used. The British Association for Bio Fuels and Oils (BABFO) reports the energy balance using this crop for biodiesel production as 1.78:1. (Richards, 2000) This is a lifecycle study which includes transportation and excludes byproduct credits.
Algae Biodiesel and Ethanol
In the future, algae may be used for refining into biodiesel. Algae have very high oil content – up to 50%. Using feedstock with higher oil content combined with improvements in processing can potentially yield much higher energy balances. (Briggs, 2004) Another advantage is that algae can be grown in desert regions where it won’t compete for land with food crops.
The closest liquid biofuel competitor to biodiesel is ethanol. The main crop for ethanol production in the USA is corn. There are many more studies on the energy balance of ethanol than there are on biodiesel. An often cited balance of 1.34:1 is reported by Shapouri, Duffield, and Wang (2002). Other studies, most notably Pimentel (2003), calculate negative energy balances for ethanol. That is, more energy is required to produce ethanol than is available in the ethanol fuel. However, subsequent studies have rebuked this claim. Yet other studies place the energy balance of ethanol between 1.11:1 and 1.67:1 (Shapouri et al., 2002 p2) “Well below that of the accepted balance for soy biodiesel of 3.2:1. Iogen Corporation’s new state-of-the-art ethanol production facility may achieve higher energy balances using their process to produce cellulosic ethanol. Table 1 on the next page provides a summary of some of the energy balances reviewed in this section.
|Process Energy||Fossil Energy|
|Ethanol (Shapouri, 2002)||–||1.34:1|
|Petroleum Diesel (NREL, 1998)||4.98:1||.83:1|
|Rape (Canola) biodiesel (Richards, 2000)||–||1.78:1|
|Soydiesel (Ahmed, 1994)||–||1.45:1|
|Soydiesel (NREL, 1998)||4.3:1||3.2:1|
|WVO Biodiesel (This Project)||–||7.8:1|
Table 1: Summary of Energy Balances Reviewed
Collection of Data
At the onset of this project, the author knew little about biodiesel. The author’s first introduction to biodiesel was at the National Grassroots Biodiesel Conference in January 2005. A review of literature on the subject provided a sense of the process of biodiesel production and the energy inputs required. Further literature review provided the framework surrounding an energy balance. The biodiesel production process and discussion on energy balances are presented in the previous section, conceptual framework. Further data on energy inputs was collected by a walk through energy audit of the refinery, conversations with co-op worker-members, and hands-on production. Inputs per 75 gallon batch are: soydiesel, 1.07 gallons; electricity, 1.73 kWh; methanol, 16.5 gallons; KOH, 3.5 pounds plus amount determined by the titration; and heat in the form of hot water.
Fossil Energy Inputs
The amount of methanol and KOH used in the process is fixed and easily determined: 22% methanol by volume of oil, and .748 oz of KOH per gallon of oil plus additional amount determined by titration. Therefore, the amount of methanol per 75 gal batch is a constant 16.5 gal. KOH will vary per batch depending on the titration results. However, at least 3.51 lb per batch will be used. The additional KOH required by titration will not be considered and should not have a large bearing on the results.
Determination of the amount of biodiesel used is likewise fairly straightforward. A distance of 40 miles is traveled in a vehicle that gets 14 MPG using a total of 2.86 gallons of soydiesel. Between 200 and 250 gallons of WVO is collected per trip. For conservative calculations 200 gal will be considered. 200 gallons of WVO will make 2 2/3 batches. Total soydiesel used per batch is then 1.07 gallons.
The greatest uncertainty is in the amount of electricity used. A survey of the facilities reveals the use of: a mixing/transfer pumps, an aquarium air pump, lights, and a few other miscellaneous appliances. Other appliances not directly related to the production of biodiesel were not considered, for example, electricity used for the intern housing. Length of use for the appliances can vary considerably depending on say, the impurities in the biofuel and thus the time needed for washing. Therefore, the author allowed for a 20% additional power use.
After determining and quantifying all the energy inputs, the fossil energy inputs had to be identified. The fossil energy inputs were determined to be electricity, methanol, KOH and a portion of the soydiesel used. Piedmont biofuels uses grid electricity which is a mixture of coal, nuclear, natural gas, heating oil and large hydro. The commercial methanol used is most likely reformed from natural gas. A surcharge for the production of KOH from fossil fuels is added to the inputs. All the biodiesel inputs were assumed to come from commercially produced soydiesel. The NREL study (Sheehan et al., 1998) found soydiesel to have a fossil energy balance of 3.2:1, which is a fossil fuel efficiency of 320%. This data is used to determine the portion of the soydiesel input from fossil fuels. No consideration was given to the embodied energy in the solar collectors because they were made using mainly scrap materials. There was no consideration of the embodied energy in the equipment used to produce the biofuel.
In addition to the fossil energy inputs, the energy outputs had to be determined. The fuel product is biodiesel which is produced in 75 gallon batches. As mentioned, the NREL report (Sheehan et al., 1998) doesn’t assign a value to the unused glycerin byproduct. Piedmont Biofuels utilizes the glycerin and therefore it initially seemed fair to give an energy credit to glycerin byproduct. Amhed et al. (1994) identifies three ways to assign an energy credit to the glycerin product: Weight Energy Credit, Value Energy Credit, and Replacement Credit. Weight Energy Credit is based on the weight fraction of glycerin produced, Value Energy Credit is based on the market value of the products, and Replacement Credit evaluates the energy that would be displaced by not producing the replaced product. Based upon Amhed’s work, it was decided not to assign energy credits to the glycerin: its market value is very low and the products the glycerin replaces at Piedmont Biofuels (goat feed and compost) are made with little energy.
Analysis of Data
Using the data collected, a computer spreadsheet program was created to easily analyze the data and allow other homebrewers to determine the energy balance of a batch process. Figure 5 shows an image captured from the program. The program is designed such that any co-op worker-member could use the computer program to determine the efficiency of any batch produced regardless of the size of the batch. As such, comments are included in the spreadsheet to assist the user. The worksheet has been made flexible to allow for constantly changing techniques and up-sizing of the batch process.
Users work through sub-worksheets to estimate electricity and biodiesel use by entering data in the yellow highlighted fields. Other highlighted spreadsheet fields are available for users to enter amounts of methanol and KOH used in the process and biodiesel yield. Calculations of the energy balance (energy output/energy input) then will be computed automatically. Results are presented for a fossil energy balance. The automatic calculations performed by the spreadsheet are explained in the following paragraphs and a flowchart of the operations and energy inputs is presented in figure 6.
To begin analyzing the data, values first had to be attached to the energy inputs. Fuel values were easily determined from tables of fuel properties. (Properties of fuels, 2005; Estill, personal communication, 2004 March 27) BTU is the unit most often used for fuel values in America and were kept in this report. Low heating values were used in this report and other reports reviewed. Any combustion of fuel will produce carbon dioxide and water. Low heating values take into account the energy lost in vaporizing this water. Since combustion of fuel cannot occur without vaporizing water, it is just to use this lower heat value. (Sheehan et al., 1998) Energy values are summarized in the first two columns of the table 2 on the next page.
|Energy Source||Energy Value||Lifecycle Surcharge||Total Assigned Value|
|Methanol||56,800 Btu/Gal||11,634 Btu/Gal Based on 83% efficiency||68,433 Btu/Gal|
|Electricity||3,415Btu/kWh||7,968.3 Btu/kWh Based on 30% efficiency||11,383.3 Btu/kWh|
|KOH||N/A||11,275 Btu/lb||11,275 Btu/lb|
|Soydiesel||128,300||-88,206 Btu/Gal Based on 320% fossil efficiency||40,094 Btu/Gal|
A surcharge needs to be added to the energy value to account for the embodied energy of these inputs. Surcharges were taken from previously published reports. Analysis of inputs from the NREL report (Sheehan et al., 1998) show electricity is calculated at 30% efficiency. Embodied energy surcharges for methanol were taken from a report from the Institute for Local Self-Reliance. (Ahmed et al., 1994) Natural gas is reformed into methanol at 83% efficiency for this report. Soydiesel is the only energy input to have a positive fossil energy balance. So only a portion (31%) of the biodiesel used is counted as a fossil fuel input. (Sheehan et al., 1998) The process examined in the Ahmed et al. (1994) report used NaOH rather than KOH as a catalyst and assigns a value of 11,275 Btus per pound to its manufacture. This number is accepted as a close representation of the manufacture energy value of KOH. The last two columns of table 2 above lists surcharge value, and total embodied energy.
Total fossil fuel inputs are then the sum of the fossil inputs multiplied by the embodied energy value of the fuel. Table 3 below summarizes the fossil energy inputs.
|Fossil Input||Quantity per Batch||Embodied Value of Fuel||Fossil Fuel Value|
|Methanol||16.5 Gal||68,434 BTU/Gal||1,129,161 BTU|
|Soydiesel||1.07 Gal||40,094 BTU/Gal||42,900 BTU|
|KOH||3.51 lb.||11,275 BTU/lb||39,575 BTU|
|Electricity||1.73 kWh||11,383 BTU/kWh||19,352 BTU|
|Total Fossil Fuel Value||1,230,988|
The fossil energy balance is then the fuel product energy value divided by the total fossil fuel input values. At 128,300 BTU/Gal, total fuel product value is 9,622,500 BTU for a 75 gallon batch. Fossil energy balance is then 7.8:1.
Findings and Discussion
Calculations show the fossil energy balance of Piedmont Biofuels biodiesel processing facility to be 7.8:1. This is a large energy balance which proves the highly renewable nature of production of biodiesel from WVO. The greatest contribution to the fossil energy balance comes from the methanol which contributes 91.7% of the fossil energy inputs. The result is greater than any energy balance reviewed in the Review of Literature section. This is not an unexpected result. There are two main reasons why we would expect a greater fossil energy balance than WVO biodiesel’s closest competitor, soydiesel.
Firstly, as mentioned there is a difference in the lifecycle of the two feedstocks. Soydiesel from soybean oil must count all energy inputs to: produce the soybeans, transport soybeans to a soy crushing facility, and recover soybean oil at the crusher. In the NREL study, (Sheehan et al., 1998) these operations account for 47.8% of total fossil energy required to produce soydiesel. Biodiesel made from WVO does not need to include these energy inputs. Eliminating these inputs from the soydiesel fossil energy balance found in the NREL study (3.2:1) increases the fossil energy balance to 6.2:1. The only operations that need to be included for WVO are transportation from the restaurant to Piedmont Biofuels, and Conversion of WVO into biodiesel.
The second reason we would expect a greater fossil energy balance is that Piedmont Biofuels attempts to use renewable energy in its process wherever it can: solar thermal for process heat and biodiesel for WVO collection. The NREL study assumes energy inputs come from fossil inputs where there is the option of using either fossil or renewable energy.
Biodiesel made from WVO is an excellent alternative to gasoline. Besides its excellent fossil energy balance as determined in this project, there are many other benefits of diesel as a transport fuel. Though there’s not enough WVO to meet demand, an energy source with such a great energy balance shouldn’t be discarded by restaurants. The author hopes that the result of this project will help Piedmont Biofuels promote and develop the use of all biodiesel. Piedmont Biofuels is currently in the process of designing a biodiesel processor for the North Carolina Zoo in Asheboro. When this processor begins operating, it will use food court waste vegetable oil to produce biodiesel to be used in zoo trams.
This study can serve as a baseline for comparing refineries and improving on operations. The internet can serve as a forum where homebrewers from around the world can compare techniques and energy balances.
The author also recognizes the limits of single solutions to problems. Biodiesel alone can not solve the energy problems much of the world is facing today. More should be done to promote increased fuel efficiency of cars such as smaller, more aerodynamic, lighter, diesel hybrid cars.
Throughout the research many misuses and misunderstandings of energy balances were encountered. It seems that fossil energy balances should play a more critical role in decision making – perhaps even at the same level as financial decisions. Hydrogen as a transportation fuel is a case in point: It is highly inefficient (.66:1 fossil energy balance) from an energy standpoint when hydrogen is reformed from fossil fuels. (Spath & Mann, 2001)
To substantially increase the fossil energy balance, methanol recovery techniques could be employed at Piedmont Biofuels to recapture excess methanol in the biodiesel and recycle it. Methanol can also be made from biomass. This biomethanol would be largely renewable and only a fraction would count towards the fossil energy balance.
There seems to be limited verification of the accuracy of some values. Further areas of study such as the efficiency of producing methanol from natural gas and of producing electricity as it is used from the grid would provide useful data.
And finally, the computer spreadsheet program that can provide a record of biodiesel production at Piedmont Biofuels can be used and adopted by other producers and researchers for wider applications.
Ahmed, I., Decker, J. & Morris, D. (1994). How much energy does it take to make a gallon of soydiesel? Institute for Local Self-Reliance. Minneapolis, Minnesota. Retrieved May 16, 2005 from http://www.carbohydrateeconomy.org/library/admin/ uploadedfiles/How_Much_Energy_Does_It_Take_To_Make_A_Gallon_.pdf.
Alovert, M. (2005). Biodiesel homebrew guide: Everything you need to know to make quality alternative diesel fuel out of waste restaurant fryer oil (10th ed.). Unpublished.
Biodiesel: A cleaner, greener fuel for the 21st century (2003). Retrieved May 16, 2005, from http://www.sustainablebusiness.com/features/feature_template.cfm?ID=915
Briggs, M. (2004). Widescale biodiesel production from algae. Retrieved May 16, 2005 from UNH Biodiesel Group: http://www.unh.edu/p2/biodiesel/article_alge.html.
Estill, L. (2004, spring/summer). Biodiesel: How “grow your own” has taken on a whole new meaning. Private Power Magazine, 1(1), 16-22.
Estill, L. (2004, March 27). Energy Balance.
Pimentel, D. (2003). Ethanol fuels: Energy balance, economics, and environmental impacts are negative [Electronic version]. Natural Resources Research, 12(2) 127-134. Retrieved May 16, 2005 from http://acfa.org.sg/library/Ethanol%20fuelsEnergy%20Balance,%20Economics,%20and%20Environmental
Properties of Fuels (2005). Retrieved May 25, 2005 from http://www.methanol.org/pdfFrame.cfm?pdf=fuels.pdf
Richards, I.R. (2000). Energy balances in the growth of oilseed rape for biodiesel and of wheat for bioethanol. Retrieved May 17, 2005 from http://www.biodiesel.co.uk/Levington.htm.
Shapouri, H., Duffield, J. A., & Wang, M. (2002). The energy balance of corn ethanol: An update. U.S. Department of Agriculture, Office of the Chief Economist, Office of Energy Policy and New Uses. Retrieved May 16, 2005 from http://www.usda.gov/oce/oepnu/aer-814.pdf
Sheehan, J., Camobreco, V., Duffield, J., Graboski, M., & Shapouri, H. (1998). An overview of biodiesel and petroleum diesel life cycles. National Renewable Energy Laboratory. Golden, Colorado. Retrieved May 16, 2005 from http://www.nrel.gov/docs/legosti/fy98/24772.pdf
Spath, P., & Mann, M. (2001). Life Cycle Assessment of Hyfrogen Production via Natural Gas Steam Reforming. National Renewable Energy Laboratory. Golden, Colorado. Retrieved June 18, 2005 from http://www.nrel.gov/docs/fy01osti/27638.pdf
Latest Study Shows Ethanol Generates More Energy Than It Takes to Produce (2005). Retrieved May 16, 2005 from http://www.greenbiz.com/news/news_third.cfm?NewsID=27864
Leue, T. (2004, March 09). Energy Balance [Msg 32608]. Message posted to: http://groups.yahooo.com/group/biofuelmessage/32608