Abstract

Unit process and system boundary

The unit processes described in Figure 1 are within the system boundary for the cradle-to-gate LCI analysis of wood pellet manufacturing (“pelletization”). The dotted and solid lines cover pelletization and cumulative emissions, respectively. The pelletization system boundary covered the cradle-to-gate emissions, including emissions generated for material and energy produced off-site, such as grid electricity that is used on-site, but excludes cradle-to-gate emissions for wood residue production. Life-cycle data for wood residue production were provided on a mass allocation basis. About 50 percent of wood leaving the wood flooring manufacturing process was wood residue; therefore, the wood residue carries 50 percent of the burden from the wood flooring manufacturing process. Cumulative emissions (the solid line) cover the cradle-to-gate emissions generated for material and energy produced for both wood residue and pelletization production. The functional unit was 907 kg (1 ton) of bagged wood fuel pellets, which is the standard unit in commerce. The following describes each of the manufacturing processes.

View Figure

• Download as Powerpoint
• Download Figure

• Raw material (waste) collection: Raw material is collected on-site from a connected but separate wood processing facility such as a hardwood flooring mill. Approximately half of the hardwood lumber raw material is converted to flooring; the remainder (sawdust, planer shavings, trim blocks, and edging strips) is available for the pellet process. The raw material is stored in a dry facility on-site.

• Drying: Raw materials, when taken from green feedstock (i.e., roundwood), must be uniformly dried to a low moisture content (9% to 11% on an ovendry basis). The mills surveyed for this study all use waste residue from adjoining facilities, which already has a low moisture content level; however, additional drying is sometimes used even for dry residues. One facility in this study did use a gas boiler, which significantly impacted their energy input. One other facility used wood waste (sawdust) to dry 80 percent of their raw material. Weighted averages of both of these facilities' inputs were used in the inventory analysis.

• Hammer mill: Once the material is collected, it is broken down into small, uniform particles (∼2 mm) using a hammer mill. The hammer mill is operated by electric motors.

• Pelletizer: Pellets (∼6 mm in diameter and 25 mm long) are extruded using machinery that is similar to the equipment used to form feed pellets for the agriculture industry. Pelletizers use large electric motors to extrude the pellets through steel dies. High pressure (∼300 MPa) and temperatures (∼90°C) soften lignin and binds the wood particles together to make uniform and consistent pellets. While no adhesives are required for this process, small amounts of lubricants and water are sometimes added to improve processing.

• Cooling: The pellets are hot when they emerge from the pelletizer, and they enter a counterflow cooler to expedite cooling and evaporation. Then they are stored in a hopper and further cooled before bagging.

• Bagging: Finished pellets are bagged, usually in small, semiautomated bagging lines that are powered by electricity. Pellets are usually fed by conveyor to a bagging station where the pellets are fed into 18.14-kg (40-lb) plastic bags.

Because we were unable to collect data for each component of the pelletization process, this LCI does not separate inputs and outputs by unit processes; instead, data were collected on the processing of wood pellets as a whole system from gate (wood residue) to gate (bagged pellet).

Assumptions

The data collection, analysis, and assumptions followed the protocol defined in Consortium for Research on Renewable Industrial Materials (CORRIM) “Research Guidelines for Life Cycle Inventories” (CORRIM 2010). Additional conditions include the following:

• All data from the survey were weight averaged for the six plants based on each mill's pellet production for 1 year (2009). Missing data were not included in the weighted averages.

• The electricity consumption (the largest input of this manufacturing process) is expressed as kilowatt hours per ton of pellets using electricity data taken from the US LCI database (USDA 2012). Most pellet manufacturers reported electricity bills that allowed us to calculate the energy (kWh) based on their production rate. One mill reported actual kilowatt hour usage.

• Only two mills reported values for oil and grease as inputs. These amounts were calculated per ton of wood pellets. The weighted average for these two reporting manufacturers assumed that the other manufacturers used similar amounts of oil and grease as lubricants.

• Only one mill reported a value for liquid petroleum gas (LPG), 5.71 liters/Mg of wood pellets (1.37 gal/ton). This mill used LPG for further drying before pelletization. This input was weighted according to production percentages.

• Two mills reported values and gave samples of bags used for the finished pellets. It was assumed that these bags were representative of bags used at all of the mills.

• One mill reported using 80 percent green wood feedstock. The mill used heat from burning wood residues to dry the feedstock, and this wood input was weighted according to production percentages and included.

• Environmental impacts associated with the pellet mill equipment and any replacement parts were not included.

LCIAs were performed using TRACI 2 (US EPA 2011). TRACI is a midpoint oriented LCIA methodology developed by the US EPA specifically for the United States using input parameters consistent with US locations. The environmental midpoint impact categories of global warming potential (kg CO₂ eq), acidification potential (H⁺ moles eq), respiratory effects (PM 2.5 eq), eutrophication potential (kg N eq), and smog potential (kg NO_x eq) were examined.

Material flows

The most significant inputs for pellet operations are wood residues and electricity. The surveyed manufacturers reported no cogeneration of electricity. Other fuels used are primarily for rolling stock (i.e., forklifts) and include diesel fuel and LPG. This machinery was usually shared between the pelletization plant and other components of the hardwood flooring operation. The survey respondents were unable to report a fuel usage for these machines specifically for the pelletization operation. Given the limited requirements for these machines in the pellet making process, these inputs were assumed to be insignificant. Other material inputs used in the manufacture of wood pellets are plastic bagging, water, oil, and grease. Water is used to adjust the moisture content during pelletizing. Some oil and grease are used for lubrication in the pelletizing process.

The process of pelletizing wood waste changes only the density of the wood residue raw material. Therefore, it takes 1 ton of wood residues to produce 1 ton of wood pellets. The manufacturing inputs determined from the surveys are summarized in Table 1.

Table 1.— Inputs to the pelletization process for 1 ton of wood pellets in the Southeast.

View Fullscreen

Life-cycle inventory

Wood pelletization primarily has energy and wood residue inputs and only one output—wood pellets. Electricity is used to operate almost all the systems, as described in the next section. Wood pelletization does not create a solid waste stream.¹ All wood residues are recycled in the pelletization process, and airborne particulate emissions (dust) are assumed to be insignificant. Because air, water, and solid waste emissions are minimal, the majority of emissions (Table 2) are those associated with pregate actions (hardwood flooring manufacture and electricity production).

Table 2.— Emissions to air, water, and soil associated with the production of 1 ton of bagged wood pellets from hardwood flooring residues.

View Fullscreen

The cumulative life-cycle emissions and wastes associated with wood pelletization are pregate (i.e., those associated with wood flooring production and electricity production). There are no emission control measures used during pelletization.

Table 2 lists the emissions to air, water, and soil for wood residues production (cradle to gate), for the pelletizing process (gate to gate), and for these two processes combined (cumulative). The data collected by our survey for the pelletizing process are listed under the heading “pelletization” in the table. The cumulative emissions include those associated with the growth, harvest, and transportation of logs and the production of the hardwood flooring that results in the creation of wood residue, and the emissions associated with the production of the electricity (calculated in SimaPro from data provided by Hubbard and Bowe [2010] and the US LCI database [USDA 2012]) that is used during pelletization. Separation of the wood residue production, pelletization, and cumulative impacts in this way enables analysis of the relative importance of the pelletizing process compared with raw materials (wood residues) production. Pelletization-associated fossil CO₂ emission of 125 kg/Mg includes emissions from generation of the electricity used to run the pelletizing equipment and burning LPG to dry the wood residue. In addition, woody biomass is burned on-site for drying the wood residue as shown by a biomass CO₂ value of 72 kg/Mg. Most of the fossil and biomass CO₂ emissions are associated with the hardwood flooring residues production.

Impact assessment

The various impact categories for the production of wood pellets, showing each input's relative contribution, are shown in Figure 2. The largest impact for each category is carried over from the wood residue material (i.e., tree growth, wood harvest, transportation, and wood flooring manufacture). However, the additional input of electricity for pelletization has an important impact on most categories. The plastic bag used to store the pellets also has a noticeable impact in some categories, e.g., respiratory effects. Other inputs used in the pellet- making process, such as the oil for lubrication, had negligible impacts.

View Figure

• Download as Powerpoint
• Download Figure

Wood pellets are a potential source of convenient renewable biomass energy, but some nonrenewable energy (i.e., electricity derived from coal, etc.) is required to make the wood residues and to transform the residues into the more convenient pellet fuel. Figure 3 depicts the relationship between total potential biomass energy in the pellets and the associated cradle-to-gate fossil fuel emissions related to producing wood residues and manufacturing pellets. Burdens assigned to the wood residues coproduct are allocated by mass (left) or by value (right) relative to the primary wood flooring product. Wood pellets can provide a net benefit in terms of providing biomass energy, but their embodied fossil energy is important. Wood pellets require up to 5.8 MJ of fossil energy to produce 17.3 MJ of bioenergy for a net balance of 10.4 MJ/Mg.

View Figure

• Download as Powerpoint
• Download Figure

Wood processing residues are generated as a coproduct from manufacturing hardwood flooring. Because the residues' impacts are allocated by mass (a 50%:50% split) from the production of hardwood flooring, pellets are assumed to carry significant environmental impacts (see Fig. 3, left). A change in allocation procedure for wood flooring manufacturing would affect the conclusions considerably (Kim et al. 2009, Luo et al. 2009) because wood residues are of low value compared with the primary product (i.e., hardwood flooring). The total cradle-to-gate cumulative allocated energy (i.e., fossil energy and bioenergy) required to make a wood pellet is 13.4 MJ/Mg, based on mass-based allocation of embodied energy to the wood residues produced during hardwood flooring manufacture. If value-based allocation is used, the total embodied energy (fossil energy and bioenergy) for a ton of wood pellets is reduced to 3.0 MJ/Mg. Biomass energy (i.e., wood combustion energy) makes up about 53 and 12 percent of the total embodied energy in mass- and value-based allocation scenarios, respectively.

For wood flooring manufacturing, if allocation were done by economic value instead of mass, the burdens assigned to the (low-value) wood residues would be reduced. For example, if we assume that wood flooring has a wholesale value of $2/ft² ($1,764/Mg) and that wood residues have a value of $20/ton ($22/Mg; at the mill gate), then the allocation of burdens would be 98.8 percent to the flooring product and 1.2 percent to the residues. In that scenario, the fossil energy associated with producing the residues would be reduced from 3.4 to 0.08 GJ (Fig. 3, right). If the wood flooring residues were assumed to have no environmental impacts associated with them (i.e., the burdens were allocated entirely to the flooring), then only the additional inputs required to convert the hardwood flooring residues to pellets would be considered, and only the pelletization-specific inputs and emissions described above would apply to the pellet product. Allocating all the burdens to the wood flooring would be consistent with the premise that wood flooring manufacturers are in the business to make flooring not wood residue. However, because there is (relatively small) economic value in wood residues, economic allocation may be the most appropriate method (Jungmeier et al. 2002, Luo et al. 2009, FPInnovations 2011).

Because the amount of nonrenewable energy used to produce pellets is less than the potential bioenergy they contain (Fig. 3), the global warming potential (GWP) of pellet fuel is much less than fossil fuels such as natural gas (Fig. 4). This comparison to natural gas is intended to provide a baseline analysis. Transportation of the fuels to the combustion was not considered for this analysis and combustion efficiencies of 83 and 80 percent were assumed for pellets and natural gas, respectively (Forest Products Laboratory 2004). Transportation of pellets can have a major impact on their energy balance, as shown by Magelli et al. (2009); however, an LCA that included transportation was beyond the scope of this study.

View Figure

• Download as Powerpoint
• Download Figure

Allocation of the burdens (i.e., fossil energy inputs) from wood residues production is again an important variable in the analysis. If embodied fossil energy inputs are allocated to the hardwood flooring residues by mass, then the GWP of pellets is less than half that of gas; however, if the residues are considered to be a low-value coproduct, then the GWP of pellet fuel becomes negative. If the wood flooring residues are simply considered a waste and are assumed to carry no embodied fossil energy, the GWP calculation becomes even more favorable (data not shown); however, the additional apparent benefit is minor given the assumed low monetary value of the wood residues (1.2% of the total value).

Limitations

As with all LCIs and LCAs, the conclusions that can be drawn are influenced and constrained by the underlying assumptions. This study focused on a particular type of pellet operation (operations that use hardwood flooring residues) in a particular location (the US Southeast). Results for mills that use other resources—particularly those that require additional drying as part of the pellet-making process—will be dramatically different (Hagberg et al. 2009). The additional production that is forecast for the pellet industry globally may well involve roundwood (nonresidue) raw materials. The results for this study were also heavily influenced by the local electrical generation source. For example, locations that rely on hydropower for electrical generation (Sjølie and Solberg 2011) will produce different GWP results from those calculated here.

This study included a cradle-to-gate analysis, but environmental impacts beyond the mill gate could be important, for example when transportation to markets in Europe is considered (Magelli et al. 2009). However, this study can provide a building block for such “cradle-to-grave” analyses.

Only 25 percent of the mills in the region responded to the survey (representing about 50% of the production). The data collected from these mills may not be more broadly representative; however, as noted above, the electrical energy and other inputs were consistent with other reports.