Climate impact of diverting residual biomass to cement production

Co‐firing residual lignocellulosic biomass with fossil fuels is often used to reduce greenhouse gas (GHG) emissions, especially in processes like cement production where fuel costs are critical and residual biomass can be obtained at a low cost. Since plants remove CO2 from the atmosphere, CO2 emissions from biomass combustion are often assumed to have zero global warming potential ( GWPbCO2 = 0) and do not contribute to climate forcing. However, diverting residual biomass to energy use has recently been shown to increase the atmospheric CO2 load when compared to business‐as‐usual (BAU) practices, resulting in GWPbCO2 values between 0 and 1. A detailed process model for a natural gas‐fired cement plant producing 4200 megagrams of clinker per day was used to calculate the material and energy flows, as well as the lifecycle emissions associated with cement production without and with diverted biomass (supplying 50% of precalciner energy demand) from forestry and landfill sources. Biomass co‐firing reduced natural gas demand in the precalciner of the cement plant by 39% relative to the reference scenario (100% natural gas), but the total demands for thermal, electrical, and diesel (transportation) energy increased by at least 14%. Assuming GWPbCO2 values of zero for biomass combustion, cement's lifecycle GHG intensity changed from the reference (natural gas only) plant by −40, −23, and − 89 kg CO2/Mg clinker for diverted biomass from slash burning, forest floor and landfill biomass, respectively. However, using the calculated GWPbCO2 values for diverted biomass from these same fuel sources, the lifecycle GHG intensities changes were −37, +20 and +28 kg CO2/Mg clinker, respectively. The switch from decreasing to increasing cement plant GHG emissions (i.e., forest floor or landfill feedstocks scenarios) highlights the importance of calculating and using the GWPbCO2 factor when quantifying lifecycle GHG impacts associated with diverting residual biomass to bioenergy use.

clinker, an intermediary solid product of cement production (Nhuchhen et al., 2021;Worrell et al., 2001). Other emission sources include the combustion of fossil fuels to achieve process temperature requirements as well as the use of electrical energy to grind raw materials and operate other auxiliary devices in clinker making (Ayer & Dias, 2018;Nhuchhen et al., 2021).
Cement plants have generally used coal and pet coke to meet thermal energy demand (Ayer & Dias, 2018, Nhuchhen et al., 2021, but these energy sources have high GHG emission intensity, ranging from 95 kg CO 2 e/ GJ LHV for coal to 98 kg CO 2 e/GJ LHV for pet-coke (Pardo et al., 2011). Although a lower carbon (C) fuel such as natural gas (NG) (56 kg CO 2 /GJ LHV ) could reduce the emissions associated with fuel combustion, only a few cement plants, where NG is available at a lower cost, have used it as an alternative to coal or pet-coke (Nhuchhen et al., 2021;Pardo et al., 2011), which can reduce clinker production costs as well as fossil C emissions.
A recent study (Nhuchhen et al., 2021) reported a 5.8% reduction in the GHG emission intensity relative to an NG-fired plant when wood dust was co-fired with NG to generate one megagram of clinker in a cement plant. However, the study also noted a 14% increase in thermal energy intensity of the co-firing scenario relative to the reference scenario (NG only). The observation was attributed to higher air demand (3% oxygen concentration [O 2 ]) in the flue gas required for the complete combustion of solid fuels in the co-firing scenario compared to that (1% [O 2 ]) required for the combustion of NG in the reference scenario.
Since Nhuchhen et al. (2021) focused on direct emissions from the thermal energy supplied to the cement plant, they did not account for upstream emissions associated with the production and pipeline distribution of the NG. Indirect emissions from electricity supplied to run the plant such as those used to grind raw materials were also not accounted for in the study. Nevertheless, Nie et al. (2020) estimated that the upstream emissions are about 18% higher than the combustion emissions of NG. Likewise, emissions from electricity could be high, especially when sourced from a C-intensive grid system (Canada Energy Regulator, 2021;Nhuchhen et al., 2021).
In addition, fossil emissions associated with the collection, transportation and processing of the co-fired biomass were not included in the analysis. Although these GHGs sources have been reported to have minimal impact on the net emissions associated with bioenergy production (Liu & Rajagopal, 2019), they could also be a main source of GHGs depending on several factors such as the distance of the biomass location to the bioenergy facility, the mode of transportation, and the type of fossil energy used (Zhang et al., 2010). Furthermore, the study did not account for biogenic CH 4 and N 2 O emissions associated with the bioenergy systems (Cherubini et al., 2009). Although diverting residual biomass from slash burning into bioenergy avoids GHG emissions by preventing open-air combustion of biomass (Pingoud et al., 2012;Ter-Mikaelian et al., 2016), industrial combustion of biomass also releases GHGs into the atmosphere (Ministry of Environment British Columbia, 2014).
Like most bioenergy-related studies, Nhuchhen et al. (2021) followed the assumption of C neutrality and did not account for the impact of biogenic CO 2 emissions when determining the net GHG impact of the co-fired scenario relative to the reference scenario. Bioenergy is usually considered C-neutral because CO 2 emitted to the atmosphere during biomass combustion is assumed to be offset by CO 2 that is removed from the atmosphere through photosynthesis during plant growth Ragauskas, 2006). Therefore, the global warming potential assigned to this biogenic CO 2 is zero (i.e., GWP bCO 2 = 0 g CO 2 e/g b CO 2 ) compared to a value of one for fossil CO 2 (i.e., GWP fCO 2 = 1 g CO 2 e/g f CO 2 ).
In 2011, Cherubini et al. (2011) analyzed crops and forests grown as a bioenergy resource and published a method for calculating the global warming potential of biogenic CO 2 (GWP bCO 2 ) released when combusted for energy use. It showed GWP bCO 2 for a 100-year time horizon ( 100 GWP bCO 2 ) values ranging from 0 to 0.43 g CO 2 e/ b CO 2 , depending on the rotation period of biomass production.
More recently, Adetona and Layzell (2023) reported on a method to calculate GWP bCO 2 for residual lignocellulosic biomass that is diverted from agricultural or forest ecosystems, or from landfill sites. Values ranged from 0 to 1 g CO 2 e/ b CO 2 , depending on the fate of the residual biomass if it was not diverted to energy use. The longer the residual biomass would have remained in the biosphere before CO 2 enters the atmosphere, the higher the negative impact of the decision to divert and combust the biomass for energy.
While natural decomposition releases C gradually without getting any energy recovery, combustion releases biomass C instantaneously to the atmosphere  but can provide useful thermal energy for industries. If the residual biomass is destined to be burnt (e.g., forest slash) diverting it to energy could deliver a valuable energy service with no incremental increase in the atmospheric CO 2 load.
This study uses a detailed process model for an NGfired cement plant to calculate the material and energy flows, as well as the lifecycle emissions associated with cement production without and with accounting for diverted biomass from forestry and landfill sources. The purpose of the study is to assess the relative importance of the calculated GWP bCO 2 values from diverted residual biomass in reducing the GHG intensity of clinker production.
Three sources of residual biomass are assessed (forest slash destined to be burnt, forest floor residues and municipal wood waste destined for a landfill site). In addition to the GWP bCO 2 factors, life cycle assessment (LCA) includes the (a) avoided fossil fuel emissions (including upstream) when the NG is displaced with bioenergy, (b) avoided biogenic non-CO 2 emissions (CH 4 and N 2 O), (c) fossil fuel emissions associated with biomass collection, processing and transport, and (d) biogenic non-CO 2 emissions (CH 4 and N 2 O) from burning biomass for bioenergy.

| Energy demand for clinker production in the reference and co-firing scenarios
The cement plant modelled in this study utilizes a dry clinker-making process that produces 4200 megagrams of clinker (Mg clk) per day (Nhuchhen et al., 2021). A multi-stage cyclone-based preheating system is used to bring the raw materials to calcination temperature before feeding them into the precalciner of the cement plant. The calcined raw materials from the precalciner enter a rotary kiln where clinkers are formed. Typically, fuels are combusted in the precalciner (about 60%) and rotary kiln (40%) (Nhuchhen et al., 2021).
With NG as a fuel (reference scenario), complete combustion in the precalciner can be achieved when the flue gas [O 2 ] is only 1% (Nhuchhen et al., 2021). In the co-firing scenario, the plant allows up to 50% of the precalciner energy demand (i.e., 30% of total plant thermal energy demand) to be supplied by solid fuels such as biomass. However, to ensure complete combustion of the solid fuel, more excess O 2 is needed to be supplied, thereby resulting in a higher than 1% [O 2 ] in the flue gas. Here, this study adopted a case of excess air supply that resulted in 3% [O 2 ] in the flue gas at the exit of the precalciner (Nhuchhen et al., 2021). Table 1 summarizes the thermal and electrical energy requirements for the two scenarios in the production of 1 Mg of clinker as reported by Nhuchhen et al. (2021). Note that the co-firing scenario requires 14% more thermal energy (Table 1, Item 4), primarily resulting from the need to heat the additional air associated with delivering the excess O 2 (Table 1, Item 9). As a result, sensible heat losses in the flue gas of the cement plant were estimated to be 71% higher in the co-firing scenario compared to the reference scenario (Table 1, Item 14). The electricity demand in the co-firing scenario was also estimated to be 16% higher (Table 1, Item 12) because more gas volume was required to be handled by various fans of the clinker-making process (Table 1, Items 5-8).

| System boundaries for calculations of lifecycle emission
A consequential lifecycle assessment methodology was used to assess the impact on system-level GHG emissions of diverting residual biomass for clinker production. The system boundaries for the NG and co-firing scenarios are summarized in Figure 1, and include the following assumptions: • The non-energy sources in the clinker-making process (i.e., process emissions), set at 556 kg CO 2 /Mg clk (Nhuchhen et al., 2021) for both scenarios. • The total thermal energy demand for clinker production is 3.29 and 3.79 GJ LHV /Mg clk, for the reference and cofiring scenarios, respectively (Nhuchhen et al., 2021). The C emissions associated with the combustion of NG is 57 kg CO 2e /GJ LHV (Nhuchhen et al., 2021) and those from the recovery and upgrading of NG are estimated as 18% of combustion emissions (Nie et al., 2020). • The non-CO 2 emissions associated with the combustion of biomass in a cement plant are 0.010 kg CH 4 /GJ LHV dry weight (dw) and 0.0067 kg N 2 O/GJ LHV dw (Ministry of Environment British Columbia, 2014). • Table 2 summarizes emission factors for electricity, NG, diesel, and biomass production and use in the cement plant, and in the production and delivery of the fuels to the plant. Also included are the emissions factors associated with the combustion of forest slash ( Table 2, Item 5) and some alternative lower emission factors for diesel, NG, and electricity in the transition to a net-zero future (Table 2, Items 2,4,8). • In the reference scenario, residual biomass is not diverted to the cement plant, but left to be combusted (forestry slash burning) or decomposed (on the forest floor or landfill site). For wood residue decomposition in landfill, anaerobic conditions were assumed for the 100year period, so 60% of the biomass C loss was assumed to be as CH 4 , with 40% as CO 2 (Micales & Skog, 1997). A 100-year GWP CH 4 factor of 25 g CO 2 e/g CH 4 was assumed. Forest floor decomposition was assumed to be aerobic so there would only be CO 2 emissions. • Material lost during the collection, transport, and processing stages of residual forestry or landfill biomass is considered negligible (Morris, 2017). • A rapid and complete transfer of biogenic C to the atmosphere occurs as CO 2 when residual biomass is diverted to bioenergy use rather than being combusted in the field (slash burning) or being left to decompose. The impact of the accelerated release of biomass CO 2 leads to a calculation of GWP bCO 2 Liu et al., 2020), which is described in more detail in Section 2.4.

| Sources and logistics of biomass supply and processing
The residual lignocellulosic biomass assessed in this study includes forest residues that are typically burnt as slash or left on the forest floor. Residues destined for landfill sites were assumed to be municipal wood wastes, predominantly from demolition and construction wastes.

| Forest residues
At collection, the forest residues are assumed to have a water content of 47% (Thakur et al., 2014). The biomass is forwarded to the forest roadside using a Caterpillar 322 L excavator with a fuel efficiency of 0.650 L diesel/Mg ww (Thakur et al., 2014). At the roadside, the biomass is airdried to 30% moisture content before being chipped (fuel T A B L E 1 Summary of data from Nhuchhen et al. (2021) on energy flows (GJ LHV /Mg clk) and greenhouse gas emissions (kg CO 2 /Mg clk) associated with the production of one megagram of clinker (clk) in natural gas-fired (reference-scenario) versus biomass and natural gas co-fired (co-firing scenario) plant. 4. Sum of Items 1-3.

5-8. Fan
A provides air for combustion and clinker (clk) cooling, Fan B moves hot air from the clinker cooler through the exhaust vent cyclone separator, Fan C moves flue gas through the preheater tower, and Fan D moves flue gas and air to the stack through the main baghouse. The details of the calculation are presented in the Supplemental Material using data obtained from Nhuchhen et al. (2021). Hot air is also known as exhaust vent air (Item 6).
10. For the reference scenario, calculated as Item 9 × the average specific electricity demand for fans (0.00374 kilowatt-hours per normal cubic meter [kWh/ Nm3]) × 0.0036 GJe/kWh. For the biomass, the additional air requirement was provided by speeding up the fans, so electricity demand was calculated by using the fan affinity law (Item 9 [−] ÷ Item 9 [reference]) × Item 10 [reference] (AXAIR, 2021).
11. For the reference scenario, the other electricity demands (e.g., grinding raw materials and kiln operation) were determined by subtracting the electrical energy required by the fans from the total electricity required by the plants, which is 90 kWh/Mg clk as reported by Nhuchhen et al. (2021). The energy was converted from kWh/Mg clk to GJ/Mg clk using the conversion factor of 0.0036 GJ/kWh. For the co-firing scenario, we assumed that the other energy required is the same as that of the reference scenario. The details of the calculation are presented in the Supplemental Material using data obtained from Nhuchhen et al. (2021).
14. Sensible thermal energy carried away from the cement plant by the hot flue gas (Nhuchhen et al., 2021,  16. Other losses in clinker making, including radiation, convection, and conduction losses (Nhuchhen et al., 2021;Worrell et al., 2001 At the central processing facility for forest residues (Figure 1), the wood chips are air-dried to 10% moisture content and ground to a particle size of about 2 mm (99.8 kWh/Mg ww, Rezaei et al., 2016;Thakur et al., 2014) to meet the requirement of the cement plant for combusting solid fuels in the precalciner (Mokrzycki & Uliasz-Bochen, 2003;Nhuchhen et al., 2021). The wood dust is then trucked 400 km to the cement plant (10 Mg ww/truck, 0.285 L diesel/km), returning empty. The smaller payload on the truck reflects the lower bulk density of wood dust (0.143 Mg ww/m 3 , Stasiak et al., 2019) compared to wood chips (0.240 Mg ww/m 3 , Sultana & Kumar, 2011). At the cement plant, the wood dust is assumed to be dried to 0% moisture using waste heat.

| Landfill waste
For residual lignocellulosic biomass diverted from landfills, the energy and emissions associated with collecting and transporting the biomass to a central processing facility for landfill residues (not the same as that used for forestry residues) are attributed to the BAU waste management processes. At the central processing facility, biomass (with a water content of 10%) is chipped (0.735 L diesel/Mg ww) and ground to a particle size of about 2 mm (99.8 kWh/Mg ww, Rezaei et al., 2016;Thakur et al., 2014) The wood dust is then trucked 100 km from to the cement plant (10 Mg ww/truck, 0.285 L diesel/km), returning empty. At the cement plant, the wood dust is assumed to be dried to 0% moisture using waste heat.

| Calculation of the global warming potential of biogenic CO 2 (GWP bCO 2 )
The decision to divert residual woody biomass for use as bioenergy in clinker production accelerates the time it takes for biomass C to be released into the atmosphere, compared to the BAU alternatives Liu et al., 2020). Therefore, this decision may have an anthropogenic impact on the global climate.
The methodology developed in Adetona and Layzell (2023) was used to calculate values for GWP bCO 2 (based on a 100-year time horizon) that was applied to the CO 2 emissions from biomass combustion. This methodology requires an understanding of the fate of biomass C under BAU conditions so it can be compared with the destiny of biomass C if diverted to use as a bioenergy F I G U R E 1 System boundaries for the two cement-producing scenarios analysed in the study. Scenario (a) involves the production of clinker using only natural gas (NG) as a fuel source, while scenario (b) involves residual lignocellulosic residual forest or landfill biomass (50% of the pre-calciner energy required) with natural gas. The study accounted for the flows of NG (blue font and lines), biomass (green font and lines), and electrical (orange font and lines) energy as well as associated emissions (grey font and lines) that are within the system boundary, including biogenic emissions (CO 2 , CH 4 , and N 2 O) from slash burning or decomposition. Energy and emissions associated with the production of raw materials for the clinker are outside the system boundary, thus not considered in this analysis. resource. Figure 2 summarizes the range (shaded area) and assumed average (lines) profiles for biomass C remaining in the biosphere for 100 years following a decision regarding how that biomass will be used: • Combustion for industrial heat and power generation leaves little biochar (Lehmann et al., 2006), thus we assumed all the biomass C was converted to CO 2 ( Figure 2, Orange line). • Slash burning of residual forest biomass leaves behind 1.2%-5.1% of the biomass C as biochar (Finkral et al., 2012;Ter-Mikaelian et al., 2016), which is resistant to subsequent degradation (Lehmann et al., 2006). Only 3%  to 25% (Hammes et al., 2008) of the biochar C is lost within 100 years of the formation while the remainder of the C contributes to long-term soil C storage (Hammes et al., 2008;. Therefore, for this study, we assumed that an average value of 3.1% of biomass C was left as biochar and that it declined by 14% over the subsequent 100 years ( Figure 2, purple line). The details of the calculations are presented in Material S1. • Residual biomass left on the floor of temperate forests decomposes slowly over time (Zell et al., 2009) as shown in Figure 2. The average mass of biomass C remaining in the biosphere ( Figure 2, the green line) was used in this study for the subsequent analysis. The method used in determining the mass of biomass C remaining is presented in Adetona and Layzell (2023). • Biomass in landfill sites decomposes more slowly than biomass in the forest due to a lower microbial decomposition in landfill sites (Micales & Skog, 1997;Wang et al., 2011;Ximenes et al., 2008). The average mass of biomass C remaining in the landfills (Figure 2, blue line) was used for the subsequent analysis. The method used to quantify the mass of biomass C remaining is presented in Adetona and Layzell (2023).
For each dataset represented by the solid lines in Figure 2, the following calculations were carried out: • The annual reduction in biosphere C was assumed to be transferred to the atmosphere as annual pulses of CO 2 with the following exceptions. In the slash burning scenario, about 0.5% ( Table 2, Item 5D) was assumed to enter the atmosphere as methane (CH 4 ). In the landfill scenario, the woody biomass was assumed to be under anaerobic conditions so 60% of the C lost from the biomass was assumed to enter the atmosphere as CH 4 with the balance as CO 2 . The CH 4 emissions were handled separately as avoided emissions when residual biomass is diverted to energy. Only the CO 2 emissions were considered in the calculation of GWP bCO 2 . • The atmospheric decay of each annual pulse of CO 2 was calculated over 100 years according to the Bern 2.5 CC equation (equation 1 in Adetona & Layzell, 2023). The C as CO 2 removed from the atmosphere was assumed to return to the biosphere in oceans and the terrestrial ecosystem. • For each year, the pulses were summed for each scenario from 0 to 100 years to calculate the 100-year atmospheric CO 2 load resulting from that biomass combustion/decomposition. • The total atmospheric CO 2 load for each scenario over the 100 years ( 100 AC load ) was calculated as the area under the curve, so it has units of Mg C-years.
The four values for 100 AC load were then used to calculate the 100-year GWP bCO 2 (g CO 2 e/g biogenic CO 2 released in bioenergy use) associated with residual biomass diversion to bioenergy use for each BAU scenario (Adetona & Layzell, 2023

| Net GHG emissions
The net biogenic CO 2 emission changes were then calculated by subtracting the emissions associated with cofiring residual biomass with NG in a cement plant from the emissions associated with firing only NG (i.e., the reference scenario).

| Global warming potential of biogenic CO 2 associated with diverting residual biomass to bioenergy
When a pulse of CO 2 enters the atmosphere as a result of a combustion process, the Bern 2.5 CC model can be used to project its removal (decay) from the atmosphere over the subsequent 100 years ( Figure 3). If the initial pulse is 1 Mg C and the CO 2 is not removed from the atmosphere, the atmospheric CO 2 load would be 100 Mg C-years. However, the Bern 2.5 model reveals an atmospheric CO 2 load of 48.5 Mg C-years (blue-shaded area in Figure 3). The CO 2 leaving the atmosphere during this time returns to the biosphere as inorganic or organic C in oceans or on land (green-shaded area in Figure 3). When residual biomass decomposes over many years (e.g., forest floor biomass decomposition, Figure 4b), we assume many smaller annual 'pulses' to the atmosphere for many years and each pulse is expected to decay in the atmosphere following a profile similar to that shown in Figure 3. However, because new pulses are entering the atmosphere every year, the sum of all pulses creates a profile similar to Figure 4b (blue-shaded area), and the atmospheric CO 2 load is lower (e.g., 31.1 Mg C-years) than that resulting from a combustion process (48.5 Mg C-years; Figure 3).
The difference between the atmospheric profile in Figures 3 and 4b is provided in Figure 4e and shows an area of 17.2 Mg C-years. This area represents the impact on atmospheric CO 2 load that is associated with the decision to divert residual forest residues to bioenergy use.
Using a similar approach, the atmosphere CO 2 load associated with diverting biomass from slash burning to energy use was calculated to be 1.40 Mg C-yrs (Figure 4a,d), and from landfill to energy use was calculated to be 47.3 Mg C-yrs, respectively (Figure 4c,f). Table 3 shows these values are used to calculate GWP bCO 2 values of 0.029, 0.354, and 0.976 g CO 2 e/g b CO 2 for the decision to divert residual biomass to energy use from slash burning, forest floor decomposition and landfill residues, respectively ( Table 3, Item 3).

| The energy required for biomass collection, processing, and transportation
About 0.044 GJ LHV diesel/Mg dw ( ÷ energy content (16.06 GJ∕Mg dry biomass) × 1000 (kg dw biomass∕Mg dw biomass) × carbon content (0.43 kg C∕kg dry biomass) × 3.67 kg CO 2 ∕kg C × 100 GWP bCO 2 of each BAU scenario F I G U R E 3 The fate of one megagram of C as CO 2 (orange area) released to the atmosphere through combustion. At time 0 the CO 2 enters the atmospheric pool (the blue-shaded area) and then a portion is transferred to the biosphere carbon pool (the green-shaded area) over 100 years The atmospheric CO 2 load in megagram-carbon years (Mg C-years, i.e., the blue-shaded area) was calculated for a 100-year period and the value is shown on the chart. ). About 0.069 GJ LHV diesel/Mg clk was supplied to transport the forest biomass from the central processing unit for forest residues to the cement plant (Table 4, Item 38).
The same amount of electricity (0.030 GJe/Mg clk) that was needed to grind the forest residues was required to grind the residues diverted from landfills (Table 4, Item 31). However, lower diesel energy (0.002 GJ LHV diesel/Mg clk) was required to chip the landfill residues (Table 4, Item 27) relative to the forest residues (0.003 GJ LHV diesel/ Mg clk) due to the higher efficiency of the chipper that was utilized at the central processing unit for landfill residues.
In addition, lower diesel energy (0.017 GJ LHV diesel/Mg clk) was required to transport the landfill residues from their central processing unit to the cement F I G U R E 4 The effect of diverting biomass to energy on the atmospheric CO 2 load compared to business-as-usual. The fate of one megagram of residual woody biomass carbon over 100 years burned as slash on the forest floor (a), left to decompose on the forest floor (b), or to decompose in a landfill (c). The orange-shaded area represents biomass carbon (see Figure 2). With combustion or decomposition, biomass C is transferred to the atmosphere as CO 2 (blue-shaded area) and over time the CO 2 is removed according to the Bern 2.5 climate change model (Cherubini et al., 2011) and moves back to the biosphere (oceans and terrestrial ecosystems [green-shaded area] over 100 years from the start of the decomposition). The values for Mg C-years (gray-shaded area in panels d-f) represent the atmospheric CO 2 load from the biogenic CO 2 entering the atmosphere. In the landfill scenario, cumulative methane emissions are identified on the chart and their contribution to global warming will be considered separately.   T A B L E 4 Energy use and emission associated with diverting residual biomass to energy use in a cement plant.

Item Parameters Units
Co-fired biomass diverted from:

Forest floor Landfill
Biomass characteristics  (Table 4, Item 38) since the processing unit was four times closer to the cement plant than the central processing unit forest biomass, where the forest residues were processed into wood dust (Table 4, Item 32). Nevertheless, the total energy required to collect, process, and transport the residues diverted from forests or landfills (Table 4) was only 3.2% and 1.3%, respectively, of the energy that was needed to process one Mg clk (Table 1, Item 4).

| Avoided fossil fuel GHG emissions associated with clinker production
About 145 kg CO 2 e/Mg clk (Table 5, Item 4) of GHGs associated with NG was emitted into the atmosphere when biomass was co-fired with NG in the cement plant. Approximately 75 kg CO 2 e/Mg clk of these emissions are from the kiln (Table 5, Item 2), and 69 kg CO 2 e/Mg clk is from the precalciner (Table 5, Item 3) of the cement plant.
Note: 1. For the co-fired biomass diverted from burning and forest floor, data were obtained from Thakur et al. (2014). For the landfills, the data were obtained from Hossain and Poon (2018

T A B L E 4 (Continued)
T A B L E 5 Energy flows (GJ LHV /Mg clk) and greenhouse gas emissions (kg CO 2 /Mg clk) associated with the production of one megagram of clinker (clk) in natural gas-fired (reference-scenario) versus biomass and natural gas co-fired (co-firing scenario) plant.

Item Parameters
Reference scenario Thus, relative to the reference scenario (188 kg CO 2 e/Mg clk), about 43 kg CO 2 e/Mg clk (Table 5, Item 4) emissions were avoided when biomass was co-fired with NG in the plant.
The upstream emissions associated with the production and upgrading of NG accounted for 33.6 and 25.9 kgCO 2 /Mg clk in the reference and co-firing scenarios, respectively (

| Avoided biogenic CH 4 and N 2 O emissions associated with residual biomass diversion
Compared to the controlled, industrial combustion of biomass, when biomass is burnt in the field, the non-CO 2 emissions tend to be much higher ( Table 2, Items 5 and 6). Therefore, diverting residual biomass from slash burning to industrial heat production delivers GHG benefits.
In the current study, about 18 kg CO 2 e/Mg clk of biogenic GHG emissions were avoided (

| Residual biomass collection, processing, and transportation
Emissions associated with biomass collection, processing, and transportation, collectively called pre-combustion, ranged between 7.75 kg CO 2 e/Mg clk and 15.2 kg CO 2 e/ Mg clk for municipal wood waste diverted from landfill and forest biomass diverted from slash burning or decomposition in the forest, respectively (Table 5, Item 25). These emissions include those generated from diesel fuels used in forwarding, chipping, and transporting the biomass (Table 4). Diesel energy required to collect and process biomass diverted from landfill was lower than that required by other feedstocks because the energy required for collecting the biomass was assigned to waste management. Also, emissions associated with transporting landfilldestined biomass were 80% lower than those required to transport biomass that was diverted from slash burning or decomposition on the forest floor because the location of landfills is closer to the cement plant relative to the other biomass sources (Table 4, Item 27).

| Industrial biomass combustion
Approximately 2.88 kg CO 2 e/Mg clk was emitted when the residual biomass was combusted in the precalciner of the co-fired plant, 89% of the emissions were N 2 O and the remainder were CH 4 ( Table 5, Items 8-10).

| Grid power demand
For the reference scenario, the electricity required by fans to support flue gas and airflow as well as other activities such as kiln operation (Table 5, Item 7) was responsible for 6.5% of the total GHG emissions at the cement plant (Table 5, Item 11).
The electricity emissions increased by 26% for the cofiring scenario relative to the reference scenario due to increases in the air required for the complete combustion of the co-fired biomass in the precalciner (Table 5, Item 7), emissions associated with the combustion of NG and coal used in generating the electricity (Table 5, Item 17), and the extra energy required to grind the biomass in the processing units (

| The atmospheric impact of biogenic CO 2
In the co-firing scenarios, the residual biomass requirement was 1.22 GJ LHV /Mg clk (Table 1, Item 3), equivalent to 76 kg dry biomass/Mg clk (Table 4, Item 8). Given a C content of 43.1% (Table 4, Item 6), the biogenic CO 2 released into the atmosphere was estimated at 120 kg CO 2 / Mg clk (Table 5, Item 29). Applying the GWP bCO 2 values previously calculated for each scenario (Table 3, Item 3), the accelerated CO 2 load is calculated to be 3.5, 42.5, and 117 kg CO 2 /Mg clk for diverting to energy use, residual biomass from slash burning, forest residues, and landfill, respectively (Table 5, Item 30).

| The lifecycle impact of diverting residual biomass to clinker production
Summing all sources and reductions in GHG emission showed the reference (NG only) scenario producing 838 kg CO 2 e/Mg clk was associated with the reference scenario (Table 5, Item 34), with 556 kg CO 2 e/Mg clk were process emissions (Table 5, Item 1) that are generated when calcium carbonate embedded in the feedstock limestone is converted to calcium oxide (Nhuchhen et al., 2021;Worrell et al., 2001). The energy-related emissions include 222 (188 + 33.6) kg CO 2 e/Mg clk from NG production and use (Table 5, Items 4  and 14), and 61 (51.6 + 9.28) kg CO 2 e/Mg clk from electricity production and use (Table 5, Items 7 and 17). Figure 5 provides a summary of the results showing how the co-firing scenarios impacted the lifecycle emissions in the reference scenario (set as 'zero' in Figure 5). In all three co-firing scenarios, diverting residual biomass to bioenergy use in a cement plant reduced emissions from NG production and combustion by 51 kg CO 2 /Mg clk ( Figure 5, dark blue bars). In addition, the decision to divert residual biomass to bioenergy avoided 18 and 59 kg CO 2 /Mg clk non-CO 2 of biomass-based emissions in the slash burning and landfill scenarios, respectively ( Figure 5, light blue bars).
While reducing emissions, the decision to divert residual biomass to energy use also generated additional emissions compared to the reference scenario, including those emissions associated with collection processing and transporting the biomass in the cement plant (Table 5, Item 25), the non-CO 2 emissions from biomass combustion (Table 5, Item 10) and the emissions needed to meet the additional demand for electricity at the cement plant ( Figure 5, Items 7 and 17). In total, they led to an additional 28 kg CO 2 e/kg clk for the residues diverted from forestry operations, and 21 kg CO 2 e/kg clk for the residues diverted from landfills.
When combined with the avoided emissions, the three co-firing scenarios projected changes in GHG emissions of −40, −23, and −89 kg CO 2 e/kg clk for the residual biomass diverted from slash burning, forest floor, and landfill sources, respectively ( Figure 5). However, these calculations do not include the impact of biogenic CO 2 emissions. When that GHG contribution is considered, the net change in emissions relative to the reference cement plant is −37, +20 and +28 kg CO 2 e/kg clk for the residual biomass diverted from slash burning, forest floor, and landfill sources, respectively ( Figure 5).

| Decarbonizing fuel and electricity production: Impact on GHG emissions
Shifting to a net-zero emission electrical grid had the largest impact on the GHG intensity of clinker production, reducing it by 16-19 kg CO 2 e/Mg clk ( Table 2, Item 8, and Figure 6). In comparison, decarbonizing oil recovery and diesel production reduce emissions by 1-3 kg CO 2 e/Mg clk ( Table 2, Item 2 and Figure 6), reflecting the more substantive role for electricity than diesel fuel F I G U R E 5 The greenhouse gas impact (kg CO 2 e/Mg clk) of diverting biomass to energy for cement production. Relative to the reference scenario (zero on the y-axis), the blue bars show emission reductions, while the green bars show the offsetting emission increases beginning at the most negative values achieved by the emission reduction. Therefore, the arrows show the net impact of biomass diversion on the carbon intensity of cement production (units of kg CO 2 e/Mg clinker).

F I G U R E 6
The effect of using energy feedstocks with lower C intensity on the results of this study. If the assumed greenhouse gas (GHG) emissions associated with diesel, natural gas or electricity generation were lower than those in the default option (see Table 2), the calculations for life cycle GHG emissions would change as shown here. The default option is similar to the values presented in Figure 5. in clinker production, especially when biomass is being combusted. When we assumed NG recovery and upgrading to have only about half the emissions ( Table 2, Item 4), the cofiring scenarios showed an increase in GHG emissions by 3-4 kg CO 2 e/Mg clk relative to the reference ( Figure 6). This is because the reference scenario is dependent on NG and addressing upstream emissions dropped the emissions intensity of the reference scenario from 838 to 822 kg CO 2 e/Mg clk (data not shown).

| The effect of co-firing residual biomass on the lifecycle emissions for clinker production
Compiling the results from Table 5 reveals how precalciner co-firing of residual biomass impacts the lifecycle GHG emissions for clinker production in an NG-fired cement plant. Only the residual biomass diverted from slash burning was found to reduce the lifecycle emissions below that of the reference scenario and even then, the savings were only about 4.4% of total emissions (Figure 7).

| Biomass co-firing and energy flows in a natural gas-fired cement plant
As a gaseous fuel, the complete combustion of NG in the air is possible even when the resulting flue gas has only about 1% [O 2 ]. In contrast, to ensure the complete combustion of solid fuels such as biomass or coal typically require more excess O 2 , in the range of 3% in the flue gas (Nhuchhen et al., 2021). Therefore, when an NG-fueled cement plant replaces 50% of the precalciner fuel supply with biomass, the total energy demand increases from 3.29 to 3.76 GJ LHV /Mg clk (Table 1, Item 4). The additional thermal energy is needed to heat the additional air supply, and the larger volumes of gas flow means that more heat is lost in the process (Table 1, Items 14-16).
In addition, the need to move additional air and flue gas through the clinker plant with biomass co-firing increases the electricity demand (Table 1, Item 12). However, despite these challenges, co-firing biomass, especially lowcost, residual biomass, promises to be a cost-effective strategy to reduce the GHG emissions associated with cement production. This study provides a system-level, LCA of the impact of residual biomass co-firing for cement production. Particular focus is placed on the implications of including GWP bCO 2 values using the method described in a recent study (Adetona & Layzell, 2023).

| Life cycle GHG emissions for clinker production assuming GWP bCO 2 of zero
In the production of clinker, process emissions are the dominant source of GHG emissions accounting for 556 kg CO 2 e/ Mg clk (Table 5, Item 1). In the reference scenario (no cofiring), on-site emission from the combustion of NG leads to an additional 188 kg CO 2 e/Mg (Table 5, Item 4). Assuming a grid intensity of 570 kg CO 2 e/MWh (Table 2, Item 7), the electricity used by the cement plant contributes another 51.6 kg CO 2 e/Mg clk for a total of 795 kg CO 2 e/Mg Clk (Table 5, Items 7 and 11). The upstream emissions associated with the NG consumed at the cement plant and in the production of electricity adds another 33.6 (Table 5, Item 4) and 9.28 kg CO 2 e/Mg clk (Table 5, Item 17), respectively, resulting in a life cycle GHG emissions of 838 kg CO 2 e/Mg clk (Table 5, Item 34). Such emission levels are typical of an NG-fired cement plant (Nhuchhen et al., 2021) but lower than that for a cement plant using coal or petroleum coke as a fuel (Carbone et al., 2022;Worrell et al., 2001).
In the co-firing scenarios, where biomass provides 50% of energy demand for the precalciner, NG consumption in F I G U R E 7 The effect of diverting biomass to energy use on the lifecycle emissions of cement production. Comparison of the lifecycle emissions of a natural gas fired cement plant without (reference scenario) and with the precalciner co-combustion of residual biomass diverted from slash burning, forest floor, and landfill resources. The biomass-based emissions include the net of the avoided non-CO 2 emissions from slash burning or landfill storage, the emissions associated with collecting, processing, transporting, and combusting the biomass, and the calculated atmospheric load from biogenic CO 2 when residual biomass is diverted to energy use. (see Figure 5 for details). The dashed line denotes the reference scenario emissions.  (Table 5, Item 4 and 14), but there are increases in the emissions from electricity (Table 5,  Items 7 and 17), and some emissions from biomass combustion (Table 5, Item 10) (Government of Alberta, 2019; Morris, 2017). In addition, biomass collection, processing and transportation is projected to contribute another 7.8-15.2 kg CO 2 e/Mg clk (Table 5, Item 25). Relative to the forest residues, lower fuel energy use is required to process landfill residues due to the shorter proximity of the landfill site to the processing plant (Caserini et al., 2010). Previous studies have reported that with distances of 25-150 km, transportation emissions are not an important determinant of the net GHG impact associated with displacing fossil energy with bioenergy (Liu & Rajagopal, 2019). In the future, transportation emissions can be minimized by using low or no-C transportation fuels such as electricity or H 2 (Layzell et al., 2020;Worrell et al., 2001).
It is important to note that diverting residual biomass to bioenergy use can result in reductions in GHG emissions compared to the reference scenario. In the case of forest slash burning, diverting it to energy use should avoid about 18 kg CO 2 e/Mg clk (  (Table 5, Item 33), a 40 kg CO 2 e/Mg clk reduction compared to the reference cement plant as long as all the CO 2 emissions from industrial biomass combustion are considered to have a GWP bCO 2 of zero.
Decomposing forest floor biomass is typically an aerobic process (Kukrety et al., 2015) so virtually all of the biomass C is released as CO 2 , and there are no avoided biogenic CH 4 or N 2 O emissions (Table 5, Item 29), resulting in life cycle emission of 815 kg CO 2 e/ Mg clk (Table 5, Item 33), a 23 kg CO 2 e/Mg clk reduction compared to the reference cement plant.
Landfill biomass is known to generate CH 4 emissions (Micales & Skog, 1997) that could be avoided by residual biomass diversion to bioenergy use. Wood tends to decompose slowly in landfill sites, losing from 0% to 23% of the original biomass C over a 100-year period (Micales & Skog, 1997;. If the landfill is aerobic, the biomass decomposes more quickly, and CO 2 would be the primary product. However, many landfills are anaerobic, where decomposition leads to biogas with about 60% CH 4 and 40% CO 2 (Micales & Skog, 1997). In this study, we project a 9% loss over 100 years ( Figure 2) and assume anaerobic conditions. Therefore, after 100 years in a landfill site, 53.9 kg C as CH 4 is released to the atmosphere per kg lignocellulosic biomass stored (Figure 4c). Given biomass with a C content of 43.1% (Table 4, Item 6) and a lower heating value of 16.1 GJ/Mg dry (Table 4, Item 5) diverting residual biomass away from landfill to energy use will reduce GHGs by 58.8 kg CO 2 e/Mg clk (Table 5, Item 26 and 28). This GHG benefit could be up to three times greater if the biomass diverted from landfill was not wood waste. For example, paper waste is known to decompose more quickly and release more CH 4 under anaerobic conditions (Micales & Skog, 1997).
Diverting woody biomass to energy use results in life cycle GHG emissions of 749 kg CO 2 e/Mg clk, an 89 kg CO 2 e/Mg clk reduction in GHG emissions compared to the reference clinker plant (

| Estimating the global warming potential of biogenic CO 2
When residual forestry or landfill biomass is diverted from business-as-usual (BAU) practices (e.g., slash burning or microbial decomposition) to use as a bioenergy feedstock, it is likely to result in differences in timing of CO 2 that is released to the atmosphere. A recent study (Adetona & Layzell, 2023) described a methodology to calculate the atmospheric CO 2 load and the 100 year GWP bCO 2 associated with the decision to divert residual biomass to energy use.
Using this method, GWP bCO 2 values of 0.029-0.354 and 0.976 g CO 2 e/g b CO 2 are calculated for biogenic CO 2 release in a cement plant when the biomass is diverted from slash burning, forest floor or landfill sites, respectively (Table 3). These values are similar to, or higher than, the GWP bCO 2 values published for crops or trees grown for energy use where longer crop rotations (Cherubini et al., 2011) are associated with higher values for GWP bCO 2 (up to 0.43 g CO 2 e/g b CO 2 ). Guest et al. (2013) extended the work of Cherubini et al. (2011) by considering the effect of an interim storage period for biomass (e.g., in buildings or furniture) before it is used for energy. In their analysis, long-term storage of biomass (up to 100 years) reduced the GWP bCO 2 values and showed the potential to generate a negative value for GWP bCO 2 . Therefore, the lowest (most negative) values for GWP bCO 2 were associated with short rotation, purposegrown biomass that was stored for many years before being used as an energy resource. Although subsequent studies have used a generic GWP bCO 2 value to account for the impact of biogenic CO 2 emissions for a bioenergy system (Liu & Rajagopal, 2019), it is crucial to understand the effects of assumptions and parameters on GWP bCO 2 before applying the metric value to bioenergy analysis.

| GHG emissions from CO 2 generated by residual biomass use in clinker production
When the GWP bCO 2 values for each co-firing scenario ( Table 3, Item 3) are multiplied by the CO 2 generated in residual biomass combustion (Table 5, Item 29), the calculated atmosphere CO 2 loads range from 3.5 to 117 kg CO 2 e/Mg clk (Table 5, Item 30). Residual biomass diverted from slash burning had the lowest atmospheric CO 2 load (3.5 kg CO 2 e/Mg clk, Table 5, Item 30) because this biomass was destined to be burnt ( Figure 2). The GWP bCO 2 and atmospheric CO 2 load of this scenario would have been zero (Miner et al., 2014) had we not accounted for the formation and deposition of biochar (0.031 Mg biochar C/Mg biomass C) on the forest floor with slash burning (Lehmann et al., 2006;Ter-Mikaelian et al., 2016).
Residual biomass diverted from the forest floor had a higher atmospheric CO 2 load (42.5 kg CO 2 e/Mg clk, Table 5, Item 30) because the biomass could have remained on the forest floor for many years if the residues were left to decompose (Figure 2). Based on an average decomposition profile utilized in this analysis, up to 20% of the forest biomass C would have remained in the forest even after 100 years of decomposition if the biomass was not diverted into bioenergy use at the cement plant. In this study, we did not model how a higher probability of wildfires would reduce the forest floor C stocks (Kurz et al., 2013) in the (BAU) scenario, resulting in a lower GWP bCO 2 and creating a more favourable case for diverting residual forest biomass for energy use. Indeed, the removal of forest floor residues is used to reduce the frequency and intensity of wildfires (O'Connell, 1997).
The atmospheric CO 2 load was highest (117 kg CO 2 e/ Mg clk, Table 5, Item 30) when residual lignocellulosic biomass was diverted to energy use from a landfill. Biomass decomposition is slower in a landfill relative to a forest floor because the anaerobic condition of the landfill is not conducive for fungi that are efficient in decomposing biomass (Ximenes et al., 2008). Consequently, landfills are important C sinks, especially when environmental conditions of the site are less favourable for microbial degradation such as in cold regions of the world, including Canada (Micales & Skog, 1997;Zhang et al., 2008).

| Life cycle GHG emissions for clinker production with calculated values for GWP bCO 2
When the atmospheric CO 2 loads for the slash burning scenario are included in the LCA, little impact is observed on the calculation of the GHG benefit associated with biomass diversion to co-firing compared with the reference scenario. The calculation of the GHG benefit changed from 40 to 37 kg CO 2 e/Mg clk ( Figure 5).
However, significant differences are observed in the other two co-firing scenarios. In the forest floor scenario, a 23 kg CO 2 e/ Mg clk reduction in GHG emissions relative to the reference scenario becomes a 20 kg CO 2 e/ Mg clk increase in GHG emission ( Figure 5). In the landfill scenario, an 89 kg CO 2 e/Mg clk reduction in GHG emissions relative to the reference scenario becomes a 28 kg CO 2 e/ Mg clk increase in GHG emission ( Figure 5). These findings show that the assumption of C neutrality for biomass combustion can underestimate the impact of bioenergy decisions (Figure 7; Matuštík & Kočí, 2022;Shapiro-Bengtsen et al., 2022).
While these results position residual forest floor and landfill biomass as less attractive feedstocks for diversion to energy use, it is important to recognize that there may be advantages in doing so that are distinct from climate change. For example, residual biomass diversion could help to reduce forest fire risk, enhance forest regrowth, or reduce pressure on landfill sites. Other factors, not quantified here include impacts on biodiversity, acidification, and ecotoxicity (Cherubini et al., 2009;Weldu & Assefa, 2016).
It is also important to note that the current study explores the use of residual biomass in an NG-fired cement plant. Most cement plants in the world are fueled with coal or petroleum coke and are not as energy efficient as NG-fired plants (Cavalett et al., 2022;Naqi & Jang, 2019;Thwe et al., 2021). If the co-firing analyses carried out here were to have been applied to a coal-fired cement plant, the GHG benefits of residual biomass diversion would probably have been greater. Even though the impact of biogenic CO 2 emissions on the calculated GHG benefit would be of a similar magnitude, there is a possibility that all scenarios would show a net GHG benefit of diverting residual biomass to energy use in coal-fired cement production.

| Sensitivity analysis: Impact of assumptions on GHG intensity of the fuel and electricity used in biomass transport and clinker production
In this study, default values are assumed for the GHG intensity of the diesel, NG and electricity (Table 2, Items 1, 3 and 5-7) used in the clinker-making process. In the transition to net-zero emission energy systems, efforts are being deployed to reduce these upstream GHG emissions. To assess the impact of progress towards upstream fuel/electricity decarbonization on the conclusions of this study, lower GHG intensity values were identified for diesel, NG, and electricity (Table 2, Items 2, 4 and 8). The results showed only minor variations from model projections using default values; the decarbonization of the electrical grid had the largest impact ( Figure 6).

| CONCLUSION
Diverting low-cost residual biomass for energy use is often seen as an economically viable and environmentally friendly strategy to reduce demand for fossil fuels and the GHG emissions they generate, thereby addressing climate change. This study challenges this paradigm and shows the need for a deeper understanding of the source of the residual biomass to assess the full lifecycle climate impacts associated with the decision to divert biomass to energy use.
The energy and material flows were modeled in an NGfired cement plant without and with the co-firing of residual biomass diverted from various sources. If the residual biomass was destined to be combusted or decompose rapidly (i.e., within a year or two), diverting it to bioenergy use should have substantial climate change benefits. However, if the residual biomass was destined to be preserved for a long period on the forest floor or in a landfill site, diverting it to energy use creates an anthropogenic atmospheric load of biogenic CO 2 that tends to undermine the potential GHG benefits.
This work highlights the need for consequential LCA for any projects considering the diversion of residual biomass to use as an energy resource.