I study the life-cycle carbon footprint of an electric vehicle battery pack because the battery pack is one of the most carbon-intensive and strategically important components of an electric vehicle. As electric vehicle adoption expands, the production, use, and retirement of the electric vehicle battery pack create energy and material flows that influence both industrial competitiveness and climate performance. I therefore build a cradle-to-grave evaluation model, apply it to a lithium iron phosphate electric vehicle battery pack, identify dynamic influencing factors, and develop a software tool for repeatable assessment.

My central argument is that a static carbon footprint value for an electric vehicle battery pack is not sufficient. The carbon footprint of an electric vehicle battery pack changes with electricity structure, recycled material use, charge-discharge efficiency, mass-energy allocation, and manufacturing energy intensity. I therefore combine life cycle assessment, carbon emission factor calculation, sensitivity analysis, scenario analysis, Monte Carlo simulation, and software engineering into one evaluation framework. The framework helps me quantify not only how much carbon an electric vehicle battery pack emits, but also how that footprint can shift over time.
Life cycle assessment basis. I follow the standard life cycle assessment logic: goal and scope definition, life cycle inventory analysis, life cycle impact assessment, and interpretation. The functional unit I use is one kilowatt-hour of energy delivered by the electric vehicle battery pack to the vehicle over its service life. The reference flow is the average battery pack mass required to deliver that energy. My system boundary is cradle-to-grave and includes raw material preparation, manufacturing, use, and end-of-life recycling. I exclude transport logistics, packaging, maintenance fluids, and distribution storage under clearly stated assumptions because my main purpose is to evaluate the carbon footprint of the electric vehicle battery pack under dynamic material and energy conditions.
Carbon emission factor method. I calculate each activity’s carbon footprint by multiplying activity data by a carbon emission factor. For a process \(i\), the basic relationship is:
$$C_i = AD_i \times CEF_i$$
where \(C_i\) is the carbon footprint of process \(i\), \(AD_i\) is the activity level, and \(CEF_i\) is the carbon emission factor. This equation is simple, but it becomes powerful when I expand it into a matrix model for an electric vehicle battery pack with many materials, parts, energy types, and recycling outputs.
Raw material preparation model. I define a material mass matrix for the electric vehicle battery pack. The rows represent components and the columns represent material types:
$$M = \begin{bmatrix}
m_{11} & m_{12} & \cdots & m_{1q} \\
m_{21} & m_{22} & \cdots & m_{2q} \\
\vdots & \vdots & \ddots & \vdots \\
m_{p1} & m_{p2} & \cdots & m_{pq}
\end{bmatrix}$$
where \(p\) is the number of components and \(q\) is the number of material types. Because virgin and recycled materials have different carbon emission factors, I define two factor vectors:
$$CEF_{mv} = \left[cef_{mv,1}, cef_{mv,2}, \ldots, cef_{mv,q}\right]$$
$$CEF_{mr} = \left[cef_{mr,1}, cef_{mr,2}, \ldots, cef_{mr,q}\right]$$
I also define the recycled material ratio vector \(R_{1,j}\) and the virgin material ratio vector \(R^{*}_{1,j}\):
$$R_{1,j} = \left[r_{1,1}, r_{1,2}, \ldots, r_{1,q}\right]$$
$$R^{*}_{1,j} = \left[1-r_{1,1}, 1-r_{1,2}, \ldots, 1-r_{1,q}\right]$$
The raw material preparation carbon footprint of the electric vehicle battery pack is then:
$$C_{material} = M CEF_{mv} R^{*}_{1,j} + A M CEF_{mr} R_{1,j} + (1-A) M CEF_{mv} R_{1,j}$$
where \(A\) is an allocation coefficient between recycled and virgin material burdens. I use \(A\) to reflect market conditions and avoid double counting. When recycled material demand is high, \(A\) is lower; when supply is high and demand is low, \(A\) is higher. This formulation lets me evaluate how recycled aluminum, recycled copper, recycled graphite, and recycled plastics change the carbon footprint of the electric vehicle battery pack.
Manufacturing model. The manufacturing stage includes component production and battery system production. Component production covers cathode, anode, electrolyte, separator, shell, and battery management system. Battery system production covers cell manufacturing, module assembly, and pack assembly. For component production, I define an energy consumption matrix \(E_p\) and an energy carbon factor vector \(CEF_p\):
$$E_p = \begin{bmatrix}
e_{p,11} & e_{p,12} & \cdots & e_{p,1s} \\
e_{p,21} & e_{p,22} & \cdots & e_{p,2s} \\
\vdots & \vdots & \ddots & \vdots \\
e_{p,n1} & e_{p,n2} & \cdots & e_{p,ns}
\end{bmatrix}$$
$$CEF_p = \left[cef_{p,1}, cef_{p,2}, \ldots, cef_{p,s}\right]$$
Direct process emissions are calculated as:
$$C_{other,p} = \sum_i W_i GWP_i$$
where \(W_i\) is the mass of greenhouse gas \(i\) and \(GWP_i\) is its global warming potential. The component production footprint is:
$$C_{part} = E_p CEF_p + C_{other,p}$$
For battery system production, I use the same structure:
$$C_{battery} = E_b CEF_b + C_{other,b}$$
Thus the total manufacturing carbon footprint of the electric vehicle battery pack is:
$$C_{production} = C_{part} + C_{battery}$$
This structure allows me to separate parts, cells, modules, and pack assembly. It also helps me test how manufacturing energy intensity and electricity carbon factors affect the electric vehicle battery pack.
Use stage model. During use, the electric vehicle battery pack stores and releases electricity. Energy losses occur because charge-discharge efficiency is less than one hundred percent. Additional energy is consumed because the battery pack mass adds vehicle mass. I define total use energy as:
$$E_u = N D$$
where \(N\) is the vehicle electricity consumption per hundred kilometers and \(D\) is the total driving distance. The charge-discharge loss energy is:
$$E_{charge} = (1-\eta_{cd}) N D$$
where \(\eta_{cd}\) is charge-discharge efficiency. The mass-induced energy is:
$$E_{weight} = k \frac{m_{pack}}{m_{vehicle}} N D$$
where \(k\) is the mass-energy allocation coefficient, \(m_{pack}\) is the electric vehicle battery pack mass, and \(m_{vehicle}\) is the vehicle mass. The use-stage carbon footprint is:
$$C_{use} = (E_{charge} + E_{weight}) CEF_e$$
where \(CEF_e\) is the electricity carbon emission factor. This equation is central to my dynamic analysis because both \(\eta_{cd}\) and \(CEF_e\) change over time.
End-of-life recycling model. The end-of-life stage includes dismantling, crushing, separation, material recovery, and disposal. Energy consumption during recycling is:
$$C_{rE} = E_r CEF_r$$
Auxiliary material consumption is:
$$C_{rm} = M_r CEF_{r,aux}$$
Recovered material credits are:
$$C_{rmaterial} = \mu M CEF_{mv}$$
where \(\mu\) is the material recovery ratio. Non-recoverable disposal emissions are:
$$C_e = \sum_z Y_z CEF_z$$
The recycling stage carbon footprint after allocation is:
$$C_{recycle} = \eta \left(C_{rE} + C_{rm} + C_e – C_{rmaterial}\right)$$
where \(\eta\) is the allocation coefficient. If recovered materials offset virgin material production, \(C_{recycle}\) can become negative, meaning that recycling provides a carbon benefit to the life cycle of the electric vehicle battery pack.
Full life-cycle carbon footprint. I combine all stages:
$$C_{total} = C_{material} + C_{production} + C_{use} + C_{recycle}$$
This equation is the core of my evaluation model for the electric vehicle battery pack. Each term is dynamic because material ratios, electricity carbon factors, manufacturing energy intensity, charge-discharge efficiency, and recycling recovery rates can change.
Case study parameters. I select a lithium iron phosphate electric vehicle battery pack because lithium iron phosphate technology has become dominant in many electric vehicle markets due to safety, cycle life, and cost. The selected electric vehicle battery pack has 196 cells, a nominal cell voltage of 3.2 V, a total voltage of 627.2 V, and a capacity of 75 Ah. The main parameters are summarized in Table 1.
| Parameter | Value | Unit |
|---|---|---|
| Battery pack system mass | 485.3 | kg |
| Battery pack system capacity | 57 | kWh |
| System energy density | 95 | Wh/kg |
| Number of cells | 196 | cells |
| Vehicle mass | 1832.6 | kg |
| Vehicle electricity consumption | 13 | kWh/100 km |
| Driving distance | 150000 | km |
| Delivered energy over life | 19500 | kWh |
Raw material inventory. I organize the raw material inventory by component. The electric vehicle battery pack contains cathode materials, anode materials, electrolyte, separator, shell, coolant, and battery management system. The main material inputs per functional unit are shown in Table 2.
| Component | Material | Mass per functional unit (kg) |
|---|---|---|
| Cathode | Lithium iron phosphate | 5.44E-03 |
| Cathode | N-methyl-2-pyrrolidone | 2.46E-04 |
| Cathode | Polyvinylidene fluoride | 7.44E-04 |
| Cathode | Aluminum foil | 1.24E-03 |
| Anode | Graphite | 2.47E-03 |
| Anode | N-methyl-2-pyrrolidone | 4.92E-04 |
| Anode | Polyvinylidene fluoride | 2.46E-04 |
| Anode | Copper foil | 1.98E-03 |
| Electrolyte | Ethylene carbonate | 9.90E-04 |
| Electrolyte | Dimethyl carbonate | 9.90E-04 |
| Electrolyte | Lithium hexafluorophosphate | 4.92E-04 |
| Separator | Polyethylene | 3.69E-04 |
| Separator | Polypropylene | 3.69E-04 |
| Shell | Aluminum | 6.92E-03 |
| Shell | Polypropylene | 9.90E-04 |
| Cooling | Ethylene glycol | 1.74E-04 |
| Battery management system | Electronic and metal components | 7.38E-04 |
Manufacturing inventory. I divide manufacturing energy into electrolyte production, separator production, shell production, cell production, module assembly, and pack assembly. The electricity consumption per functional unit is shown in Table 3. Electricity dominates the manufacturing carbon footprint, so the carbon intensity of electricity is a key dynamic factor for the electric vehicle battery pack.
| Manufacturing activity | Electricity consumption (kWh) |
|---|---|
| Electrolyte production | 3.63E-03 |
| Separator production | 2.89E-03 |
| Shell production | 3.76E-03 |
| Cell production | 2.14E-01 |
| Module and pack assembly | 1.90E-02 |
| Total manufacturing electricity | 2.43E-01 |
Use-stage inventory and calculation. I use a charge-discharge efficiency of 90 percent, a mass-energy allocation coefficient of 0.56, and a total delivered energy of 19500 kWh. The charge-discharge loss energy is 0.10 kWh per functional unit, and the mass-induced energy is 0.165 kWh per functional unit. The total use-stage electricity burden is therefore 0.265 kWh per functional unit before applying the electricity carbon factor. This makes the use stage one of the largest contributors to the life-cycle carbon footprint of the electric vehicle battery pack.
Recycling inventory. For end-of-life treatment, I assume dismantling, component separation, crushing, and material recovery. I use electricity consumption values for battery system dismantling, cell and component dismantling, and material crushing and separation. I recover lithium, graphite, copper, aluminum, steel, and plastics. I also account for auxiliary chemicals such as hydrochloric acid, hydrogen peroxide, sodium hydroxide, calcium chloride, and fresh water. Table 4 summarizes the main recovered products from the electric vehicle battery pack.
| Recovered product | Mass per functional unit (kg) |
|---|---|
| Lithium chloride | 8.89E-03 |
| Recycled graphite | 2.00E-03 |
| Recycled copper | 1.63E-03 |
| Recycled aluminum | 6.25E-03 |
| Recycled steel | 2.48E-04 |
| Recycled plastic | 1.46E-03 |
Carbon emission factors. I use carbon emission factors from recognized life cycle databases, technical standards, and published inventories. The selected factors are summarized in Table 5. These factors are dynamic because electricity grid factors and recycled material factors change with technology and policy.
| Category | Name | Carbon emission factor | Unit |
|---|---|---|---|
| Material | Lithium iron phosphate | 2.93 | kgCO2e/kg |
| Material | N-methyl-2-pyrrolidone | 4.20 | kgCO2e/kg |
| Material | Polyvinylidene fluoride | 3.96 | kgCO2e/kg |
| Material | Aluminum foil | 9.12 | kgCO2e/kg |
| Material | Graphite | 5.48 | kgCO2e/kg |
| Material | Copper foil | 9.78 | kgCO2e/kg |
| Material | Ethylene carbonate | 1.37 | kgCO2e/kg |
| Material | Dimethyl carbonate | 1.30 | kgCO2e/kg |
| Material | Lithium hexafluorophosphate | 19.60 | kgCO2e/kg |
| Material | Polyethylene | 0.57 | kgCO2e/kg |
| Material | Polypropylene | 4.72 | kgCO2e/kg |
| Material | Aluminum | 11.20 | kgCO2e/kg |
| Material | Ethylene glycol | 3.38 | kgCO2e/kg |
| Material | Battery management system | 12.61 | kgCO2e/kg |
| Material | Hydrochloric acid | 0.85 | kgCO2e/kg |
| Material | Hydrogen peroxide | 0.20 | kgCO2e/kg |
| Material | Sodium hydroxide | 1.59 | kgCO2e/kg |
| Material | Calcium chloride | 3.03 | kgCO2e/kg |
| Material | Fresh water | 1.05 | kgCO2e/kg |
| Energy | Natural gas | 0.07 | kgCO2e/m3 |
| Energy | Electricity | 0.59 | kgCO2e/kWh |
| Energy | Diesel | 0.64 | kgCO2e/kg |
| Disposal | Incineration | 0.80 | kgCO2e/kg |
| Disposal | Landfill | 7.96 | kgCO2e/kg |
Base-case carbon footprint results. Using my model and inventory, I obtain the stage-level carbon footprint of the electric vehicle battery pack per functional unit. The raw material preparation stage is 0.173 kgCO2e, the manufacturing stage is 0.144 kgCO2e, the use stage is 0.155 kgCO2e, and the recycling stage is -0.00668 kgCO2e. The negative recycling value indicates a net carbon benefit from material recovery. The total life-cycle carbon footprint is 0.465 kgCO2e per kWh delivered. Table 6 presents these results.
| Life cycle stage | Carbon footprint (kgCO2e per kWh) | Contribution |
|---|---|---|
| Raw material preparation | 1.73E-01 | 37.2% |
| Manufacturing | 1.44E-01 | 31.0% |
| Use | 1.55E-01 | 33.3% |
| Recycling | -6.68E-03 | -1.4% |
| Total | 4.65E-01 | 100.0% |
I observe that raw material preparation is the largest positive contributor, followed closely by use and manufacturing. This means that reducing the carbon footprint of the electric vehicle battery pack requires simultaneous action on material sourcing, manufacturing electricity, and use-phase efficiency. A single-stage optimization is not enough.
Data quality assessment. I evaluate inventory data quality with a data quality indicator method combined with Monte Carlo simulation. I score data on temporal representativeness, technological representativeness, and geographical representativeness. I then compute an expected value and a percentage position \(Q\):
$$E = \frac{1}{n}\sum_{i=1}^{n} E_i$$
$$Q = \frac{E – E_{\min}}{E_{\max} – E_{\min}} \times 100\%$$
I map \(Q\) to a data quality indicator value. Higher values indicate better quality. I then assign beta distributions and run Monte Carlo simulation with 1000 samples. Table 7 summarizes the data quality and uncertainty results for the electric vehicle battery pack.
| Life cycle stage | Data type | DQI | Beta range | Relative standard deviation |
|---|---|---|---|---|
| Raw material preparation | Activity data | 4.5 | ±15% | 14.85% |
| Raw material preparation | Carbon factor data | 4.0 | ±20% | 14.85% |
| Manufacturing | Activity data | 4.5 | ±15% | 12.92% |
| Manufacturing | Carbon factor data | 4.5 | ±15% | 12.92% |
| Use | Activity data | 4.5 | ±15% | 12.75% |
| Use | Carbon factor data | 4.5 | ±15% | 12.75% |
| Recycling | Activity data | 4.5 | ±15% | 14.24% |
| Recycling | Carbon factor data | 4.0 | ±20% | 14.24% |
The relative standard deviations are all below 15 percent, which indicates acceptable data quality for my evaluation of the electric vehicle battery pack. However, raw material preparation and recycling show slightly larger uncertainty than manufacturing and use. This suggests that future work should focus on improving material and recycling inventory data for the electric vehicle battery pack.
Sensitivity analysis and dynamic factors. I use a sensitivity coefficient to quantify how a change in an input factor changes the carbon footprint:
$$E_i = \frac{(A_i – A_0)/A_0}{(F_i – F_0)/F_0}$$
where \(E_i\) is the sensitivity coefficient, \(F_i\) is the changed factor value, \(F_0\) is the original factor value, \(A_i\) is the changed assessment result, and \(A_0\) is the original assessment result. I combine the sensitivity coefficient with stage contribution weights to identify key dynamic factors. The stage weights from my base case are 0.36 for raw material preparation, 0.30 for manufacturing, 0.32 for use, and 0.02 for recycling. Table 8 shows the key factors and their combined influence for the electric vehicle battery pack.
| Stage | Stage weight | Key factor | Sensitivity result | Combined influence |
|---|---|---|---|---|
| Raw material preparation | 0.36 | Aluminum material use and carbon factor | High | Greater than 0.10 |
| Manufacturing | 0.30 | Electricity consumption and electricity carbon factor | High | About 0.298 |
| Use | 0.32 | Charge-discharge efficiency | Highest | Greater than 0.10 |
| Use | 0.32 | Electricity carbon factor | Moderate to high | Greater than 0.10 |
| Use | 0.32 | Mass-energy allocation coefficient | Moderate | Greater than 0.10 |
I identify four major dynamic factors for the electric vehicle battery pack: electricity carbon emission factor, recycled aluminum utilization ratio, charge-discharge efficiency, and mass-energy allocation coefficient. These factors are not fixed; they evolve with energy policy, technology learning, material recycling markets, and vehicle lightweighting. I therefore analyze each factor with forecasting or scenario methods.
Electricity structure forecasting. I collect historical annual generation data for hydro, thermal, nuclear, wind, and solar power. I standardize the data:
$$z = \frac{x-\mu}{\sigma}$$
I use polynomial regression to capture nonlinear generation trends:
$$y_i = \beta_0 + \beta_1 x_i + \beta_2 x_i^2 + \cdots + \beta_n x_i^n + \varepsilon_i$$
I select polynomial orders by k-fold cross validation and mean squared error:
$$MSE = \frac{1}{n}\sum_{i=1}^{n}(y_i – \hat{y}_i)^2$$
The selected orders are 13 for hydro, 3 for thermal, 3 for nuclear, 3 for wind, and 10 for solar. I then project the generation mix from 2025 to 2032. Table 9 shows the projected generation shares used in my dynamic analysis of the electric vehicle battery pack.
| Year | Hydro share | Thermal share | Nuclear share | Wind share | Solar share |
|---|---|---|---|---|---|
| 2025 | 0.13 | 0.62 | 0.06 | 0.11 | 0.08 |
| 2026 | 0.13 | 0.61 | 0.06 | 0.11 | 0.09 |
| 2027 | 0.13 | 0.59 | 0.07 | 0.12 | 0.09 |
| 2028 | 0.13 | 0.57 | 0.07 | 0.13 | 0.10 |
| 2029 | 0.14 | 0.55 | 0.07 | 0.13 | 0.11 |
| 2030 | 0.15 | 0.53 | 0.07 | 0.14 | 0.11 |
| 2031 | 0.17 | 0.50 | 0.07 | 0.14 | 0.12 |
| 2032 | 0.19 | 0.48 | 0.07 | 0.14 | 0.12 |
I define four electricity scenarios. The baseline scenario uses the 2022 average electricity carbon factor. The polynomial regression scenario uses my projected generation mix. The policy-expected scenario assumes non-fossil generation reaches 39 percent by 2025 and 50 percent by 2030. The strengthened policy scenario assumes 49 percent by 2025 and 60 percent by 2030. Table 10 shows the resulting electricity-induced carbon footprint for the electric vehicle battery pack.
| Scenario | Electricity-induced carbon footprint (kgCO2e) | Reduction relative to baseline |
|---|---|---|
| Baseline 2022 | 0.304 | 0.00% |
| Polynomial regression projection | 0.270 | 11.80% |
| Policy-expected scenario | 0.261 | 14.14% |
| Strengthened policy scenario | 0.217 | 28.62% |
The results show that electricity decarbonization significantly reduces the life-cycle carbon footprint of the electric vehicle battery pack. Because electricity is used in raw material preparation, manufacturing, use, and recycling, cleaner electricity creates benefits across the entire life cycle. The strengthened policy scenario gives the largest reduction, which means that accelerating renewable generation is one of the most effective strategies for lowering the carbon footprint of the electric vehicle battery pack.
Recycled aluminum utilization. Aluminum is a major contributor to the raw material preparation stage of the electric vehicle battery pack. I analyze how increasing recycled aluminum changes the raw material carbon footprint. The relationship follows the raw material model:
$$C_{material,Al} = m_{Al}\left[(1-r_{Al})CEF_{Al,v} + r_{Al}CEF_{Al,r}\right]$$
where \(r_{Al}\) is the recycled aluminum ratio. I find that increasing the recycled aluminum ratio reduces the raw material preparation carbon footprint of the electric vehicle battery pack. Table 11 summarizes selected cases.
| Recycled aluminum ratio | Change in raw material preparation carbon footprint |
|---|---|
| 0% | Baseline |
| 10% | -2.09% |
| 100% | -20.91% |
Although full replacement of virgin aluminum is not always practical, the result indicates strong potential. Increasing recycled aluminum in the electric vehicle battery pack reduces dependence on bauxite mining and high-energy primary aluminum production. This is a material-level dynamic factor that can be influenced by recycling policy, collection systems, and secondary aluminum market maturity.
Charge-discharge efficiency and mass-energy allocation. The use stage is sensitive to charge-discharge efficiency and mass-energy allocation. I construct five scenarios relative to the base case of 90 percent efficiency and 0.56 mass-energy allocation. Table 12 shows the use-stage carbon footprint and the reduction relative to the base case.
| Scenario | Charge-discharge efficiency | Mass-energy allocation coefficient | Use-stage carbon footprint (kgCO2e) | Reduction relative to base |
|---|---|---|---|---|
| Base | 90% | 0.56 | 0.155 | 0.0% |
| Scenario 1 | 90% | 0.30 | 0.110 | 29.0% |
| Scenario 2 | 95% | 0.56 | 0.121 | 21.9% |
| Scenario 3 | 95% | 0.30 | 0.078 | 49.7% |
| Scenario 4 | 98% | 0.56 | 0.100 | 35.5% |
| Scenario 5 | 98% | 0.30 | 0.059 | 61.9% |
The highest reduction occurs when charge-discharge efficiency reaches 98 percent and the mass-energy allocation coefficient falls to 0.30. This scenario reduces use-stage carbon emissions by 61.9 percent relative to the base case. I therefore conclude that improving electrochemical efficiency, thermal management, and lightweight design can substantially lower the carbon footprint of the electric vehicle battery pack.
Carbon reduction recommendations. Based on my dynamic analysis, I propose three main recommendations. First, increase the share of renewable electricity in all life-cycle stages of the electric vehicle battery pack. This affects raw material processing, manufacturing, use, and recycling simultaneously. Second, increase recycled aluminum use and other recycled materials in the electric vehicle battery pack. This reduces primary material demand and lowers raw material preparation emissions. Third, improve charge-discharge efficiency and reduce the mass-energy allocation coefficient through electrode optimization, electrolyte improvement, thermal management, and lightweight design. These measures together can reduce the life-cycle carbon footprint of the electric vehicle battery pack in a robust and technology-sensitive way.
Software system development. I develop a carbon footprint evaluation system for the electric vehicle battery pack to make the model usable, repeatable, and updatable. The system is called a power battery carbon footprint evaluation tool. I design it with three functional modules: login, front-end operation, and back-end management. The front-end operation module supports model creation, inventory analysis, result calculation, and dynamic analysis. The back-end management module maintains carbon emission factor data.
Functional requirements. The system must allow a user to create a new model, define the goal and scope, enter life cycle inventory data, calculate stage-level and total carbon footprints, view results visually, and perform dynamic scenario analysis. Table 13 summarizes the functional requirements for the electric vehicle battery pack evaluation system.
| Module | Function | Description |
|---|---|---|
| Login | User authentication | Register, log in, and protect user data |
| Front-end operation | Model creation | Enter project name, goal, functional unit, reference flow, and system boundary |
| Front-end operation | Inventory analysis | Enter input and output data for raw material, manufacturing, use, and recycling stages |
| Front-end operation | Result calculation | Compute stage-level and total carbon footprint of the electric vehicle battery pack |
| Front-end operation | Dynamic analysis | Analyze electricity carbon factor, recycled aluminum ratio, charge-discharge efficiency, and mass-energy allocation |
| Back-end management | Carbon factor database | Add, query, edit, and update material and energy carbon factors |
Non-functional requirements. The system must be easy to use, reliable, responsive, and maintainable. I design a simple interface so that users without advanced life cycle assessment training can evaluate an electric vehicle battery pack. I use a browser-server architecture and a three-layer structure: application layer, logic layer, and data layer. The application layer provides the user interface. The logic layer performs calculation and business logic. The data layer stores user data, model data, inventory data, carbon factor data, and dynamic analysis data.
Database design. I use MongoDB because it supports flexible documents and scalable data storage. The database contains tables for users, models, raw material preparation, manufacturing, use, recycling, carbon emission factors, and dynamic analysis. Table 14 summarizes the main database entities used in the electric vehicle battery pack evaluation system.
| Database entity | Key fields | Purpose |
|---|---|---|
| User information | User ID, account, password | Authentication and access control |
| Model | Model ID, project name, goal, functional unit, reference flow, system boundary | Store model definition for the electric vehicle battery pack |
| Raw material preparation | Material name, quantity, type, input type | Store raw material inventory |
| Manufacturing | Material name, quantity, type, input type | Store manufacturing inventory |
| Use | Material name, quantity, type, input type | Store use-stage inventory |
| Recycling | Material name, quantity, type, input type | Store recycling inventory |
| Carbon emission factor | Material ID, material name, factor | Store carbon factors for calculation |
| Dynamic analysis: electricity | Scenario, year, electricity use, electricity carbon factor | Analyze electricity structure effects |
| Dynamic analysis: recycled aluminum | Recycled aluminum ratio, virgin aluminum ratio, carbon factor | Analyze recycled aluminum effects |
| Dynamic analysis: efficiency and mass | Charge-discharge efficiency, mass-energy allocation parameter, carbon factor | Analyze use-stage dynamic effects |
Workflow. The workflow starts with registration or login. The user then creates a new model or opens a historical model. For a new model, the user enters the project name, research goal, functional unit, reference flow, and system boundary. The user then enters inventory data for raw material preparation, manufacturing, use, and recycling. The system calculates the carbon footprint using the life cycle model and the carbon factor database. The user can view results as bar charts. The user can then perform dynamic analysis by changing electricity carbon factors, recycled aluminum ratios, charge-discharge efficiency, or mass-energy allocation. If the result requires improvement, the user returns to the goal and scope or inventory stage, updates data, and recalculates.
Calculation logic in the software. The system applies the same equations I described earlier. For the electric vehicle battery pack, the total carbon footprint is:
$$C_{total} = C_{material} + C_{production} + C_{use} + C_{recycle}$$
For dynamic electricity analysis, the system computes:
$$C_{use} = (E_{charge} + E_{weight}) CEF_e$$
For recycled aluminum analysis, the system computes:
$$C_{material,Al} = m_{Al}\left[(1-r_{Al})CEF_{Al,v} + r_{Al}CEF_{Al,r}\right]$$
For efficiency and mass analysis, the system computes:
$$E_{charge} = (1-\eta_{cd}) N D$$
$$E_{weight} = k \frac{m_{pack}}{m_{vehicle}} N D$$
This logic ensures that the software can evaluate the electric vehicle battery pack under changing technical and policy conditions.
Result interpretation. My base-case result shows that the electric vehicle battery pack has a life-cycle carbon footprint of 0.465 kgCO2e per kWh delivered. Raw material preparation contributes 0.173 kgCO2e, manufacturing contributes 0.144 kgCO2e, use contributes 0.155 kgCO2e, and recycling provides a benefit of -0.00668 kgCO2e. The recycling benefit is small relative to the total, but it is important because it shows that material recovery improves the overall carbon balance of the electric vehicle battery pack. If recycling efficiency improves and recovered material quality increases, the negative contribution can become larger.
Dynamic interpretation. The dynamic analysis shows that the carbon footprint of the electric vehicle battery pack is not fixed. Electricity decarbonization can reduce electricity-induced emissions by 11.80 to 28.62 percent depending on the scenario. Recycled aluminum can reduce raw material preparation emissions by 2.09 percent at a 10 percent replacement ratio and by 20.91 percent at full replacement. Improving charge-discharge efficiency and reducing mass-energy allocation can reduce use-stage emissions by 21.9 to 61.9 percent. These results demonstrate that the electric vehicle battery pack should be evaluated with dynamic parameters rather than a single static value.
Uncertainty interpretation. The Monte Carlo results show relative standard deviations below 15 percent for all stages. Raw material preparation and recycling have the highest uncertainty, around 14.24 to 14.85 percent. Manufacturing and use have lower uncertainty, around 12.75 to 12.92 percent. I interpret this as acceptable for screening and policy analysis, but not sufficient for detailed product declarations without further data improvement. Better primary data for aluminum, graphite, copper, and recycling processes would improve the reliability of the electric vehicle battery pack carbon footprint evaluation.
Implications for design and policy. My model suggests that design and policy should not focus on only one stage. For the electric vehicle battery pack, raw material selection determines a large share of emissions, manufacturing electricity determines another large share, and use-phase efficiency determines the remaining share. A low-carbon electric vehicle battery pack therefore requires low-carbon aluminum, low-carbon cathode and anode materials, clean manufacturing electricity, high charge-discharge efficiency, lightweight design, and high material recovery. These factors interact, so I evaluate them together rather than separately.
Limitations and future work. I exclude transport logistics from my system boundary, so the total carbon footprint of the electric vehicle battery pack may be underestimated. I also lack measured charge-discharge efficiency degradation data over the full service life. In future work, I will extend the model to include transport distance, transport mode, packaging, distribution, and maintenance. I will also build a degradation model for charge-discharge efficiency and capacity fade so that the use-stage carbon footprint of the electric vehicle battery pack can be calculated dynamically over time. I will further improve the recycling module by incorporating different recycling technologies, recovered material quality, and market allocation rules.
Conclusion. I construct a cradle-to-grave carbon footprint evaluation model for an electric vehicle battery pack, apply it to a lithium iron phosphate electric vehicle battery pack, assess data quality, identify key dynamic factors, and develop a software tool. The base-case carbon footprint is 0.465 kgCO2e per kWh delivered, with raw material preparation, manufacturing, and use each contributing a significant share. The recycling stage provides a small net benefit. Dynamic analysis shows that cleaner electricity, higher recycled aluminum use, improved charge-discharge efficiency, and lower mass-energy allocation can substantially reduce the carbon footprint of the electric vehicle battery pack. I conclude that dynamic factors must be integrated into carbon footprint evaluation to support low-carbon design, policy making, and life-cycle management of the electric vehicle battery pack.
