Carbon Footprint of Power Transformers Evaluated Through Life Cycle Analysis

Abstract

The growing global demand for electricity is one of the crucial factors contributing to the increase in greenhouse gas (GHG) emissions. Power transformers, although essential components in electricity transmission and distribution systems, significantly impact the environment. This paper employs a Life Cycle Assessment (LCA) to evaluate the carbon footprint of two oil-immersed transformers with a capacity of 31.5 MVA and 25 MVA, manufactured in Poland in 2023, with an assumed lifetime of 40 years. The analysis follows a cradle-to-use approach and considers various scenarios, including differences in the average operating load levels, electricity generation from different sources, and a time frame spanning 2024–2063. After 2–3 years of transformer operation, even at low loads, the CO₂ equivalent (CO₂-eq) emissions associated with energy losses exceed those generated during the transformer production. These results underscore the critical importance of utilizing advanced construction materials and design solutions. Moreover, this analysis highlights the need to implement systemic actions aimed at restructuring electricity generation, especially in regions that heavily dependent on fossil fuels.

1. Introduction

The Paris Agreement, adopted in 2015, establishes an international framework to limit the increase in average global temperature to below 1.5 °C, requiring net-zero GHG emissions to the atmosphere by 2050 [1]. Within this context, the European Union has committed to reducing GHG emissions, including, among others, carbon dioxide, methane, hydrofluorocarbons, perfluorocarbons, nitrous oxide, and sulfur hexafluoride, by at least 55% by 2030 compared to the 1990 levels as part of the European Green Deal strategy. The energy sector, responsible for over 40% of global GHG emissions [2], plays a critical role in this transition due to its heavy reliance on fossil fuels. It covers various industries, including oil, gas, power generation, and petrochemicals, with electricity generation being a key component. The decarbonization of electricity production is essential for achieving climate targets, as it directly affects emissions from power generation, transmission, and distribution. Power transformers are critical devices in electrical power systems, enabling the efficient transmission and distribution of electricity [3]. They operate on the principle of electromagnetic induction and allow for electrical energy to be transmitted over long distances, first by increasing the voltage level to reduce the associated energy losses and then lowering it to the level required by receivers in industrial facilities or households. Transformers are manufactured as oil or dry types by global companies such as Schneider Electric, Siemens AG, General Electric, ABB Ltd., Mitsubishi Electric Corporation, Hitachi Energy, and Toshiba Corporation, as well as by those operating in local markets.

The number of transformers continues to grow steadily, driven by increasing energy demand, grid modernization efforts, and the integration of renewable energy sources into the power system. The global market size was valued at USD 22.83 billion in 2022 and is projected to expand at a compound annual growth rate (CAGR) of 7.1% from 2023 to 2030. By 2025, the market size is estimated to reach USD 31.01 billion, and it is expected to grow to USD 49.27 billion by 2030, with a CAGR of 9.7% during the forecast period [4]. In 2023, Poland’s transformer infrastructure consisted of 3095 high-voltage transformers (including 224 for 400 and 220 kV) and 263,996 medium-voltage transformers [5].

Transformers contribute approximately 4% of CO₂-eq emissions from the electric power system, amounting to 730 million tons per year [6,7]. CO₂-eq emissions occur throughout the entire lifecycle of a transformer, beginning with the extraction of raw materials, followed by the production of individual components, assembly, transportation, operation, and, ultimately, the disposal of its parts. Guo et al. [2] determined that the operational phase of transformers has the most significant impact on total CO₂-eq emissions, with approximately 96% attributed to power losses. Similarly, Cai et al. [8] reported a value of 98%, further highlighting the dominant contribution of no-load loss and load loss to the overall carbon footprint of transformers. This is primarily due to the continuous operation of transformers throughout the year and their long lifespan, which typically exceeds 40 years. Consequently, electricity consumption during the operational phase has the greatest influence on the overall environmental impact of transformers, surpassing emissions associated with other lifecycle stages, such as the transportation of raw materials to the factory, the manufacturing process, and the delivery of transformers to their operational sites. Under the assumption of the European electricity mix, Jorge et al. [9] estimated that a transformer rated at 500 MVA could contribute up to 90,000 tons CO₂-eq to climate change over its lifetime. Šerkinić et al. [10] conducted a carbon footprint analysis for the distribution oil-immersed transformers (400 kVA and 630 kVA) produced in Croatia, covering the cradle-to-gate phase. The results revealed that the total carbon footprint ranges from 4676 kg CO₂-eq to 6736 kg CO₂-eq, depending on the transformer capacity. The specific carbon footprint was calculated at 3.8 kg CO₂-eq per kilogram of material used for the 400 kVA transformer and 4.0 kg CO₂-eq per kilogram of material used for the 630 kVA transformer, highlighting the significant environmental impact of material usage (magnetic steel, aluminum coils and wires, and transformer oil) in transformer production. Güldurek and Esenboğa [11] analyzed the carbon footprint of Beta Energy, a transformer manufacturer in Adana Province, Türkiye, taking into account both direct and indirect greenhouse gas emissions from all company activities. The study found that in 2023, Beta Energy’s total carbon footprint reached 1.8 mln t of CO₂-eq, with 6044 transformers sold, which results in approximately 300 t of CO₂-eq per transformer. Wang et al. [12] analyzed the carbon emissions of distribution network transformers (200 kVA) manufactured in China at different life cycle stages, reporting values of 282 kg, 782 kg, 122.96 kg, 11,079.64 kg, and −88.6 kg CO₂ for the manufacturing, transportation, construction and installation, operation, and waste recycling stages, respectively, highlighting the significant impact of the operation phase. These studies show that the environmental impact of transformers depends on the specifics of production in different locations.

Current studies on the carbon footprint of power transformers, although using data from Poland, a country where 47% of electricity is generated from coal combustion [13], lead to conclusions that are universal in nature, independent of the country in which transformers are produced and operated. The main novelty of this study is that the calculations were carried out assuming both constant and time-varying CO₂-eq emission factors associated with no-load and load losses occurring during the transformer operation period. Until now, only a constant value of this factor has been considered in the calculations. Another innovative aspect of this study is the discussion of simplifications and limitations applied in the calculation of CO₂-eq emissions at various stages of the transformer’s lifecycle and their potential impact on the obtained results. In addition, this study consolidates and extends the knowledge of CO₂-eq emissions for transformers with a power of several tens of MVA. Almost all of the available literature focuses on much smaller transformers, usually with capacities in the range of a few hundred kVA.

The research hypothesis, which underlines the novelty of this study, is that even at low transformer loading, CO₂-eq emissions during the operation phase reach or exceed those from the production and transportation phases within the first few years after commissioning.

The carbon footprint is a widely recognized and systematically applied methodology for evaluating the environmental impact of human activities, with a particular focus on GHG emissions and their contribution to climate change [14]. It represents the total quantity of CO₂ emissions, both direct and indirect, produced by a specific activity or accumulated across the various stages of a product’s lifecycle, typically expressed in metric tons of CO₂ or CO₂-equivalent. The carbon footprint is most accurately assessed using the LCA methodology. LCA provides a robust and comprehensive framework for quantifying emissions at every stage of a product’s lifecycle, from raw material extraction through manufacturing, usage, and end-of-life disposal [15,16]. The LCA analysis consists of the following stages: goal and scope definition, Life Cycle Inventory (LCI), Life Cycle Impact Assessment (LCIA), and interpretation. This methodology was selected because it is standardized, internationally recognized (ISO 14040/44 [17,18]), and enables a holistic evaluation of environmental impacts across the entire lifecycle of a product or system, in contrast to methods such as Material Flow Analysis or Input–Output Analysis. This integration of LCA and CO₂-eq emission assessment enables a detailed evaluation of environmental impacts, facilitating the development of strategies to minimize emissions in alignment with global sustainability goals.

In this study, the carbon footprint analysis of oil-immersed transformers based on primary data obtained from a Polish company was conducted, including raw material acquisition, material transportation to the factory, manufacturing, transportation of the transformer to the operation site, and product use. This study considers various scenarios, including different transformer operating load levels, diverse energy source origins, and a time period spanning from 2024 to 2063. Given the complexity of such an analysis, accurately determining the carbon footprint of a power transformer throughout its production and operational lifespan presents a significant challenge. During the production phase, this complexity arises from the fact that a transformer is composed of numerous components, which are delivered to the manufacturing facility either as fully assembled or nearly finished products. Ideally, each of these components should be accompanied by a certificate specifying its associated carbon footprint. In the operational phase, the assessment of the carbon footprint requires the predictive modeling of transformer usage conditions over several decades. This prediction must account for dynamic load variations over time, the expected number of tap changer operations (if applicable), planned maintenance activities (e.g., diagnostic procedures and oil replacement or regeneration), and fluctuating environmental conditions that influence the operation of the heating elements (e.g., installed in the control cabinet), pumps, and cooling fans, assuming that the transformer is equipped with such systems. Furthermore, it is crucial to anticipate the sources and proportional contributions of electrical energy required to compensate for no-load and load losses, as well as the transformer’s auxiliary power consumption. In practical applications, determining the quantitative carbon footprint of a transformer necessitates adopting numerous simplifying assumptions, resulting in an approximate value that may either underestimate or overestimate the actual CO₂-eq emissions. A common simplification involves assuming that the carbon footprint is solely attributed to the raw construction materials rather than the processed components derived from them.

2. Materials and Methods

2.1. Life Cycle Assessment Analysis

2.1.1. Scope, Functional Unit, System Boundary

The objective of this study is to assess and quantify the environmental impact of two oil-immersed transformers with a capacity of 31.5 MVA and 25 MVA, referred to herein and after as TRF-1 and TRF2, respectively, manufactured in Poland in 2023, using primary data collected from their manufacturers. The life cycle analysis was carried out in accordance with the requirements of international standards ISO 14040 (Environmental Management—Life Cycle Assessment—Principles and Framework) [17] and ISO 14044 (Environmental Management—Life Cycle Assessment—Requirements and Guidelines) [18]. Various system boundary approaches, as shown in Figure 1, including “cradle to cradle”, “cradle to gate”, “cradle to grave”, “gate to gate”, “gate to grave”, and “cradle to use”, can be applied in LCA studies depending on the scope and objectives of the assessment. In this study, the “cradle to use” approach was selected, meaning that the analysis encompasses the entire lifecycle—from raw material extraction, material transportation to the factory, manufacturing, transportation of the transformer to the operation site, and its operational phase while excluding the end-of-life stage. Future research plans include analyzing various end-of-life scenarios for the transformer to assess their environmental impact and potential recycling or disposal strategies.

The functional unit was defined as 1 MVA of transformer capacity.

The basic technical parameters of transformers are presented in

Table 1. Characteristics of TRF-1 and TRF-2 oil-immersed transformers.

Parameters	Unit	TRF-1	TRF-2
Nominal Capacity	MVA	31.5	25
Nominal Voltage	kV	115/16.5	115/16.5
No-load losses	kW	12	10.5
Load losses	kW	140	125
Cooling system	-	ON-AN	ON-AN
Type of transformer oil	-	mineral	mineral
Total mass	t	60	49
On-load tap changer	-	Yes	Yes
Lifespan	year	40	40

.

2.1.2. Life Cycle Inventory Analysis of Power Transformers

The inventory data were collected from the Polish company in 2023 and include all materials, consumables, transportation, and electricity consumption, as presented in

Table 2. Inventory analysis of materials for power transformers.

Materials	Unit	TRF-1	TRF-2
Steel	t	16.319	12.039
Electrical steel	t	15.864	13.454
Copper	t	7.903	6.566
Cellulose insulation:
-Paper	t	0.271	0.211
-Pressboard	t	1.349	1.214
Mineral oil	t	14.000	11.300
Other materials	t	4.294	4.216

,

Table 3. Inventory analysis of material transportation to the factory.

Geography	Type of Transport	Distance [km]	TRF-1 Mass [t]	TRF-2 Mass [t]
Poland	Semi-trailer trucks	250	7.955	6.633
Germany	Semi-trailer trucks	480	0.460	0.324
Czech Republic	Semi-trailer trucks	500	0.558	-
Turkey	Semi-trailer trucks	2100	15.864	13.454
Slovakia	Semi-trailer trucks	530	2.654	2.196
Sweden	Semi-trailer trucks	360	14.231	11.540
	Cargo ship	690	14.231	11.540
Estonia	Semi-trailer trucks	1100	5.148	4.237
India	Semi-trailer trucks	360	8.324	5.827
	Cargo ship	15,130	8.324	5.827
Europe	Semi-trailer trucks	1200	4.806	4.789

and Table 4. The data collection process was one of the most time-intensive stages of LCA and had a significant impact on the accuracy of the results.

Steel is used, among others, for components such as the tank with its cover and radiators, while electrical steel is used for the transformer core. Mineral oil serves as an insulating, cooling, and impregnating medium, improving the dielectric properties and durability of the transformer [19,20], while copper is used for the windings. Cellulose insulation, including paper and pressboard, is used in transformers for electrical insulation and mechanical support of the windings. Materials classified as “other” include both those listed in Table 2, in cases where it was not possible to explicitly determine their mass from a specific structural component (e.g., the on-load tap changer and its drive mechanism), as well as those not listed, whose mass was either negligible or difficult to estimate reliably (e.g., varnishes, rubber, glass, silica gel, porcelain, and aluminum). The percentage contribution of individual structural materials in the total mass of transformers TRF-1 and TRF-2 is presented in Figure 2.

The percentage contribution of individual structural materials, as shown in Figure 2, is very similar for both transformers.

Table 3 presents the mass of construction materials and components, taking into account the transportation distance from the manufacturer to the transformer factory as well as the type of transport used.

Since materials from a given country could be sourced from multiple locations, an average transportation distance was assumed. For materials categorized under Europe, the delivery originated from various European countries, including both those explicitly listed and those not specified in Table 3, particularly in cases where the exact mass of certain components could not be identified and attributed to a specific country. It was assumed that the road transport of structural materials, finished components, and the manufactured transformer is carried out using semi-trailer trucks equipped with diesel engines. For sea transport, it was assumed that general cargo ships were used. With a few exceptions, transportation takes place within Europe.

The determination of the carbon footprint of a given product requires knowledge of the amount of electricity consumed during the manufacturing process. Transformer manufacturers use the electricity consumption factor (ECF) to estimate this electricity consumption, which represents the average amount of electrical energy required to produce 1 MVA of transformer capacity. This indicator accounts for electricity consumption not only during the production process, including the operation of various devices and systems such as winding machines, overhead cranes for transporting transformer components, lighting, blowers, and tank sandblasting machines but also in the operations of the test station and the physico–chemical laboratory, which conduct inter-process and final acceptance testing of the transformers. Additionally, this factor accounts for the electricity consumption of administrative and office structures, covering activities from design to commercial processes. In 2023, the ECF for transformer production was 768 kWh/MVA.

During the operational phase, power transformers achieve high efficiency, typically ranging from 97% to 99%. This value is influenced by no-load losses and load losses. The energy lost during operation necessitates additional power generation, leading to the emission of a corresponding amount of CO₂-eq into the environment. Additionally, depending on the transformer’s design, energy may be consumed by various auxiliary systems, such as the drive of an on-load tap changer, fan and/or pump drives, and heaters used in dehumidifiers or control cabinets. Since the analyzed transformers are characterized by natural cooling (ON-AN) and therefore do not have pump or fan drives, the installed heaters have low power consumption and operate intermittently. Moreover, due to the difficulty in determining the expected operating frequency of the on-load tap changer, the carbon footprint associated with the electricity consumption of these devices has been omitted from the assessment. Similarly, the carbon footprint related to maintenance activities necessary for transformer operation, such as diagnostic procedures, potential oil regeneration/replacement, or unforeseen repair actions, has not been considered. In summary, the analysis considers only the carbon footprint associated with the energy consumption required to compensate for no-load and load losses over the transformer’s estimated operational lifespan of 40 years. This study assumes continuous operation throughout the year (8760 h) under a constant load varying from 0% to 100% of its nominal capacity in increments of 10%.

2.1.3. Life Cycle Impact Assessment Method

In this study, the LCIA of the oil-immersed transformer was conducted using the carbon footprint methodology. The total carbon footprint was calculated using Equation (1):

CF = activity (unit) · EF_CO2-eq (t CO₂-eq/unit)

(1)

where CF—total carbon footprint of the transformer [t CO₂-eq], activity—data (e.g., electricity consumption in kWh), and EF_CO2-eq—emission factor.

Table 4 presents the emission factors for materials used in transformer construction. The table includes representative EF_CO2-eq values reported in refs. [2,6,21,22,23,24,25,26] and environmental databases such as Ecoivent 3.11. [27] (Ecoinvent Center, St-Gallen, Switzerland) and indicates the specific factor selected for subsequent calculations.

Table 4. Emission factors associated with the production of individual construction materials.

Materials	EFCO2-eq [t CO2-eq/t]	References	Selected EFCO2-eq Value [t CO2-eq/t]
Steel	0.970 ÷ 6.200	[6,21,22,23,24,25,26,27]	2.500
Electrical steel	2.000 ÷ 6.580	[6,21,22,23,24,25,26]	3.000
Copper	1.100 ÷ 8.000	[6,21,23,24,25,26,27]	4.738
Cellulose insulation	0.817 ÷ 1.760	[23,24]	0.817 (paper), 1.183 (pressboard)
Mineral oil	1.120 ÷ 3.000	[6,21,24,26]	1.210

Table 4. Emission factors associated with the production of individual construction materials.

Materials	EFCO2-eq [t CO2-eq/t]	References	Selected EFCO2-eq Value [t CO2-eq/t]
Steel	0.970 ÷ 6.200	[6,21,22,23,24,25,26,27]	2.500
Electrical steel	2.000 ÷ 6.580	[6,21,22,23,24,25,26]	3.000
Copper	1.100 ÷ 8.000	[6,21,23,24,25,26,27]	4.738
Cellulose insulation	0.817 ÷ 1.760	[23,24]	0.817 (paper), 1.183 (pressboard)
Mineral oil	1.120 ÷ 3.000	[6,21,24,26]	1.210

As shown in Table 4, multiple emission factor values are often available for the same material. Variations in steel may result from differences in steel type (e.g., carbon steel, stainless steel, cast iron), production technology (e.g., blast furnace vs. electric arc furnace), the proportion of recycled content, and the energy mix used in manufacturing, for example. For materials categorized as “other” in Table 2, the emission factor was determined as a weighted average based on the mass of the identified materials used in the transformer and their corresponding emission factors (Table 4). For TRF-1, the calculated value was 2.596 t CO₂-eq/t, while for TRF-2, it was 2.610 t CO₂-eq/t. Given the minimal difference between these values, the final emission factor of 2.600 t CO₂-eq/t was adopted for further calculations.

Table 5. Emission factors associated with materials transportation to the factory.

Transportation	EFCO2-eq [t CO2-eq/tkm]	References	Selected EFCO2-eq Value [tCO2-eq/tkm]
Road transport	0.000085 ÷ 0.000260	[26,27,28,29]	0.000130
Sea transport	0.000003 ÷ 0.000050	[26,27,28,29]	0.000025

provides emission factor values for material transportation to the factory. The listed values represent emissions over the entire life cycle of the fuel used, commonly referred to as Well-to-Wheel (WTW) emissions. In this study, “tkm” refers to ton-kilometers, a unit commonly used in LCA to quantify the environmental impact of freight transport, calculated as the product of transported mass and distance traveled.

Both analyzed transformers were installed in Poland at distances of 325 km (TRF-1) and 375 km (TRF-2) from their manufacturing locations and were delivered to the site of operation by road transport, as shown in Figure 3.

To calculate the carbon footprint associated with electricity consumption during the manufacturing process or transformer operation, it is also necessary to determine the CO₂-eq emission factor corresponding to the production of 1 kWh of electricity. This factor should account for the contribution of different energy sources in electricity generation. For the calculations, data from ref. [30], covering the years 2018–2024, were used, as presented in

Table 6. Emission factors associated with electricity production in Poland.

Year	EFCO2-eq Value [t CO2-eq/kWh] [30]
2018	0.000743
2019	0.000717
2020	0.000693
2021	0.000698
2022	0.000667
2023	0.000690
2024	0.000652

.

In this study, calculations were conducted for two scenarios. In the first scenario, it was assumed that the CO₂-eq emission factor associated with the production of 1 kWh of electricity remains constant throughout the entire operational lifetime of the transformer. The adopted value corresponds to the CO₂-eq emission factor determined for Poland in 2024, based on the assumption that transformers manufactured in 2023 are put into operation at the beginning of the following year—EF_CO2-eq (const), t CO₂-eq/kWh. In the second scenario, it was assumed that the CO₂-eq emission factor varies annually—EF_CO2-eq (var), t CO₂-eq/kWh. In both scenarios, the transformer is considered to operate continuously throughout the year (8760 h) under a constant load ranging from 0% to 100% of its nominal capacity in increments of 10%.

For calculations conducted under the second scenario, it was necessary to develop a forecast for changes in the CO₂-eq emission factor associated with electricity production for the period 2024–2063. This forecast was directly related to the observed and expected changes in the structure of the energy sources that make up the energy mix. A time series of CO₂-eq emissions was constructed using the data presented in Table 6. An analysis of the available data indicated that CO₂-eq emission trends were approximately linear. Based on this observation, the linear regression model was chosen as the most suitable approach for projection. In Poland, decisions regarding the development of renewable and nuclear energy are strongly influenced by current political choices, making it difficult to anticipate future changes. Currently, CO₂-eq emissions from electricity production are the highest in Europe, and while there is significant potential for their reduction, any flattening of the curve is likely to occur in the long term.

The forecast was developed using Microsoft Excel 2024, utilizing its built-in linear regression functionality to model future changes in the CO₂-eq emission factor. The analysis was performed under the assumption of no seasonality in the dataset, leading to a linear projection of emission trends. The results are presented in Figure 4. Additionally, to assess the reliability of the projection, two-sided confidence interval boundaries were calculated and plotted, assuming a confidence level of 0.95.

3. Results and Discussion

In this study, the carbon footprint analysis was conducted across various lifecycle stages, including raw material acquisition, material transportation to the factory, manufacturing, transportation of the transformer to the operation site, and product use, to assess its environmental impact. The Polish energy mix was used to calculate electricity consumption during manufacturing and energy losses during operation. The carbon footprint results for TRF-1 and TRF-2 during the raw material extraction phase are presented in

Table 7. CO₂-eq emissions from raw material extraction phase for TRF-1 and TRF-2.

	TRF-1 (31.5 MVA) [t CO2-eq]	TRF-2 (25 MVA) [t CO2-eq]
Steel	40.798	30.098
Electrical steel	47.592	40.362
Copper	37.444	31.043
Cellulose insulation	1.817	1.609
Mineral oil	16.940	13.673
Other materials	11.164	10.998
Total	155.755	127.783

.

A comparative analysis of the percentage contribution of CO₂-eq emissions associated with the individual construction materials, as shown in Figure 5, demonstrates a high degree of similarity between the two transformers.

The carbon footprint associated with raw material extraction amounts to 4.945 t CO₂-eq/MVA of transformer TRF-1 and 5.111 t CO₂-eq/MVA of transformer TRF-2. The largest contribution to emissions comes from electrical steel (30.6% for TRF-1, 31.6% for TRF-2), structural steel (26.2% for TRF-1, 23.6% for TRF-2), and copper (24.0% for TRF-1, 24.3% for TRF-2), which is due to their energy-intensive production processes and high demand for primary raw materials. This is particularly evident in steel production, where blast furnaces are traditionally coal-fired, leading to substantial CO₂-eq emissions. A more sustainable alternative is Bluemint^® steel by Thyssen-Krupp, which is produced using hydrogen-fired furnaces powered by electricity from renewable sources, significantly reducing the carbon footprint to 0.6 t CO₂-eq/t [31]. Mineral oil also significantly contributes to total emissions (10.9% for TRF-1, 10.7% for TRF-2) as its production is linked to petroleum processing. A key strategy for reducing the environmental impact of transformers is replacing traditional mineral oils with synthetic or natural esters [32,33]. Mineral oils pose risks such as low flash points, poor biodegradability, and potential soil contamination in case of leaks or accidents. Rózga et al. [34] demonstrated that transformers filled with natural esters exhibit significantly better fire resistance than those using mineral oil, thereby reducing fire-related emissions and environmental damage.

It is important to emphasize that CO₂-eq emissions should not be assessed solely based on raw materials but rather on the components manufactured from them. For instance, while the calculations account for the mass of steel required to produce radiators, the factory does not receive raw steel but rather fully manufactured radiators. Consequently, the authors argue that the carbon footprint associated with transformer production is likely underestimated.

The CO₂-eq emissions related to the transportation of materials to the factory are presented in

Table 8. CO₂-eq emissions from material transportation to the factory.

	TRF-1 [t CO2-eq]	TRF-2 [t CO2-eq]
Road transport	7.379	6.226
Sea transport	3.394	2.403
Total	10.773	8.629

.

The process of assembling a transformer in a factory is represented by the amount of electricity needed to carry it out. The CO₂-eq emission values corresponding to this stage of transformer production are given in

Table 9. CO₂-eq emissions associated with electricity used during transformer production.

	TRF-1 [t CO2-eq]	TRF-2 [t CO2-eq]
Electricity consumption	16.692	13.248

.

Figure 6 further illustrates the comparison of the percentage contributions from material production, transportation of materials to the factory, and electricity consumption involved in the production of transformers TRF-1 and TRF-2.

Analyzing Figure 6, it can be seen that the highest emissions come from construction materials, accounting for about 85% of the total emissions related to transformer production (this corresponds to 155.755 t CO₂-eq for TRF-1 and 127.783 t CO₂-eq for TRF-2). In contrast, the transport of materials to the factory makes the smallest contribution, equal to about 6% of the total emissions related to transformer production, depending on their weight and transport distance.

The CO₂-eq emissions associated with transformer transportation from the factory to the operational site are presented in

Table 10. CO₂-eq emissions from transformer delivery to the operational site.

	TRF-1 [t CO2-eq]	TRF-2 [t CO2-eq]
Road transport	2.535	2.389

.

Taking into account the weight of each transformer, the modes of transport used, and the associated CO₂-eq emission factor (as presented in Table 5), as well as the distance between the manufacturing facility and the installation site, the estimated carbon footprint resulting from this stage is 0.080 t CO₂-eq/MVA of TRF-1 and 0.096 t CO₂-eq/MVA of TRF-2, respectively. It is worth noting that transportation emissions can vary significantly depending on the logistics chain, including factors such as fuel type, vehicle efficiency, and load optimization.

Table 11. CO₂-eq emissions results for the use phase of TRF-1 and TRF-2.

	TRF-1 [t CO2-eq]		TRF-2 [t CO2-eq]
	Scenario 1	Scenario 2	Scenario 1	Scenario 2
Electricity consumption (assuming an average load of 50% over a 40-year period)	10,737.658	6963.973	9538.238	6172.232

presents the CO₂-eq emissions results for the use phase of transformers, corresponding to the average annual load of 50% for scenario 1 (constant EF_CO2-eq) and scenario 2 (variable EF_CO2-eq) over a 40-year period. The table provides numerical values of CO₂-eq emissions at a 50% load, selected as a representative value based on our observations, which indicate that typical transformer loads in Poland range from 30% to 70% on average. For comparison, Das et al. [7] reported that the average annual loads for distribution transformers vary between 21% and 50% across different countries.

The analysis of the results in Table 11 indicates that the CO₂ emissions per unit of transformer capacity are 340.878 t CO₂-eq/MVA for TRF-1 in scenario 1 and 221.079 t CO₂-eq/MVA in scenario 2. For TRF-2, the emissions are 381.530 t CO₂-eq/MVA in scenario 1 and 240.810 t CO₂-eq/MVA in scenario 2.

The impact of load variation on CO₂-eq emissions across its entire range is presented in Figure 7a,b.

Figure 7 illustrates the cumulative CO₂-eq emissions over a 40-year lifespan, resulting from electricity consumption required to compensate for no-load and load losses for transformers TRF-1 and TRF-2 in both analyzed scenarios. The emissions of CO₂-eq are expressed in tons. Load losses, associated with transformer windings, vary depending on the load level. In this study, load losses were modeled under a constant load ranging from 0% to 100% of the nominal capacity, increasing in 10% increments. No-load losses, on the other hand, result from the magnetization current required to energize the transformer core. Unlike load losses, they remain constant regardless of the transformer’s load and are assumed to be 12 kW for TRF-1 and 10.5 kW for TRF-2. It is worth noting that Figure 7 also presents the CO₂-eq emissions at a 0% load, which results from the no-load losses of the transformer, occurring due to core magnetization even in the absence of an external load. An analysis of the results presented in Figure 7 underscores the necessity of initiatives to reduce GHG emissions from electricity generation, highlighting the importance of increasing the proportion of renewable energy sources in the energy mix.

Figure 8 shows the CO₂-eq emissions after 40 years of the TRF-1 transformer operation phase for scenario 2, incorporating the projected variation in the CO₂-eq emission factor. In case of a 50% load, the projected CO₂-eq emissions amount to 6777.218 t CO₂-eq, with an accuracy of ±10%, meaning that the emissions may vary between 6081.911 t CO₂-eq and 7472.525 t CO₂-eq. The results highlight the importance of taking uncertainty into account in long-term environmental impact assessments, as this directly affects companies’ emissions costs. Underestimating or overestimating emissions can lead to poor investment decisions, misallocation of resources, and suboptimal emission reduction strategies.

A particularly interesting aspect is the comparison of CO₂-eq emissions from different life cycle stages of the analyzed oil-immersed transformers, specifically the production phase (cradle-to-gate) and the operational phase. For the operational phase, the analysis is limited to the first two years—2024 and 2025. The comparison was conducted under scenario 1, which assumes a constant EF_CO2-eq, and scenario 2, which assumes a variable EF_CO2-eq over time, as shown in Figure 9a,b.

It can be seen from Figure 9 that after 2–3 years of transformer operation, even at low loads, the CO₂-eq emissions associated with energy losses exceed those generated during transformer production. The timeframe indicated here may vary for individual countries, depending on the characteristics of their energy mix. This observation should be a very strong impulse to take action in two areas. The first should be related to reducing no-load and load losses and concerns transformer manufacturers. The second is related to the absolute necessity to reduce CO₂-eq emissions during electricity production and concerns joint action by policy makers and electricity producers.

Table 12. Summary of results of CO₂-eq emissions for TRF-1 and TRF-2 for both scenarios.

	TRF-1 [t CO2-eq]		TRF-2 [t CO2-eq]
Lifecycle Stage	Scenario 1	Scenario 2	Scenario 1	Scenario 2
Raw materials extraction	155.755	155.755	127.783	127.783
Transportation of materials to the factory	10.773	10.773	8.629	8.629
Manufacturing	16.692	16.692	13.248	13.248
Transportation to the operational site	2.535	2.535	2.389	2.389
Usage (40 years, 50% load)	10,737.658	6963.973	9538.238	6172.232
Total CO2-eq emissions (40 years, 50% load)	10,923.413	7149.728	9690.287	6324.281

presents the summary and comparison of CO₂-eq emissions at each lifecycle stage, as well as the total emissions for both analyzed scenarios for TRF-1 and TRF-2.

In scenario 1, the total carbon footprint for a predefined functional unit was 346.775 t CO₂-eq/MVA for TRF-1 and 387.612 t CO₂-eq/MVA for TRF-2. In scenario 2, where the CO₂ emission factor varied over time, the total carbon footprint was 226.976 t CO₂-eq/MVA for TRF-1 and 252.971 t CO₂-eq/MVA for TRF-2.

The results in Table 12 indicate that accounting for changes in the energy mix over time can significantly reduce the overall environmental impact of transformer operation. Consequently, the operational phase of transformers accounts for 94% to 99% of the total CO₂ emissions, depending on the load level, which is consistent with findings from other studies [2,8,11,12]. The obtained results highlight the critical importance of using high-quality construction materials and modern design solutions to minimize no-load and load losses. For the core, materials with a narrow hysteresis loop, such as silicon steel or amorphous steel, should be used, while high-purity copper with low resistivity is recommended for the windings. Design improvements include the use of a laminated core composed of thin, insulated sheets, magnetic shielding, and toroidal or circular core shapes. Additional loss reduction for a transformer of the same rated power can be achieved by increasing its dimensions, which involves the use of a larger core and windings. However, it is important to recognize that implementing these solutions increases the purchase cost of the transformer. Moreover, increasing the size of the core and windings or incorporating magnetic shielding requires a larger quantity of materials, which in turn increases the carbon footprint associated with transformer production and transportation to the installation site.

This analysis highlights the need to implement systemic actions aimed at restructuring electricity generation, especially in regions that are heavily dependent on fossil fuels, particularly in countries such as Poland, where the majority of energy is still generated from the combustion of hard coal and brown coal. Achieving a transition to cleaner energy sources requires targeted political initiatives and strategic policy decisions. Therefore, policymakers should prioritize the development of regulatory frameworks and incentive mechanisms that support this transformation. The findings of this study serve as a valuable reference for both policymakers and industry stakeholders in optimizing transformer efficiency and reducing carbon emissions throughout their life cycle.

4. Conclusions

In this study, the carbon footprint analysis was performed for two oil-immersed transformers manufacturing in Poland across various lifecycle stages, including raw material acquisition, manufacturing and assembly, transportation, and product use, to assess their environmental impact. Furthermore, this study examined various scenarios incorporating both constant and time-varying CO₂-eq emission factors, specifically addressing emissions related to no-load and load losses throughout transformer operation.

Based on the performed analysis, the following key conclusions were formulated:

• The highest contribution to CO₂-eq emissions at the raw material extraction and manufacturing stages—80.8% for TRF-1 and 79.5% for TRF-2—originates from metal production, particularly electrical steel, structural steel, and copper. This highlights the need for targeted metal management strategies and effective recycling methods. However, the widespread adoption of metal recycling in transformers faces significant technological and economic barriers, including low technological readiness of existing recycling methods, high operational costs, and degraded quality of recovered metals.
• Power losses during the operational phase account for 94% to 99% of the total environmental impact, depending on the load level. Within just the first few years of operation, even at low loads, CO₂-eq emissions associated with energy losses exceed those generated during the entire production process.
• High-quality construction materials and modern design solutions should be applied to minimize no-load and load losses. Recommended materials: for the core, materials with a narrow hysteresis loop (e.g., silicon steel or amorphous steel); for the windings, high-purity copper with low resistivity. Design improvements include the use of a laminated core composed of thin, insulated sheets; magnetic shielding; and toroidal or circular core shapes.
• Electricity generation must be restructured, especially in coal-dependent regions, where the majority of energy is still generated from hard coal and brown coal combustion. Transitioning to cleaner energy sources requires targeted political initiatives, strategic policy decisions, regulatory frameworks, and incentive mechanisms to drive transformation.

Future research should focus on enhancing the accuracy of transformer carbon footprint calculations by improving data quality. From a practical perspective, requiring manufacturers of transformer components and assemblies to provide detailed carbon footprint data would significantly improve the precision of lifecycle assessments. Also, the use of actual transformer load curves (seasonal, monthly, or even daily) should be considered when predicting CO₂-eq emissions resulting from load and no-load losses. Additionally, greater emphasis should be placed on the end-of-life phase, particularly in the context of the circular economy, resource sustainability, and societal impact. A comprehensive economic and social assessment is also needed to evaluate the financial feasibility and broader benefits of low-emission solutions. Such an integrated approach will provide a holistic understanding of the environmental, economic, and societal trade-offs associated with transformer technologies.