Numerical assessment of the influence of a R744 cooling unit design and components on the overall carbon footprint of a light refrigerated vehicle

Numerical assessment of the influence of a R744 cooling unit design and components on the overall carbon footprint of a light refrigerated vehicle Francesco Fabris(a), Wasim Shah(b), Sergio Marinetti(a), Silvia Minetto(a), Antonio Rossetti(a) (a) Consiglio Nazionale delle Ricerche, Istituto per le Tecnologie della Costruzione, Corso Stati Uniti, 435127 Padova – Italy (b) Università degli Studi di Padova, Doctoral Program in Industrial Engineering, Via Venezia, 1 – 35131 Padova - Italy ABSTRACT To meet the upcoming challenge of environmental sustainability, refrigerated transport is required to address energy efficiency and use sustainable fluids, namely natural refrigerants, at the same time taking into consideration the typical constraints of mobile systems, i.e. space and weight, system complexity, reliability and ease of servicing and maintenance. In this study, the design of a R744 refrigeration system employed on a light commercial vehicle used for urban delivery of fresh products (4-5 kW at 0°C) is considered. The effects of different system architectures, controls and components are numerically evaluated focusing on the trade-off between the increase of the unit complexity, leading to higher energy efficiency of the sole cooling system, and the decrease of the unit overall encumbrance and weight, leading to lower carbon footprint of the vehicle. Numerical results show that the use of an ejector can enhance the unit COP up to +28.6% for high ambient temperature conditions (40°C). A compact unit design based on the use of a light and efficient variable-speed scroll compressor leads to a reduction in carbon emissions of an average delivery mission between -17.9% and -34.4%, depending on climatic conditions. The effects of evaporator sizing on the system performance and emissions are discussed. The overall carbon footprint of the units over their entire life cycle is assessed, highlighting that more than 95% of the equivalent CO2 emissions are related to system operation. The use of natural refrigerant R744 reduces the direct leakage emissions to negligible values. Keywords: Refrigerated Transport; Carbon Dioxide; Energy Efficiency; Weight; Design Optimization; Carbon footprint 1. INTRODUCTION Refrigeration plays a crucial role in ensuring quality and safety of perishable products, such as food or pharmaceuticals. Over 1.12 billion people in the world face immediate risks related to lacking refrigeration, causing significant food loss and preventing a proper fight to hunger and malnutrition (Baha et al., 2025). In 2024, approximately 5.4 billion pieces of refrigeration equipment were estimated to be in operation worldwide and the market is expected to grow especially in underdeveloped areas, with a sevenfold increase of size in Africa and fourfold increase in South Asia by 2050 (Baha et al., 2025). On the other hand, the substantial growth of the refrigeration sector worldwide will determine an increase in carbon emissions linked to the operation of refrigeration systems. According to UNEP and FAO (2022), the current cold chain already contributes up to 4% of the global emissions, highlighting the crucial importance and need of developing new sustainable refrigeration solutions with limited impact on the environment. Within the cold chain, refrigerated transport represents an important yet impactful step: it has been estimated that it accounts for nearly 25% of the total cold chain emissions (IIR, 2021). This sector has seen significant growth in recent years, reaching a total of approximately 5.7 million refrigerated road vehicles worldwide (Baha et al., 2025), and with further increase foreseen for the next decades. Demanding operating conditions like vehicle speed, vibrations and accelerations (Tassou et al., 2009), together with highly variable thermal load, are essential boundaries to be considered for system design, which often results in oversized units (Maiorino et al., 2021) and consequent non optimized partial load operation. Moreover, refrigerant leakages represent a major issue in transport applications, with reported leakage rates which can be as high as 165% of the Transport Refrigeration Unit (TRU) initial change over the system lifetime (United Nations, 2017). Current market is almost entirely based on synthetic refrigerants (HFCs and HFOs): while R404A and R134a have been the most commonly used fluids in the last years, the phase down, enforced by the F-gas regulation in EU and the Kigali amendment worldwide, led to the identification of retrofitting fluids, like R452A, R449A and R442A (Minetto et al., 2023), still characterized by a significant Global Warming Potential (GWP). Overall, there is therefore a critical need for a comprehensive evaluation of the total carbon footprint of road refrigerated transport units to reduce emissions associated with both the unit operation and the refrigerant leakages. Vapour compression systems still represent the vast majority of road TRU market, as documented by Fertel et al. (2023), who reported that 95.3% of refrigerated vehicles in France in 2022 still relied on vapour compression systems. Given the dominant share of the market covered with this technology, research studies need to be focused on the development of more efficient vapour compression units which, at the same time, should be designed to operate with natural refrigerants, to reduce their overall carbon footprint. An in-depth review on the use of natural working fluids in transport refrigeration by Minetto et al. (2023) highlighted that, while industrial manufacturers have already developed and put in the market natural refrigerants-based solutions for specific sub-sectors such as marine transport (Ladam, 2021; Söylemez et al., 2022) or intermodal containers refrigeration (Carrier, 2017), only a few solutions can be found regarding road transport refrigeration. Hydrocarbon-based TRUs have been investigated both by research studies (Colbourne et al., 2017) and industrial manufacturers (Natural Refrigerants, 2022; Cooling Post, 2022), thanks to the simple refrigeration cycle structure and efficiency allowed by the hydrocarbons thermodynamic properties. However, almost all the mentioned systems employed hydrocarbons in a primary circuit kept outside of the refrigerated body and employed a secondary fluid to deliver the cooling effect inside the refrigerated compartment, to avoid potential flammability issues inside the confined space of the insulated box. On the other hand, CO2 (R744) is a non-toxic and non-flammable fluid, but its thermophysical properties require specific and more complex system design to achieve a high efficiency. A few prototypes have been proposed by industrial manufacturers (Carrier, 2016; Cooling Post, 2020), although being developed for mobile trailers which can be transported on the road. A few interesting examples of employment of R744 for refrigerated trucks have been reported in the last couple decades by another industrial manufacturer (Micheletto and Rosso, 2004: R744 system with eutectic plates for frozen food transportation; Cold Car, 2015: R744 system with photovoltaic panels for fresh food transportation) demonstrating good potential; however, such prototypes have never been further developed and launched on the market. To the best knowledge of the Authors, no R744 TRU for light commercial vehicles is currently available on the market. Numerical research activities carried out before the present study (Artuso et al., 2020; Fabris et al., 2021; Fabris et al., 2024a) assessed that a R744 transport refrigeration system developed for a light commercial vehicle can achieve COP values higher than commercially available HFC systems even at high ambient temperature, but at the same time highlighted that the increased weight of the refrigerated unit due to its schematic complexity and components bulkiness can lead to slightly higher fuel-related emissions. Consequently, significant reductions in the carbon footprint can be done through proper materials choice and component design. The influence of components choice, insulating materials and refrigerant on the cost and energy performance of TRUs has been recently investigated by RamirezQuintero et al. (2025), although an already phased-out HCFC refrigerant was considered in that study. The implementation of photovoltaic panels on the roof of the refrigerated vehicle was also investigated (Rossetti et al., 2022; Maiorino et al., 2024), showing great potential to enhance the energy balance of refrigerated trucks. This study focuses on the optimal design of a R744 TRU for light commercial vehicles and numerically evaluates different R744 unit configurations, to identify the best trade-off between COP and unit weight, with the overall goal of minimizing the global carbon footprint. To find the optimized conditions, numerical simulations are performed for different layouts, entailing different complexity level in the circuit arrangement, evaporator size and control logic. In addition, the influence of different climatic conditions is also accounted for in the evaluation of the overall carbon footprint of the proposed solutions. 2. COOLING UNIT LAYOUTS The objective of this study is the design and performance evaluation of a transport refrigeration unit based on the use of natural working fluid R744 and designed to fulfil the cooling needs of a light commercial vehicle, employed for last mile delivery of chilled goods in urban environment. The design cooling power is equal to 4-5 kW of Medium-Temperature (MT) cooling capacity, at a set point internal air temperature equal to 0°C. The internal temperature of 0°C has been chosen in accordance to the value prescribed in the procedure for determining the efficiency of class A mechanically refrigerated equipment, as defined in the ATP agreement (United Nations, 2024). In this study, two different TRU layouts are proposed. Each of these layouts then presents the possibility to operate in different configurations, modifying the refrigerant circuit to include or exclude certain components. 2.1 Modulated orifice unit Firstly, a modulated orifice (MO) unit, based on the results of previous research studies (Artuso et al, 2020; Fabris et al., 2021; Fabris et al., 2024a), is considered, and its simplified layout is presented in Figure 1. It is a traditional back-pressure concept, with a fixed-speed semi-hermetic compressor (operating at the rated velocity of 1450 rev min-1) and a modulating high-pressure valve (HPV). Figure 1: Modulated orifice (MO) unit layout. Different configurations are available by including or excluding the ejector and the auxiliary evaporator acting on the opening of solenoid valves, as presented in Figure 2, where red colour is used to represent high-pressure level, green represents low-pressure level and light green, when present, represents an intermediate pressure level. Dashed lines are used to indicate the portions of the unit not used in the considered configuration. The back-pressure (BP) configuration occurs when excluding both the ejector and the auxiliary evaporator, as displayed in Figure 2a. Figure 2b presents the schematic of the ejector (EJ) configuration, where a two-phase ejector is added in parallel with the HPV, for exploiting part of the available expansion work at the outlet of the gas cooler to provide a pressure lift to the refrigerant from evaporation to liquid receiver pressure level, thus increasing the compressor suction pressure and consequently reducing the compressor power consumption required to provide a fixed cooling effect. The compressor size is chosen to always choke the ejector motive nozzle, while the excess mass flow rate is expanded in the HPV which controls the system high-pressure. Figure 2c presents the ejector and auxiliary evaporator (EJ,AUX) configuration, where an auxiliary evaporator is engaged between the ejector-HPV and the liquid receiver, providing cooling effect in the same chilled space where the main evaporator operates. The auxiliary evaporator helps extending the range of conditions in which the ejector can be used towards low ambient temperatures. A more detailed discussion on the role and the effects on the auxiliary evaporator can be found in Artuso et al. (2020). The components included in this layout, selected according to market availability, and their main characteristics and weight are reported in Table 1. (a) (b) (c) Figure 2: MO unit: (a) back-pressure (BP) configuration; (b) ejector (EJ) configuration; (c) ejector and auxiliary evaporator (EJ,AUX) configuration. (Red: high pressure; green: low pressure; light-green: intermediate pressure; dashed lines: portions not used). Table 1. List of components used in MO unit and their main characteristics Component Main characteristic Weight Compressor Displacement: 21.6 cm3; Speed: 1450 rev min-1 77.3 kg Gas cooler External heat exchange area: 16.9 m2 18.0 kg Evaporator External heat exchange area: 19.7 m2 35.4 kg Auxiliary evaporator External heat exchange area: 15.9 m2 28.2 kg Liquid separator Internal volume: 10 L 5.0 kg 2.2 Fixed orifice unit As a simpler and less complex alternative to the MO unit layout, a fixed orifice (FO) unit layout is proposed in this study, and its schematic is presented in Figure 3. The high-pressure is controlled through a variable-speed scroll compressor (operating between 1200 and 6000 rev min-1), while the electronic high-pressure valve (HPV) is replaced by a fixed expansion valve (FEV). Even for this layout, different configurations can occur. Figure 4a illustrates the fixed expansion valve (FEV) configuration while Figure 4b presents the ejector (EJ) configuration, where the two-phase ejector is used as the only expansion device. The meaning of the colours and of the dashed lines used in Fig. 4 is the same as in Fig. 2. Being the main goal of the FO unit layout the reduction of the system complexity compared to the MO unit layout, the auxiliary evaporator is not included in this layout. Further discussion on the exclusion of the auxiliary evaporator in the FO unit will be provided in the results section. The components included in this layout, selected according to market availability, and their main characteristics and weight are reported in Table 2. Figure 3: Fixed orifice (FO) unit layout (a) (b) Figure 4: FO unit: (a) fixed expansion valve (FEV) configuration; (b) ejector (EJ) configuration. (Red: high pressure; green: low pressure; light-green: intermediate pressure; dashed lines: portions not used). Table 2. List of components used in FO unit and their main characteristics Component Main characteristic Weight Compressor Displacement: 6.7 cm3; Speed: 25-100 rev s-1 13.8 kg Gas cooler External heat exchange area: 16.9 m2 18.0 kg Evaporator External heat exchange area: 19.7 m2 35.4 kg Liquid separator Internal volume: 10 L 5.0 kg 3. NUMERICAL EVALUATION 3.1 Cooling unit modelling approach The numerical approach employed to simulate the cooling unit steady-state and dynamic performance is extensively described in previous studies by the same authors (Artuso et al., 2020; Fabris et al., 2021; Fabris et al., 2024b). For the sake of brevity, the main characteristics of the modelling approach are here reported, while a thorough description of the numerical model of each system component, including the formulation of differential equations and of the empirical correlations used to define the heat transfer coefficients can be found in the above-mentioned studies. The numerical models are developed using the multi-physics commercial software Simcenter Amesim (Siemens, 2024). The numerical approach is based on discretizing real components into lumped parameter elements, mutually connected to represent the entire system. Each element is defined by nonlinear time-dependent differential equations, which are coupled in a system and integrated over time to solve the system dynamic behaviour. The compressor performance is modelled through the calculation of volumetric and overall compression efficiency according to the compressor performance data (mass flow rate and power input) declared by the compressor manufacturer. The performance is provided in the form of third-order polynomial equations, as a function of suction and discharge pressures, based on ten coefficients declared by the compressor manufacturer and defined according to the EN12900 standard. Heat exchangers are modelled discretizing the total heat exchange surface into a series of longitudinal lumped volumes, in a perfect counter-flow configuration. This modelling approach has been validated comparing the computed cooling effect against nominal data of heat exchangers reported in technical datasheets (Fabris et al., 2021). For each lumped volume, three elements representing the internal refrigerant flow, the pipe and fins material, and the external air flow, respectively, are considered. Differential equations accounting for internal convection, conduction through the pipe walls and fins and external convection are used. To accurately reproduce the sharp property change in supercritical conditions, a higher number of lumped volumes (18) is used for the gas cooler than for the evaporator (6 volumes). The ejector model is based on the performance data of an ejector specifically designed for MT transport refrigeration application (Fabris et al., 2024b). The performance maps of the ejector in different operating conditions are implemented into the cooling unit model: for each time step, the boundary conditions at the ejector ports are used to determine the performance of the ejector (motive mass flow rate, entrainment ratio) through linear interpolation of the ejector performance maps. The high-pressure control is performed by a Proportional-Integral (PI) controller as a function of the gas cooler outlet temperature, to set it to the optimal heat rejection pressure as defined by Liao et al. (2000) in transcritical operation, while to a saturation pressure leading to a temperature approach equal to 3 K at the gas cooler outlet in subcritical operation. In the MO unit layout, the PI controls the opening of the HPV valve, while in the FO unit layout it controls the compressor speed. 3.2 Carbon footprint evaluation A quasi-steady-state numerical approach is used to evaluate the equivalent CO2 emissions associated with the TRU operation when employed on a refrigerated van in a short-range multi-drop delivery mission in urban environment. The programming environment used to estimate energy consumption and emissions is MATLAB (Mathworks, 2024). Four different climatic conditions (EnergyPlus, 2025) are considered with one hour time step for the carbon footprint evaluation: Helsinki (humid continental, warm summer – Dfb under the Köppen-Geiger system; Kottek, 2006), Strasbourg (temperate oceanic climate - Cfb), Athens (hot-summer Mediterranean climate - Csa) and Phoenix (hot desert climate - BWh). 3.2.1 Average delivery mission The short-range delivery mission is defined setting a daily travelled distance equal to 150 km, with an average vehicle velocity equal to 18.9 km h-1 (based on the WLTP Class 3a Urban Cycle definition). Over the delivery mission time, 20 stops with the duration of 5 minutes each are considered, in which the box doors are open and external air infiltration occurs. A 1-hour additional stop is considered to account that drivers have the obligation to stop at least for one hour after four consecutive hours of driving. The resulting mission duration is equal to 10.6 hours. To account for the thermal loads variability over the day, it has been considered that the delivery mission can occur between 08:00 and 22:00. The thermal loads are therefore calculated for each hour within this span, but then proportionally reduced to account for the actual mission duration (10.6 hours) within the considered 14-hours span. The insulated box external dimensions are 3.1m×2.2m×2.1m, representative of a light commercial vehicle. The operating internal air temperature set-point is 0°C, for fresh products preservation. The insulated box global heat transfer coefficient K is 0.60 W m-2 K-1. Given the numerical model inputs (ambient conditions and mission definition), the main thermal loads in the insulated box are computed. The pulldown load corresponds to the energy required to lower the box temperature from an initial stationary temperature to the operational set-point and it is estimated based on the experimental characterization reported by Artuso et al. (2019). The stationary equilibrium temperature was assumed to be the mean value between the average ambient temperature for the given day and location and the internal air set-point (0°C). The conduction-convection heat infiltration is modelled according to the global heat transfer coefficient (K), as defined in the ATP agreement (United Nations, 2024). The sensible and latent infiltration load is evaluated based on the conservative assumption that, during each of the 5-minute door openings, the entire air volume inside the insulated box is replaced by external air (characterized by temperature and relative humidity of the ambient), which needs to be pulled down to the temperature set-point. It has to be pointed out that previous studies demonstrated that radiation load is significant for refrigerated vehicles only if long stops are considered (Artuso et al., 2019). In fact, when the truck is in motion, the external convection between outer surface of the insulated body and the ambient air is sufficient to maintain the surface temperature close to the ambient level. In this study, it is assumed that pulldown operations take place in the warehouse where the food products to be delivered are stored, with the vehicle kept in shade and not exposed to direct solar radiation. For these reasons, solar radiation is not included in the model. Given the hourly ambient temperatures, the TRU performance (cooling effect and COP) is calculated through the numerical evaluation described in Section 3.1. However, a COP degradation factor is applied following the EN 14825 approach to account for partialisation, as the hourly thermal load might result lower than the nominal cooling effect of the unit in those operating conditions. From the values of thermal loads and COP, the power consumption of the unit for each hour of the year can be calculated. This value is then averaged over the year and, considering the mission duration, it leads to the yearly average mission energy consumption. The TRU is assumed to be powered by the vehicle main engine through a belt connection to the driveline. Accordingly, a carbon conversion factor equal to 651 gCO2,eq kWh-1 (Grigoratos et al., 2019) has been used. The CO2 emissions linked to the extra fuel consumption needed to transport the weight of the TRU is also considered with a conversion factor equal to 37.5 gCO2,eq ton-1 km-1 (Fabris et al., 2024a). 3.2.2 Life cycle evaluation A simplified assessment of the carbon footprint of the considered TRUs over their life cycle is conducted based on the structure described in Fabris et al. (2024a). The average age of light commercial vehicles in use in Europe, equal to 12.0 years (ACEA, 2023), is assumed as the TRU lifetime. The yearly operation is assumed to be characterized by repetitions of the average delivery mission, calculated as described in Section 3.2.1, for 250 days in the year (Li, 2017). Being the total refrigerant charge of unit 4.5 kg of R744, direct leakage emissions are calculated assuming a yearly leakage rate of 15% (Tassou et al., 2009; Li, 2017; Wu et al., 2022) and assuming that, during the TRU dismantling at end of life, the whole R744 charge is released in the atmosphere, given its low GWP (equal to 1), almost negligible compared to traditionally used refrigerants such as R404A (GWP100 = 3922) and R134a (GWP100 = 1530) or to retrofitting fluids used in latest years such as R452A (GWP100 = 2141), and its harmfulness towards environment. CO2 refrigerant is primarily recovered from industrial processes via Carbon Capture and Storage (CCS). The average carbon footprint associated with the CCS process, accounting for various technological pathways, is estimated at approximately 0.14 kgCO2,eq per kg of CO₂ captured (Rubin et al., 2015). Additional compression and purification steps are required to achieve purity and avoid excessive H₂O and non-condensable gases, and the associated carbon footprint is estimated in 0.04 kgCO2,eq per kg of CO₂ processed (Posch and Haider, 2012). Electrical energy consumption factors equal to 4.22 kWh/kg and to 2.69 kWh/kg are assumed for the TRUs manufacturing and recycling processes, respectively (Wu et al., 2022). The carbon footprint linked to these processes is then evaluated considering the average GHG emission intensity of the electricity generation in Europe, equal to 0.21 kgCO2,eq/kWh (EEA, 2025). 4. NUMERICAL RESULTS 4.1 Cooling unit performance Firstly, the numerically evaluated steady-state performance of the different cooling unit layouts and configurations is presented under different ambient temperature conditions. 4.1.1 Modulated orifice unit The steady-state performance of the MO unit, in each of its possible configurations, is presented in Figure 5. Figure 5a and 5b present the cooling unit thermodynamic COP and overall COP, respectively, defined as: 𝐶𝑂𝑃𝑡ℎ =𝑄𝑐 𝑃 𝑐𝑜𝑚𝑝 Eq. (1) 𝐶𝑂𝑃 = 𝑄𝑐− 𝑃 𝑓𝑎𝑛𝑠,𝑒𝑣 𝑃 𝑐𝑜𝑚𝑝 + 𝑃 𝑓𝑎𝑛𝑠,𝑒𝑣 + 𝑃 𝑓𝑎𝑛𝑠,𝑔𝑐 Eq. (2) where 𝑄𝑐 represents the cooling effect provided by the refrigeration system, 𝑃 𝑐𝑜𝑚𝑝 the compressor power input and 𝑃 𝑓𝑎𝑛𝑠 the power consumption of fans, which for the evaporator represents an additional thermal load and power consumption at the same time. The cooling effect 𝑄𝑐 is presented in Figure 5c. When using the ejector (EJ and EJ,AUX), only 𝑇𝑎𝑚𝑏 > 20°C is considered since, at lower ambient temperatures, the motive energy of the refrigerant flow at the outlet of the gas cooler (and, therefore, at the ejector motive nozzle) is too low to entrain mass flow rate at the ejector suction nozzle. Moreover, a fixed opening ratio of the expansion valve EV (Figure 1) is assumed, corresponding to the value resulting in the pressure lift maximizing COP at design conditions (𝑇𝑎𝑚𝑏 = 30°C). (a) (b) (c) Figure 5: MO unit steady-state performance for 𝑻𝒊 = 0°C and different 𝑻𝒂𝒎𝒃: (a) thermodynamic COP; (b) COP; (c) cooling effect Figure 5a shows that the thermodynamic COP of the unit operating in ejector configurations is always significantly higher than in BP configuration. Moreover, the auxiliary evaporator is more beneficial towards lower ambient temperature conditions (for 𝑇𝑎𝑚𝑏 = 20°C the 𝐶𝑂𝑃𝑡ℎ improvement between EJ,AUX and EJ configurations is equal to +5.7%), while at higher ambient temperatures the auxiliary evaporator becomes counterproductive (-2.1% in 𝐶𝑂𝑃𝑡ℎ for 𝑇𝑎𝑚𝑏 = 40°C), in agreement with the results presented in Artuso et al. (2020). However, when accounting for the additional thermal load and additional power consumption of the evaporator fans (Figure 5b), the EJ,AUX configuration (two evaporators, therefore two fans) presents a lower overall COP than the EJ configuration for almost every considered operating condition. Moreover, considering the cooling effect (Figure 5c), the EJ,AUX configuration delivers the maximum cooling effect increase, compared to the BP configuration, for low ambient temperatures, when the actual cooling demand is lower. On the contrary, the EJ configuration presents the maximum increase in cooling capacity for high ambient temperatures, when it is needed the most due to higher thermal loads. Following the numerical results described above, and at the same time considering the resolution of achieving a less complex cooling unit layout for the FO system compared to the MO system, the EJ,AUX configuration is not considered among the possible configurations in the FO unit layout. 4.1.2 Fixed orifice unit The steady-state performance of the FO unit, in each of its possible configurations, is presented in Figure 6, in which the overall COP, as defined in Eq. (2), is reported in Figure 6a, while the cooling effect is reported in Figure 6b. takes-delivery-of-world-first-carrier-transicold-natural-refrigerant-trailer-unit.html Carrier, 2017. https://www.carrier.com/container-refrigeration/en/worldwide/products/ContainerUnits/naturaline/ Colbourne, D., Solomon, P., Wilson, R., De Swardt, L., Nosbers, R., Schuster, M., 2017. Development of R290 Transport Refrigeration System. Institute of Refrigeration, Carshalton, UK, 2017. Available online: https://secure.toolkitfiles.co.uk/clients/33346/sitedata/files/MACP-News-Article-10.03.1720170310141830.pdf Cold Car, 2015. “Cold Car all’avanguardia con il progetto Vector 2015”. Available online: https://coldcar.com/cold-car-allavanguardia-con-il-progetto-vector-2015/ Cooling Post, 2020. Greencold goes large with CO2 mobile refrigeration. Available online : https://www.coolingpost.com/features/greencold-goes-large-with-co2-mobile-refrigeration/ Cooling Post, 2022. Electrically-driven TRU uses R1270 and CO2. Available online: https://www.coolingpost.com/products/electrically-driven-tru-uses-r1270-and-co2/ EnergyPlus, 2025. EnergyPlus Weather Data. Available online: https://energyplus.net/weather European Environment Agency (EEA), 2025. Greenhouse gas emission intensity of electricity generation in Europe. Available online: https://www.eea.europa.eu/en/analysis/indicators/greenhouse-gasemission-intensity-of-1 Fabris, F., Artuso, P., Marinetti, S., Minetto, S., Rossetti, A., 2021. Dynamic modelling of a CO2 transport refrigeration unit with multiple configurations. Applied Thermal Engineering, 189(February), 116749. https://doi.org/10.1016/j.applthermaleng.2021.116749 Fabris, F., Fabrizio, M., Marinetti, S., Rossetti, A., Minetto, S., 2024a. Evaluation of the carbon footprint of HFC and natural refrigerant transport refrigeration units from a life-cycle perspective. International Journal of Refrigeration, 159, 17–27. https://doi.org/10.1016/j.ijrefrig.2023.12.018 Fabris, F., Bodys, J., Marinetti, S., Minetto, S., Smolka, J., Rossetti, A., 2024b. Numerical modelling of a single-compression multi-temperature ejector-supported R744 refrigeration unit for last mile delivery. 160(by 2030), 65–75. https://doi.org/10.1016/j.ijrefrig.2024.01.014 Fertel, C., Cavalier, G., Engelmann, P. (2023). Situation and perspective of evolution of the French refrigerated transport fleet. Proceedings of the 26th International Congress of Refrigeration, 21-25 August 2023,Paris, France. http://dx.doi.org/10.18462/iir.icr.2023.0919 Grigoratos, T., Fontaras, G., Giechaskiel, B., Zacharof, N., 2019. Real world emissions performance of heavyduty Euro VI diesel vehicles. Atmospheric Environment, 201(December 2018), 348–359. https://doi.org/10.1016/j.atmosenv.2018.12.042 International Institute of Refrigeration (IIR), 2021. 7th Informatory Note on Refrigeration and Food, The Carbon Footprint of the Cold Chain. http://dx.doi.org/10.18462/iir.INfood07.04.2021 Kottek, M., J. Grieser, C. Beck, B. Rudolf, and F. Rubel, 2006: World Map of the Köppen-Geiger climate classification updated. Meteorol. Z., 15, 259-263. https://doi.org/10.1127/0941-2948/2006/0130 Ladam, Y., 2021. CO2 transcritical systems for marine. International webinar: CO2 as refrigerant in fishing vessels. Available online: https://www.sintef.no/globalassets/sintefocean/coolfish/5_ladam.pdf Li, G., 2017. Comprehensive investigation of transport refrigeration life cycle climate performance. Sustain. Energy Technol. Assess. 21, 33–49. https://doi.org/10.1016/j.seta.2017.04.002 Liao, S.M., Zhao, T.S., Jakobsen, A., 2000. A correlation of optimal heat rejection pressures in transcritical carbon dioxide cycles. Applied Thermal Engineering 20 (9), 831–841. https://doi.org/10.1016/S1359-4311(99)00070-8 Micheletto, A., Rosso, G., 2004. La "ECO2 refrigerated body" della Cold Car. Available online: https://www.centrogalileo.it/nuovaPA/Articoli%20tecnici/Co2/Cold%20Car.htm Minetto, S., Fabris, F., Marinetti, S., Rossetti, A., 2023. A review on present and forthcoming opportunities with natural working fluids in transport refrigeration. International Journal of Refrigeration, 152(April), 343–355. https://doi.org/10.1016/j.ijrefrig.2023.04.015 Maiorino, A., Petruzziello, F., Aprea, C., 2021. Refrigerated transport: state of the art, technical issues, innovations and challenges for sustainability. Energies 14 (21), 7237. https://doi.org/10.3390/en14217237 Maiorino, A., Petruzziello, F., Cilenti, C., Llopis, R., Aprea, C., 2024. Performance evaluation of a hybrid photovoltaic-vapor compression system serving a refrigerated van. International Journal of Refrigeration, 168, 720-729. https://doi.org/10.1016/j.ijrefrig.2024.10.021 Mathworks, 2024. Matlab 2024b. https://it.mathworks.com/ Natural Refrigerants, 2022. ‘First’ R290 Refrigeration Unit for Electric Vehicles to Launch at IAA Transportation Conference. Available online: https://naturalrefrigerants.com/first-r290-refrigerationunit-for-electric-vehicles-to-launch-at-iaa-transportation-conference/ Posch, S., Haider, M., 2012. Optimization of CO2 compression and purification units (CO2CPU) for CCS power plants. Fuel 101, 254–263. https://doi.org/10.1016/j.fuel.2011.07.039 Ramirez-Quintero, D. A., Silva, F. L., Miranda, M. H. R., Silva, L. C. A., Eckert, J. J., 2025. Influence of different types of insulating materials, compressors, and refrigerants on the cost and energy performance of refrigerated transport systems. Applied thermal engineering 269, Part A, 125994. https://doi.org/10.1016/j.applthermaleng.2025.125994 Rossetti, A., Marinetti, S., Artuso, P., Fabris, F., Minetto, S., 2022. Implementation of a solar aided refrigration unit for refrigerated trucks employing photovoltaic generators. Energy Reports, 8, 77897799. https://doi.org/10.1016/j.egyr.2022.05.284 Rubin, E.S., Davison, J.E., Herzog, H.J., 2015. The cost of CO2 capture and storage. Int. J. Greenhouse Gas Control 40, 378–400. https://doi.org/10.1016/j.ijggc.2015.05.018 Siemens, 2024. Simcenter system simulation - Simcenter Amesim software. Available online: https://plm.sw.siemens.com/en-US/simcenter/systems-simulation/amesim/ Söylemez, E., Widell, K. N., Gabrielii, C. H., Ladam, Y., Lund, T., Hafner, A., 2022. Overview of the development and status of carbon dioxide (R-744) refrigeration systems onboard fishing vessels. International Journal of Refrigeration, 140(May), 198–212. https://doi.org/10.1016/j.ijrefrig.2022.05.007 Tassou, S. A., De-Lille, G., Ge, Y. T., 2009. Food transport refrigeration - Approaches to reduce energy consumption and environmental impacts of road transport. Applied Thermal Engineering, 29(8–9), 1467–1477. https://doi.org/10.1016/j.applthermaleng.2008.06.027 UNEP and FAO, 2022. Sustainable Food Cold Chains: Opportunities, Challenges and the Way Forward. Nairobi, UNEP and Rome, FAO. https://doi.org/10.4060/cc0923en United Nations, 2017. Methodological Tool: Calculation of Baseline, Project and leakage Emissions from the use of Refrigerants. Clean Development Mechanism. Available online: https://cdm.unfccc.int/methodologies/PAmethodologies/tools/am-tool-28-v1.pdf/history_view United Nations (2024). Agreement on the International Carriage of Perishable Food-stuffs and on the Special Equipment to be Used for Such Carriage (ATP). UNECE Transport Division, Geneva, Switzerland, 2024. Available online: https://unece.org/text-and-status-agreement Wu, J., Li, Q., Liu, G., Xie, R., Zou, Y., Scipioni, A., Manzardo, A., 2022. Evaluating the impact of refrigerated transport trucks in China on climate change from the life cycle perspective. Environ. Impact Assess. Rev. 97, 106866. https://doi.org/10.1016/j.eiar.2022.106866