Switchable Heat Pipes for Eco-Friendly Battery Cooling in Electric Vehicles: A Life Cycle Assessment

Abstract

Battery thermal management systems (BTMSs) ensure that lithium-ion batteries (LIBs) in electric vehicles (EVs) are operated in an optimal temperature range to achieve high performance and reduce risks. A conventional BTMS operates either as an active system that uses forced air, water or immersion cooling, or as a complete passive system without any temperature control. Passive systems function without any active energy supply and are therefore economically and environmentally advantageous. However, today’s passive BTMSs have limited cooling performance, which additionally cannot be controlled. To overcome this issue, an innovative BTMS approach based on heat pipes with an integrated thermal switch, developed by the Fraunhofer Cluster of Excellence Programmable Materials (CPM), is presented in this paper. The suggested BTMS consists of switchable heat pipes which couple a passive fin-based cold plate with the battery cells. In cold state, the battery is insulated. If the switching temperature is reached, the heat pipes start working and conduct the battery heat to the cold plate where it is dissipated. The environmental benefits of this novel BTMS approach were then analysed with a Life Cycle Assessment (LCA). Here, a comparison is made between the suggested passive and an active BTMS. For the passive system, significantly lower environmental impacts were observed in nearly all impact categories assessed. It was identified as a technically promising and environmentally friendly approach for battery cooling in EVs of the compact class. Furthermore, the results show that passive BTMS in general are superior from an environmental point of view, due their energy self-sufficient nature.

1. Introduction

Electric vehicles (EVs) are a major player in the ecological transformation to a net-zero emission society. In the European Union (EU), the transport sector is responsible for about 25% of total greenhouse gas emissions [1]. The main advantage of EVs is the ability to directly use electrical energy, which ideally should be harvested using sustainable methods. However, to achieve system sustainability, not only drive energy but also the production of the electric vehicle must be sustainable. In addition, EVs are often more expensive than traditional vehicles, for example due to the battery technology cost and limited production scale. European countries are making efforts to decrease their price to make electric driving possible for everyone [2].

Today’s EVs use lithium-ion batteries (LIBs) due to their superior energy density, lack of memory effect and high lifespan [3]. The safety and performance of LIBs, such as a high capacity, efficiency and lifespan, demand storage and operation in a small temperature range [4]. Lower temperatures decrease battery performance by reducing diffusion rates and ionic conductivity as well as increasing the viscosity of the electrolyte and the internal resistance. Moreover, the risk of lithium plating, i.e., the formation and deposition of metallic lithium at the anode, increases along with the irreversible loss of active material and the formation of dendrites. These dendrites can cause an internal shortcut and a thermal runaway (catastrophic chain reaction) in the worst case. Temperatures above the limit reduce lifespan and, in the worst case, also lead to a thermal runaway [5]. For this reason, cell manufacturers limit the maximum permissible temperature range for LIBs during discharging to, e.g., −20 to 60 °C and during charging to, e.g., 0–40 °C [6]. The risk of irreversible damage and exothermic decompensation reactions begins at around 70 °C and increases during charging in the range of the maximum battery voltage [7]. Due to this high thermal sensitivity, a battery thermal management system (BTMS) is mandatory for EVs.

Generally, a distinction is made between active and passive BTMSs. Passive BTMSs are inexpensive, energy self-sufficient and often maintenance free, while active BTMSs offer higher cooling power and the ability to control the cooling power. By using electrical energy to cool the battery, the cooling power can easily be controlled and enhanced. Active cooling can be achieved with forced convection, e.g., by a fan or a compressor-based cooling system like a chiller combined with a forced indirect or immersion liquid cooling system [8]. The demand on cooling power increases with environment temperature, charging or discharging power and C-rate, due to higher thermal losses. Therefore, active BTMSs are inevitable for EVs with huge batteries and a high charging/discharging power. However, smaller EVs with small batteries and lower charging power also need less cooling power. Here, passive BTMSs are sufficient and can reduce cost, energy demand and the environmental impact of the EV. Passive BTMSs, in general, are based on the utilization and optimisation of heat capacity, conductivity and convection. Neglected in many publications, but a decisive basis of every battery BTMS is the capacity of the battery system itself (cells, housing, etc.), which is used to store heat, to flatten temperature peaks and to homogenize the temperature distribution. This can be improved with phase change materials (PCMs). PCMs can store large amounts of heat in a small temperature range, due to their large latent heat. Hence, much research has been conducted to integrate and utilize PCMs in BTMSs [9,10]. However, the stored heat needs to be dissipated into the environment, eventually. Here, a good heat transport is mandatory and can be achieved by an improved conductivity, convection, and a combination of both. Like the heat capacity, the right design and choosing the suitable materials for the battery system is the base for a good thermal conductivity. Nevertheless, a thermally well-designed battery housing is not sufficient. This is shown, e.g., by the 2019 62 kWh Nissan LEAF, which suffered from the rapid gate effect (reducing speed and charging power) [11]. To increase the temperature exchange with the environment, heat pipes are a great option. Heat pipes are available in various designs and are used, among other things, for efficient heat transport and for homogenizing temperature fields [12]. Therefore, several studies regarding the use of heat pipes in BTMSs have been performed [9,10]. Another option for a rapid temperature exchange with the surroundings is natural convection. Here, an air flow channelled from the outside of the moving vehicle is used for heat dissipation. However, only little attention from the research community was given to this topic [13] and only in combination, e.g., with PCMs [14].

Using the described passive methods, BTMSs can be developed that are simple, inexpensive, and energy-self-sufficient to name but a few advantages. However, these BTMSs have no option to control temperature. Hence, state-of-the-art passive BTMSs need to make a design compromise between insulation for cold conditions, and good conductivity for hot conditions or fast charging. The performance therefore depends on the weather conditions. Additionally, fast charging is limited to a certain level. If the threshold is reached, the BTMS regulates the charging power and the maximum speed of the EV down (rapid gate effect [11]). These issues currently limit the use of passive BTMSs and are the reason why active BTMSs are chosen for nearly all modern EVs. Here, recent developments in passive thermal switches combined with elements with high thermal conductivity might be a game-changer.

As summarized in [15], various passive thermal switches have been developed so far. Contact thermal switching is an easy and effective approach for thermal switches, that is based on switching the conduction by simply bringing materials in and out of contact. In combination with shape memory alloys (SMAs), passive thermal switches can be designed. SMAs are able to recover their shape after deformation by heating. This can be used to apply a displacement and/or a force when a certain temperature is reached. SMA-based contact thermal switches have been investigated for the use in BTMS systems. These investigations showed promising results and are based on SMA wires that switch in-between a good contact of battery and heat sink, if warm, and no contact, if cold [16,17,18]. Nevertheless, transferring this method to large battery systems for EVs, where hundreds of cells have to be moved, does not appear to be feasible. In contrast, heat-pipe-based approaches offer a steady position of both, the heat pipe and the battery cell. Hence, switchable heat pipes are an excellent choice for passive BTMSs. Switchable heat pipes, such as the one developed by the Fraunhofer Cluster of Excellence Programmable Materials (CPM) [19], described in detail in Section 2.1, show a good switching ratio as well as a good thermal conductivity and allow lightweight constructions and a small installation space. The absence of moving parts allows a tight stacking and makes them maintenance-free. Despite these advantages, BTMSs with switchable heat pipes have not yet been investigated, to the authors’ knowledge.

A life cycle assessment (LCA) quantifies the environmental impacts of a product or a process and is an ISO standardised method [20,21]. It can assist in pointing out possibilities for improving the environmental impact and the results can be used to inform decision-makers in the industry as well as government or non-government organisations. To minimise the consequences of climate change, it is becoming increasingly important that products and processes are environmentally friendly. Particularly when developing new solutions, LCA can provide information at an early stage on how the new idea compares to existing options. To the authors’ knowledge, an analysis of the environmental impact of active and passive BTMSs has not yet been published.

In this paper, a passive system with temperature control by the abovementioned heat-pipe-based thermal switches and cooling plate is presented and investigated. In contrast to conventional systems, where either an active system utilizing water cooling or immersion cooling or a completely passive system without any temperature control (other than limiting the power of the battery) is applied, the novel concept allows for temperature control as well as a good balance between performance, cost and environmental impact. The concept combines advantages of both approaches. On the one hand, it enables control of the heat flux between the cooling plate and battery cell, despite its passive nature. This enables the battery to be cooled when it is hot and to be thermally insulated when it is cold. On the other hand, the system is simple, maintenance free, energy self-sufficient and potentially has a smaller environmental as well as economic impact compared to active systems with water-cooled cooling plates, vapor-compression refrigerators, pumps, etc. This paper first presents the concept of these novel thermal switches (Section 2.1). The innovative thermal management concept is then described (Section 2.2). This is followed by a life cycle assessment (LCA) in which the environmental impact of the presented concept is compared to a conventional active thermal management system (Section 3).

2. Design of Thermal Management System

2.1. Thermal Switch Based on an Adsorption Material in a Heat Pipe

Switchable heat pipes have numerous advantages for passive BTMSs, as described in the introduction. Moreover, non-switchable heat pipes for battery thermal management have been investigated in a significant number of experimental and numerical research publications during the past decade. The current status was recently summarized in a review article by Weragoda et al. with over 240 references [22]. Besides conventional tubular heat pipes, niche heat pipe types such as pulsating heat pipes, loop heat pipes and mini/micro heat pipes have also been reviewed. Despite the field of research being highly active, the authors conclude that heat pipe-based BTMSs are yet to be commercialized since questions like failure criteria, understanding of limitations, etc., are still open.

In this work, the focus is on heat pipes exhibiting a thermal switch functionality. For this special type of heat pipe, Schneider et al. reviewed the state of the art and proposed battery cooling in electric vehicles as a possible application [23]. An overview of the presented switching concepts is given in

Table 1. Overview of existing variably controllable heat pipe concepts, based on [23] with additional updates. Working principle: external or internal working principle (i.e., switching mechanism). Actuation: actuation taking place with necessity (active) or without necessity (passive) for an additional energy source.

Concept	Working Principle	Actuation	Ref.
Heat pipe with moving plug	Internal	Passive	[24,25]
Heat pipe with storage tank	Internal	Passive	[26]
Variable conductance heat pipe (VCHP)	Internal	Passive	[27]
Heat pipe with magnetic valve	Internal	Active	[28]
Heat pipe with valve flap	Internal	Active	[28]
VCHP with piston	Internal	Active	[29]
Externally moved heat pipe	External	Passive	[30]
Heat pipe with shape memory alloy actuator	External	Passive	[23,31]
Heat pipe with thermal coupling	External	Active	[28]
Heat pipe with integrated adsorbent	Internal	Passive/Active 1	[19,32,33], this work

¹ Thermal switch functionality can be activated either passively or actively [33].

.

In this work, we focus on the innovative passive thermal switch concept shown in Table 1 that is based on a heat pipe with integrated adsorbent material, containing the working fluid in adsorbed form (Figure 1). The switching temperature of the adsorbent is programmable and can be defined and adjusted by the material composition in line with requirements. The concept summarized below was presented and realized as switchable thermosiphon in previous publications [19,32,33].

While heat pipes and thermosiphons transport huge amounts of heat using latent heat transport, only heat pipes can work with and against gravity. This is achieved by a so-called wick (capillary structure). At the heat source, a working fluid is vaporised. The vapor travels through the pipe to the heat sink where the working fluid condensates. Here, the integrated adsorbent is binding the working fluid, which is water in this case, below a certain temperature threshold, the switching temperature T_S. Thus, the latent heat transport is switched off. If the temperature exceeds this threshold, the absorbent releases significant amounts of water by desorption. Hence, the latent heat transport is switched on and the heat pipe’s thermal resistance drops significantly. The described approach enables a passive thermal switch which contains no moving parts. As no external forces such as mechanical power or electricity are required, the switch can be operated cost-effectively.

While in [19,32] the concept is presented using a thermosiphon, in this paper, the concept is presented using a heat pipe, which can work against gravity. In Figure 2, exemplary experimental results for the described thermal switch concept are shown, based on a demonstrator consisting of a copper heat pipe with sintered wick, an outer diameter of 10 mm, a total length of 40 cm and a vertical orientation with the evaporator at the bottom. TAPSO-34 was used as the adsorbent [19] and water was used as the working fluid. The aim of this demonstrator was to show the basic capabilities of the thermal switch concept independently of its specific application. For each target application (for example, electromobility), the thermal switching properties can be adapted as described below.

The characteristic parameter of a thermal switch is its thermal resistance, i.e., the temperature difference between the hot and cold side divided by the transported heat flow. This should be as high as possible in the “off state” and as low as possible in the “on state”, i.e., the ratio of the thermal resistances in the off state to the on state (switching ratio) should be as high as possible. These quantities can be determined experimentally by applying a constant heat power of, e.g., 10 W on the heat pipe evaporator (hot side) and dissipating the heat on the heat pipe condenser (cold side) via water cooling (at a cooling water temperature of 20 °C). In the special experimental implementation discussed here, the sorbent was in a separate reservoir with a dedicated heater, allowing different temperatures to be actively set for the sorbent independently of the heat power at the hot side.

Initially, there was a high thermal resistance in the off state of ca. 7 K/W, which essentially corresponds to the low heat conduction across the thin metal wall and wick of the heat pipe. Due to the high thermal resistance, the hot side temperature reached high levels. To demonstrate the switching effect, the temperature at the sorbent reservoir was then increased, while the heat power on the hot side was kept constant at 10 W. As soon as the switching temperature of approx. 65 °C was exceeded at the reservoir, the working fluid was released, the heat pipe activated and the thermal resistance dropped very sharply. As a result, the temperature on the hot side dropped sharply (not shown in Figure 2). For testing the limits of the system, the heat power on the hot side was increased, leading to an even better latent heat transport due to the increased activity of the working fluid. For example, at a heat output of 59 W and in the on state, a thermal resistance as low as ca. 0.13 K/W was measured, which is on par with commercial heat pipes of the same dimensions. In conclusion, a switching factor of ca. 50 was demonstrated.

We emphasize that the demonstrator presented here and in references [19,32,33] only serves as a proof of concept, corresponding to a technology readiness level of 3–4 (similar to the other switchable heat pipe concepts as listed in Table 1). The technology has not yet been optimized or adapted for battery thermal management. The dimensions and shape of the thermal switch as well as other properties can and must be adapted for this and potential other applications. Furthermore, at a constant cold side temperature, the switching temperature is determined by the type and quantity of sorbent used and must also be adapted according to the target application.

2.2. Passive Thermal Management Concept Based on Thermal Switch

The innovative thermal management concept, which is theoretically investigated in this paper, is based on the presented heat-pipe-based thermal switch. Several of these thermal switches are placed between a passive cooling plate and the battery modules or cells, as can be seen in Figure 3. Thus, a passive control of the heat flux between battery and cooling plate is achieved. To meet the temperature requirements for LIBs, it is assumed that the switching temperature T_S is at about 25 °C. Currently, experimental and theoretical research to identify and validate an adsorbent suitable for this temperature range is under way, and potential candidates have been identified.

A good thermal coupling between heat source (battery cell) and adsorbent is important to reduce switching time to a minimum. Therefore, the adsorbent is placed directly at the battery module bottom. Furthermore, this ensures that the switching is dominated by the battery temperature and the impact of ambient temperature is limited. The heat pipe condenser is joined with the cooling plate. The cooling plate is designed as a sandwich structure with cooling fins between two aluminium plates. This allows the cooling plate to function additionally as underride protection. The integration of the underride protection has the advantage that aluminium, building space and weight are saved. Hence, it has a positive environmental and economic impact on the BTMS design.

The type, geometry and number of heat pipes to be used are dependent on the actual thermal management system design, heat pipe orientation, temperatures and the heating power to be dissipated. Possible implementations of a thermal management system using heat-pipe-based thermal switches are shown schematically in Figure 4. For the LCA in this work, it was assumed that standard tubular heat pipes with an outer diameter of 10 mm, a length of 320 mm, a wall thickness of 0.5 mm, a wick thickness of 0.9 mm and water as the working fluid are used and the heat pipes are oriented horizontally (Figure 3, right side). For these given boundary conditions, the maximum heat power that can be transferred before the heat transfer limits are reached can be estimated by considering the capillary, entrainment, sonic, viscous and boiling limit (see, e.g., [12,34]). The limit values also depend on the temperature. Around room temperature, the dominating limit is typically the capillary limit, limiting the maximum heat transfer per heat pipe to below 100 W. Considering these theoretical calculations and additional experiments, it was estimated that each heat pipe can transport up to 50–100 W of heat power. For the LCA in this work, it was calculated with 50 heat pipes, which results in a total cooling power of up to 5 kW. This is enough for a small EV to allow for multiple fast charging.

The cooling performance of the passive cooling plate depends on the airflow caused by driving. Therefore, it only offers a small amount of cooling while parked. Hence, the heat which is generated by charging needs to be buffered in the thermal capacity of the battery until driving. This is a common method and part of every BTMS design. The cooling power of up to 5 kW provided by the airflow while driving unloads the heat buffer. Thus, fast charging can be used at the next stop again and no rapid gate effect occurs.

3. Life Cycle Assessment of the Innovative BTMS and Comparison with the Active System

To quantify the environmental impact of the innovative BTMS with switchable heat pipes and to compare it to a conventional system depending on the air conditioning of the EV, a life cycle assessment (LCA) was conducted. LCA is a management tool for the evaluation and comparison of potential environmental impacts of a product or service analysed from raw material acquisition, through manufacturing, use and end of life phases. This technique consists of accounting for input flows (e.g., materials’ and energy consumption) and output flows (e.g., emissions to air, water and soil) over the entire life cycle of the product or service. LCA is standardised according to ISO 14040 [20] and 14044 [21].

3.1. Goal and Scope of the Case Study

The goal of this case study was the assessment of the environmental impacts of the innovative passive heat-pipe-based system developed by Fraunhofer CPM used for the cooling of batteries in EVs and its comparison with a conventional active cooling system. The intention is to show differences regarding the environmental impacts in various categories and to answer the question whether the innovative passive cooling system with heat pipes has the potential to be more environmentally friendly than the conventional one. This case study is addressed to scientists and engineers in the field of battery BTMSs to evaluate the possible implementation of the innovative cooling system.

For the active BTMS used for comparison, a minimal additional production effort for the battery cooling system is assumed. Hence, the production phase is only considered with an additional aluminium cooling plate in the LCA, which is sent to a recycling plant at the end-of-life. The production of the cooling system, consisting of water pump, compressor, cooling fan, valves, etc., was not taken into account for the LCA. Here, it was considered that the air conditioning system can be used for the battery cooling as well. There are two reasons for this minimal approach. First, the results show the minimum environmental benefit of the passive systems. Hence, the results of LCAs with real EVs would in most cases show an even greater ecological benefit. Second, the battery cooling system differs for every car. There is no standard and selecting a specific EV would compromise the generality of the results. Furthermore, there are no generic data available for the production of the components of the cooling system. However, the additional energy consumption of the cooling system in the use phase cannot be neglected and was therefore considered in the LCA. An overview of the active cooling system is shown in Figure 5 together with the corresponding system boundary.

The innovative passive system consists of a cooling plate, and several heat pipes exclusively responsible for the cooling of the battery, as described in more detail in Section 2.2. Small amounts of water and adsorption material TAPSO-34 [35] are filled into the heat pipes and electricity is required during vacuuming of the heat pipes and sintering of the internal wick. Due to the passive nature of the BTMS, no energy is needed in the use phase of the EV to cool the battery. Heating in very cold weather conditions was not considered, but it is the same for the passive and active system. The cooling plate and the heat pipes are sent to recycling plants in their end-of-life phase. An overview of the passive cooling system is shown in Figure 6 together with the corresponding system boundary.

In both cases, it was a cradle-to-cradle approach, i.e., from resource extraction to end-of-life of the materials. The production of the parts for the battery cooling, the energy needed in the use phase of the EV for battery cooling and the recycling of the components were considered. The metalworking from aluminium sheets and copper pipe to cooling plate and heat pipe was excluded from the LCA of the system. Components containing first and foremost metal (cooling plate and heat pipes) were recycled. The recycling rates are indicated in the assumptions for the respective system. The assembly of the sub-components and the installation of both systems in the vehicle were neglected.

The function of the system is the cooling of the battery of an EV of a compact class type. The functional unit is the mileage of the vehicle, which is assumed to be 150,000 km. The results are related to this functional unit, unless otherwise stated. Environmental Footprint Indicators (EF 3.1) were chosen to evaluate the environmental impact of both systems. EF is an initiative of the European Commission aimed at harmonizing the assessment of the environmental performance of products [36,37]. Focus was given to global warming potential (GWP), which captures emitted greenhouse gases and is a very commonly used indicator in LCA studies to quantify the impact of a product or process on the impact category climate change. Normalisation and weighting of the results were performed based on the EF 3.1 impact categories (EF 3.1 normalisation (person equivalents) and EF 3.1 weighting) [38,39].

The innovative passive system is a case study, as it is not yet used in EVs. Information about the input and output data needed for the life cycle inventory were provided by Fraunhofer CPM. Some data were calculated based on this information. The source for the data is given in

Table 2. Mass flows for the active and passive system for battery cooling. In both cases, the input and output flows are listed. The values are primary data (p) from Fraunhofer CPM or calculated (c) based on their information. The calculation of data is given in the appendix. The data refer to the functional unit (150,000 km mileage).

	Input (I)/Output (O) Flow	Value and Unit	Type of Data
Active battery cooling	Aluminium sheet for cooling plate (I)	15 kg	p
	Cooling plate for recycling (O)	15 kg	p
Passive battery cooling	Aluminium sheet for cooling plate (I)	11 kg	c (Appendix A)
	Copper pipe (I)	3.2 kg	p
	Water deionised (I)	0.2 kg	p
	TAPSO-34 (I)	1.2 kg	p
	Cooling plate for recycling (O)	10.8 kg	p
	Copper for recycling (O)	3.2 kg	p
	Disposal of filling of heat pipes (O)	1.4 kg	p

and

Table 3. Energy flows for the active and passive system for battery cooling. There are only energy input flows. The values for the active battery cooling were calculated (c) based on the assumptions described in the main text. The calculation of data is given in the appendix. The values for the passive battery cooling are primary data (p) from Fraunhofer CPM. The data refer to the functional unit (150,000 km mileage.).

	Input Flow	Value and Unit	Type of Data
Active battery cooling	Electrical energy for water pump	316 kWh	c (Appendix B)
	Electrical energy for air conditioning compressor	32.6 kWh	c (Appendix C)
	Electrical energy for cooling fan	2.7 kWh	c (Appendix D)
Passive battery cooling	Electrical energy for vacuuming of the heat pipe	20.8 kWh	p
	Electrical energy for sintering of the heat pipe	4.0 kWh	p

containing the mass and energy flows for the two systems and in the assumptions. The LCA was completed with the LCA for Experts software [40] together with the corresponding databases [41]. Data sets representing Germany were used.

Information about the exact energy demand for only cooling the battery with the active system was not available. Furthermore, the passive system for battery cooling is still part of research work, so the following assumptions were made together with the corresponding experts from Fraunhofer CPM:

valid for both systems

• The total mileage of the vehicle was conservatively assumed to be 150,000 km. Many manufacturers provide a warranty of 160,000 km or 8 years on the battery [42]. To show the influence of the mileage of the electric vehicle, a scenario was calculated representing 200,000 km instead of 150,000 km in the base case.
• The energy consumption of the EV while driving is 18 kWh/100 km [43]. So, for the total mileage, the energy demand is 27,000 kWh. The influence of the outside temperature and the state of charge of the battery were not considered.
• It is assumed that the EV is only very rarely fast charged. It is only performed in 1% of the cases. Fast charging (performed with an electrical power of 50 kW) is accompanied by cooling in the active system, whereas slow charging with 7 kW electrical power is not accompanied by cooling.
• An average speed of 60 km/h was assumed.
• The battery cooling system does not need to be replaced over the life of the vehicle.
• It was assumed that the weight of the cooling system is similar in both cases, and it has no impact on energy consumption. It was further assumed that aerodynamic effects are also similar and therefore negligible.

assumptions for the active systems

• The impact of the control system was considered negligible.
• The water pump runs continuously while driving and charging with 50 W.
• The air conditioning compressor switches on at an outside temperature of 20 °C in the case of fast charging and at an outside temperature of 30 °C in the case of driving mode. For fast charging, it needs a power system of 1 kW at 20 °C and 4 kW at 40 °C, and for driving mode it needs a power system of 1 kW at 30 °C and 2 kW at 40 °C. A linear increase in power demand based on the power system was assumed between the two points.
• The cooling fan is only needed for fast charging. Its power system is 500 W.
• The recycling rate for the cooling plate is 95%; 5% is disposed of.

assumptions for the passive systems

• The cooling plate was not designed in this investigation. Due to its dual use as cooling plate and underride protection, it is hard to exactly calculate the isolated impact of the cooling plate. Therefore, it was assumed that the cooling plate is made of 11 kg of aluminium.
• The recycling rate for the cooling plate is 95%; 5% is disposed of.
• The copper of the heat pipes can be recovered to 90%.

The mileage of the EV is one of the relevant parameters that have an impact on the results of the environmental assessment of the active and passive battery cooling systems. The effect of a mileage of 200,000 km compared to 150,000 km was assessed. The corresponding input flows are listed in

Table 4. Energy flows for the active system for battery cooling for a mileage of 150,000 and 200,000 km. The values are calculated according to the appendices given.

Input Flow	150,000 km	200,000 km	Appendix
Electrical energy for water pump	316.0 kWh	421.6 kWh	Appendix B
Electrical energy for air conditioning compressor	32.6 kWh	43.3 kWh	Appendix C
Electrical energy for cooling fan	2.7 kWh	3.6 kWh	Appendix D

.

3.2. LCA Results and Discussion

The active and passive system for battery cooling in an EV of the compact class were compared regarding their environmental impacts. The results for the impact category climate change are shown in Figure 7. It is obvious that the passive system for battery cooling has a significantly lower impact than the active system. In the case of a mileage of 150,000 km, the impact is more than eight times lower; in the case of 200,000 km, more than eleven times. The passive system with heat pipes benefits from the fact that it has no impact in the use phase. At the same time, the impact of the additional component is low compared to the cooling plate, at least in terms of climate change. From the very first kilometre driven, the passive system has an environmental advantage in the impact category climate change. And the higher the total mileage of the EV, the more the use phase comes into play where the passive system has no impact. Therefore, longer use phases increase the environmental advantage of the passive system.

In the production phase and in the end-of-life, the impact on climate change of the active and passive system is in a similar order of magnitude. The passive system has a slightly lower impact in the production phase but also has a lower benefit through recycling. Most of the impact comes from the cooling plate made from aluminium. It accounts for 86%, while only the remainder is attributable to the heat pipes. Even if it is considered that the recycling rate of the active system is 95% as assumed and the recycling rate of the passive system is 0%, the total impact of the passive cooling technology is about 33% lower than the active technology regarding climate change in the case of a mileage of 150,000 km.

The other EF 3.1 impact categories are shown in Figure 8 for the cases where the passive system has a lower impact and in Figure 9 for the only case where the passive system has a higher impact than the active system. In nearly all these impact categories the passive system has a lower impact than the active system. It accounts for only 3–43% of the comparative system with two exceptions: in the impact category, human toxicity non-cancer, the values for both systems are comparable; in the impact category resource use of minerals and metals, the impact of the passive system is 33 times higher. Even after normalisation (relating the results of the individual impact categories to a reference value) and weighting (converting the results of the individual impact categories using numerical factors) of the results, the impact of the passive system on resource use of minerals and metals is high: it is 1.2 × 10⁻³ compared to 3.5 × 10⁻⁵ for the active system with 150,000 km (Appendix E). The overall result for the EF 3.1 normalised and weighted is about four times lower for the passive system than for the active system with a mileage of 150,000 km. In terms of its overall effect, the passive system is therefore preferable from an environmental point of view.

The copper material used for the tube is responsible for the high impact of the passive system on resource use of minerals and metals. It was assumed that the recycling rate of the copper for the heat pipes is 90%. Improving this rate through an efficient recycling strategy could lower the impact not only for the resource use of minerals and metals. Furthermore, the active cooling system can require additional pumps, vents, fans, etc., that contain copper or other materials but have been neglected in the LCA. This course of action was taken not to present the passive system in an unjustifiably positive light, but rather to look at the worst-case scenario. Finally choosing another material for the heat pipe, for example aluminium, can reduce the impact in the category resource use of minerals and metals. But there might be negative effects on the overall results as well.

In the active system, the impacts in the single categories are mainly dominated by the cooling plate and the energy for the water pump. In the passive system, the cooling plate and the copper have the main impact in most of the impact categories. However, recycling can compensate the effort for the components of the two systems.

The LCA was based on many assumptions and some aspects had to be neglected due to a lack of information. For example, the weight of the cooling plate for passive battery cooling is one factor that could influence the results. The dimensions might be different for a real EV and therefore its environmental impacts. But as long as the recyclability is high, the overall impact remains low even if the cooling plate must be bigger than expected. The assumptions about the active BTMS were chosen conservatively. Aspects like larger or additional cooling components were not considered. Therefore, it is very likely that the benefit of the passive BTMS is even greater than the results of the LCA indicate, even for the aspect where the active system achieved comparable results (human toxicity non-cancer) and better results (resource use of minerals and metals).

4. Conclusions

In this paper, the environmental impacts of a novel passive and a conventional active battery thermal management system (BTMS) of electric vehicles (EVs) were investigated using a life cycle assessment (LCA). In preparation for the LCA, the new theoretical concept of the passive BTMS was first presented. The innovative aspect here is the use of passive switchable heat pipes, which enable the control of the heat flux between the battery cells and a passive fin-based cold plate. The heat switch is based on a programmable absorption material developed by the Fraunhofer Cluster of Excellence Programmable Materials (CPM). When the heat pipe is cold, the working fluid is bound and the conductivity is low. In the warm state, the working fluid is free, and the conductivity is high. The detailed concept and measurements are presented in the paper.

For the LCA, it was difficult to obtain reliable, accurate production data for the active BTMS. Therefore, the assumptions of the active BTMS were very conservative. Nevertheless, the LCA showed significantly lower environmental impacts of the novel passive BTMS with switchable heat pipes compared to the active BTMS in almost all environmental impact categories (EF 3.1). The overall impact after normalisation and weighting is about four times lower for the passive system than for the active system with a mileage of 150,000 km. Considering the conservative assumptions of the active BTMS, it is very likely that a more detailed LCA would show an even larger environmental benefit.

Although the cooling power is limited due to the passive nature of the BTMS, it is sufficient to enable fast charging and avoid temperature-related speed limitation for small to medium-sized EVs. On the other hand, price, environmental impact, energy consumption and thus travel range are advantages compared to EVs with active BTMSs. For these reasons, the novel passive cooling system based on switchable heat pipes is a promising technical and an environmentally friendly approach for use in compact class EVs. Nevertheless, it should be noted that the presented concept and LCA are theoretical, as an experimental study was not feasible due to budget and technological readiness. Therefore, further work should focus on improving heat pipe switches to a level where a battery module or system can be set up and experimentally validated. Improvements to the heat pipe switches should focus on the dimensions and shape of the heat pipes, their fabrication and the sorbent to be used. Research on BTMSs should focus on the cold plate design and a good coupling to the heat pipe as well as to the battery.