Introduction

The automotive industry is one of the biggest industries worldwide, as vehicles are indispensable for modern life. Worldwide automobile ownership exceeded 1.1 billion in 2010, of which the EU and USA accounted for 50 %, comprising 270 million and 240 million units, respectively [1]. It was also reported that vehicle ownership in China reached 140 million in 2012 [2]. In Japan, automobile ownership was 75 million [1]. Meanwhile next-generation vehicles such as hybrid electric vehicles (HEVs), plug-in hybrid electric vehicles (PHEVs), and electric vehicles (EVs) have been developed and rapidly introduced, and more than 7.5 million HEVs had been sold worldwide as of 2013. In Japan, the ownership of next-generation vehicles reached 2.9 million in 2012 [3].

On the other hand, global generation of end-of-life vehicles (ELVs) was estimated at 40 million in 2010, which accounts for 4 % of total automobile ownership [4]. Increasing attention is given to the management of end-of-life (EoL) HEVs, because approximately two decades have passed since they were first introduced to the market. The Ministry of Economy, Trade and Industry, Japan (METI) forecast that the numbers of EoL HEVs in Japan would reach approximately 0.1 million in fiscal year (FY) 2020, and 0.7 million as of FY2025 [5].

Critical metals, particularly rare earth elements (REEs), are essential constituents of many industrial products, including some vehicle components. Hoenderdaal et al. [6] indicated that, even with their limited market share, electric vehicles (HEVs, PHEVs, EVs, and electric bikes) already accounted for 23 % of worldwide demand for Dy in 2011. The authors also forecast that electric vehicles would account for 65–78 % of total Dy demand in 2050. Seo et al. [7] estimated the Dy demand for products, including vehicles, in Japan until 2030, and showed that Dy demand for driving motors in HEVs would account for 56 % of 1700 t-Dy, the total demand in 2030.

From the viewpoints of resource security and stable supply, it is important to recycle any critical metals, including REEs, that are contained in a waste stream [8]. Global in-use stock (not limited to the automobile industry) of REEs was estimated to be 448,000 tons, including 144,000 tons of Nd and 8600 tons of Dy as of 2007 [9]. Seo et al. [7] indicated that total Dy demand in Japan could be reduced from 1700 tons to 1570 tons by promoting recycling of products that contained NdFeB magnets. Sprecher et al. [10] estimated the potential for recycling Nd from computer hard disk drives. As mentioned previously, a HEV would be one of the largest consumers of REEs, and hence represents the greatest potential for REE recovery in future. REEs are utilized in HEV-specific components, particularly hybrid transmission and secondary batteries. Alonso et al. quantified the REE contents in conventional and HEVs [11]. They showed that a HEV contained a total of 3.5 kg REEs in NiMH (nickel metal hydride) battery and 0.6 kg in hybrid transmission (driving motor and generator), with only 0.4 kg in the remaining components. Both forecasting future demand and clarifying REE recovery potentials from EoL HEVs are useful for considering future ELV management.

The purpose of this study is to clarify the present and future recovery potential of REEs that are disposed of as EoL HEVs. Hybrid transmission and NiMH battery units were considered as HEV-specific components in this study. Other components containing REEs—such as motor magnets in air conditioning compressor, power steering, and speaker magnets—were excluded, because they are not exclusive to HEVs but are also used in conventional vehicles. Therefore, evaluation of the potential to recover REEs from these components should include analysis of conventional vehicles. As clarified by Alonso et al. [11], HEV-specific components were the dominant source of REE consumption. This study focuses on the increasing consumption of REEs, and their potential recovery derived from the future proliferation of HEVs. In this study, REE recovery potential was defined as the amounts of REEs contained in EoL HEV-specific components: the maximum amounts potentially recovered if all EoL components were successfully collected and 100 % of REEs recovered from the components.

Materials and methods

Firstly, the number of EoL HEVs that will be disposed of during FYs 2010–2030 was estimated. Then, the REE contents of the two HEV-specific components were analyzed. Finally, the amounts of REEs contained in EoL HEV-specific components and HEVs in use were estimated as REE recovery potential and domestic stock, respectively.

Estimation of the number of end-of-life HEVs

Number of first-registered HEVs until 2030

Hybrid electric vehicles were considered as HEVs in this study: PHEVs were excluded owing to their short history. Exported HEVs were not included, and only domestic HEVs were considered. The number of EoL HEVs in Japan between 2010 and 2030 was first estimated using the population balance model, which is a dynamic estimation model based on the mass balance between input, stock, and output of materials or products that have a defined lifespan. In this study, the lifespan of an HEV was defined as the period from the fiscal year in which it was first registered in Japan to the fiscal year in which its registration was finally canceled. Namely, the period of use by subsequent owners was also included in the lifespan.

The annual number of HEVs first registered in FY1997–2012 was derived from statistical data reported by the Automobile Inspection and Registration Information Association (AIRIA) [12] and the Next-generation Vehicle promotion center (NeV) [3]. The prospective number of HEVs in FY2013–2020 was estimated using the number of vehicles predicted to be registered during the period [13] and the share of HEVs among all types of vehicle [14]. Based on the target introduced by METI [14], the ratio of the annual number of first-registered HEVs to all types of first-registered vehicles is assumed to reach 20–30 % (representing the minimum (Min.) and maximum (Max.) cases) as of FY2020, and 30–40 % as of FY2030. This follows a linear increase from FY2012 (16.5 %) onward, which is shown in the electronic supplemental materials. Figure 1 shows the assumed number of HEVs first registered in FY1997–2030.

Fig. 1
figure 1

Number of HEVs first registered in FY1997–2030

Full size image

Remaining rate of HEVs

To estimate the number of HEVs in use, the lifespan distribution with respect to the first registration year was considered as a Weibull distribution, which has frequently been used in previous studies [15–17]. The remaining rate function is shown in Eq. 1.

$$ F(y, t) = exp\left\{ { - \left( {\frac{t + 0.5}{\eta }} \right)^{m} } \right\} $$
(1)

\( F(t) \) is the remaining rate of HEVs first registered in FY y, \( {\text{y}} \) is the first registered year (=FY1997, 1998, …, 2030), \( t \) is the Vehicle age, \( m \) is the Shape parameter, \( \eta \) is the scale parameter.

It was assumed that the lifespans of HEVs varied by their first registration years. The shape parameter m and scale parameter η were estimated by the least squares method using statistical data. Compared with conventional vehicles, statistical data on HEVs are limited because they have a shorter history. Therefore, statistical data were used for the Prius model produced by the Toyota Motor Corporation, which is the most common and longest-established HEV in Japan [12]. Table 1 shows the estimated shape and scale parameters.

Table 1 Estimated shape and scale parameters of the remaining rate function
Full size table

HEVs have undergone several technological upgrades, categorized as three generations: first-generation HEVs, produced in FY1997–2002; second-generation HEVs, produced in FY2003–2008; and third generation and onward HEVs, produced after 2009. Statistical data from AIRIA indicate that first-generation HEVs tended to have a shorter lifespan than the second generation. Figure 2 shows estimated remaining curves obtained using averaged statistical data for first-generation (and averaged statistical data for FY2003–2005 for second generation) HEVs; data for FY2006–2008 was excluded owing to lack of data. The average duration of use (vehicle age) at a remaining rate of 50 % among second-generation HEVs was longer than that among first-generation HEVs: 20.2 years among the second generation, and 14.5 years among the first generation. Although it is difficult to clearly identify the reason for this, it may be attributed to a change in vehicle size (from small-sized first-generation to standard-sized second-generation vehicles), the extended duration of use period by second users, and user awareness. The parameters m and η obtained from averaged data between FY2003 and FY2005 were applied to FY2006–2030, assuming that this longer lifespan would be inherited by future HEVs.

Fig. 2
figure 2

Estimated remaining rate curve of HEVs

Full size image

Number of end-of-life HEVs and hybrid transmissions

Using the number of HEVs first registered during FY1997–2030 and every remaining rate as explained previously, the numbers of EoL HEVs during FY2010–2030 were estimated as described in Eqs. 2–4. It was assumed that hybrid transmissions only required disposal at the end of the vehicle’s life: no hybrid transmissions reached EoL or required replacement during the lifespan of the HEV.

$$ N_{\text{use}} (y,\; t) = N_{{1{\text{st}}\_{\text{rgstd}}}} (y, \;0) \times F(y, \;t) $$
(2)
$$ N_{\text{ELV}} (y,\;t) = N_{\text{use}} (y,\;t) - N_{\text{use}} (y,\;t + 1) $$
(3)
$$ {\text{Total}}\_N_{\text{ELV}} (Y) = \mathop \sum \limits_{{y = {\text{FY}}1997}}^{Y} N_{\text{ELV}} (y,\; t = Y - y) $$
(4)

\( Y \): Counted fiscal year (=FY2010, 2011,…, 2030). \( N_{\text{use}} (y, \;t) \): Annual number of HEVs in use that were first registered in FY y of vehicle age t. \( N_{{ 1 {\text{st}}\_{\text{rgstd}}}} (y, \;0) \): Annual number of HEVs that were first registered in FY y (\( = N_{\text{use}} (y, \;0) \)). \( N_{\text{ELV}} (y,t) \): Annual number of EoL HEVs that were first registered in FY y of vehicle age t. \( {\text{Total}}\_N_{\text{ELV}} (Y) \): Annual total number of EoL HEVs in FY Y.

Number of end-of-life NiMH battery units

As of 2013, most HEVs have a NiMH battery as their secondary battery, whereas most EVs have a Li-ion (lithium-ion) battery. In this study, all HEVs until FY2030 were assumed to have a NiMH battery because the future share of NiMH and Li-ion batteries for HEVs was difficult to forecast.

The lifespan of NiMH battery units is shorter than that of HEVs, and therefore the units must be replaced several times during the vehicle’s lifetime, such as during automobile inspections. Therefore, it should be noted that EoL NiMH battery units were disposed of during the HEV lifespan and also when the EoL HEV was disposed of.

An internet questionnaire survey of HEV owners was used to estimate the replacement probability for NiMH battery units. The questionnaire survey was conducted during March 2013, resulting in 604 valid responses. Based on the results, it was assumed that NiMH batteries could be replaced a maximum of two times during the HEVs lifespan. The replacement probability was assumed to be the same for all generations of HEVs using NiMH batteries. The replacement probability was estimated using the maximum likelihood method, as shown in Eqs. 5–9. Replacement probabilities \( P_{1} (y, t) \) and \( P_{2} (y, \;t) \) were assumed to be independent of each other and identically distributed. Figure 3 shows the estimated replacement probabilities of EoL NiMH battery units. The estimated shape parameter \( m' \) and scale parameter \( \eta^{'} \) were 1.95 and 15.6, respectively. When the HEV age reaches 12.9 years, the replacement probability (at least once) for the NiMH battery unit is 50 %.

Fig. 3
figure 3

Estimated replacement probability (P 1 and P 2 ) of EoL NiMH battery units

Full size image
$$ M_{\text{EoL}} (y,\; t) = N_{\text{ELV}} (y, \;t) + RB_{1} (y, t) + RB_{2} (y, \;t) $$
(5)
$$ RB_{1} (y, \;t) = N_{\text{use}} (y,\; t) \times \left( {P_{1} (t) - P_{1} (t - 1)} \right) $$
(6)
$$ RB_{2} (y, t) = N_{\text{use}} (y, t) \times \left( {P_{2} (t) - P_{2} (t - 1)} \right) $$
(7)
$$ P_{1} \left( t \right) = 1 - { \exp }\left\{ { - \left( {\frac{t + 0.5}{{\eta^{'} }}} \right)^{m'} } \right\} $$
(8)
$$ P_{2} \left( t \right) = 1 - \mathop \smallint \limits_{0}^{\infty } P_{1} (t - t') \times p_{1} \left( {t^{'} } \right)dt' $$
(9)

\( M_{\text{EoL}} (y, t) \) is the Annual number of EoL NiMH battery units from HEVs first registered in FY y of vehicle age t. \( RB_{1} (y, \;t) \), \( RB_{2} (y, \;t) \) is the Annual number of EoL NiMH battery units from HEVs first registered in FY y of vehicle age t by replacement (\( RB_{1} \) is the first time, \( RB_{2} \) is the second time). \( P_{1} (t) \) is the Replacement probability for NiMH battery unit equipped with HEV of vehicle age t (at least once). \( P_{2} (t) \) is the Replacement probability for NiMH battery unit equipped with HEV of vehicle age t (twice). \( p_{1} (t) \) is the Probability density function of \( P_{1} (t') \). \( t' \) is the Vehicle age when NiMH battery unit was replaced first time.

Chemical analysis of REE contents

Motor magnets in hybrid transmission

A hybrid transmission comprises two motors: the driving motor and the generator. These motors were collected during dismantling surveys conducted in November 2012 [18] and November 2013. The hybrid transmissions of three generations, produced in 1997 (first generation), 2003 (second generation), and 2009 (third generation), were collected as samples. Then demagnetization was conducted at 500 °C and motor magnets could be manually separated. After powdering treatment, the powdered motor magnets were dissolved using H2SO4 and HNO3. Alkali fusion was not needed because no residues remained after acid dissolution. Finally, the solutions underwent chemical analysis by ICP-MS. The REE contents were then quantified by multiplying the REE concentration by the weight of the motor magnets.

Battery cells in NiMH battery unit

It was indicated that secondary battery cells contained REEs as misch metal in the negative electrodes [11]. The dismantling surveys conducted in November 2012 collected a NiMH unit used in first-generation HEVs produced in 1997 [18]. That battery unit comprised 252 battery cells weighing 43.4 kg. The elements contained in one of the battery cells were quantified through chemical analysis by ICP-MS in a similar way to motor magnets in hybrid transmission. The REE contents were then quantified by multiplying the REE concentrations by the weight of the battery cells.

It was assumed that the REE concentrations per cell did not differ between generations. However, there is a trend for miniaturization of HEV battery units through successive generations. The NiMH battery unit produced in 1997 contained 42 modules, compared with 38 modules in 2003 (second generation) and 28 modules in 2009 (third generation). Assuming that a module comprised 6 cells, the first-, second-, and third-generation units were assumed to contain 252, 228, and 168 cells, respectively. The weight of the NiMH battery unit was assumed to be proportional to the number of cells, giving a reduction of 9.5 % in the second generation and 33.3 % in generations three and later, compared with the first generation.

Estimation of REE recovery potentials from end-of-life HEV-specific components

Using the number of EoL HEVs and NiMH battery units, and the REE contents of a hybrid transmission and a NiMH battery unit, the annual amounts of REEs from EoL HEVs during FY2010–2030 were estimated as recovery potential from the downstream automobile flow. In similar fashion, annual demand for REEs was also estimated for new HEVs produced during the year, and for the stock of HEVs already in use.

Results and discussion

Estimated end-of-life HEVs

As shown in Fig. 4a, the number of HEVs in use in Japan (domestic stock) is estimated to increase by 13–17 times, from 1.4 million in FY2010 to 18–24 million in FY2030. Considering that total automobile ownership in Japan is presently 75 million [4], HEVs in use would account for 23.5–31.5 % of domestic automobile ownership in FY2030.

Fig. 4
figure 4

Estimated a HEVs in use and b end-of-life HEVs generated in FY2010–2030

Full size image

The estimated number of EoL HEVs that will be generated during FY2010–2030 is shown in Fig. 4b. As of FY2030, 0.51–0.65 million HEVs will have reached the EoL stage; only 11,000 HEVs had reached the EoL stage as of FY2010. Compared with the forecast by METI [5], the present study estimates a much lower number of EoL HEVs, which may be attributed in part to the longer assumed lifespan.

Chemical analysis of REE contents

The analysis detected Pr, Nd, and Dy REEs in the motor magnets used in the hybrid transmission. As shown in Table 2, the Pr content of a hybrid transmission produced in 1997 was 71 g, while in 2009 it was 1.0 g. The Nd contents of a hybrid transmission were not so different between generations. On the other hand, the Dy content dramatically decreased between successive generations. Total REE content decreased from 490 to 340 g (31.1 % reduction) across the two model changes between 1997 and 2009. Alonso et al. [11] also showed a similar result: REE content in the hybrid transmission (motor/generator) was 0.6 kg.

Table 2 REE contents in hybrid transmissions produced in 1997, 2003, and 2009 [g/component]
Full size table

As seen in Table 3, a NiMH battery unit produced in 1997 contained a variety of REEs, such as 1200 g Ce and 870 g La, totaling approximately 2.6 kg of REEs. This amount was less than the 3.5 kg of REEs reported by Alonso et al. [11]; however, if considering other components of the NiMH unit, such as the printed circuit board, this gap would be reduced.

Table 3 REE contents of a NiMH battery unit produced in 1997
Full size table

Estimation of REE recovery potentials from end-of-life HEV components

Hybrid transmission

This section presents and discusses the results based on the maximum scenario. The results for the minimum scenario are shown in the electronic supplemental materials. Figure 5 shows the estimated REE demands and recovery potentials of hybrid transmissions in HEVs during FY2010–2030. In Fig. 5, “demand” refers only to the annual demand for REEs in hybrid transmissions of newly registered HEVs: Demands for other components or other products (non-HEVs) were not included. Demand increased from approximately 150 tons (0.4 tons Pr, 120 tons Nd, and 31 tons Dy) as of FY2010, to 460 tons in FY2020 and 620 tons in FY2030. REE recovery potential also increased, from 5.0 tons (0.6 tons Pr, 3.1 tons Nd, and 1.3 tons Dy) as of FY2010, to 62 tons (1.4 tons Pr, 48 tons Nd, and 13 tons Dy) as of FY2020, and 220 tons (1.5 tons Pr, 170 tons Nd, and 45 tons Dy) as of FY2030. In other words, REE recovery potential equaled 35.4 % of the REE demand for new HEV transmissions as of FY2030 if all EoL HEV hybrid transmissions were collected.

Fig. 5
figure 5

Estimated REE demand and recovery potentials for hybrid transmissions for FY2010–2030 (Max. case) Rec. recovery potential; Demand annual demand for REEs in hybrid transmissions of newly registered HEVs

Full size image

Table 4 compares the estimated domestic demand and recovery potentials of Nd and Dy reported in previous studies. Seo et al. [7] estimated that domestic Dy demand for HEV and EV use would reach approximately 510 tons as of 2020, and 950 tons as of 2030. The estimates in the present study are lower because Seo et al. [7] assumed 70–160 g of Dy per HEV/EV, whereas this study assumed 68 g for vehicles of the third and subsequent generations, based on chemical analysis. Furthermore, this study focused only on HEVs, whereas Seo et al. [7] included EVs. They also considered the REE demand for exported vehicles. With regard to recovery potentials, METI estimated that Nd and Dy recovery potentials from next-generation vehicles would reach approximately 30 tons and 7 tons as of 2020, and 250 tons and 42 tons as of 2025. The estimated potentials as of 2025 in this study were approximately half those reported by METI, possibly because this study also estimated half the number of EoL HEVs owing to their longer lifespan.

Table 4 REE demand for automobile industry including HEVs in Japan: comparison with previous studies
Full size table

NiMH battery unit

Figure 6 shows estimated REE demand and recovery potentials of NiMH battery units during FY2010–2030 (Max. case). The REE demand of approximately 780 tons as of FY2010 increased to 2400 tons in FY2020, and 3200 tons in FY2030. REE recovery potential also increased, from 73 tons as of FY2010, to 970 tons as of FY2020, and 2900 tons as of FY2030. REE recovery potential increased more rapidly than demand because EoL battery units were generated from not only EoL HEVs but also by replacement during the HEV lifespan. REE recovery potential equaled 92.1 % of REE demand for NiMH battery units in new HEVs as of FY2030 if all EoL NiMH battery units were collected.

Fig. 6
figure 6

Estimated REE demand and recovery potentials for NiMH battery units during FY2010–2030 (Max. case) Rec. recovery potential. Demand annual demand for REEs for NiMH batteries in newly registered HEVs

Full size image

The REE recovery potentials of each source in Fig. 7 indicate that the REE recovery potential from replacement NiMH battery units exceeded those from the other sources, namely NiMH battery units and motor magnets from EoL HEVs. This was because the number of HEVs in use is at least ten times greater than the number of EoL HEVs. This finding suggests that it is important to develop a NiMH battery-recycling scheme that considers NiMH battery units at both the replacement and EoL stages.

Fig. 7
figure 7

REE recovery potentials of each disposed source (Max. case)

Full size image

Table 5 shows the domestic REE stock as HEVs in use. The stock increased 11- to 15-fold from FY2010 to FY2030, and was roughly proportional to the number of HEVs in use, although REE content per component tended to decrease through successive generations of HEVs. Table 5 also shows that an estimated 37,000–49,000 tons of REE stock in FY2030 would be disposed of after FY2030.

Table 5 Domestic REE stocks as HEVs in use during FY2010–2030
Full size table

It should be noted that the intended functions of REEs differ between motor magnets in a hybrid transmission and NiMH battery cells. For instance, Dy in a Nd magnet reinforces coercive force and heat resistance, whereas REEs contained in the negative electrodes of a NiMH battery play a role in hydrogen storage. Therefore, REE recovery potentials from the two components were not comparable: even if the potential to recover REEs from NiMH battery cells exceeds that of motor magnets, it is important to promote the collection and recycling of both components. A more desirable approach would entail developing recycling loops such as “motor magnets to motor magnet,” and “NiMH battery cell to NiMH battery cell.”

Uncertainties

There were some uncertainties derived from the assumptions employed in this study. For instance, the lifespan distribution of second-generation (average over FY2003–2005) vehicles was passed on to the third generation and onward HEVs. Figure 8 shows the estimated proportion of EoL HEVs of each generation. Although third-generation HEVs were assumed to have longer lifespan, they dominated the total EoL HEVs and accounted for 89.1 % in FY2020 and 97.8 % in FY2030. This was because the number of newly registered HEVs rapidly increased after FY2009, as shown in Fig. 1.

Fig. 8
figure 8

Proportion of end-of-life HEVs of each generation (Max. case)

Full size image

The proportions of recoverable REEs reflect the proportions of EoL HEVs from each generation of hybrid transmission, as shown in Fig. 9; the exception was Pr, which is minimally present in third generation and onward vehicles. A similar tendency was found for NiMH battery units, as shown in the electronic supplemental materials. Therefore, it is concluded that promoting miniaturization and alternative technologies that reduce the consumption of REEs for the hybrid transmission could greatly contribute to future REE recovery potentials, even if HEVs achieve longer lifespans. Considering the fact that HEV technologies have changed approximately every 6 years, there would be three model changes and opportunities to adopt such developments between FY2014 (present) and FY2030.

Fig. 9
figure 9

Proportions of REEs recoverable from each generation of hybrid transmission (Max. case)

Full size image

It was assumed that the REE composition of NiMH battery cells remained constant until FY2030. The REE composition of motor magnets in the hybrid transmission was also assumed to be constant after the third generation. Other types of secondary battery, such as Li-ion, might be utilized in future HEVs. Such technical developments would affect the REE recovery potentials.

It is also to be noted that REE demand and recovery potentials from other types of next-generation vehicles were not considered in this study. Furthermore, the collection efficiency within the EoL HEV management stream, and recovery efficiency during the recycling processes, should also be considered when estimating the recovery amounts of REEs.

Indicators of resource recovery efficiency from products are useful for developing sustainable systems. In addition to REE recovery potentials, this study showed the REE recovery potentials as a proportion of total demand for each year (see Figs. 5 and 6). The findings indicate the amounts of REEs that might be recovered from EoL HEV-specific components to thereby offset the demand for domestic production. However, it should be noted that most REE stocks remain in the form of vehicles in use (See Table 5). When evaluating recovery efficiency, it is also important to consider the product lifetime (time lag). For instance, what percentage of HEV-specific components of the first-generation HEVs was finally collected relative to the total amounts used? Future management of ELVs requires an evaluation system that combines several indicators.

In addition to REEs, HEVs contain other critical metals. For instance, NiMH battery cells contain Co and Li. A recycling scheme is therefore required for recovery of critical metals including REEs. In addition, common components in both HEVs and conventional vehicles also contain critical metals such as alloys, printed circuit boards, and catalytic converters. Vehicles also contain hazardous substances such as Pb and Hg. From the perspectives of both resource value and minimizing the risk of environmental pollution, robust systems are required to manage ELVs, including next-generation vehicles.

Conclusion

This study first estimated the numbers of EoL HEVs during FY2010–2030, and then clarified the potential for recovery of REEs from two HEV-specific components—the hybrid transmission and the NiMH battery unit. Motor magnets in the hybrid transmission and NiMH battery cells were considered as the parts containing REEs. The results suggest that 0.51–0.65 million HEVs will reach the EoL stage in FY2030, whereas only 11,000 HEVs reached the EoL and disposal stage as of FY2010. REE recovery potentials will increase from FY2010 to FY2030, from 5.0 tons to 220 tons for EoL hybrid transmission, and from 73 tons to 2900 tons for NiMH battery units. As of FY2030, a total of 49,000 tons of REEs will be contained within HEV-specific components of HEVs still in use. Moreover, as of FY2030, the potential for recovery of REEs from EoL hybrid transmissions and NiMH battery units is estimated to equal 35.4 and 92.1 % of respective demand.

This estimation only considered REEs derived from HEV-specific components. Conventional vehicles also contain critical metals; however, such vehicles contain hazardous substances. From the perspectives of both resource value and managing the risk of environmental pollution, it is therefore necessary to develop an appropriate system for managing ELVs, including next-generation vehicles.