The described analysis does not take into account the potential future changes in the processing of end-of-life cars; as such, disassembly before shredding will not be discussed in detail.
Using minimum and maximum values for the Nd content in shredder feed (light duty vehicle/LDVs, household appliances, other sources) as described by us previously [10], two different scenarios were considered in our forecast for the next 20 years: (1) Hybrid vehicle (HEVs) and electric vehicles (EVs) are not shredded with conventional LDVs and household appliances and (2) HEVs and EVs are shredded with conventional LDVs and household appliances. This distinction is important to arrive at accurate conclusions due to the much higher Nd content of HEVs and EVs when compared to conventional vehicles [11].
Several significant variables that affect the final Nd content of these vehicles (Nd content in conventional vehicles; Nd content in HEVs and EVs; market penetration of HEVs and EVs; and Nd content in household appliances) have been approximated as detailed below for the years 2019, 2024, 2029, and 2034 (see SI for detailed calculations).
The calculated average weight of conventional LDVs produced between 2004 and 2013 is 1834 kg [12]. This average weight was used in calculations for the weight of LDVs and was assumed to remain constant between 2014 and 2034.
The weight of Nd in conventional LDVs over time has been considered using two scenarios (a) the Nd content remains constant at 303 g per LDV [10] after 2013 due to a continuous high price of Nd; or (b) the Nd content increases linearly until 2034, continuing the trend before 2014 [10]. These two approximations are illustrated in Fig. 1. The Nd content in conventional LDVs further depends on the age distribution of shredded LDVs, which have been approximated using previously documented age distributions of cars shredded in 1995 and 2007 [10].
Fig. 1Minimum (Nd content constant; red) and maximum (Nd content increases linearly; blue) assumptions used to calculate the total weight of Nd in conventional LDVs
Full size image
The weight of HEVs/EVs and the weight of Nd in HEVs/EVs need to be approximated in order to arrive at the Nd content of ferrous scrap from HEVs/EVs. The weight of HEVs was assumed to be equal to the average weight (1483 kg) of the four highest selling HEV models in the U.S. (Toyota Prius, Honda Civic, Ford Focus and Chevy Malibu) [13]; similarly, the weight of EVs was assumed to be equal to the average weight of 1570 kg. These average weights of HEVs and EVs as well as the amount of Nd (683 g in HEVs; 756 g in EVs)2 after removal of the NiMH battery [14] are further postulated to remain constant until 2034 for two major reasons: First, the calculated average weights of the four most common models on the American market produced between 2009 and 2014 remained constant [15]; thus we assume a continuation of this trend. Second, even if lighter vehicles will be produced in the future, it is unlikely that such a weight reduction stems from reduction of functional, moving parts containing motors and Nd magnets. Thus, the overall Nd content of a vehicle per kg ferrous scrap might actually increase; therefore, our assumption of no change in vehicle weight and Nd content is indeed a conservative approximation.
Various forecasts of EV and HEV market penetration have been made and are illustrated in Fig. 2 [16, 17]. The lowest market share of 8 % HEVs/EVs in 2034 is predicted by the U.S. Energy Information Administration [16], which provides several reasons for its forecast. First, HEVs/EVs have higher purchase prices than conventional LDVs and are more fuel efficient during the time of ownership. Thus, more fuel-efficient conventional LDVs, which are currently developed, and an expected rising price for electricity will result in consumers deciding against the purchase of HEVs/EVs. Second, battery technology is not expected to improve substantially in the near future with respect to battery cost, safety, and performance; all these elements would favor purchases of conventional vehicles.
Fig. 2a Minimum and b maximum market penetration of HEVs and EVs
Full size image
In contrast, the highest market penetration of 81 % HEVs/EVs in 2034 is predicted by Breker and coworkers [17]. Breker's forecast considers the various beneficial impacts of EVs on the economy (lower oil imports, improved trade deficit, more jobs in battery industry, healthcare cost savings due to less pollution). Since this approach only predicts market shares until 2030, market shares were linearly extrapolated until 2034. Figure 2 graphically illustrates these minimum and maximum market penetrations of HEVs and EVs, which were used for our calculations.
As described in detail elsewhere [10], the Nd content in all LDVs (conventional LDVs, HEVs, and EVs) was calculated according to Eq. (1), using the variables x v,i , the weight of Nd in ferrous scrap from shredding type v LDVs manufactured in year i, w v,i , the percentage of type v LDVs manufactured in year i, s v,i , the market penetration of type v LDVs manufactured in year i, and \( \bar{x} \), the average Nd content in ferrous scrap.
$$ \bar{x} = \frac{{\mathop \sum \nolimits w_{v,i} x_{v,i} s_{v,i} }}{{\mathop \sum \nolimits w_{i} }} $$
(1)
w v,i and x v,i were calculated as described previously [10]; s v,i is the market penetration described above. All values for x v,i , w v,i , s v,i , and \( \bar{x} \) have been tabulated for the years 2014, 2019, 2024, 2029, and 2034 in the SI.
The Nd content in household appliances was assumed to increase or stay constant at the same rate with which the Nd content in conventional LDVs increases or stays constant as assumed suggested by a prior analysis (see SI for detailed calculations) [18]. Calculations were performed using 0.70 g Nd/kg ferrous as the maximum Nd content (as reported for U.K. appliances) [10, 19] and 0.61 g Nd/kg ferrous as the minimum Nd content (adjusted for major appliances) [10, 20].
Initially, we surveyed all metals for which secondary production data are available through the United States Geological Survey (USGS) [21]. These data are exclusively U.S. data, meaning that only U.S. prices and secondary production in the U.S. are analyzed. Our analysis focused on inflation-adjusted prices, as these prices are a better measure of real value than raw price data. The further below illustrated rate of secondary production for the analyzed metals is reported in percentages as the ratio of secondary production divided by the apparent consumption in the U.S. for the same year. Secondary production is defined differently for each metal [21], but in general is the creation of new metal from metal scrap. Apparent consumption is a calculation that takes into account primary and secondary production, imports, and exports. The apparent consumption totals are also calculated by the USGS. We treated secondary production rates as analogous to recycling rates, because secondary production rates compare the amount of new metal created from recycled scrap to the total amount used in the United States. Secondary production and apparent consumption data are both reported in metric tons per year and the axes scales are chosen to suit data and to align the zero points for both graphs; maxima of secondary production rate and inflation-adjusted price in the relevant time period are set to 100 %.
Our study aims to examine the effect of one economic system’s response to recycling in relation to its total usage of a material. While prices may be radically different in other markets, we aim to investigate the response of US markets to US recycling rates. Although our study will only analyze the recycling rate in the US, we postulate that the response would be analogous to the way other free markets could react to a locally changing recycling rate.
Our analyses of secondary production and price development focus on the time after 1939. We postulate, based on historical events (such as the first World War and the Great Depression), that the period of price-control during WWII (1939–1943) [22] can be used as a starting point for later price development in a free market, which is the focus of our analyses. This restriction focuses our analyses on modern and currently relevant economic conditions.
Next, a criticality analysis was performed for each of the 16 metals with available data for secondary production, inflation-adjusted prices, and apparent consumption [21]. We did not consider data available for iron and steel as iron is not rare per se (iron is the fourth most abundant element in the earth’s crust) [23]. Steel was not considered as it consists of a mixture of elements [24]; thus its price can depend on the criticality of each of its components.
The elements of the performed analyses were adopted from the U.S. Department of Energy Critical Materials Strategy (see SI) [6]. This analysis was necessary since the availability of data on secondary consumption is not a factor in previous studies. Based on this analysis, we concluded that Co, Zn, Sb, Ta, W, and platinum group metals (PGMs, including Pt, Pd, Ir, Os, Rh, and Ru) [21] could be considered to be critical materials, while Mg, Al, Cr, Ni, Cu, Ag, Sn, Au, Hg, and Pb were not deemed critical. In order to further elucidate the influence of recycling on price development, we tabulated historical events and the resulting change in price for all 16 metals (see SI). This was necessary as, to the best of our knowledge, other comparative and comprehensive analyses of a series of metals are not available in the prior literature, while more focused analyses (e.g., for Co and REs) are well documented [6, 25]
In our analyses of price volatility as discussed in the next sections, low price volatility was defined as a change in price of less than 25 % of the overall maximum price during a decade. High price volatility was defined to exist when the price changes were larger than 25 % of the overall maximum, inflation-adjusted price. As an example, Cobalt (Co) prices change between 20 and 100 % of the maximum inflation-adjusted historical price during the 10 years between 1973 and 1983; during the two decades prior to 1973, the price of Co changes only between 14 and 28 % of the maximum inflation-adjusted historical price [26]. The first scenario (1973–1983) is thus a period with high price volatility (and low price stabilization), while the latter examples (1953–1963 and 1963–1973) represent decades of low price volatility.
In order to elucidate how recycling rates influence the price of critical materials, we first analyzed the available literature on this topic, which is surprisingly limited. Kirchain and coworkers found through simulations that price volatility is more pronounced when recycling rates are low (25 % and lower) [27]. High recycling rates, in contrast, contribute to price stabilization [28]. Despite this prior computational study, the conclusion has not been verified using historical price and recycling rates. Therefore, we decided to analyze historical data for critical metals with a special consideration for the influence of recycling rates on price stability.
If you have any questions on Scrap Shredder. We will give the professional answers to your questions.