Recycling of metals in solid household waste is an integral part of sustainable waste management in Europe. The EU member states are required to establish collection schemes for different recyclable waste fractions including metal by 2015, and reuse or recycle 50% of the recyclable materials from households by 2020. The EU Packaging Directive 94/62/EC [1] requires that the metal fraction of household waste must have a recycling ratio of at least 50% by the end of 2008. In addition, by the end of 2008 packaging waste was to be reduced by 5% relative to the Gross Domestic Product (GDP).

In 2007, the waste management system of the Helsinki metropolitan area covered the cities of Helsinki, Espoo, Vantaa and Kauniainen, consisting of households with all together around 906,000 inhabitants [2]. In this area the collection of metal scrap is arranged according to the unit size of the metal product. Within this area, the large metal scrap fraction and the small sized metal fraction of municipal solid waste (MSW) have separate collection systems. The small size metal fraction can contain tins and beverage cans, aluminum cooking moulds and foils, metal lids and caps, metal cups for tea-light candles, empty and dry paint tins and empty aerosol bottles [3]. Large metal scrap often contains items such as bicycles and gutters, metal parts of furniture, metal tubes and cables, wood-heated sauna stoves, metallic machines and devices, clean and open barrels [3]. The cities do not have an obligation to organize collection for metal fraction of household waste from residential buildings. Consequently, the regional drop-off collection with fixed collection points, often located at public spaces such as shopping centers, has been the most common way to collect metal fraction of packaging waste.

In addition to regional drop-off collection systems, the households and properties negotiate metal collection contracts with private waste collection providers independently. The metal fraction is then collected separately from other solid household waste fractions. This leads to a situation that the adjacent properties might have a contract with a different collector, which makes the collection system economically as well as environmentally ineffective.

Producer responsibility also applies to metal packaging. The producers and importers of metal packages are required to arrange the disposal of the packages at their end of life. This has been defined in the Finnish Waste Act (1072/1993) [4], Chapter 3a, as a mandatory obligation. Producers can meet their obligations by joining a producer community. In Finland the relevant producer community is Mepak-Recycling Ltd. The shareholders of this organization are metal packaging manufacturers, the packing industry and retail-wholesale trade organizations in Finland, altogether twelve shareholders [5]. Mepak organizes recycling by providing collection points with containers and contracting companies to recover the metal from the packaging waste.

The target material of this study was the MSW collected from residential properties, in this article referred to in short by term solid household waste. The expression metal fraction of solid household waste is used to describe the small sized metal waste fraction of this solid household waste. Mixed household waste refers to the unsorted fraction of residential MSW, which is delivered to the landfills. The term source sorting is used in order to describe the waste sorting performed by the households prior to collection. With property collection we refer to the collection of solid household waste from residential properties of different sizes. The solid household waste is collected in containers accepted or provided by the municipality. Regional collection describes the drop-off collection method, located at central locations, where the collection system is not as close to properties as in a property collection scheme and each household delivers the source-separated metals to the collection points. Recycling means processing of the used metal fraction of solid household waste as raw material for new metal production.

The total amount of solid household waste in the Helsinki metropolitan area in 2007 was 339,060 t, which is 336 kg/capita [6]. Research carried out by the Helsinki Region Environmental Services Authority [7] shows that the amount of metal fraction in unsorted mixed household waste is 5.4 kg/capita/annum, which is about 1.6 wt.-% of the total solid household waste produced. The metal fraction in unsorted mixed household waste contains around 0.78 kg/capita/annum of aluminum packaging and 2.08 kg/capita/annum of other metal packaging and 2.55 kg/capita/annum of other metals. The metals in unsorted mixed household waste mainly consist of metal packaging waste, such as steel and aluminum packing and cans, aluminum foil, paint tins, spray cans, but also electrical wires, keys, tools, screws and bolts, crown caps, closures, steel bands and straps [7]. The total amount of mixed household waste delivered to landfills was 158,556 tons in 2007 [7]. Estimation for collected metal fraction of solid household waste is 1.1 kg/capita/annum [8].

Metals are often considered to be 100% recyclable. In theory this is true for pure metals, but in practice it is unrealistic for several reasons. Recyclability is strongly affected by the collection efficiency and the quality of the sorted metal fraction and the separation processes chosen and available for treatment of the metal fractions. In the Helsinki region solid household waste forms about half of the total MSW [6], and solid household waste forms in fact a significant metal sink. Considerable amounts of metal scrap could be recovered from solid household waste by more effective collection systems. The remaining metals in the mixed household waste have mostly ended up in landfills. In the future, the mixed waste will be incinerated and some of the metals can be recovered either before or after the municipal solid waste incineration (MSWI) process. The recoverability however is limited, since in the MSWI a large part of aluminum is melted and oxidized, and steel scrap quality is reduced [9].

Currently, there is a debate whether more effective collection schemes of metal fraction of solid household waste would be sustainable in terms of the emitted greenhouse gases (GHG). It is however possible that material recycling produces more GHG emissions than what can be credited by substituted material in primary material production [10].

The Waste & Resources Action Program report (WRAP) [11] evaluated nine Life Cycle Assessment (LCA) studies on the metal fraction of solid household waste. In most of these studies, recycling is more environmentally profitable than land filling or collecting metals from municipal waste incineration plants’ bottom ash. This result is also supported by other studies [12,13,14]. In addition, Dahlén, et al.’s (2007) [15] study demonstrated that the result of extended property collection is that more metal and other recyclables were source-separated and recycled. Yet, according to Tanskanen and Kaila [16], property collection of additional fractions of recyclables and reduction of on-site obligation is limited because of increased costs of solid household waste collection. Their study also shows that the greatest increase in costs occurs among properties with the smallest waste accumulations. According to their research, the volume of bins needed per ton of waste and the pick-up times show a non-linear pattern of increase when the amount of waste collected per pick-up is decreased [16]. For the same reason, changes in the separation strategy among small properties have the greatest effects on the costs of waste collection [16].

The purpose of this research was to analyze and compare the changes in the GHG emissions reduction potential of six alternative metal fractions of solid household waste collection schemes in the Helsinki metropolitan area.

The research questions were as follows:

Life Cycle Inventory Assessment (LCIA) was carried out in order to compare the environmental impacts of the alternative collection scenarios of the metal fraction of household packaging waste. Data from the literature and from the local waste management authority was used for the assessment. The scenarios with alternative separate collection systems for metal were assessed by means of a life cycle assessment. The article is based on the results of the research completed by the first author [17].

In order to compare the collection scenarios for the metal fraction of solid household waste, it was necessary to consider the overall GHG emissions generated by metal collection systems and the effect their utilization has on the overall GHG emissions. In other words, the entire process from metal collection, separation, mixed household waste collection, metal separation from MSWI bottom ash, and substitution effect of primary metals production was studied. The emissions from MSWI were assumed to remain unchanged between the different scenarios and were not analyzed separately.

The six scenarios of the study are described in Table 1. Scenarios were chosen so that differences from different collection methods could be studied and compared. In all the scenarios with source separation of metals the metals were collected as a separate fraction and not commingled with other recyclables fractions. In property collection, the glass and metal fraction of solid household waste were assumed to be collected by the same collection truck. The truck had two different compartments with pressuring mechanism to maximizing the bulk density of the collected material. The different collection methods chosen were property collection and regional collection. There is a common assumption in all the scenarios that the metals, which are not promoted to be source sorted, end up in the mixed household waste. These metals further end up to MSWI, where metals are collected from bottom ash. The source separation efficiency was based on the estimate of an earlier study by Vares and Lehtinen (2007) [18]. The regional collections source sorting efficiencies were estimated by using an earlier study by Kaila [19,20] that showed dependency of recycling rate on population of the regional collection points.

In Scenario 0, all the solid household waste is incinerated, forming a benchmark for comparison of the other scenarios. In this scenario, all the metals end up in to the MSWI process. In Scenario 1, the separate metal collection is organized using only 135 regional collection points. Scenarios 2–4 were residential property collection scenarios, where the metals are separately collected from residential properties of over 20, over 10 and over five households, respectively. In Scenarios 2–4, there are also 135 regional collection points, which are targeted for properties that do not have residential property collection. Scenario 5 represents the other end of the spectrum, where the metals are source-separated for recycling with known source sorting efficiency in all the properties and there is no regional collection. In the Scenarios 1–5, the metals that are not source-separated, end up into the incineration process.

The collection services were optimized according to the accumulation of the metal fraction of solid household waste. The collection services for suburban single-family houses and suburban apartment blocks were assumed to be similar. The source-separated metal fraction of solid household waste was collected directly from properties or collected with a regional collection scheme with centrally placed regional drop-off containers. Table 2 shows the number of inhabitants in different collection schemes that are covered by regional collection and by the property collection scheme.

The modeled scenarios were formulated in collaboration with the Helsinki Region Environmental Services Authority (HSY) and took into account the HSY requirements for future collection systems.

The metal fraction of solid household waste was considered to be used as substitute raw materials to produce new metal products. The method used in this research was a life cycle assessment in terms of emitted GHG. In this assessment, the goal was to describe the relevant parts of the technological system of recycling of solid household waste metal fraction. The assessment is described in the following section.

The GHG reduction potential was calculated in terms of GWP. GWP describes total warming impact of equal masses of different greenhouse gases on atmosphere over chosen period of time. The gases that contribute to GWP are mainly CO₂, CH₄, and N₂O. The values used in this study are from the IPCC 2007 Fourth Assessment Report [21], where CO₂ = 1, N₂O = 298, CH₄ = 25, CF₄ = 7390 and C₂F₆ = 12200, expressed in kg CO₂-equivalent (CO₂-eq)/t collected metal fraction of solid household waste.

The GHG emissions were computed both from direct emissions and indirect emissions. The direct emissions taken into account were caused by motor vehicles, which were used to transport and move the materials from collection sites to recycling facilities and further to material recovery processing plants.

The indirect emissions taken into account in this study were:

Indirect emissions based on construction of buildings, machinery and equipment for provision of the above-mentioned products and services were not included in the life-cycle inventory.

For this study, the consequential life cycle assessment approach (described in Ekvall & Weidema [22]) was chosen. The consequential LCI methodology aims at describing how the environmentally relevant physical flows to and from the technological system will change in response to possible changes in the life cycle [22].

The functional unit, the base unit of assessment, was defined as one ton of the mixed metal fraction of solid household waste. GWP per one ton of collected metal fraction of solid household waste was linked to this functional unit. GWP values were expressed in kg CO₂-equivalent (CO₂-eq)/t collected metals in different scenarios. The GHG reduction potential was calculated by comparing the GWP of different scenarios.

The GHG reduction potential for the different collection scenarios compared to the situation, where no metals were recovered, ranges from -8000 to -9000 t CO₂-eq, when source sorting efficiency was 30% and from -12,000 to -14,000 t CO₂-eq with 50% source sorting efficiency. In Scenario 0 the GHG reduction potential is -3400 t CO₂-eq with a 50% yield from MSWI bottom ash and -5500 t CO₂-eq with a 80% yield from MSWI bottom ash. The GHG reduction potential of the regional collection Scenario 1 is -4600 t CO₂-eq. Results show that separate collection from residential properties, or “property collection”, clearly produces more environmental benefits in terms of GHG emissions compared to the scenario without source sorting. One of the reasons for this is the choice made of not recovering the aluminum from MSWI bottom ash as well as the loss of metal quality due to the incineration process.

The major cause for higher GWP of regional collection, compared to property collection, is that the recycling rate is rather low at regional collection points. The reason for the low recycling rate for regional collection is likely that collection is less controlled at regional collection points compared to property collection and that consumers are more likely to separate at source rather than travel to a collection point. Kaila [19,20] shows that the regional collection recycling rate is less than 20% if the population sharing the collection point is more than 900 people. That means that in the Helsinki area there should be at least 1000 regional collection points.

In terms of GWP, the best scenario was the “separate collection for all”- scenario, Scenario 5. The second best practice is to collect metals from properties that include five or more apartments and from regional collection points (Scenario 4). The GWP of Scenario 4 is about 3–5 percentage units higher than that of Scenario 5. Sensitivity analysis shows that the effect of the metal yield or grade on the GWP does not affect the scenario rank order. In addition, the sensitivity analysis shows that not performing a metal yield from MSWI bottom ash results in a significantly higher GWP. This indicates that separation from MSWI bottom ash is important, but cannot replace source separation. A larger amount was saved elsewhere in the system due to recycling. Furthermore, taking into account the fossil fuels with regard to energy content instead of mass and scarcity does not change the tendency observed in the results. The results presented here have been weighted with regard to the scarcity of the resources, where the weighting criterion of the EDIP1997 method [40] is used. All scenarios show consistent savings of metal resources. Note that these savings are proportional to the sorting efficiency for metal packaging.

It is clear that a sorting or separation efficiency of 100% is theoretical only, but efficiency could be increased by targeted initiatives such as better guidance and education as well as providing a better infrastructure for citizens. As an example, the deposit-return system of beverage cans and bottles reaches a sorting efficiency near 100%. Such an efficiency, however, is not realistic to expect from the recovery of the metal fraction of solid household waste. Nevertheless, a future target of 50% reuse or recycling can be set and be expected to be realizable. This will depend on the development in waste composition, amount of waste as well as recycling technology. Therefore, development in waste amounts and efficiency of collection scenarios should be carefully monitored in the future.

Reduction of emissions from collection and transport could make recycling more beneficial, but reduction of emissions from waste treatment is needed as well. Merrild [41] demonstrates that in general aluminum and steel could be transported considerable distances without compromising the benefit of recycling when compared to incineration with energy recovery. This means that even export of recyclables for example from Europe to Asia could be justified from an environmental point of view. Optimization of collection routes and effective utilization of truck capacities are two other suggestions. For instance, Tanskanen and Kaila [16] demonstrate that the introduction of additional source-separated fractions in property-close collection scenarios increases the fuel consumption, but this is to some extent mitigated by more fuel-efficient methods with combined or commingled collection.

The net GHG emissions reduction potential of different collection scenarios, relative to the benchmark scenario reduction potential, is shown in Figure 3. The figure shows that the cost of transportation in terms of GHG emissions is small compared to the GHG credit resulting from metal recovery from waste. Material losses in base case mean that the metals are lost in pre-treatment processes. Assumption for material loss in pre-treatment is 1% of the total input, and it is the same for all the metals [38]. In reality, the losses can be much higher. The losses in pre-treatment must be replaced with primary production. The effects of these losses are represented as negative values in Figure 3.

The total GHG emissions of the waste management system in the Helsinki metropolitan area in 2009 was 87 000 t CO₂-eq [42]. The results show that source-sorting and collection at property and regional levels have a substantial impact on GHG emission. Efficient sorting and collection can result in a decrease of GHG emission of nearly 10%.

The collection systems for recyclables in the various scenarios were defined with respect to the organizational and technical boundaries in the waste management system at hand, and only easily recyclable fractions were considered. It is recommended that municipalities enhance the recycling rate by establishing new collection schemes for the metal fraction of solid household waste. Within the scenarios studied, Scenario 5 provides the highest recycling rate and the highest environmental benefits. Scenarios 2, 3 and 4 were also environmentally beneficial compared to Scenarios 0 and 1. The main distinctions between Scenario 5 and the other scenarios are that waste is collected from the property and not regionally from collection points. In Scenario 0, collection is also done at the property, but without source-separation and all the mixed solid household waste is incinerated. Hence, a combination of collection directly from the properties and source-separation shows the greatest benefit, measured in GHG reduction potential emission.

Incineration of all waste is not recommended because the recycling rate would decrease due to quality decrease of metals in the MSWI process. This quality decrease leads to the loss of credited GHG emissions from recovered metals and thus to lower environmental benefits for the scheme. The magnitude of changes in environmental impact, resource consumption and costs may be relatively small, but these changes should be compared to the effect of other initiatives within the waste management system before judging their importance. In this study, it is assumed that all aluminum is not recovered after incineration, although a non-negligible amount of metallic aluminum may in practice be recovered from MSWI bottom ash. The purpose of the study presented in this paper is however to investigate the differential impact of various scenarios on the reduction of GHG emission through a collection scheme strategy. Sensitivity analysis shows that recovery of aluminum from MSWI bottom ash does change GHG emission for each scheme, but does not change the relative ranking of scenarios.

If the decision makers conclude that the alternative collection systems are all acceptable with regard to environmental impact and costs, they would have the opportunity of putting weight on other aspects such as improvement of sorting efficiency and level of service. It is important to note that the results of this work are sensitive to our assumptions regarding the potential utilization of the recycled metals. The calculations were done for pure metal fractions, and the recycling processes were assumed to substitute for virgin production of similar materials. The actual options for recycling and utilization depend strongly on the amount of metal scrap collected and its quality, and it is recommended that these issues be thoroughly assessed in each case. Further economic assessment is also needed in order to determine the economic ranking of these different scenarios.