Radioactivity of residues from waste incineration facilities in Finland

Waste incineration in Europe has been increasing in the past few decades as there is a need to reduce the burden on landfills and their associated environmental concerns. While incineration reduces the volume of the waste, the volume of slag and ash is still substantial. To find out potential radiation risks that incineration residues could set to workers or the public, the levels of radioactive elements in these residues were investigated from nine waste incineration plants in Finland. Natural and artificial radionuclides were detected in the residues, but in general the activity concentrations were low. This study shows that the level of Cs-137 in the fly ash from municipal waste incineration follows the pattern of 1986 fallout zones in Finland, although the levels are significantly lower than in ash from bioenergy production from the same areas. Am-241 was also detected in many samples, although the activity concentrations were very low. Based on the findings in this study, the typical ash and slag residues from municipal waste incineration do not need radiation protection measures for workers or the public even in regions that received up to 80 kBq m−2 of Cs-137 fallout in 1986. The further use of these residues need not be restricted due to radioactivity. Hazardous waste incineration residues and other special cases need to be considered separately, depending on the original waste composition.


Introduction and background
The purpose of this study was to characterise the radioactivity of typical waste incineration residues in Finland in order to examine if there is a need for protective measures or restrictions due to radioactivity in the recycling or utilisation of the residues. Due to the global need to advance sustainable development and the circular economy, and mitigate the effects of climate change, measures are needed to reduce waste, increase reuse, recycling and utilisation of residues, and reduce the burning of fossil fuels (e.g. United Nations 2030 Agenda for Sustainable Development [1], the European Green Deal [2] and the European Circular Economy Action Plan [3]). One result of these efforts is that the incineration of municipal waste in Europe has doubled from 1995 to 2019 [4], although the long-term goal in Europe is the prevention and reduction of waste [3] rather than increasing waste incineration further. Previously there have been only a few studies of the radioactivity of waste incineration residues internationally [5][6][7][8], and no studies in Finland. Due to the levels of heavy metals in the residues [9,10] the beneficial use of waste incineration residues can be limited, but as advanced forms of reuse and recycling are developed, such as more efficient metals recovery [11], it is also important to investigate and document the levels of radionuclides in the residues. For this purpose, in this study the residues from nine waste incineration plants in Finland were characterised for natural and artificial radionuclides.

Legal background
In Finland, the directive 2013/59/Euratom (EU-BSSD) has been implemented with the Radiation Act 859/2018, and notification and exposure assessment are required if the activity concentrations of natural radionuclides exceed exemption and clearance levels (naturally occurring radioactive material, NORM). Therefore, the general exemption and clearance levels of 1 kBq kg −1 for U-238, Th-232 and their progeny and 10 kBq kg −1 for K-40 apply also to residues from waste incinerators, even though the Radiation Act does not explicitly mention the operations of waste incineration plants. There are also regulations in place concerning the radioactivity of building materials including natural radionuclides and fallout Cs-137 [25]. Due to the presence of fallout Cs-137 in the Finnish environment [26] originating from the Chernobyl nuclear accident in 1986 and atmospheric nuclear weapons testing in the 1950s and 1960s, the general exemption and clearance level of 0.1 kBq kg −1 for Cs-137 from the EU-BSSD is not practical in Finland. For example, almost all ash from bioenergy production in Finland exceeds this level for Cs-137, often by a large margin [27,28]. In Finland, the exposure from fallout Cs-137 is considered as part of natural radiation and thus fallout Cs-137 is included in the regulations for natural radiation [25]. The reference levels used for natural radiation in Finland are 1 mSv a −1 for workers and 0.1 mSv a −1 for the public (excluding radon, building materials and background). For the exposure of the public to external gamma radiation from building materials, the reference levels are 1 mSv a −1 for indoor exposure in buildings (additionally 0.1 mSv a −1 from Cs-137 alone) and 0.1 mSv a −1 for outdoor earthworks such as roads, streets and field structures.
Based on evaluations made by the Radiation and Nuclear Safety Authority (STUK), specific exemption from exposure assessment is given to materials (excluding building materials) containing fallout Cs-137 in activity concentrations less than 5 kBq kg −1 and 1 kBq kg −1 for workers and public, respectively [25]. In EU-BSSD Annex VIII a screening index is used for the indoor gamma radiation emitted by buildings, and this is implemented in the Finnish regulations [25]. This index can also be applied to screening the constituents of mixed building materials (e.g. mixing ash into concrete), taking into account that for the final mixed material the proportions of the mixture need to be considered in the screening. Public exposure from concrete needs to be assessed separately if the Cs-137 activity concentration in concrete exceeds 50 Bq kg −1 [25]. In addition to the screening index of buildings, there are other similarly structured screening tools in the Finnish regulations for the gamma radiation emitted from materials used in the construction of roads, streets and yards, and for other earthworks [25].

Materials and methods
Incineration residue samples were collected by the plant operators from nine different waste incineration plants around Finland (table 1, figure 1) and sent to the STUK laboratory. During the time of sampling there was a total of 11 operational waste incineration plants in Finland. In one location there were separate boilers for hazardous waste (plant 8) and municipal waste (plant 9), and these were counted as separate facilities because of the different fuel types (table 1). Six of the studied plants incinerated mainly MSW. Additionally, some of these facilities may also incinerate smaller side streams of wastewater sludge, waste wood, medical waste, industrial waste or hazardous waste, so the waste content can vary also temporally at one facility. Plant 7 was a smaller pilot plant for incineration of wastewater sludge, and plant 1 incinerated solid recovered fuel including waste wood (table 1).
Plant operators were asked to send samples from different ash and slag material batches that are the final residues of the incineration process. Two temporally different samples were requested for each ash and slag sample type: one that would represent a shorter period batch (1 week to 1 month) and another a longer period combined sample (6 months to 1 year). Altogether we received 63 samples of 0.5-1 litre volume. For the purposes of data analysis, APC residue, fly ash and boiler ash samples were grouped together as a fly ash residues group (fly ash, 39 samples), because all these materials are from the flue gases and all these three fractions are not consistently forming or reported from every facility. Respectively, bottom slag and bottom ash were grouped together as a bottom ash residues group (bottom ash, 21 samples) representing the materials remaining at the grate after combustion. Additionally, one facility had a separate gas fired power boiler out of which three residue samples were analysed. The concentrations in these three samples were below or close to the limits of detection for all nuclides and they are not considered further in the results, leaving 60 samples for statistical analysis.
Samples were dried at 105 • C and all concentrations were calculated to dry weight. For gamma-ray measurements, the dried samples were compressed into 100 ml cylindrical containers (STUK 'T-jar'). Containers were packed in airtight bags that were heat sealed under vacuum. Samples were then let to stand for 3 weeks before gamma-ray measurements so that radon progeny would reach secular equilibrium. Some of the samples were also chosen to be additionally measured immediately after sealing, in order to analyse possible short-lived artificial radionuclides in the residues.
Gamma-ray measurements were made at the STUK laboratory with multiple Canberra and Ortec high-purity germanium (HPGe) detectors of the type suitable for low energies (e.g. Canberra BEGe with a carbon epoxy window). These detectors are cooled with liquid nitrogen and data were collected with digital multi-channel analysers such as Ortec DSPEC 502. The laboratory has a low-background environment, and the detectors are housed in low-background lead shielding with 15 cm of lead in most cases. The software used for peak search, peak area determination, baseline and background subtraction, nuclide identification and activity calculations including coincidence corrections was UniSampo-Shaman [29][30][31]. Efficiency calibrations were made using certified reference sources of 20 mm thickness for one geometry (STUK 100 ml 'T-jar'). Corrections for other sample geometries, sample heights, density and chemical compositions were made using efficiency transfer software Efftran [32] in combination with UniSampo-Shaman and the STUK laboratory information system [33].
In the gamma-ray analysis Pb-210 was determined from the 46.5 keV peak, Cs-137 from the 661.7 keV peak and K-40 from the 1460.8 keV peak. Ra-226 was determined as a weighted average of the radon progeny Bi-214 and Pb-214 (295, 351, 609, 1120 and 1764 keV peaks). Th-232 was determined as a weighted average of Ac-228 and Tl-208 (338, 583, 911 and 2614.5 keV peaks, corrected for branching at Bi-212) since these decay products representing Ra-228 and Th-228 were in or close to equilibrium. U-238 was determined from the 1001 keV peak of Pa-234 m, or the 63 keV peak of Th-234, if Pa-234 m was below the level of minimum detectable concentration (MDC). Additionally, for the subset of samples that were analysed with inductively coupled mass spectrometry (ICP-MS; N = 27), the ICP-MS-result for U-238 was selected (see [33] for a comparison of gamma-ray spectrometry with ICP-MS).
U-238 analyses were performed using ICP-MS for a subset of the ash samples (N = 27) at the STUK laboratory. Approximately 0.5 g of ash samples were accurately weighed and treated with aqua regia (3:1 HCl:HNO 3 ) overnight on ice and then digested with the UltraWAVE ECR microwave digestion system Table 1. Information on the waste incineration plants included in the study, Information is from the annual reports of the plant operators, from a joint website of the environmental administration in Finland.    (Milestone, Italy) using a temperature of 165 • C for 20 min following 190 • C for 10 min. After digestion the samples were diluted with ultrapure distilled water to 100 ml and filtered to avoid insoluble solids. For ICP-MS analysis the samples obtained from microwave dissolution were further diluted to 1:100 with 1% nitric acid. Analyses of U-238 concentrations were performed by ICP-MS (iCAP Q, Thermo Fisher Scientific, USA) using calibration standards of 0, 1, 10 and 100 µg l −1 for U-238 (Romil Ltd, UK) to generate the calibration curve. Bismuth was used as an internal standard to compensate for matrix effects (Romil Ltd, UK).  than two orders of magnitude ( figure 2(B)). In contrast, there is little difference between fly ash and bottom ash Ra-226 concentrations ( figure 3(A)) and there are also no large differences between different facilities for Ra-226 ( figure 3(B)).

Naturally occurring radionuclides and Cs-137
The activity concentrations of natural radionuclides obtained in this study are of the same level as in the other available studies on MSWI residues from the United States, Japan and Germany [5][6][7][8]. In the Japanese plant, Cs-137 was also analysed [7] and the median Cs-137 concentration in fly ash was ten times higher in Finland.
The difference in Cs-137 arises from the Chernobyl fallout (figure 1) which is discussed in section 3.1.1.
Higher concentrations of Pb-210 and Cs-137 in the fly ash group versus bottom ash are typical of high-temperature combustion processes such as burning of coal or biofuels, reflecting the volatility of lead and caesium at relatively lower temperatures when compared with uranium, thorium and radium. The more volatile components such as polonium, lead and caesium are enriched in the fly ash fractions when compared with the bottom ash residues (e.g. [5,[34][35][36]). This same feature is visible in the results of this study, although the concentrations of Pb-210 and Cs-137 in MSWI residues (median 107 and 37 Bq kg −1 in fly ash) are significantly smaller than in residues from bioenergy production in Finland [27,37], where the median Pb-210 and Cs-137 concentrations in fly ash were 570 and 1300 Bq kg −1 , respectively [27].

Cs-137 and the fallout zones
It is known that Cs-137 is enriched in the bark of, for example, pine, spruce and birch, when compared with the wood itself [21,38,39]. Waste incineration plants include wood in some cases, either as industrial waste or other types of wood waste (table 1), and MSW can also include wood. A previous study on ash from bioenergy production in Finland showed that significantly higher Cs-137 concentrations in fly ash were found in the regions with the highest original fallout levels [27]. For waste incineration facilities included in this study, there is no evident increase in Cs-137 concentration in the fly ash with increasing percentage of wood in the fuel (tables 1 and 3), but there is a clear increase of Cs-137 concentration according to the fallout zone of the facility, which is in contrast with nuclides such as Ra-226 ( figure 4). The results of Cs-137 for both MSWI fly ash from this study and biofuel fly and grate ash [27] are shown in figure 5. The increasing trend of Cs-137 concentration with the fallout zone is very similar for both fuel types but the median concentration of Cs-137 in biofuel ashes is systematically one order of magnitude larger. This is perhaps understandable considering the difference in the fuels, as in this study the highest percentage of wood in fuel was 60%, and wood that has been used in industrial construction rarely contains significant amounts of bark. In the previous study on ash from bioenergy production in Finland [27], bioenergy facilities burned 70%-100% wood. It was found that increasing the amount of peat in the fuel would reduce the concentration of Cs-137 in the fly ash, but the highest Cs-137 concentrations were found in the facilities that used 100% wood. In bioenergy production it is typical to burn woodchips, which also contain significant amounts of bark. Also, the left-over bark from sawmills is used in bioenergy production. Therefore, a combination of original regional fallout levels and the amount of bark in fuel is suggested as a possible explanation for the different Cs-137 concentration levels in fly ash from waste incineration versus bioenergy production (figure 5). The existence of the trend with fallout zones also for MSWI fly ash suggests that despite the centralisation of production and logistics for many household goods there is still a local component in MSW which is preserving some of the Cs-137 signature of the local fallout zone. Table 3.

Artificial radionuclides other than Cs-137
Besides Cs-137, other artificial radionuclides, I-131 (3.6-36 Bq kg −1 , three samples), Lu-177 (3.6-15 Bq kg −1 , two samples) and Am-241 (0.4-60 Bq kg −1 , 23 samples) were also detected in the ash samples. I-131 and Lu-177 were detected only in fly ash, but Am-241 was also detected in bottom ash. I-131 and Lu-177 are commonly used in hospitals in radiotherapy treatments of cancer patients and can possibly end up in MSW among hospital waste or waste from patients' households. These nuclides are also continuously detected from sewage sludge that might end up in the incineration facilities in some localities. In Helsinki, STUK regularly monitors the sewage sludge from Viikinmäki wastewater treatment plant as a part of its environmental radiation surveillance programme. The activity concentrations of these nuclides in dried sludge in 2021 were 210-430 Bq kg −1 for I-131 and 950-1700 Bq kg −1 for Lu-177 [24].
Since Am-241 was detected in almost half of the samples in multiple locations, a mass balance calculation was made to investigate whether it is plausible that the Am-241 found in the waste incineration residues in this study could be from 33 kBq household smoke alarms. These alarms are exempted by the radiation legislation but they should be taken into recycling and not end up in MSW. However, these alarms can end up in municipal waste when not disposed of appropriately. For MSW grate incineration, bottom ash (bottom ash and bottom slag) typically represents 16%-25% by mass of the original waste on a wet basis, and fly ash (boiler ash, fly ash and APC waste combined) 3%-15% [13,14]. For approximate mass balance calculations mass fractions of 20% and 10% can be used for bottom ash and fly ash groups, respectively. If the amount of  incinerated waste was 300 000 t a −1 in fresh weight (table 1), 60 000 t a −1 of bottom ash and 30 000 t a −1 of fly ash could be formed. Since the median activity concentration for Am-241 is approximately 1 Bq kg −1 for both ash types (table 2), these can be combined to a total of 90 000 t a −1 of residues with 1 Bq kg −1 of Am-241. Only two samples showed higher concentrations than the average, but they are from 1 week and 1 month samples representing short 'spikes' compared with the long-term level of 1 Bq kg −1 (figure 6). To reach 1 Bq kg −1 activity concentration in 90 000 t a −1 of ash from 33 kBq smoke alarms would require more than 2700 items per year at one MSWI plant. This does not seem plausible since the recycling of appliances has been in place for a long time already. The highest Am-241 value from a 1 week sample (figure 6) could be from, for example, industrial smoke alarms but some other sources are probably also contributing to the general Am-241 level in the residues. For example, waste containing Am-241 from radiation practices can be appropriately disposed of as mixed waste and end up in incineration, if the activity is lower than the exemption level (10 kBq and 1 kBq kg −1 Am-241 for <1000 kg of material). It is also known that a small amount of Am-241 can be found in the environment due to the fallout from atmospheric weapons testing [40]. For example, the measurement database of STUK contains results of ash samples from peat combustion. These samples contain 1-3 Bq kg −1 of Am-241, when the measurement has been made with a small enough MDC of <1 Bq kg −1 . A similar value of a few Bq kg −1 can be typically measured in sediments from the 1950s to 1980s [41]. It is unclear how much of this fallout Am-241 could end up in MSW but at least a small contribution is possible from, for example garden waste containing peat or soil. The exact origin  Screening index values for the external gamma radiation emitted by waste incineration residues in Finland if they were used as building materials in the construction of buildings (A), or for roads, streets and yards (B). An index value <1 indicates that material can be used without further exposure assessment. Index formulae can be found in [25].
of the Am-241 in MSWI residues remains unclear but the measured level is of no concern in terms of radiation protection.

Screening of building materials
For the purposes of assessing radiation risk of waste incineration residues as part of building materials, the screening index values for the ashes of this study were calculated, and the results are shown in figure 7 for building construction and road construction. All the analysed residues have index values <1 which indicates that they could be used as building materials without further exposure assessment. The same is also true from the point of view of Cs-137 in concrete for buildings, as the maximum Cs-137 concentration in residues in this study is 150 Bq kg −1 , and in the STUK guidance it is calculated that for typical amounts of ash mixed into concrete, the Cs-137 activity concentration in ash could be up to 300 Bq kg −1 before the exposure assessment limit of 50 Bq kg −1 in the final concrete would be reached. It is important to remember that this evaluation is only from the point of view of radiation and the waste incineration ashes could contain chemical hazards such as heavy metals which could prevent their use as building materials. However, for the purposes of facilitating waste reduction and the circular economy, it is also important to state that radioactivity is not limiting the potential beneficial use of MSWI residues.

Effects from incinerated waste type
Nearly all the operational waste incineration plants in Finland were included in this study (9 out of 11). Most of the plants used MSW for fuel, therefore the results can be considered at least nationally representative for MSWI residues. The other fuel types, wood, hazardous waste, industrial waste and wastewater sludge were much less commonly used (table 1). It is still worthwhile to investigate whether there are any compositional differences in the residues which would be correlated to the fuel type. When comparing the reported fuel types with the activity concentrations in ash, the most significant differences arose in the case of wastewater sludge. The ashes from wastewater sludge had the lowest Cs-137 concentrations and highest U-238 and Th-232 concentrations among the studied residues. The mean Cs-137 concentration from wastewater sludge fly ash was 1 Bq kg −1 whereas for fly ash from other fuel types it was 50 Bq kg −1 . The mean U-238 and Th-232 concentrations were 70 and 60 Bq kg −1 for wastewater sludge ashes, and 35 and 20 Bq kg −1 for ashes from other fuel types, respectively. Also, there were no detections of Am-241 from the wastewater sludge ashes. However, the number of wastewater sludge ash samples was small (n = 3, tables 1 and 3) so not too much significance can be attributed to the non-detection of Am-241, as it could also be a result of the variability of MDCs between the different detectors used. For Ra-226, K-40 and Pb-210, the concentrations in wastewater sludge ash were within the variability of the MSWI residue compositions. No clear correlations were detected in the activity concentrations compared with fuel type for industrial waste, hazardous waste or wood, but the number of samples for these categories was too low for meaningful comparisons.

Conclusions
Based on the radiological characterisation of 60 residue samples (bottom slag, bottom ash, boiler ash, fly ash and APC residue) from nine waste incineration facilities in Finland, it is concluded that the contents of natural and artificial radionuclides in the studied waste incineration residues are low from the radiation protection point of view. This is the case even for the geographical areas which received up to 80 kBq m −2 of Cs-137 fallout in 1986. Therefore, there are no restrictions to the reuse, recycling or utilisation of these residues originating from radioactivity. However, this conclusion is only generally applicable to incinerated municipal waste, as most of the studied facilities used MSW as fuel, and only restricted materials are meant to be disposed of as municipal waste. In contrast, the composition of hazardous waste can potentially be highly variable depending on the origin of the waste. If hazardous waste containing significant amounts of natural or artificial radionuclides is incinerated, there could be a need for radiation protection measures when handling the residues, as radionuclides are concentrated in the incineration residues compared with the original waste.

Data availability statement
All data that support the findings of this study are included within the article (and any supplementary files).