1 Introduction

According to available statistical data (Eurostat 2024), in 1995 about 61% of municipal waste was transferred directly to the landfill; however, in 2019 only 24% was accumulated in that way. Such a change resulted from the introduction of new regulations in the European Union (EU), which obliged member countries to reduce biodegradable waste fractions disposed to the landfill. As a consequence of this action, there was a clear change in municipal waste treatment. The most popular waste recycling methods include composting, mechanical sorting, and recovery of waste, as well as incineration.

The development of the latter method was also influenced by a change in the approach to the promotion of renewable energy sources, as energy production from waste became an important element of waste and energy policy. So far, the European leaders in this area include Germany, France and the United Kingdom (Scarlat et al. 2019). In Poland, the implementation of the above-mentioned legislative changes has contributed to the start of the construction of several municipal waste incineration plants.

Cement plants in which alternative fuels are used for clinker burning also turn out to be an important support in such use of waste. Mainly it is refuse-derived fuel (RDF), a municipal solid waste (MSW) that has undergone complex processes of transforming it into a fuel of a certain quality and calorific value. Its production takes place in specialized plants, where the waste is segregated, and its size is reduced by shredding, chipping and grinding. After that, it is screened, mixed, dried and then packed. In 2020, the consumption of RDF fuel in Poland was estimated at over 1.7 million tonnes, of which about 90% were incinerated during the cement production process (Nowak 2023).

The presence of biodegradable materials in RDF and their moisture content mean that its biological decomposition naturally occurs, and these processes, associated with the growth of microorganisms, result in bioaerosol emission (Weichgrebe et al. 2022). The research carried out by Mahar et al. (1999) showed that in a RDF production plant, the mean concentration of total bioaerosol was 6.8 × 105 organisms m−3, and bacterial endotoxins—29 EU (endotoxin unit) m−3. The occurrence of microorganisms and their toxins during the production of RDF was related to the presence of airborne organic dust particles in both the inhalable and respirable fractions (Mahar 1999). Majority of hitherto published RDF studies have been focused mainly on the analysis of their calorific value, moisture content, and the presence of chemical impurities (Hemidat et al. 2019; Kimambo and Subramanian 2014). Hence, there are limited data on microbial contamination of RDF. Also, the data describing the microbiological cleanliness of the work environment in a cement plant are very scarce, and only a few analyses of settled dust samples have been performed (Gutarowska et al. 2018). So far, researchers have shown the presence of dust-borne bacteria, including strains from e.g. Bacillus, Acinetobacter, Staphylococcus, and Pseudomonas genera.

Taking the above into account, the purpose of this study was the quantitative and qualitative analysis of airborne bacteria at workplaces in cement plants, where RDF is used for burning clinker. In addition, RDF samples supplied for combustion were subjected to microbial analysis.

2 Methods

2.1 Sampling Sites and Stationary Measurements

This study was carried out in 4 cement plants (CP1–CP4) in Poland, where the heat recovery from RDF required for burning clinker ranged so far between 60 and 71% of the conventional fuels used.

In total, 24 sampling points were established in the cement plants including different stages of technological process, such as: RDF inbound area (2), RDF storage hall (6), RDF sampling point (6), RDF loader cabin (1), weight operator room (1), RDF conveyor belt (3), laboratory rooms for RDF sample analysis (5). Additionally, external background samples at each plant (in total 4 measurement points) were collected approximately 100 m from their boundaries (Table 1).

Table 1 General characteristics of studied cement plants
Full size table

Bioaerosol samples were stationary and taken using a single-stage MAS impactor (100NT, MBV AG, Switzerland) placed about 1.5 m above the ground. Two bacterial samples were collected at every studied point for 30 s each. Impactor was operated at a flow rate of 100 L min−1. For the measurement of total bacteria, the sampler was loaded with Petri plates containing Tryptic Soy Agar (TSA) with 5% additive of sheep blood (bioMérieux, France). In total, 56 samples were taken for microbiological analysis.

Simultaneously with bioaerosol measurements, at each sampling point, the temperature and relative humidity were measured in duplicates with the use of portable thermo-hygrometer (model Omniport 20, E + E Electronic GmbH, Germany).

In the studied cement plants, 14 RDF samples were also collected, directly into 50 mL sterile Falcon tubes and transported to a laboratory for further microbiological analysis (Table 1).

2.2 Identification of Microorganisms

In the laboratory, RDF samples were subjected to extraction in saline solution. From these suspensions, 3 subsequent tenfold dilutions were made, which were then plated in 1 mL volumes on Petri plates filled with microbiological medium, the same as for bioaerosol sampling.

The incubation conditions for all samples were: 1 day (37 °C) + 3 days (22 °C) + 3 days (4 °C). The extended incubation for microbial samples was applied to allow the growth of slow-growing strains at low temperatures (Skórska et al. 2005). The final microbial concentration was expressed in colony forming units (CFU) present in 1 cubic meter of sampled air (CFU m−3) or in 1 g of RDF (CFU g−1).

Bacteria were analyzed based on their ability to degrade organic substrates enzymatically and the subsequent detection of the appropriate metabolites generated by these reactions. For this purpose, biochemical API 20 Staph, 20 Strep, 20 NE, 20 E, 50 CHB/E, and Coryne tests (bioMérieux) allowing identification of the clinically essential strains were applied. The biochemical test results were interpreted to assess species or genus affiliation, with an identification percentage above 80 accepted as reliable.

2.3 Statistical Analysis

Data was analysed with Statistica, ver. 10 (StatSoft, Inc., USA), using Mann–Whitney (M–W) and Kruskal–Wallis (K–W) tests as well as Spearman’s rank correlation coefficient to confirm significant relationships between observed bioaerosol concentrations and microclimate parameters, as the selected data were not normally distributed. The Chi-square test with Bonferroni correction was also used to assess differences in microbial diversity at workplaces. A significance level of 0.05 was used in all tests.

3 Results

3.1 Analysis of RDF Samples

The average bacterial concentration found in RDF samples was at the level of was at the level of 2.0 × 1011 CFU g−1; (SD = 7.6 × 1011). The lowest level of bacterial contamination was found in RDF samples from the CP1 cement plant (1.9 × 106 CFU g−1;SD = 2.6 × 106), while the highest was in the sample from CP4 (2.9 × 1012 CFU g−1), and these differences were statistically significant (Table 2).

Table 2 Bacterial concentrations in RDF samples from cement plants
Full size table

The taxonomic diversity in RDF showed 51 bacterial species belonging to 27 genera (Table 3). Of the 149 bacterial strains identified in RDF samples, more than 36% were Gram-positive cocci of Enterococcus, Staphylococcus, Streptococcus, Micrococcus and Kocuria genera. Gram-negative rods (e.g. Enterobacter, Escherichia, Klebsiella, Proteus, Pseudomonas, Serratia and Shigella genera) were also abundant. Moreover, there were found: Gram-positive bacilli, non-sporing rods of Cellulomonas, Brevibacterium, and Corynebacterium genera were also identified in RDF samples, as well as both mesophilic (e.g. Actinomyces, and Streptomyces genera) and thermophilic (Thermobifida genus) actinomycetes (Fig. 1).

Table 3 Qualitative characteristics of bacteria present in cement plant samples
Full size table
Fig. 1
figure 1

Percentage contribution of bacterial genera in RDF, and bioaerosol samples depending on the sampling site

Full size image

3.2 Analysis of Bioaerosols

The analysis of bioaerosol samples from 4 cement plants showed that the average concentration of total bacteria amounted to 4.2 × 104 CFU m−3 (SD = 4.5 × 104). The concentration of bacteria in the background samples was two orders of magnitude lower (4.2 × 102 CFU m−3; SD = 4.1 × 102) and the difference was statistically significant (Z = 3.97; p < 0.001). The highest bioaerosol concentrations were found in CP3 (9.8 × 104 CFU m−3; SD = 2.5 × 104), and the lowest in CP4 (5.0 × 103 CFU m−3; SD = 4.6 × 103). Statistical analysis showed that the difference in bacterial concentrations between cement plants was significant (H = 30.6; p < 0.001) (Table 4).

Table 4 Bacterial aerosol concentrations in cement plants
Full size table

The comparison of the seven types of workplaces (Fig. 2) did not show significant differences between them. However, the highest median concentration was found in the loader cabin, in which the concentration of bacteria reached the level of 1.1 × 105 CFU m−3. The lowest contamination was noted at workplaces located near the conveyor belt that transferred RDF to the cement rotary kiln (4.6 × 103 CFU m−3). Moreover, the analysis did not show any statistical differences in concentrations between the samples collected in the laboratory rooms and the workplaces located in the technological line.

Fig. 2
figure 2

Median concentrations of bacterial aerosols depending on the sampling site. (1)—TLV for mesophilic bacteria in “working premises contaminated with organic dust”; (2)—TLV for mesophilic bacteria in “public premises” (Pośniak and Skowroń, 2022); K–W test—Kruskal–Wallis test

Full size image

The air temperature at the workplaces was between 9.0 and 28.6 °C, and the relative air humidity ranged from 36.4 to 78.3%. The airborne bacteria concentrations in cement plants depended on the microclimate parameters. They were significantly correlated with the air temperature (R = − 0.62; p < 0.001) and relative humidity (R = 0.48; p < 0.001).

Qualitative analysis of bioaerosol samples showed high taxonomic diversity of microbiota at workplaces (Table 3). There were identified 62 species of airborne bacteria belonging to 31 genera. Of the 382 bacterial strains, about 30% were Gram-positive bacteria of Bacillus, Aneurinibacillus and Paenibacillus genera. Species of Staphylococcus (including S. aureus), Streptococcus, Enterococcus, Kocuria, and Micrococcus genera were also numerous. In addition, the presence of Gram-negative rods of Aeromonas, Enterobacter, Ewignella, Escherichia, Klebsiella, Ochrobactrum, Pantoea, Pseudomonas (including P. aeruginosa), Proteus, Ralstonia, and Serratia genera was also revealed. Moreover, Gram-positive non-sporing rods, as well as mesophilic actinomycetes were identified (Fig. 1). The study found that 58% of the identified bacterial species occurred in both RDF and bioaerosol samples. In addition, 17 bacterial species that may pose a threat to human health were identified in the air at workplaces. The analysis showed that pathogenic strains were significantly more often (p < 0.01) isolated during RDF sampling activities than in other workplaces (Fig. 3). In turn, the atmospheric air was free of pathogens (except Streptomyces genus), and dominated by Gram-positive bacteria, mainly of Aerococcus, Micrococcus, Staphylococcus, and Bacillus genera. Furthermore, only a few Gram-negative rods of Aeromonas genus were found in background samples.

Fig. 3
figure 3

Contribution of bacterial pathogens in bioaerosol samples collected at workplaces in cement plants; # – workplaces with significance difference (p < 0.01) according to Chi-square test with Bonferroni correction

Full size image

4 Discussion

The analyses of microbial contamination at workplaces in cement plants combined with the analysis of RDF samples in connection with their combustion during cement production are, to the authors' knowledge, one of the first studies of this type. It should be noted that previously only Gutarowska et al. (2018) analysed several dust samples that settled in the RDF delivery hall and near the finished clinker conveyor belt. That study showed nearly five orders of magnitude higher concentrations of bacteria in dust from RDF than that collected during the transport of ready-made cement (2.8 × 107 CFU g−1 vs. 3.7 × 102 CFU g−1). The qualitative analysis performed at that time showed significant taxonomic diversity of bacteria, of which the dominant strains included Bacillus, Aerococcus, Corynebacterium, Lactobacillus, Acinetobacter, Staphylococcus, and Pseudomonas genera. However, the limited scope of that study did not allow to fully assess the risk caused by microbiological contaminants at workplaces in this type of plants.

Our study showed that from the stage of RDF delivery to the cement plant until combustion in the cement kiln, harmful airborne bacteria may be present. Their concentrations varied, but not significantly, depending on the activities performed by employees, with the highest levels occurring inside the loader operator's cabin, where they reached 1.1 × 105 CFU m−3. This small space turned out to be the most contaminated, which may indicate that the air-conditioning system installed there is ineffective, which probably requires its filter replacement. As it was shown in the study of Marchand et al. (2024), rarely replaced cabin filters, e.g. in garbage trucks, may become an additional source of microbiological contamination for drivers of these vehicles.

This study showed that the entire technological process of burning alternative fuels may pose a real threat to workers, especially in closed premises with limited air exchange. Among all studied workplaces, the lowest concentrations were found in RDF inbound areas, which were usually located outdoors to ensure access by trucks with containers filled with waste. The other workplaces were usually located inside buildings and the bioaerosol concentrations indirectly indicated dust emissions into the air. Indeed, dust levels is not usually monitored as a routine element of occupational exposure in many plants of this type, but a large-scale study from Norwegian cement plants showed that among production workers, whose many responsibilities also included the control of alternative fuels dosing process, the concentrations of the thoracic dust fraction were at the highest levels (0.6–7.4 mg m−3) (Notø et al. 2015). In turn, Mahar's (1999) study showed that dust particles released into the air during the production and transport of RDF served as carriers for microorganisms and bacterial endotoxins. Therefore, it might be concluded that organic dust was emitted at those workplaces. It seems that such a phenomenon can also be observed in cement plants.

Regarding the measured bioaerosol concentrations, the hygienic assessment of the examined workplaces was performed using the permissible concentrations of microorganisms, proposed by the Expert Group on Biological Agents (EGoBiA) (Pośniak and Skowroń, 2022). The proposed limit values for the bacterial and fungal concentrations in the air were developed based on the results of occupational volumetric measurements of bioaerosols, concerning the potential harmfulness of a specific biological agent, and should be treated as optional standard or auxiliary reference values. According to it, for "working rooms contaminated with organic dust", the threshold limit value (TLV) proposed for mesophilic bacteria was set at the level of 1 × 105 CFU m−3. On their basis, it can be concluded that the TLV for mesophilic bacteria was exceeded in 21% of the collected samples, some of which were obtained during RDF sampling for laboratory analysis. It should be noted that these activities were also characterized by the highest load of pathogenic strains, which should be an additional incentive to implement appropriate measures to protect workers.

A special group of the examined workplaces were the laboratory rooms where the analysis of RDF fuel samples was conducted. This is an unusual type of workplace where our study has shown that bioaerosol emissions may occur. However, because of the laboratory work performed there and the limited human access, greater cleanliness is usually required. So far, both in Poland and in the world, no reference values have been developed for this type of premises. Nevertheless, it seems that the most appropriate TLV values are those established by EGoBiA for workers' protection in public premises, including TLV for mesophilic bacteria at the level of of 5 × 103 CFU m−3 (Pośniak and Skowroń, 2022). Based on the above, it can be concluded that the recommended reference value for mesophilic bacteria was exceeded in 60% of the collected samples. This situation resulted from specific activities such as grinding, weighing and portioning of RDF samples, with the simultaneous lack of local air extraction that would eliminate dust emissions at these workplaces.

The bioaerosol concentrations at all workplaces depended on the microclimate parameters. At one-third of the examined workplaces, relative humidity exceeded 60%, which favours the growth of bacteria, especially Gram-negative rods, which were absent in atmospheric air. In turn, the negative correlation between airborne bacteria and air temperature values indicates that its increase reduced the amount of moisture in the air, which contributed to limiting the bacterial growth. Considering this, ventilation installations' operation should be checked, especially their efficiency for a given volume of rooms. Its proper operation will allow control of the humidity level at the studied workplaces, which can help to limit microbial pollution.

Microbiological air contamination at workplaces in cement plants undoubtedly resulted from the presence of refuse-derived fuels there. The quantitative analysis of RDF fuel samples was characterized by a wide range of concentrations, which may most likely result from different levels of moisture in the fuels themselves. This parameter was not monitored in the currently collected samples, but according to Nowak (2023), the moisture content in RDF burned in Polish cement plants ranged from 5.8 to 36.4%. Moreover, RDF may differ in their morphological composition and production process, because such fuel supplied to cement plants is composed, depending on the needs of a given plant. Such routine may create variable conditions for the growth of microorganisms, which was confirmed by our analyses of samples from cement plants i.e. CP3 and CP4. Both plants were characterized by a similar size and amount of incinerated RDF; however, RDF from different suppliers contributed to a significant difference in bioaerosol concentrations at workplaces. Researchers from Japan also found similar conclusions when comparing microbiological contamination of RDF produced in Japanese and German plants (Sakka et al. 2006).

The results of the qualitative analysis of RDF samples turned out to be highly consistent with the analysis of bioaerosol samples at workplaces. The presence of bacteria from Aerococcus, Aeromonas, Aneurinibacillus, Bacillus, Microbacterium, Paenibacillus, Rhodococcus, and Staphylococcus genera was probably the result of their migration from atmospheric air into the cement plant buildings. However, the presence of other genera was undoubtedly related to the burning of RDF in cement plants. This applies in particular to strains that may pose a threat to the health of employees. In total, the presence of 21 bacterial pathogens classified in risk group 2 according to Directive 2019/1833 was confirmed in both bioaerosol and alternative fuel samples (European Commission 2019).

Some species found in cement plants, i.e. Enterococcus faecalis, E. faecium, as well as strains belonging to Enterobactericeae family indicate faecal contamination of the delivered fuels derived from waste and may have adverse effect on workers' health (Krajewski et al. 2002). These bacteria in immunocompromised people can cause, infections of the urinary tract, respiratory system and skin. Detection of Klebsiella pneumoniae, K. oxytoca, Pseudomonas aeruginosa, and Staphylococcus aureus in both bioaerosol and alternative fuel samples may indicate the presence of particles from e.g. used handkerchiefs, plasters or personal hygiene products. As “medical” waste in nature (Hossain et al. 2013), usually end up in mixed waste these bacteria are known to have high virulence and resistance to antibiotics. They can also contribute to systemic infections and produce various toxins (Kot et al. 2023; Liao et al. 2022; Nowakowicz-Dębek et al. 2016). In turn, Shigella bacteria identified in RDF samples are very contagious and virulent and as such causing dysentery, a special type of bacterial diarrhoea with fever, cramping abdominal pain and frequent passing of small amounts of bloody stools mixed with mucus and pus. Such infection may occur through direct contact, as well as contaminated water or food (Mattock and Blocker 2017).

In cement plants attention should also be paid to the presence of mesophilic and thermophilic actinomycetes, which indicate the processes of decomposition of organic matter and may cause severe allergies in employees (Lacey 1997). It seems that in the future, the research in this working environment should also be extended to include the analysis of anaerobic bacteria, the presence of which has been demonstrated in settled dust (Gutarowska et al. 2018) and which may have a potentially harmful effect on human health (European Commission 2019).

The numerous Gram-negative bacteria may suggest the presence of bacterial endotoxins at workplaces. Their concentrations were not assessed in the current study; however, in several samples of the inhalable dust fraction from only one cement plant where RDF was burned, the average endotoxin concentration was 46 EU m−3 (SD = 68) (Cyprowski et al. 2023). Therefore, taking into account the currently demonstrated high variability of bacterial aerosol concentrations, it cannot be ruled out that endotoxin concentrations could also vary significantly. It should be noted that, in high concentration, endotoxins may adversely affect the efficiency of the human respiratory system (which is most often manifested by a reduction in spirometric parameters) and result in increased secretion of pro-inflammatory mediators (Liebers et al. 2020). Clarifying this issue would require additional research taking into account various dust fractions to assess the possibilities to penetrate the airways by these particles. Available data from cement plant workers from 8 European countries indicated that the thoracic dust fraction significantly contributed to lung function decline, which was reflected in reduction in forced expiratory volume in 1 s (FEV1) (Nordby et al. 2016).

5 Conclusions

Our study showed that refuse-derived fuels, which are an important source of energy for clinker burning, significantly affect the air quality at studied workplaces. Almost the entire process of handling RDF in cement plants was contaminated with bacterial aerosols. Our study showed that concentrations of airborne bacteria were generally within acceptable limits, but periodic exposure of workers to high concentrations cannot be ruled out. This particularly concerned the closed space of the RDF loader cabin and laboratory rooms, where the threshold limit concentrations for mesophilic bacteria were exceeded. Both RDF and bioaerosol samples showed the presence of numerous bacterial pathogens that may pose a threat to workers due to their infectious and allergenic properties.

Due to the obtained results, employers should consider applying different technical measures to protect workers, e.g. hermetization of the technological process, the efficient ventilation system in confined spaces, local dust extractors limiting the spread of bioaerosols in laboratory rooms or replacement filters in air-conditioning devices in all vehicles used when working with RDF. When performing certain professional activities, such as collecting RDF samples for analysis, employees should use personal protective equipment that protects the respiratory tract against biological agents from the risk group 2. Employers should also provide workers with regular training on biological agents to help develop appropriate work behaviors.