Eco-labeled composts reduce microplastic contamination and mitigate heavy metal bioavailability in agricultural ecosystems

Abstract

This study assessed the abundance, size, weight, shape, and polymer composition of microplastics (MPs) in 11 commercial compost products using ATR-FTIR spectroscopy. MPs were present in all samples, with an average abundance of 137.65 ± 6.01 items/kg_dw, and concentrations up to 631.14 mg/kg_dw. Eco-labeled composts showed significantly lower MPs abundance, size, and concentration than non-eco products. The most contaminated compost was a blonde peat substrate, while an algae-based humus showed the lowest MPs load. MPs larger than 1 mm were predominant, and films were the most common shape, likely resulting from plastic bags and packaging materials. A total of 15 polymer types were identified, with chipboard/agglomerate, modified cellulose, and polyethylene being the most frequent. Polymer diversity was greater in commercial universal composts and positively associated with anthropogenic activities. Physicochemical analysis revealed significant correlations between MPs concentration and compost quality. The presence of MPs was negatively correlated with pH and nitrogen content, but positively with organic matter. Vermicompost exhibited higher nitrogen and pH levels compared to other composts. MPs also influenced the distribution of trace elements: significant negative correlations were found between MPs levels and several elements, including As, Cd, Cr, and Pb. These findings suggest MPs may adsorb heavy metals, reducing their bioavailability. A “diversity index” (DI) of petroleum-based polymers, excluding cellulosic particles, was developed and showed significant correlations with Pb and Cd concentrations, indicating that human-related plastic waste contributes to metal contamination in compost. The results underscore the need for improved waste separation and composting processes to reduce MP and heavy metal contamination. Future regulations mandating separate collection of biowaste and textiles may mitigate these impacts.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-17034-w.

Introduction

Microplastics (MPs) are defined as small plastic particles and fibers with a diameter of 1 µm–5 mm, that originate in the environment not only by fragmentation and weathering of larger plastic debris and fabrics, but also MPs manufactured with this size for industrial purposes; these last referred as primary MPs in contrast to secondary MPs^1,2. These pollutants have been shown to present a significant ecological threat due to their harmful effects, which are linked to their tiny particle size, toxic additives, and the organic and inorganic contaminants they adsorb³, thereby posing a threat to the health of both animals and humans via the food chain⁴. MPs have been detected ubiquitously in aquatic and terrestrial ecosystems⁵, emerging as an increasing environmental concern.

Organic fertilizer, as the byproduct of composting organic matter from different origins⁶, is the best resource for maintaining a healthy and moist soil, together with large quantities of organic matter and nutrients, especially in areas where rainy periods are scarce, and drought threatens the survival of crops⁷. Composts are rich in plant nutrients and organic carbon, enhancing soil fertility by improving soil structure and soil microbial activity, nutrient supply, or soil pH regulation⁸. At the same time, reusing organic waste as a fertilizer is one method to reduce the use of mineral fertilizers and minimize waste disposal in landfill⁶. However, this byproduct has proved to contain a large number of MPs accumulated from human activities, lying a potential threat to general health and ecosystems⁹. Unfortunately, improper waste management combined with the extensive use of plastic leads to significant pollution of compost products with plastic items, which gradually degrade into MPs through physical and mechanical processes and weathering¹⁰. Braun et al.¹¹ have confirmed compost as a major input pathway for MPs into agricultural soil, with the effect still visible after 11 years.

Besides, plastic manufacturing involves the use, not only of polymers or copolymers, but also chemicals intentionally added for specific functions, including additives such as plasticizers, stabilizers, pigments for color, fillers and extenders, flame retardants, blowing agents, antioxidants, impact modifiers, lubricants, and antimicrobial agents^5,12, with a demand of about 5% of the weight of plastic products¹³. Besides, toxic metals such as As, Cd, Cr, Ni, Pb, Sb, Sn, or Zn may remain in plastics as catalytic or reaction residues, i.e., Sb contaminating polyethylene terephthalate bottles¹⁴. As a result, compost enriched with plastic in the form of MPs offers a great variety of additives and toxic metals to the soil environment, also acting as an adsorption medium of organic and inorganic compounds, and microorganisms.

Only a few studies have discussed the potential pollution of farmland MPs resulting from the application of compost^9,15,16. Besides other input pathways of MPs into soil, like the application of sewage sludge from WWTPs, plastic mulching, atmospheric deposition or irrigation previously discussed¹⁷, this paper examines how MPs are distributed across eleven horticultural composts and measures their concentrations, that might affect the soil’s physical and chemical properties. The presence of trace elements, which pose a risk to the ecosystem and human health, is also discussed, together with the effects on soil–plant interactions due to physicochemical characteristics.

This study addresses a critical knowledge gap by systematically analyzing MPs contamination in eleven horticultural compost products. Using a standardized analytical method capable of detecting and quantifying MPs across a wide range of sizes, the study provides novel data on their concentration, composition, and associated trace elements. In the meantime, this work sheds light on the role of eco-labeled composts in mitigating plastic and metal pollutants, while introducing a novel metric, the “diversity index” (DI), to assess polymer complexity. Furthermore, the research explores the potential effects of MPs on soil physicochemical properties and soil–plant interactions. Compared to previous studies (Supplementary Information, Table S1), the inclusion of multiple compost types, trace element analysis, and the focus on both environmental and agronomic implications highlight the unique and comprehensive nature of this work.

Materials and methods

Compost products

Eleven types of Spanish compost brands were used in this study: (CB1) Commercial universal substrate (price: 0.10 €/kg); (CB2) Commercial universal substrate (price: 0.70 €/kg); (CB3) Commercial universal substrate (price: 0.26 €/kg); (CB4) Eco-vermicompost Massó (price: 1.66 €/kg); (CB5) Bokashi seaweed eco-fertilizer (price: 2.40 €/kg); (CB6) Commercial universal substrate for transplanting (price: 0.40 €/kg); (CB7) Commercial universal substrate (price: 0.16 €/kg); (CB8) Commercial universal substrate (price: 0.12 €/kg); (CB9) Blonde peat compost (price: 1.16 €/kg); (CB10) Eco-commercial universal substrate for orchards and fruit trees (price: 0.60 €/kg); (CB11) Eco-substrate for cactus (price: 1.52 €/kg). Compost samples were selected to represent a diverse range of feedstocks, production methods, and intended horticultural applications. Selection criteria included geographic origin, brand diversity, eco-certification, and market availability. Products ranged in price from 0.10 to 2.40 €/kg, encompassing both conventional and eco-labeled substrates (Table S2 and Fig. S1).

Pre-treatment and analysis

Fertilizer samples were first sieved through a 5 mm stainless steel screen to remove coarse particles that could hinder microplastic analysis and then dried at 60 °C for approximately 16 h, or until a constant weight was achieved, in a forced air stove FD 23, to ensure consistent moisture removal across all samples. After that, three independent replicates of each of the eleven compost samples were processed by centrifugation, which offers a convenient and time-saving way to separate materials with different densities¹⁸. Average amounts of 9.0404 ± 0.2699 g of each replicate were processed in three separate aliquots suspended with 10 mL NaCl solution 120 g/L (ρ = 1.08 g/cm³), and agitated on an orbital shaker (250 rpm, 15 min). Following this step, samples were centrifuged in a Z-383 K centrifuge, using a fixed-angle rotor in glass centrifuge tubes (15 mL) (3000 rpm, 10 min, 20 °C), and the supernatant was decanted and placed onto a 120-mm glass Petri dish. The NaCl extraction process was repeated twice, and total supernatants for each sample on petri dishes were placed in the forced air stove previously indicated at 60 °C until total dryness, because higher temperatures could lead to melting or chemical degradation of common polymers¹⁹.

Measurement of MPs and mass calculations

Possible microplastic particles were recorded under an Olympus SZ-61TR Zoom Trinocular Microscope coupled to a Leica MC 190 HD digital camera and an image capturing software Leica Application Suite (LAS) 4.8.0, used for the analysis and recording of color and shape of each microparticle (Fig. S2). Size in its longest dimension was recorded with the assistance of the ridge detection plugin of the open-source ImageJ software before they were isolated in muffled 60-mm glass Petri dishes for further ATR-FTIR analysis. The size and shape variables included fibers, with a length to width ratio > 3, and fragments (generally angular), films (one dimension at least one tenth lower than the other two), or bead (two similar axis), all of them with a length to width ratio ≤ 3²⁰.

All isolated microparticles were analyzed by means of a Thermo Nicolet 5700 Fourier transformed infrared (FTIR) spectrometer, provided with a deuterated triglycine sulfate, DTGS, detector and KBr optics (Fig. S3). The spectra collected by attenuated total reflectance (ATR) were an average of 20 scans with a resolution of 16 cm⁻¹ in the range of 400–4000 cm⁻¹. Spectra were controlled and evaluated by the OMNIC software without further manipulations, and polymers were identified by means of different reference polymer libraries, containing spectra of all common polymers, i.e., Hummel Polymer and Additives (2011 spectra), Polymer Additives and Plasticizers (1799 spectra), Sprouse Scientific Systems Polymers by ATR Library (500 spectra), and Rubber Compounding Materials (350 spectra). A hit quality percentage match > 60% was used as the threshold for polymer identification. Cellulose or cotton-based materials with non-natural colors, i.e., black, blue, or orange, were included in the study, as they evidenced some type of anthropogenic processing.

All MPs that were positively identified were characterized individually by measuring their two orthogonal projected dimensions, referred to as d₁ and d₂. For fragments and films, these correspond to length and width, while for fibers, they represent length and diameter. For nearly isometric particles, it has been shown that the volume-equivalent diameter (d_v) of a sphere, which has the same volume as the particle, can be approximated from projected images using the following formula²¹.

For films, the third smallest dimension was assumed to be one-tenth of the smallest of the two recorded dimensions. Considering the average density value for each analyzed polymer, we then estimated the mass of each individual particle.

Physicochemical parameters, trace element content analysis, and diversity index

Total carbon (TC), total organic carbon (TOC), total nitrogen (TN), and total inorganic carbon (TIC) contents were analyzed via combustion (900 °C) using a MULTI N/C 3100 analyzer (Fig. S4). Total inorganic carbon was calculated based on the difference between total carbon and total organic carbon. Measurements of pH were performed with a Crison GLP22 pH-meter, calibrated by means of standard solutions, and the elemental analysis of compost products was carried out using an CHNS/O 628 Series Elemental Analyzer (Fig. S5).

Compost samples were also analyzed for eighteen trace elements that are proxies for important additives in plastics or potentially toxic contaminants (As, Ba, Bi, Cd, Co, Cr, Cu, Fe, Mo, Ni, Pb, Sb, Sn, Sr, Ti, V, Zn)²². A microwave oven digestion procedure was carried out by an UltraWAVE ECR Microwave Digestion System (Fig. S6) by weighing 0.5033 ± 0.0009 g of the sieved compost (0.50 µm mesh) and placing it into a Teflon microwave digestion vessel, with 3 mL of concentrated HNO₃ (69%) and 1 mL of concentrated HCl (37%) (reverse aqua regia), after a final extract volume of 50 mL with ultrapure water (Type I) (See Table S3 in Supplementary Information for cycles).

Digested samples were then filtered and 1:5 dilutions were made prior to the analysis of trace element content using inductively coupled plasma mass spectrometry (ICP-MS) (Fig. S7). The ICP-MS was equipped with standard nickel sampling and skimmer cones, a standard glass nebulizer fitted onto a Peltier-cooled Scott-type spray chamber, and a 2.5 mm diameter injector of a one-piece quartz torch. UHMI (Ultra High Matrix Introduction) dilution was automatically performed with argon gas, before reaching the interface, thus avoiding clogging problems.

The “diversity index” (DI) was calculated as the total number of distinct petroleum-based polymer types identified in each compost sample, excluding cellulosic-based microplastics (i.e., cellulose, cellophane). Formally, the index is defined as:

where P_i = 1 if the ith petroleum-based polymer type is present in the sample, and P_i = 0 otherwise, and n is the total number of petroleum-based polymer types considered in the study.

Quality assurance and quality control (QA/QC)

Basic contamination control measures were employed to reduce the risk of pollution, including the use of clothes made of natural fabric and clean cotton lab coats by the analysts. All stainless steel and glass materials were thoroughly rinsed with Type II (RO/DI) water, wrapped with aluminum foil, and heated at 150 °C for 5 h before each experiment, in order to remove all possible rests of plastic contamination. Type I ultrapure water, obtained from a MilliQ Integral water purification system equipped with a 0.20 µm end-filter, was used for analysis (Fig. S8), and NaCl solutions were prepared inside a laminar flow bench class II to avoid cross-pollution from the lab environment.

Procedural sampling controls were used during the whole study in order to control MPs airborne pollution, also acting as negative control samples or procedural blanks. Eleven 60-mm glass Petri dishes were kept open during all analytical processes, one for each compost sample, as well as two solvent blanks of NaCl solution 120 g/L (ρ = 1.08 g/cm³), that were analyzed throughout the entire study. MPs identified in control samples were used as background data to efficiently calculate the MPs content. Only two fibers were collected and subtracted from the corresponding samples, corresponding to modified cellulose (170 µm) and acrylate (540 µm) (Fig. S9).

Statistical analysis of the experimental data

Statistical treatment of the data was accomplished using the SPSS (Statistic Package for Social Science) 26.0 statistic software. The fitting performance of one-way analysis of variance (ANOVA) was computed with an F-test, and Fisher’s Least Significance Difference (LSD) test was applied when F-test reported rejection of null hypothesis (H₀) to compare paired data and identify statistically significant differences. The interquartile range (IQR) was used as a measure of statistical dispersion, defined as the difference between the 75th and the 25th percentiles of the data. Prior to running an ANOVA, data were tested for normality through Kolmogorov–Smirnov’s (K–S) test. Since all datasets met the assumptions of normality, no data transformations were necessary. All data were reported as the mean ± standard error (SE). Possible intercorrelations between different variables were assessed using Pearson’s correlation coefficient (r). All statistical analyses were considered statistically significant at a 95% confidence interval (p < 0.05).

Results and discussion

Abundance, size, weight, and type of MPs in compost samples

Microparticles selected and identified using ATR-FTIR resulted in 59.06% MPs, 26.17% as non-plastic particles, such as bentonite or silk, and 14.77% as unidentified particles. MPs were observed in all compost products, with an average abundance of 137.65 ± 6.01 items/kg_{dry weight (dw)} (IQR 50.17 items/kg_dw). These figures were within the range reported by Weithmann et al.²³ in organic fertilizer converted from biosolid (14–895 items/kg_dw) and lower than those proposed by Gui et al.²⁴ in composting of rural domestic waste (2533 ± 457 items/kg_dw), or Iswahyudi et al.²⁵ in commercial compost from Indonesia (800 items/kg). Braun et al.¹⁵ found concentrations of MPs ranging from 12 to 46 items/kg_dw in eight different composts, after sample separation in ZnCl₂ and subsequent peroxidation. In our study, the highest average abundance for MPs was for the blonde peat (CB9, 338.69 ± 4.69 items/kg_dw), and the lowest one was for an ecological humus obtained from agarophytic algae through a composting process (CB5, 73.62 ± 2.43 items/kg_dw) (see Table S4 for post-hoc analysis). In fact, the eco-products proved to have a lower average abundance of MPs (98.79 ± 5.81 items/kg_dw) than non-eco ones (154.85 ± 7.28 items/kg_dw) (F-test = 23.245, p = 0.000) (Fig. 1a). Zimmermann et al.²⁶ reported a release of up to 9 times less MPs under environmental stress for plant-based than conventional plastics, and Courtene-Jones et al.²⁷ found that eco-textile products shed fewer fibers than synthetic polyester, although even eco-products may release MPs to the environment²⁸.

Fig. 1.

Comparison of (a) average microplastic concentration, (b) average microplastic size, and (c) compost prices (mean ± standard error of the mean) (red: non-eco compost products, green: eco compost products). Different letters mean significant differences (p < 0.05).

In terms of concentration, the calculated average value of MPs in compost samples was 67.89 ± 13.09 mg/kg_dw (IQR 57.85 mg/kg_dw), with a maximum value of 631.14 mg/kg_dw corresponding to CB9. Bläsing and Amelung²⁹ reported an abundance of 2.38–180 mg/kg of plastic fragments > 0.5 mm in organic fertilizer. Our commercial universal substrate CB3 contained the highest average microplastic concentration (269.03 ± 112.99 mg/kg_dw), followed by blonde peat compost (CB9) (178.42 ± 151.63 mg/kg_dw), almost twice as high as the third average microplastic concentration (CB1: 77.41 ± 20.41 mg/kg_dw). The lowest one (10.86 ± 2.67 mg/kg_dw) was found in CB10, the eco-commercial universal substrate for orchards and fruit trees. Eco products displayed a statistically significant lower microplastic concentration (26.45 ± 15.23 mg/kg_dw) than non-eco ones (85.56 ± 17.07 mg/kg_dw) (F-test = 4.445, p = 0.038), as previously indicated for microplastic abundance. Although microplastic abundance and concentrations in compost samples were positively correlated (r = 0.370, p = 0.000), the high variability on plastic densities meant they were not exactly coincident.

The smallest microplastic size was observed in CB5, the seaweed eco-fertilizer (70 µm), similar to that described by Gui et al.²⁴, i.e., 0.05 mm, with also the lowest average size (794.29 ± 215.34 µm). Moreover, eco-fertilizers proved to have a statistically significant lower microplastic size (905.77 ± 104.75 µm) than non-eco products (1641.48 ± 110.20 µm) (F-test = 16.260, p = 0.000) (Fig. 1b). The highest average size corresponded to CB9 (2222.50 ± 595.69 µm), the compost product with also the highest abundance of microplastic (Fig. 1a). MPs with a size > 1 mm accounted for more than 60% of the total MPs, being between 1 and 2 mm the most common size (42.50%) (Fig. 2a) (see Table S5 for post-hoc analysis).

Fig. 2.

Accumulated percentages across the compost samples by: (a) size categories based on Spanish Environmental Ministry classification; (b) MPs shapes; (c) colors; (d) polymer types identified by ATR-FTIR (outer rings: commercial universal substrates, inner ring: all others).

There was also a statistically significant bivariate correlation between compost price and abundance of MPs, indicating a low average concentration of MPs for compost with high prices (r =  − 0.278, p = 0.009) (Fig. 1c). While this suggests a general trend in which lower-priced composts tend to contain higher levels of microplastics, the strength of the correlation indicates that price alone is not a reliable predictor of compost quality in terms of MPs contamination. Interestingly, significantly discounted prices combined with improved compost quality have been shown to increase the likelihood of purchase³⁰. This highlights a potential trade-off between affordability and environmental safety that consumers may face. To address this, future regulations could consider requiring the disclosure of MPs content on compost packaging, enabling more informed decision-making and encouraging cleaner production practices.

Furthermore, commercial universal composts made from biowaste, i.e., CB1, CB2, CB3, CB6, CB7, and CB8, displayed a higher average abundance of MPs than the other compost products, this is 141.99 ± 3.96 items/kg_dw versus 129.66 ± 15.50 items/kg_dw, although differences were not statistically significant (p = 0.330). This fact could be due to a high intra-specific and inter-specific variability for the compost samples, together with an improper waste disposal of plastic materials in biowaste, especially in Spain, where a mixed waste organic output is still produced at home³¹. Our future waste regulation will modify the garbage collection models, since from 2026 the separate collection of organic fraction (biowaste) and textiles will be mandatory. With the implementation of clean kitchen waste collection systems, which ensure that organic waste is separated from plastics and other non-biodegradable materials, the contamination of compost by MPs is likely to decrease³².

The MPs shape varied among different brands, with average percentage values of 36.36% for films, 26.14% for fragments, 26.14% for fibers, and 5.68% for beads (Fig. 2b) (Fig. S10), and a proportion of colored MPs in compost products that exceeded 60% (Fig. 2c). The presence of various shapes of MPs in compost underscores the variety of plastic waste being processed. It is likely that films are secondary MPs resulting from the breakdown of packaging materials and plastic bags, especially from torn garbage bags in mixed waste³³. In fact, the amount of film was higher for commercial universal substrate composts (36.36%) than for the other ones (26.63%). As reported by Wan et al.³⁴, films significantly increase the rate of soil water evaporation, which highlights the potential risk of these MPs when compost is used as fertilizer.

Polymer identification in compost products

A total of 15 different types of MPs polymers were identified using ATR-FTIR (≥ 20 µm) (Fig. 2d), including acrylate (ACR), chipboard or agglomerate (AG), modified cellulose (CEL), cellophane (CP), tetrafluoroethylene hexafluoropropylene copolymer (FEP), melamine-urea–formaldehyde resin (MUF), polycaprolactone (PCL), polyethylene (PE), polypropylene-ethylene copolymer (PEP), polyester (PES), polyethylene terephthalate (PET), methacrylate (PMMA), polypropylene (PP), polyurethane (PUR), and polyvinyl chloride (PVC). Figure 3 depicts distinctive absorption peak band examples for six identified polymers, and Table S6 (Suppl. Inf.) collects the identification of vibrational modes in the ATR-FTIR spectra of 15 specific polymers.

Fig. 3.

Representative ATR-FTIR spectra of different polymers found in compost products: (a) CB1: acrylate (ACR) (match: 70.80%), (b) CB5: polyethylene (PE) (match: 83.70%), (c) CB8: polypropylene-ethylene copolymer (PEP) (match: 79.68%), (d) CB1: polyester (PES) (match: 80.78%), (e) CB1: polyethylene terephthalate (PET) (match: 81.30%), (f) CB8: polypropylene (PP) (match: 92.96%).

The most abundant polymers were AG and CEL, especially in non-eco products (25% and 12.5%, respectively), compared to eco composts (2.27% and 10.23%, respectively). The occurrence of chipboard or agglomerate polymer may be attributed to human-induced activities, such as residential construction and furniture production³⁵. While these materials are not conventional thermoplastics, they are increasingly found in compost due to the disposal of furniture waste, construction debris, or laminated packaging. Therefore, 83.33% of this polymer was isolated in commercial universal substrates, while other polymers such as CEL (55%), CP (50%) or ACR (50%) were unevenly distributed throughout all the compost products. Together with PE, these five polymers accounted for 77.27% of the plastic material found in compost samples. CP is widely used in food packaging products, labels, cigarette wrappers, and as a release agent in the manufacture of fiberglass and rubber products, being reported as the most common MPs in Chinese table salts from lakes, rocks, and wells³⁶ and in the marine waters of Qatar³⁷. Figure 2d displays the distribution of polymer types for commercial universal substrates and all others, and Fig. S11 (Suppl. Inf.) depicts polymer distribution according to MPs shape. CEL was mainly represented in fibers, and the highest variety of polymers was displayed in fragments, including MUF, PES, PET, PUR, and PVC, not shown in the other three shapes of MPs. They are probably due to weathering fragmentation of mesoplastics and larger fragments of plastic litter into the compost. The inadequate separation of plastic materials from green and food waste before composting may lead to the presence of PVC-containing items, such as irrigation pipes, synthetic leather, flooring remnants, and garden tools to enter composting systems. The large number of colored fragments found in compost products would come from worn or broken common food and drink packaging applications, like bottles, jars, tubes, blisters, shopping bags, snack food wrappers, films from frozen foods, bakery products, caps, cups, covers, container labels, stirrers, cutlery, yogurt containers, and packaging for eggs, among others^23,29,38,39, that over time can break down into smaller fragments in organic waste streams, including compost. Iswahyudi et al.²⁵ mainly found PET in the analysis of a single compost purchased from a local agricultural store in Indonesia, concluding its extensive use and hypothesizing that PET in compost came from village household garbage.

Effect of MPs on compost physicochemical characteristics and trace elements

The main physicochemical parameters of compost products are shown in Table 1. A statistically significant decrease in compost pH values were observed with an increase in MPs concentration (r =  − 0.583, p = 0.000), as previously shown by different authors^40–43. Boots et al.⁴⁴ reported a significant decrease in soil pH exposed to PE that could be attributed to a mineralization process, because MPs can improve the O₂ supply⁴⁵. Moreover, a pH reduction implies a reduction in the negative charge of clay minerals, with a decrease on heavy metals adsorption on their surfaces and increased mobility⁴¹. Besides, organic matter in compost proved to increase with MPs abundance (r = 0.647, p = 0.000), and compost pH was found to be negatively correlated with the content of organic matter (r =  − 0.756, p = 0.000). Meng et al.⁴⁶ also found that the addition of MPs to the soil linearly increased organic matter, determining the contribution of plastic to soil carbon¹⁵.

Table 1.

Physicochemical parameters of the compost products.

Compost	pH	C (%)	H (%)	N (%)	IC (%)	TOC (%)	TC (%)	OM (%)	CaCO3 (%)	DI
CB1	7.50	23.68	2.80	1.14	4.02	19.66	23.68	33.89	33.48	10
CB2	5.75	41.45	5.36	1.21	3.62	37.82	41.44	65.21	30.18	3
CB3	7.00	30.89	4.57	1.12	4.24	26.65	30.89	45.94	35.32	2
CB4	8.70	18.33	2.89	1.62	0.86	19.19	20.05	33.07	7.16	4
CB5	7.10	11.47	1.78	1.93	0.86	10.61	11.47	18.28	7.18	4
CB6	5.80	40.15	4.41	1.05	3.84	36.31	40.15	62.60	31.96	3
CB7	7.00	25.81	3.45	1.00	3.14	22.67	25.81	39.09	26.14	3
CB8	7.00	22.43	2.77	1.33	2.87	19.56	22.43	33.72	23.90	4
CB9	4.80	48.06	5.91	1.13	0.13	47.93	48.06	82.64	1.08	2
CB10	5.75	37.98	3.79	1.23	7.05	45.03	52.08	77.62	58.75	4
CB11	6.60	37.42	4.19	0.53	16.18	21.24	37.42	36.61	34.86	1

IC, Inorganic carbon; TOC, Total organic carbon; TCN, Total carbon; OM, Organic matter; DI, Diversity index.

There was also a statistically significant inverse correlation between MPs concentration and total nitrogen in the compost (r =  − 0.482, p = 0.000). The average percentage of nitrogen proved to decrease from 1.54 ± 0.07% with low MPs concentrations in the compost (lower than 100 MP/kg_dw) to 1.13 ± 0,01% in samples with the highest abundance of MPs (more than 300 MP/kg_dw) (F-test = 23.390, p = 0.000). Similar results have been reported in our research group for wastewater samples¹⁷. Xiang et al.⁴⁷ also indicated a significant decrease in soil nitrate concentrations with MPs abundance, as they have proved to exacerbate soil C:N imbalance. MPs, largely composed of carbon, increase the carbon pool in soils, although this carbon does not break down in the same way organic matter does, which leads to an imbalance in the C:N ratio. Qin et al.⁴⁸ also reported the adsorption of NO₃^–, NO₂^–, and NH₄⁺ by PVC MPs, especially after a photoaging process. This process has been previously discussed, because of an easy hydration of carbonyl groups in plastics, that could adsorb and hence sequester available nutrients⁴⁹. In addition to these mechanisms, it is important to consider that MPs may exert inhibitory effects on soil microbial activity. The presence of MPs has been shown to alter the structure and diversity of microbial communities, affecting key processes such as nitrogen mineralization⁵⁰. This microbial inhibition could reduce the availability of inorganic nitrogen forms, thereby contributing to the observed negative correlation. Therefore, the combination of dilution, adsorption, C:N imbalance, and microbial inhibition provides a more comprehensive explanation of the phenomenon.

The percentage of nitrogen was statistically higher by 26% in vermicompost (CB4) compared to the rest of the products (F-test = 24.287, p = 0.000), with also a statistically significant higher average pH, i.e., 8.70 versus 6.74, respectively (F-test = 138.786, p = 0.000). The key factors contributing to this increased nitrogen content include: (1) a lower C:N ratio than traditional compost⁵¹; (2) physical fragmentation of organic matter by earthworms, which increases its surface area, thereby enhancing microbial access and accelerating decomposition⁵¹; (3) nitrogen transformation by diverse microbiota of earthworms’ gut, that convert organic nitrogen into more readily available forms for plants⁵²; and (4) the excretion of nitrogenous materials, such as mucus and castings⁵³. Although these mechanisms are well supported in the literature, no direct measurements of earthworm activity were conducted in this study, and their contribution remains inferred. Nitrogen transformation and decomposition of organic matter increases the pH of the vermicompost compared to traditional compost⁵⁴. On the contrary, a statistically significant lower Ca content in vermicompost (CB4) than in the other products (more than 400%) could be caused by the accumulation of Ca into the Eisenia fetida tissue⁵⁵.

The composition of trace elements is shown in Table 2. Statistically significant inverse bivariate correlations were established between most trace elements and MPs concentration in compost samples, i.e., As (r =  − 0.625, p = 0.000), Ba (r =  − 0.463, p = 0.000), Bi (r =  − 0.344, p = 0.001), Cd (r =  − 0.510, p = 0.000), Co (r =  − 0.654, p = 0.000), Cr (r =  − 0.739, p = 0.000), Cu (r =  − 0.314, p = 0.040), Fe (r =  − 0.543, p = 0.000), Mn (r =  − 0.555, p = 0.000), Ni (r =  − 0.661, p = 0.000), Pb (r =  − 0.344, p = 0.001), Sb (r =  − 0.345, p = 0.001), Sr (r =  − 0.271, p = 0.011), Ti (r =  − 0.477, p = 0.000), V (r =  − 0.548, p = 0.000), and Zn (r =  − 0.291, p = 0.006). MPs have proved to play a significant role as an adsorbent of toxic metals, decreasing their availability in soils^9,56. Dong et al.⁵⁷ found that As bioavailability in soil decreased because of heavy metal interaction with polystyrene and FEP, and Premarathna et al.⁵⁸ reported As uptake by plants decreased when As and MPs coexist. It was also observed, for non-essential anthropogenic metals like Cd, a significant correlation with the main morphological characteristics of MPs, especially for size (r =  − 0.257, p = 0.016). Cadmium enters the air and bind to small particles, atmospheric deposition being a common source of pollution in the general environment^16,59.

Table 2.

Trace element concentrations in the compost products (mean ± standard deviation).

Compost	As (ppb)	Ba (ppb)	Bi (ppb)	Cd (ppb)	Co (ppb)	Cr (ppb)	Cu (ppb)	Fe (ppm)	Mn (ppb)
CB1	5.99 ± 3.70	137.32 ± 1.40	0.23 ± 0.01	0.43 ± 0.09	3.98 ± 1.00	21.19 ± 1.00	65.06 ± 0.80	9.29 ± 0.18	247.35 ± 2.80
CB2	2.31 ± 8.60	43.88 ± 0.80	0.14 ± 0.50	0.66 ± 0.51	1.74 ± 0.38	12.44 ± 2.40	44.62 ± 2.10	2.89 ± 0.13	250.99 ± 2.50
CB3	3.79 ± 1.90	129.68 ± 0.30	0.21 ± 0.15	0.83 ± 0.49	3.27 ± 0.10	18.47 ± 1.20	88.12 ± 1.10	7.29 ± 0.70	397.72 ± 1.40
CB4	8.18 ± 1.00	822.16 ± 3.10	0.35 ± 0.23	1.51 ± 0.43	7.67 ± 0.80	27.01 ± 1.20	62.48 ± 1.60	17.17 ± 1.50	708.42 ± 2.10
CB5	13.50 ± 2.30	124.78 ± 0.90	0.13 ± 0.06	1.64 ± 0.50	4.39 ± 1.40	20.96 ± 1.50	38.73 ± 0.70	7.57 ± 1.10	562.39 ± 2.40
CB6	4.33 ± 2.60	128.32 ± 0.60	0.16 ± 0.14	0.53 ± 0.07	2.22 ± 1.30	15.60 ± 0.60	53.25 ± 1.50	7.78 ± 0.90	264.10 ± 1.00
CB7	5.32 ± 2.00	159.22 ± 2.10	0.21 ± 0.18	0.58 ± 0.08	4.10 ± 0.90	19.28 ± 2.60	76.30 ± 1.80	9.46 ± 1.60	603.36 ± 1.10
CB8	6.96 ± 1.30	587.02 ± 1.30	0.76 ± 0.07	0.60 ± 0.09	4.98 ± 1.80	23.81 ± 0.90	114.22 ± 0.80	7.82 ± 1.20	388.00 ± 130
CB9	1.60 ± 0.50	25.88 ± 0.90	0.08 ± 0.04	0.30 ± 0.06	0.47 ± 0.30	2.54 ± 1.50	3.67 ± 2.20	1.32 ± 0.40	46.46 ± 0.50
CB10	2.92 ± 0.55	86.06 ± 1.50	0.18 ± 0.04	0.49 ± 0.07	2.96 ± 2.20	14.97 ± 1.60	26.51 ± 2.00	6.05 ± 2.50	286.28 ± 1.40
CB11	3.42 ± 2.40	50.76 ± 0.80	0.12 ± 0.03	0.08 ± 0.08	1.79 ± 1.60	11.78 ± 1.20	9.11 ± 1.70	5.36 ± 0.60	87.30 ± 0.90

Compost	Mo (ppb)	Ni (ppb)	Pb (ppb)	Sb (ppb)	Sn (ppb)	Sr (ppb)	Ti (ppb)	V (ppb)	Zn (ppb)
CB1	1.65 ± 0.35	11.32 ± 0.80	29.06 ± 2.00	0.80 ± 0.70	2.17 ± 2.10	578.28 ± 0.80	108.48 ± 1.80	20.67 ± 2.50	181.95 ± 0.50
CB2	1.98 ± 0.23	4.62 ± 0.90	34.80 ± 0.60	0.82 ± 0.49	2.82 ± 1.60	135.68 ± 1.20	103.19 ± 2.90	33.43 ± 0.60	118.16 ± 2.10
CB3	3.34 ± 2.20	12.22 ± 0.40	9.88 ± 0.60	0.41 ± 0.32	2.65 ± 2.40	234.09 ± 1.90	248.19 ± 0.80	22.22 ± 0.50	555.91 ± 1.20
CB4	5.52 ± 1.40	17.60 ± 2.00	53.43 ± 0.60	1.08 ± 0.26	4.12 ± 2.00	137.50 ± 1.40	291.77 ± 0.40	58.75 ± 0.80	513.67 ± 1.10
CB5	0.69 ± 0.53	13.99 ± 0.90	28.28 ± 1.20	0.48 ± 0.22	2.60 ± 0.60	732.22 ± 1.30	175.01 ± 1.80	30.39 ± 1.20	293.13 ± 0.70
CB6	10.05 ± 0.80	6.35 ± 1.10	26.24 ± 1.70	0.59 ± 0.44	3.69 ± 2.10	75.86 ± 1.60	130.57 ± 2.60	15.68 ± 2.00	119.49 ± 2.10
CB7	4.98 ± 1.00	10.01 ± 1.00	12.18 ± 1.90	0.42 ± 0.42	3.40 ± 1.60	239.56 ± 0.70	233.01 ± 1.10	26.31 ± 0.70	623.94 ± 1.60
CB8	2.96 ± 0.70	11.81 ± 1.00	147.01 ± 2.40	1.92 ± 0.22	4.20 ± 1.70	252.33 ± 1.50	292.73 ± 2.00	17.51 ± 1.40	399.03 ± 0.70
CB9	0.63 ± 0.42	1.54 ± 1.50	14.68 ± 1.50	0.46 ± 0.20	3.02 ± 2.50	57.11 ± 1.10	53.60 ± 1.80	3.19 ± 0.60	23.89 ± 2.40
CB10	2.04 ± 1.60	7.43 ± 2.70	33.63 ± 0.20	0.82 ± 0.23	3.82 ± 1.80	93.95 ± 1.10	168.93 ± 0.90	19.27 ± 1.60	122.12 ± 1.00
CB11	1.29 ± 0.19	7.11 ± 1.40	5.28 ± 0.90	0.31 ± 0.41	1.51 ± 0.27	30.94 ± 0.70	117.21 ± 1.80	12.71 ± 1.70	34.71 ± 1.00

The “diversity index” (DI) was defined to assess the variety of petroleum-based polymers in compost products, while excluding contamination from the widely prevalent cellulosic-based MPs¹⁶. Values of DI are depicted in Table 1, with a maximum of 10 for CB1 and a minimum value of 1 for CB11. A statistically significant log correlation was found between Pb and Cd concentrations in compost samples and the diversity index, as depicted in Fig. 4. After the exclusion of ubiquitous contaminants, such as cellulosic MPs, the correlation between these two heavy metals and DI suggest the importance of human activities in MPs pollution, as they can be found in trace amounts in soil, water, and air. The correlation between DI values and Pb and Cd pollution indicates the heavy metals origin from general environmental contamination, although specific components of MPs, such as PVC, may also be involved, as Pb has commonly been used as a stabilizer to prevent thermal and UV degradation in PVC products⁶⁰. The DI does not incorporate effects of polymer aging or additive leaching, which may influence metal adsorption and this limitation should be considered when interpreting the results.

Fig. 4.

Diversity index of MPs versus the concentration of (a) lead, and (b) cadmium in the eleven analyzed compost products. The best fit logarithmic line through the data are shown, together with the F-tests.

While this study provides valuable insights into the presence and characteristics of microplastics in compost products, it is important to acknowledge that all analyzed samples were sourced from Spanish brands. This geographic limitation may affect the generalizability of the findings to other regions with different composting practices, waste management systems, and regulatory frameworks. To enhance the global relevance of the results, future research should consider expanding the sample set to include compost products from a broader range of countries and production systems. Comparative studies across diverse geographic and socio-economic contexts would help determine whether the observed trends, such as the lower microplastic content in eco-labeled composts, are consistent internationally. Additionally, an assessment of the representativeness of the selected Spanish composts in terms of market share, production methods, and feedstock diversity would further clarify the extent to which these findings can be extrapolated.

Conclusion

This study confirms the widespread presence of microplastics (MPs) in Spanish compost products, with eco-labeled composts showing significantly lower levels of MPs in terms of abundance, concentration, and particle size. The most common polymer types were chipboard, modified cellulose, and polyethylene, with films and fragments being the dominant shapes. MPs were found to influence compost quality by reducing nitrogen content and pH, while increasing organic matter. Additionally, MPs were inversely correlated with trace element concentrations, suggesting their role in adsorbing heavy metals.

To mitigate these impacts, we recommend the following management strategies:

1. Establish regulatory thresholds for allowable MPs content in diverse types of compost products, particularly those intended for agricultural use.

2. Strengthen source control by improving waste separation practices and excluding plastic contaminants from organic waste streams.

3. Implement stricter quality control during compost production, including routine monitoring of MPs and associated trace elements.

4. Promote eco-certified composts, which demonstrated lower contamination levels and better physicochemical profiles.

5. Incorporate MPs monitoring into eco-label certification criteria, ensuring that environmental claims are backed by measurable standards.

6. Encourage public awareness and education on the environmental risks of plastic pollution in compost and the benefits of using certified products.

7. Support further research on the long-term effects of MPs in soil ecosystems, including their interaction with microbial communities and nutrient cycling.

8. Develop standardized analytical methods for MPs detection in compost to ensure comparability across studies and regulatory compliance.

These measures will help reduce the environmental risks associated with MPs in compost and support the development of safer and more sustainable organic fertilizers.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: All authors; Methodology: J.B. and J.L.-C.; Writing: J.B.; Data acquisition: J.L.-C., M.D.-M. and S.O.; Visualization: All authors: All authors reviewed the manuscript.

Data availability

For ethical and proprietary reasons related to the nature of the data and our collaborators’ interests, we are unable to share the raw dataset. Nevertheless, extensive information on average values for all parameters is provided in Tables S2, 1 and 2. Point of contact: javier.bayo@upct.es.

Declarations

Competing interests

The authors declare no competing interests.