Evaluation of different composting systems on an industrial scale as a contribution to the circular economy and its impact on human health

ABSTRACT Due to the production of volatile organic compounds (VOCs), large-scale composting can cause air pollution and occupational health issues. Due to this, it is necessary to determine if the amount generated poses a health risk to plant workers, which can be a starting point for those in charge of composting plant facilities. As a result, the goal of this work is to conduct a thorough analysis of both the physicochemical features and the VOC generation of three large-scale systems. For ten weeks, the three different composting plants were monitored weekly, and VOC identification and quantification were performed using GC-MS gas chromatography. It has been observed that the biggest risk related with VOC formation occurs between the fourth and fifth weeks, when microbial activity is at its peak. Similarly, it has been demonstrated that xylenes and toluene are the ones that are produced in the greatest quantity. Finally, after ten weeks of processing, it was discovered that the material obtained complies with the regulations for the sale of an amendment. Implications: The evaluation and monitoring of the composting processes at an industrial scale is very important, due to the implications they bring. VOCs are produced by the operation of composting facilities with substantial amounts of solid waste, such as the companies in this study. These may pose a health risk to those working in the plants; thus, it is critical to understand where the VOCs occur in the process in order to maintain workers’ occupational health measures. This form of evaluation is rare or nonexistent in Colombia, which is why conducting this type of study is critical, as it will provide crucial input into determining when the highest levels of VOC generation occur. These are the ones that may pose a risk at some point, but with proper occupational safety planning, said risk may be avoided. This work has evaluated three composting systems, with different types of waste and mixtures. According to reports, while composting systems continue to produce VOCs and their generation is unavoidable, the potential risk exists only within the plant. These findings can pave the way for the implementation of public policies that will improve the design and operation of composting plants. There is no specific legislation in Colombia for the design and execution of this sort of technology, which allows the use of organic waste


Introduction
Solid urban wastes sustainable operation is a target aspect for global environmental planning and management as prevents negative impacts of improper waste disposal such as surface and underground water pollution, soil contamination, gas and odors emission, generation of leachate, proliferation of vectors and the effects on human health (Cheremisinoff 2003).The global generation of solid urban waste reached approximately 9050 million tons in 2018, and by the year 2050, an increase of 70% is expected (Guo et al. 2012;Tiseo 2022).Approximately 60% of the waste generated corresponds to the organic fraction, however, most of this fraction ends up in final disposal sites such as sanitary landfills, which is not desired.When organic waste is dumped in sanitary landfills, it undergoes anaerobic digestion mainly where a mixture of gases composed mainly of carbon dioxide and methane is released along with traces of other compounds such as hydrogen sulfide, nitrogen, and moisture (Ghosh et al. 2020;Gupta et al. 2022;Sahota et al. 2018).
Organic waste separation and recycling methods are the most accessible and cost-effective options in developing countries (Buratti et al. 2015;Kharola et al. 2022;Lin et al. 2018;Ng, Yang, and Yakovleva 2019).This is due to the lack of capacity and the inconsistency to build CONTACT Juan F. Saldarriaga jf.saldarriaga@uniandes.edu.coDepartment of Civil and Environmental Engineering, Universidad de Los Andes, Carrera 1Este # 19A-40, Bogotá 111711, Colombia.Supplemental data for this paper can be accessed online at https://doi.org/10.1080/10962247.2023.2235299.
more complex systems for the solid waste recovery and valorization (Westerman and Bicudo 2005).
Composting is one of the main alternatives for integrative urban solid waste management and the organic fraction treatment for presents advantages due to its technical and economic feasibility, can be developed at different scales, generates a useful product, and increases the final disposal sites life cycle (Díaz, de Bertoldi, and Bdlingmaier 2007).It is an economically feasible and environmentally sustainable technique that can be used in the effective treatment of solid waste worldwide (Duan et al. 2022;Mahapatra, Ali, and Samal 2022).However, there is less information about co-composting different types of municipal waste sources on a large industrial scale.It is still important to improve the industrial composting process to reduce its environmental impact and align it with the goals of the circular economy.As a result, industrial-scale examinations of composting methods are required, in which not only the physicochemical properties and how these effect the final product are studied, but also how they favor or not the formation of VOCs (Zhou et al. 2022).
Tracing operating parameters is critical for monitoring and controlling the composting process on an industrial scale, where it is necessary to develop studies that allow simple parameter selection over time and help predict the behavior of the process's other physicochemical and microbiological variables (Saldarriaga et al. 2019).Regarding the emission of environmental pollutants, one of the negative aspects of composting is attributed to the generation of volatile organic compounds (VOCs), which represent a potential risk to health and possible effects on air quality at different stages of process (Diaz, de Bertoldi, and Bdlingmaier 2007;Nadal et al. 2009;Saldarriaga, Aguado, and Morales 2014).These environmental consequences are typically regarded as prospective since their occurrence is more visible when process management and controls are inadequate, emphasizing the significance of identifying key operating factors linked with VOC creation (Saldarriaga, Aguado, and Morales 2014).It has been found that the greatest number of VOCs is generated during the stages of greatest microbial activity in the process of composting solid urban waste (Saldarriaga, Aguado, and Morales 2014).Also, VOC emissions are mainly related to deficiencies in aeration, characteristics of the starting material, temperature, and moisture content in the system (Beck-Friis et al. 2001;Getahun et al. 2012;Saldarriaga, Aguado, and Morales 2014).The study of VOCs is of special interest, since there are hundreds of compounds classified as such that generate harmful effects on human health and the environment, like BTEX (benzene, toluene, ethylbenzene, and xylene) (Hester, Harrison, and Derwent 1995).Most investigations have concentrated on identifying and quantifying VOCs, with BTEX standing out as a risk to human health; however, few studies have evaluated large-scale composting operations (Domingo and Nadal 2009;Eitzer 1995;Nadal et al. 2009;Saldarriaga, Aguado, and Morales 2014).
It is therefore fundamental to conduct investigations that allow for the correlation of microbial activity with physicochemical parameters and VOC generation in industrial facilities.This permits for the identification of the sort of VOC that can be formed based on the waste being processed and the type of composting method used (Chen, Kim, and Jiang 2018).Industrialscale composting processes must guarantee physicochemical and microbiological parameters to obtain a final product useful as an amendment, which can be marketed.As a result, the aim of this research is to evaluate three solid waste composting companies by analyzing the process's physicochemical and microbiological properties, in the same way, the VOCs produced during the composting operations will be identified and quantified.Both the VOCs and the physicochemical and microbiological properties will be correlated, to identify which of these may influence the generation of VOCs.Finally, the potential health risk that these VOCs may pose has been determined.

Compost systems
Three companies located in Bogotá, Colombia were selected, in which composting processes are developed on an industrial scale.Composting plants are distinguished by the fact that they handle an average of 20 tons of waste every week.The three facilities differ in that they treat a variety of wastes, but they all process solid urban garbage from the plants' influence areas.The first facility processes rumen waste (RUM), which is blended with pine sawdust to minimize moisture, as well as solid urban garbage.The second plant treats chicken manure mixed with solid urban garbage from a poultry plant (CHM).The third plant handles municipal solid waste, 60% of which is derived from pruning, fruit, and vegetable waste, and is mixed with municipal solid trash.The composition of the starting material for each system (RUM, CHM and USW), percentage of composition and the amount of waste treated is described in Table 1.Most of the piles are 20 m × 3 m x 1 m in size, the procedure is mechanized using specialized technology, and the piles are turned every two days.

Sampling, physicochemical, microbiological, and phytotoxicity analysis
Samples were taken weekly, for a total of 10 samples for each system (RUM, CHM, and USW).A composite sampling was carried out with an amount of approximately 1 kg of material for each sample, later all the subsamples were mixed to obtain the composite sample for each system for each week.The samples were transported to the laboratory of the Environmental Engineering Research Center of the Universidad de Los Andes, to carry out the physicochemical and microbiological analyses, phytotoxicity tests and the extraction and quantification of VOCs.The collected samples were processed according to the requirements of the Colombian Technical Standard NTC 5167 (NTC-5167 2022), which is a Colombian Standard based on ISO 7851:1983 for fertilizers and soil conditioners, as well as on other standards on ICS 65.080 category.The cation exchange capacity (CEC), density, total organic carbon (C), moisture content and water retention capacity (WHC) were evaluated according to ISO 7851:1983.Total nitrogen (N) by means of the Kjeldahl method according to ISO 11,261:1995.The phosphorus (P) content was determined according to ISO 11,263:1994.The pH and electrical conductivity (EC) were measured according to ASTM D4972-19, for this, 10 g of sample have been taken and 90 mL of water has been added, it has been shaken vigorously, then left to rest for a while, and both the pH and the EC have been measured.The methodology used for the analysis of microorganisms was previously described in previous works of the research group, in which the amount of microorganisms present in the process has been determined (Saldarriaga, Gallego, andLópez 2018a, 2018b;Saldarriaga et al. 2019).One gram of compost was diluted in 9 mL of peptone solution to prepare the sample.The first sample was serially diluted up to 10 −7 by taking one milliliter and diluting it in 9 mL of peptone solution.Sowing was done in solid culture, with 1.0 mL of the corresponding dilution transferred.After 24 hours, the individual colonies in the Petri dishes were counted.A phytotoxicity test was performed on radish seeds using an aqueous extract (1:10 compost/ distilled water mixture) (Acosta-Luque et al. 2022).The compost was filtered off after the mixture was agitated for 1 hour at 100 rpm.In a 90 mm diameter Petri dish, 9 ml of filtrate was used to moisten filter paper.The test was performed in triplicate, with 10 radish seeds planted in each Petri dish.As a control, petri plates with filter paper soaked with deionized water were employed.The boxes were kept at room temperature (25°C) in the dark.The number of germinated seeds was determined at 24 hours to calculate the germination percentage (G%), and the radicle length was assessed at 48 hours using digital image analysis (Acosta-Luque et al. 2022).The germination index (GI) was computed according to the method described by Guo et al. (2012).
The physicochemical parameters of the compost obtained in the three systems were compared with the standards established in NTC 5167 for products for the agricultural industry and organic products used as fertilizers and soil amendments.

VOCs analysis
The determination of VOCs was carried out in parallel with the physicochemical and microbiological sampling.The VOCs samples were obtained through the solid phase microextraction process (SPME) according to the ASTM D6889-03 standard.A compost sample of 2 g were extracted and carried to equipment Purge and Trap, Model 7000 CDS Analytical, Inc. (Archon, Purge and Trap AutoSampler, Varian).The sample was subjected to heating for 30 min at 60°C.The desorption process was carried out in an Agilent Technologies 6890N GC systems gas chromatograph with a 1:1 Split injector coupled to an Agilent Technologies 5975B VL-MSD with Triple-Axis Detector mass detector with a DB-624 capillary column (60 m x 0.32 mm x 1.80 μm).
The initial temperature of the column was 40°C for 5 min, then increased to 64°C at 8°C/min and held for 8 min, finally increased to 172°C at 12°C/min and finally increased to 172°C at 12°C/min and kept at this *The composition values of each system have been previously evaluated in each company; these data are not available to the public since they are protected by trade secret.
temperature for 25 min.The injector and detector had temperatures of 230°C and 160°C, respectively.To carry out the quantification of VOCs, fluorobenzene was used as internal standard and a calibration curve was obtained for a list of ten VOCs (Table 2).Following VOC quantification, the level of risk to health was estimated using the reference doses and exposure time, in accordance with other works published in the literature (Durmusoglu, Taspinar, and Karademir 2010;Saldarriaga, Aguado, and Morales 2014).

Data analysis
The data represent the mean (n = 3).As maturity indicators of the final product, the C/N ratio, the CEC/C ratio, the G% and the GI were used.Enterobacteria were selected as an indicator of the sanitization process of the material.The JMP® software was used for the statistical analysis of the data.Multivariate analyzes (correlations and principal components) were used to determine the possible relationships between the physicochemical and microbiological parameters and/or the generation of VOCs during the composting process in the three systems, as well as to select the operating variables that best explained the system variability.
A stepwise (forward and backward) regression analysis was performed to select the operating variables that predicted the compost maturity and sanitization parameters.The variables were selected based on the maximization of the adjusted R 2 .Regression analysis using a least squares model was used to generate process models considering the variables selected in the stepwise regression.

Physicochemical parameter evolution
The behavior of the physicochemical operating parameters in the three composting processes on an industrial scale are observed in Figure 1.During the first two weeks (weeks 1 and 2) the temperature range in the three systems was maintained between 30-54°C.For weeks 3, 4 and 5, the highest increase in temperature was presented in the three systems, in a range between 61-70°C.Later, between weeks 6 and week 10, the temperature decreased to ranges between 34-55°C.
The changes in temperature during the composting process for the three systems are an indicator of the different stages of the process: i) mesophilic stage, ii) thermophilic stage and iii) mesophilic stage or curing stage (Huang et al. 2004;Saldarriaga et al. 2019).The increases in temperature correspond to microbial activity, once the easily transformable materials are exhausted, the composting temperature decreases, entering a second mesophilic stage or curing stage.

Moisture content evaluation.
Moisture is one of the most significant parameters for the composting process, since it affects microbial activity, aeration, and system temperature (Saldarriaga et al. 2019).The range of moisture values recommended by other authors is 60-80% (Guo et al. 2012).For the three systems evaluated in the study, the initial moisture value was between 38-76%, with CHM being the process with the lowest initial moisture values and RUM with the highest value.The RUM system was the one that presented the highest moisture contents (62-78%) in a large part of the process (week 1 -week 6), compared to the CHM and USW systems.Moisture decreased over time for the three systems, as expected in a composting process, due to the increase in temperature, especially in the thermophilic stage of the process (between weeks 3 and 5), microbial activity and the loss of water by leachates.
pH evaluation.The pH is another of the important parameters in the composting process because it is a factor that limits the activity of microorganisms (Pan, Dam, and Sen 2011).In the first weeks there was a pH value of alkaline tendency in the three systems (8.1-8.5)(Ameen, Ahmad, and Raza 2016).Subsequently, from weeks 3 and 4, a decrease in the pH values is observed in the three systems (7.5-7.6), which may be associated with the formation of low molecular weight organic acids in the early stages of transformation of organic matter and nitrogen transformation rate (Ameen, Ahmad, and Raza 2016;Pan, Dam, and Sen 2011;Zailani and Hamid 2023).In the CHM system it is observed that the pH tends to increase (8.5-9.1).Increases in pH toward the alkaline trend are mainly due to the formation of ammonium from the breakdown of proteins.For the RUM and USW systems, at the end of the process, the pH trend is toward neutral values (6.8-7.9)(Meng et al. 2016;Sundberg, Smårs, and Jönsson 2004).Carbon and nitrogen evaluation.Carbon and nitrogen content in composting systems will depend on the initial raw materials and the mixing ratios.In the three systems evaluated, it was observed how C decreased over time, due to the mineralization of organic matter (Saldarriaga et al. 2019).However, the proportion of C decrease was different in the three systems.On average, the loss of C was 6, 16, and 20% for the RUM, CHM, and USW systems, respectively.This is due to the recalcitrance of organic matter, moisture, and the microbial activity of each of the systems (Saldarriaga, Gallego, and López 2018a).Specifically, for RUM, a lower loss of C was observed compared to the CHM and USW systems.This can be associated with a higher initial moisture content (76%) and during the composting process in RUM, which could decrease oxygen in the compost bed and with this the microbial activity associated with the transformation of organic matter.
The increase in N content could be associated with the accumulation of organic nitrogen with respect to the loss of carbon and the reduction of the composting pile (pile volume).However, in the RUM, CHM and USW systems during the first weeks a decrease in N was observed, which may be the result of volatilization losses as ammonia due to a C/N ratio at the lower end of the recommended range, in case of the CHM and USW systems (El-Mrini, Aboutayeb, and Zouhri 2022).These losses can also be associated with the thermophilic stage of the process, where the loss of N by volatilization is favored (Huang et al. 2004), as well as the initial alkaline pH conditions, since these pH conditions favor the loss of N when transformed into ammonia (Huang et al. 2004).

Cation exchange capacity (CEC) evaluation.
The CEC presented a growing trend in the three systems.The behavior of this parameter reflects the degree of development of the transformations of the organic matter in the system, especially the presence of carboxyl and hydroxyl groups because of the resynthesis of the initial materials in charge of the microbial activity.The increase in CEC means that it increases the number of materials like humus (resynthesis materials) and the ability of these to interact with the cations dissolved in the aqueous phase.In the CHM and USW systems, the CEC value was greater than 35.0 c mol (+) /kg, which guarantees the formation of highly oxidized compounds (Saldarriaga et al. 2019).

Ash content evaluation.
The ash content at the end of the composting process corresponds to salts, inorganic oxides and mineralized carbon that is not available for microbial activity (Saldarriaga et al. 2019).This parameter should rise with time because of CO 2 and water loss, as well as an increase in the fraction of inorganic material as a result of the mineralization of organic matter.This parameter's behavior is as expected in a composting process since it has increased in all systems.The volatile solids content is calculated from the ash content.While the ash content indicates the proportion of inorganic material, the volatile solids content is related to the system's organic material content.The observed trend and adjustment are a good indicator of the kinetic process of organic CO 2 loss and material resynthesis (Bernal, Alburquerque, and Moral 2009;Mathur et al. 2012;Sundberg, Smårs, and Jönsson 2004).Like the ashes, the electrical conductivity increased in all composting systems; this rise is related to an increase in dissolved ions in the aqueous phase, which results from the mineralization of organic waste.
For all the evaluated systems, the density has shown an increasing trend.This parameter indicates the material's structure as the initial material transformation and resynthesis of new compounds occurs, resulting in the development of porosity and a rise in moisture retention capacity.The latter has allowed the WHC for the three systems to increase over time.At the end of the process, the WHC value is greater than 100%, indicating that porous systems have formed, causing the material to store more water than its own weight (Khan et al. 2014;Mahapatra, Ali, and Samal 2022).

Microbiological parameter evolution and enterobacteria reduction
In the three composting systems, the total microorganism's presence was evaluated (Figure 2).Enterobacteria shows detail of the microorganism's behavior from week 5 to 10 (Figure 2).In the USW system, a higher number of microorganisms has been found in week 5 when the temperature reaches its maximum peak, with an amount of 6.5 × 10 9 CFU/g.Then, in the final week of the process (week 10) there was a decrease until it reached 644 CFU/g.The CHM system is the one with the least number of microorganisms, with a peak of 1.7 × 10 7 CFU/g in week 4 and decrease until it reaches 98 CFU/g in week 10.RUM system started with 3.0 × 10 8 CFU/g, but by week 4, when the highest temperatures are reached, it was observed peaks of 6.2 × 10 8 CFU/g, subsequently on week 10, the number of microorganisms in RUM system decreases to 321 CFU/g.
Respirometry was determined as an indicator of the evaluation of microbial activity in composting piles.In all systems, it was observed that the highest production of CO 2 (ppm) was in accordance with the weeks where higher temperature and a high presence of total microorganisms were measured, which can be associated with the rapid decomposition of easily degradable organic material present in the starting raw materials and the growth of microbial populations (Bermudez, Saldarriaga, and Osma 2019;Saldarriaga, Gallego, and López 2018a;Saldarriaga et al. 2019).
The final values of enterobacteria (Figure 2) show that in all operations, a product that conforms with Colombian regulation in terms of sanitization was obtained.The decrease in microorganisms is due to the association that microorganisms have with variables such as temperature, humidity, C, C/N ratio, and pH, according to the stepwise regression study.However, the increase in temperature during the thermophilic stage is required for compost sanitization (Saldarriaga et al. 2019).The rise in temperature in the composting pile correlates with an increase in microbial activity as measured by respirometry, and the rise in other microbial populations creates competition for space and substrate (mainly fungi and yeasts), potentially aiding in the reduction of enterobacteria populations.Changes in the C/N ratio show that the transformation of organic matter in the composting process relates to a decrease in enterobacteria, which could be related to the availability of food supplies for microorganisms and the transformation of the substrate (Bermudez, Saldarriaga, and Osma 2019;Li et al. 2022;Zailani and Hamid 2023).Similarly, changes in the pH of the process could influence microbial activity since the chemical environment may or may not favor microbes present in the composting process.As a result, it is critical to maintain control over these factors throughout the process to promote sanitization in the final product.Other authors argue that the compost sanitization process is influenced not only by temperature increases, but also by the dynamics of microbial populations and changes in the physicochemical environment, such as pH (Li et al. 2022).The least squares regression model indicates that these variables can predict the behavior of the enterobacteria content with an R 2 adjustment of 0.90 (Eq.1).

Maturity of the final product
Figure 3 shows the behavior of the maturity indicators selected for the composting process.The final values of the C/N and CEC/C ratio are important indicators of the composting process maturity.C/N is related to the organic matter transformation rate and is used as an indicator of compost maturity (El-Mrini, Aboutayeb, and Zouhri 2022; Saldarriaga et al. 2019).At the end of the composting process, a mature   (Mathur et al. 2012;Raj and Antil 2011).In this study, the CHM system had a C/N ratio value less than 20 and the RUM and USW systems had C/N ratio values less than 10, which is indicative of maturity in the final product obtained (Bernal, Alburquerque, and Moral 2009;Saldarriaga et al. 2019).Regarding the CEC/C ratio, it has been widely used to determine the degree of humification (resynthesis of humus-like substances) and compost maturity (Bernal, Alburquerque, and Moral 2009;El-Mrini, Aboutayeb, and Zouhri 2022;Raj and Antil 2011).A CEC/C ratio of 1.7 has been proposed as the lower limit to describe the humification process in compost of animal origin material, this value can be considered as an acceptable indicator of the maturity of a compost (El-Mrini, Aboutayeb, and Zouhri 2022;Raj and Antil 2011).The CEC/C ratio at the end of the process varies greatly between the three systems, with only the USW and CHM exhibiting values larger than 1.7, indicating an acceptable level of humification.While the RUM system meets all the requirements of an amendment in accordance with the regulations, do not demonstrate the level of maturity of the final product (Raj and Antil 2011).
Regarding other maturity indicators such as phytotoxicity test variables: G% and the GI (Bernal, Alburquerque, and Moral 2009;Guo et al. 2012;Xu et al. 2020).It has been suggested that G% value higher than 90% indicates phytotoxicity absence in the final product, therefore it is considered a good quality and mature compost (Bernal, Alburquerque, and Moral 2009).As for the GI, it is proposed that a value higher than 80% is indicative of compost maturity (Guo et al. 2012).For the evaluated systems, the G% was higher than 90%, finding values between 90-100% of germination with respect to the control.As the same, the GI value was higher than 80% in all three systems, since values between 90-100% were obtained.The results indicate that final compost obtained in the process has no phytotoxicity activity, which is related to the maturity of the final product.

Selection of physicochemical parameters and important compost maturity drivers
Figure 4a-d shows the analysis of principal components for the physicochemical operating parameters of the three composting systems and for the set of all data, without discriminating by system.According to the principal component analysis, the EC, WHC, density, moisture and ash variables accumulate the highest weight of component 1 (Table S1), which explains more than 50% of the variability of the RUM, CHM, and USW systems.This suggests that these can be used as operating variables in the process, regardless of the type of starting material  (Saldarriaga et al. 2019).These variables are easy to measure compared to other parameters that mean higher costs and more specialized infrastructure for their measurement.In practical terms this helps to develop easier process monitoring on an industrial scale.
Regarding the maturity indicators, the GI and CEC/C show higher weight in component 1 compared to the weight of the C/N ratio for the three systems (Figure 4).When performing the analysis independently of the system, using all the data, the behavior is similar, indicating that the GI and CEC/C ratio has the greatest weight in explaining the variability of the data (Table S1).The CEC/C is significantly correlated (p < 0.05 or 0.001) with 13, 17, 15 parameters measured in the composting process for RUM, CHM and USW, respectively.When performing the correlation analysis with the entire data set of the three systems, the CEC/C ratio correlates significantly (p < 0.05) with 11 physicochemical parameters.This maturity indicator is related with more than half of the total physicochemical parameters measured in the study (18 physicochemical parameters), which confirms that it is a maturity indicator that broadly explains the behavior of the systems.The PCA analysis for the maturity indicators (Figure 4e) indicates that the CEC/C ratio has a higher weight in component 1, which explains 64.5% of the variability of the maturity indicators data set ripe for all three systems.Along the same lines, the correlation analysis shows that the CEC/C ratio is significantly correlated with C/N and phytotoxicity (G% and GI) (Figure 4f).The transformation processes of the materials in the systems, such as organic carbon mineralization and resynthesis in materials like humus, can explain a strong relationship between the evaluated maturity criteria.In addition, the biodegradation of potentially harmful substances contained in the process, which is related to the phytotoxicity test findings.According to the study's results, low C/N ratio values are required to attain CEC/C values equal to or higher than the permitted limit.The C/N value is then indicated to be less than 10.Similarly, when the GI is larger than 90%, the appropriate CEC/C values are obtained (Figure 4).The CEC/C ratio could be used as a maturity indicator for the industrial-scale composting systems studied in this work (Mathur et al. 2012;Raj and Antil 2011;Xu et al. 2020).Figure 5 depicts the GI values, with the color bar indicating the GI values, blue low values, and red high values.The GI values are indicated by the size of the markers; smaller markers correspond to lower C/N values, while larger markers correspond to higher C/N values.
The step-by-step regression analysis showed that the physicochemical parameters EC, WHC, and ash are significant (p < 0.05) in predicting the behavior of CEC/C ratio, whether performed for each system or independently.This suggests that these three factors are important for process control since they are highly associated with the other physicochemical variables assessed, as previously explained (Saldarriaga, Gallego, and López 2018a;Saldarriaga et al. 2019).According to the least squares regression model, these variables can predict CEC/C behavior with an R 2 adjustment of 0.85 (Eq.2).

Generation of VOCs during the process
Parallel to the physicochemical and microbiological evaluation of the three composting systems, samples were taken for VOC analysis (Figure 6).Initially, the identification of the different VOCs present in each sample and their relative abundance was obtained, those with the highest reliability of 90 were selected.Using the fluorobenzene calibration curve, the quantification of BTEX (Benzene, Toluene, Ethylbenzene and Xylenes) was obtained, which have been found in all composting processes and which have been reported by other authors and which represent a risk to health (Nadal et al. 2009;Saldarriaga, Aguado, and Morales 2014).For the RUM system, between 321 and 394 VOCs have been identified, of which 21 have been identified with a reliability higher than 90 (Table S2).In the identified VOCs, toluene, ethylbenzene, o-xylene, 1, 3 dimethyl benzene and D-limonene stand out, which are registered in all the samples analyzed.In the case of the CHM system, between 404 and 452 VOCs were recorded for each sample analyzed, 15 have been identified with a reliability higher than 90.In the identified VOCs, toluene, ethylbenzene, xylenes, D-limonene, 1,3-dimethyl benzene, bicyclo [3.1.1]heptane 6,6-dimethyl,2-methylene (1S) and 1-methyl-2(1-methyl-ethyl) benzene stand out, which are identified in all samples analyzed.In the case of the USW system, between 154 and 388 VOCs were present, in total 58 VOCs with a quality greater than 90 have been identified.Within the VOCs identified, as in the RUM and CHM systems, toluene and D-limonene stand out, compounds such as benzene fluoride, caryophyllene, pinene, copaene and cyclohexadiene were also identified, which are characterized by being present in all analyzed samples.
The stepwise regression analysis showed that the predictor variables for the generation of VOCs in the three systems together were C, EC, respirometry.This is related to the degradation of the initial organic materials due to microbial activity, transforming organic substances (Saldarriaga, Aguado, and Morales 2014).The least squares model (Equation 3) shows that only C, EC, and respirometry are the variables that fit as predictors of benzene generation, one of the VOCs of highest interest and control due to its harmful effects on human health (Saldarriaga, Aguado, and Morales 2014).The model including the two mentioned variables has an R 2 of 0.89 when adjusting the real data with the predicted ones.The analysis of correlations between the physicochemical and microbiological parameters and the generation of VOCs, shows that the highest correlation of VOCs is associated with C, which supports the findings of the stepwise regression analysis and the predictive model.

Health risk index associated with the generation of VOCs
Various VOCs have been identified as being produced during the three composting processes, some of which have been reported to be hazardous to health and the environment (Nadal et al. 2009;Saldarriaga, Aguado, and Morales 2014).From the quantification obtained for the VOCs of interest, the distribution between phases has been estimated to determine the number of VOCs present in the aqueous, adsorbed, or gaseous phase, respectively, and thus identify the potential risk associated with their emission.The associated risk of these substances has been determined using the approach developed by Saldarriaga, Aguado, and Morales (2014) for estimating the hazard ratio (HR).This hazard ratio is calculated by comparing the daily consumption of the researched substance to the reference level at which no negative effect on health occurs.The calculation model is presented below: Where I, is the daily intake (mg/kg/day), RfD is the reference dose (mg/kg/day), C is the substance concentration μg/m 3 , CF is the conversion factor (mg/m 3 ) to perform the substance concentration conversion.IR is the inhalation velocity (m 3 /day), which, for this study, has been taken as 20 m 3 /day.EF is the exposure frequency of workers (day/year), which has been taken as 74 days ( = 52×6/3-30).ED is the occupational exposure time (years), which can be taken as 20 years, the average occupational life of a worker.BW is body weight (kg), which according to the US EPA, the recommendation is 70 kg.And finally, the AT, which is the average time (days), an average exposure value of 70 years lifetime (or 25,500 days) is recommended.The RfD for xylenes is 0.029 mg/kg/day, toluene is 1.43 mg/kg/day, and ethylbenzene is 0.29 mg/kg/day.In the case of benzene, which is proven to be a carcinogen, the cancer risk is calculated according to the following equation: Where CPF is the carcinogen potency factor.According to IRIS (International Resource Information System), the CPF is 0.029 (mg/kg/day).
Figure 7 shows the hazard ratio (HR) index associated with the three systems evaluated.When examining non-carcinogenic effects, some authors argue that the risk is negligible when the HR is less than or equal to one (Durmusoglu, Taspinar, and Karademir 2010;Duan et al. 2022;Saldarriaga, Aguado, and Morales 2014;Wu et al. 2018).The RUM and CHM systems then show that the HR is high throughout the processes, with weeks 4, 5, and 6 being the most dangerous.This could be related to microbial activity and organic matter decomposition.In the case of the USW system, the risk is limited to xylenes and a trace amount of ethylbenzene in the early stages of the process.As different authors have shown, the generation of VOCs depends on the starting material, the weather conditions, the average temperature, the height above sea level, among others (Domingo and Nadal 2009;Hester, Harrison, and Derwent 1995;Nadal et al. 2009;Saldarriaga, Aguado, and Morales 2014).However, it was possible to demonstrate in this study that the starting material could be the primary factor in the generation of VOCs.These VOCs could be linked to animal wastes, which in the case of RUM and CHM are primarily formed of cow rumen and chicken dung, respectively.These wastes could increase the generation of these compounds, so occupational safety measures for all operators must be optimal during the compost bin operation process.
As the same HR, for the risk of cancer with benzene, it is estimated that if this value is less than 1, there is no significant risk.The risk prediction must be greater than or equal to 1 (Durmusoglu, Taspinar, and Karademir 2010).This study found that the risk of cancer is negligible for all systems (Figure 7).However, for the CHM system in weeks 3 and 5, there are values close to 1, which require extreme caution.Moreover, further research into this type of waste and its degradation process is required.Since it has been discovered that the degradation of chicken manure produces compounds such as ammonia, among others, which can be degraded and release compounds such as benzene, which can endanger the health of composting plant workers.
Even though there was a risk relationship with TEX and that there is a high risk of cancer with benzene in the CHM system.The volume of solid waste treated in these composting plants, as well as the combination of solid waste, means that the procedure poses no significant dangers to the health of those who reside nearby.To lower the volume and concentration of these chemicals, aeration processes must be carried out more regularly, as commonly advocated in the literature (Domingo and Nadal 2009;Nadal et al. 2009;Saldarriaga, Aguado, and Morales 2014).That the plants have proper ventilation, and that the personnel have adequate protection systems.

Evaluation of the final product for all systems
Table 3 shows that all the systems evaluated comply with the Colombian technical standard (NTC-5167).
Among the most important parameters are enterobacteria, which must be below 1000 CFU/g.That guarantees the sanitation of the process, and that this product can be used as amendment or fertilizer.With respect to phosphorus, an increasing trend over time was found in all the composting processes evaluated.This element is limiting for different biological functions, so it is considered significant and must be reported when it is found in amounts greater than 1%, as well as the N content.The C/N ratio affects the maturity of the final product, which has a direct effect on the potential use of compost as a soil amendment (El-Mrini, Aboutayeb, and Zouhri 2022; Saldarriaga, Aguado, and Morales 2014).The initial and final C/N ratio values obtained suggest that the three processes can be useful for the generation of amendments.The EC values (Figure 1) are below 4 dS/m for all three systems, although the Colombian technical standard does not

Conclusion
The solid waste composting evaluated in this study is viable on an industrial scale if the physicochemical operating variables are controlled.This study suggests that the ash content, moisture retention capacity and electrical conductivity are the variables that best explain the variability of the three composting processes evaluated.These variables are easy to measure and represent a practical technical tool that is easy to apply in the industry.The CEC/C ratio is the maturity parameter selected in this study, which is significantly correlated with the C/N ratio and with the phytotoxicity tests.After ten weeks, the material was sanitized, and this was demonstrated to be connected to microbial activity and pH, which are important variables for both the final product and the formation of VOCs in the case of microbial activity.Among the VOCs identified, the BTEX, which are classified as a risk to human health, stand out.It has been revealed that an increase in its production is related to the microbial activity required for the sanitization and decomposition of the material.
According to the implications in the composting facilities it can be concluded that: • During the composting process, the evaluation of properties such as moisture content, ash content, pH, and electrical conductivity allows adequate control of the process and allows adequate degradation of solid waste.• The increase in temperatures in the process is important for its sanitization; however, it is in this stage where the greatest release of VOCs occurs.Adequate aeration, where temperatures are controlled and the generation of VOCs does not increase, is most recommended.• According to this study, the BTEX that have been identified in all the systems will be present in low or high concentrations; however, this risk can be reduced by implementing strategies within the facilities, such as generating air currents within the facilities, and by ensuring that the operators of the machines with which the aeration of the piles is carried out have the appropriate safety equipment, such as special face mask for VOCs, gloves, and clothing.
In accordance with the above, it is necessary that the composting facilities be open, where the ambient air can flow easily and dilute the gases released by the compost piles.In the same way, the operators of the turning machines must have, as mentioned above, safety equipment.Likewise, turning should be done at specific times of the day when there is a greater amount of wind in the area where the plant is located; the most recommended times are in the morning and afternoon.In general, useful amendments for soil application were obtained by following to Colombian technical regulations.

Disclosure statement
No potential conflict of interest was reported by the author(s).

C
/N ratio evaluation.The C/N ratio is an essential parameter when starting and developing a composting process (El-Mrini, Aboutayeb, and Zouhri 2022;Guo et al. 2012).It is advised that the initial C/N ratio be between 20 and 30 because high ratios can result in a slow decomposition process and hence a longer composting time.Similarly, low ratios might result in N loss due to ammonium release (El-Mrini, Aboutayeb, and Zouhri 2022;Xu et al. 2020).The three systems studied in this work presented initial values of C/N between 21-40.RUM presented a higher C/N ratio, outside the recommended range.In the three systems, the C/N ratio decreased over time, associated with a decrease in the C content and an increase in the N content throughout the composting process.After weeks 6 and 7 of the composting process, the C/N ratio tends to stabilize in the 8-11 range, which may be an indication of the maturation phase, which is characterized by humification and the formation of high-molecular-weight compounds (El-Mrini, Aboutayeb, and Zouhri 2022).At the end of the composting process, the three systems showed adequate C/N ratio values (10-11), indicating an adequate organic matter transformation rate, nutrient load, and composting process time(Saldarriaga et al. 2019).
compost is expected to have C/N values less or equal to 25 (TMECC 2002).Other authors propose C/N values less than 20 indicate mature compost and less than 10 very mature compost

Figure 5 .
Figure 5. Relationship between CEC/C, C/N and GI for the three composting systems.

Figure 6 .
Figure 6.Generation of VOCs during the composting process.

Figure 7 .
Figure 7. Hazard ratio associated with the generation of VOCs during the composting process.

Table 1 .
Composting systems and starting material.

Table 2 .
Calibration for quantification of VOCs.

Table 3 .
Quality of the final product according to the NTC-5167 standard for its use as an amendment.requirecompliance with EC values, EC values higher than 4 dS/m are not adequate for soil amendments since they can generate negative effects on plants.