Influence of aeration rate and reactor shape on the composting of poultry manure and sawdust

ABSTRACT

To achieve successful composting, all the biological, chemical, and physical characteristics need to be considered. The investigation of our study was based on various physicochemical properties, i.e., temperature, ammonia concentration, carbon dioxide concentration, pH, electrical conductivity (EC), carbon/nitrogen (C/N) ratio, organic matter (OM) content, moisture content, bacterial population, and seed germination index (GI), during the composting of poultry manure and sawdust for different aeration rates and reactor shapes. Three cylindrical-shaped and three rectangular-shaped pilot-scale 60-L composting reactors were used in this study, with aeration rates of 0.3 (low), 0.6 (medium), and 0.9 (high) L min⁻¹ kg⁻¹ DM (dry matter). All parameters were monitored over 21 days of composting. Results showed that the low aeration rate (0.3 L min⁻¹ kg⁻¹ DM) corresponded to a higher and longer thermophilic phase than did the high aeration rate (0.9 L min⁻¹ kg⁻¹ DM). Ammonia and carbon dioxide volatilization were directly related to the temperature profile of the substrate, with significant differences between the low and high aeration rates during weeks 2 and 3 of composting but no significant difference observed during week 1. At the end of our study, the final values of pH, EC, moisture content, C/N ratio, and organic matter in all compost reactors were lower than those at the start. The growth rates of mesophilic and thermophilic bacteria were directly correlated with mesophilic and thermophilic conditions of the compost. The final GI of the cylindrical reactor with an airflow rate of 0.3 L min⁻¹ kg⁻¹ DM was 82.3%, whereas the GIs of the other compost reactors were below 80%. In this study, compost of a cylindrical reactor with a low aeration rate (0.3 L min⁻¹ kg⁻¹ DM) was more stable and mature than the other reactors.

Implications: The poultry industry is growing in South Korea, but there are problems associated with the management of poultry manure, and composting is one solution that could be valuable for crops and forage if managed properly. For high-quality composting, the aeration rate in different reactor shapes must be considered. The objective of this study was to investigate various physicochemical properties with different aeration rates and rector shapes. Results showed that aeration rate of 0.3 L min⁻¹ kg⁻¹ DM in a cylindrical reactor provides better condition for maturation of compost.

Introduction

The poultry industry of South Korea has continued to expand, with significant increases in production. Korea’s chicken production is projected to increase 6.5% to 905,000 metric tons (MT) in 2018 from 850,000 MT in 2017 (United States Department of Agriculture [USDA] 2017). As a result, a large amount of manure is produced, which can cause serious environmental pollution. Composting of the manure is a widely used technique for dealing with this problem. Composting has been recognized as a natural process for stabilizing organic wastes. Proper composting effectively reduces odor emission and nitrogen (N) loss, and the organic waste becomes a relatively stable product for use as fertilizer. After composting, handling and transport of waste become easier and safer. To achieve good compost, factors such as the moisture content, carbon/nitrogen (C/N) ratio, aeration rate, temperature, and pH of the compost, as well the selection of the composting materials and the applied methodology, should be appropriately controlled.

In comparison with laboratory- or pilot-scale experimentation, which allows for a high degree of control and replication, full-scale or windrow composting experimentation can be both expensive and difficult to control (Mason and Milke 2005). Among the influencing factors, aeration rate and reactor shape are key for sustaining a thermophilic phase. The aeration rate affects microbial activity, degradation rate, and temperature variations during the composing process (Kuter, Hoitink, and Rossman 1985). Too much aeration can lead to cooling and the prevention of thermophilic conditions; however, too little aeration can lead to anaerobic conditions (Ahn, Richard, and Choi 2007). Several previous studies have investigated aeration methods and rates, with different aeration rates being recommended, e.g., 0.44 L min⁻¹ kg⁻¹ dry matter (DM) in the composting of maize stalks and cow feces (Nada 2015), 0.62 L min⁻¹ kg⁻¹ volatile solids (VS) in the composting of vegetable and fruit wastes (Arslan, Ünlü, and Topal 2011), 0.5 L min⁻¹ kg⁻¹ organic matter (OM) in the composting of chicken manure and sawdust (Gao et al. 2010), 0.25 L min⁻¹ kg⁻¹ OM in the composting of dairy manure with rice straw (Li, Zhang, and Pang 2008), 0.43−0.86 L min⁻¹ kg⁻¹ OM in the composting of food waste (Lu et al. 2001), 0.04–0.08 L min⁻¹ kg⁻¹ OM for swine manure composting (Lau et al. 1992), and 0.87–1.07 L min⁻¹ kg⁻¹ OM in the composting of dairy manure with crop and forest residues (Hong, Matsuda, and Ikeuchi 1983). The range of aeration rates varies according to the compost material and composting process.

To ensure that the thermophilic phase can be attained and sustained during composting, reactor shape and volume must be considered during reactor design (Phillip 2010). The amount of heat produced in a compost reactor is proportional to the volume (V) of the substrate, and the heat loss is proportional to the surface area (SA) of the reactor. Therefore, the SA/V ratio must be considered when building a compost reactor (Magalhaes et al. 1993). The volumes of compost reactors reported in the literature range from 0.4 to 1780 L. Large-volume reactors are able to achieve higher temperatures owing to a lower SA/V ratio (Phillip 2010). In previous studies, both cylindrical and rectangular shape reactors have been used to study various composting processes, such as a 30-L self-heating cylindrical reactor with a SA/V ratio of 18 m² m⁻³ (VanderGheynst and Lei 2003), a 119-L self-heating cylindrical reactor with a SA/V ratio of 12.1 m² m⁻³ (Barrington et al. 2003), a 33.3-L self-heating cylindrical reactor with a SA/V ratio of 17.6 m² m⁻³ (Schloss, Chaves, and Walker 2000), a 985-L self-heating rectangular reactor with a SA/V ratio of 6 m² m⁻³ (Pecchia, Beyer, and Wuest 2002), and a 840-L self-heating rectangular reactor with a SA/V ratio of 6.5 m² m⁻³ (Leth, Jensen, and Iversen 2001).

For high-quality composting, the aeration rate in different reactor shapes must be considered. The objective of this study was to investigate various physical and physicochemical properties, i.e., temperature, ammonia concentration, carbon dioxide concentration, pH, electrical conductivity (EC), C/N ratio, organic matter content, moisture content, bacterial population, and seed germination index (GI), during the composting of poultry manure and sawdust at different aeration rates in cylindrical and rectangular reactors.

Materials and methods

Composting setup and design

Six pilot-scale composting reactors with two different shapes (three rectangular reactors and three cylindrical reactors) were used in this study. The reactors were made from high-density polyethylene. The cylindrical reactors each had a total volume of 60 L (0.66 m height and 0.34 m diameter) with a SA/V ratio of 14.8 m² m⁻³, and the rectangular reactors each had a total volume of 60 L (height: 0.66 m; length: 0.3 m; width: 0.3 m) with a SA/V ratio of 16.3 m² m⁻³ (Figure 1). The reactors were insulated with a layer of polyurethane foam (2.54 cm thickness). A vertical rotating axis with blades ensured complete mixing of the composting mass prior to sampling. The reactors were equipped with a valve for removing the leachate.

Figure 1. Schematic of the cylindrical and rectangular reactors with experimental apparatus. 1, air compressor; 2, solenoid valve; 3, timer; 4, gas washing bottle (NaOH solution); 5, gas washing bottle (distilled water); 6, airflow meter; 7, reactor with insulation; 8, perforated plate; 9, digital thermometer; 10, ammonia sensor; 11, carbon dioxide sensor; 12, leachate outlet.

A plastic grid was installed 0.08 m above the bottom of the reactor to support the composting bed and ensure uniform air distribution. The air was input at the bottom using an air compressor (Air Land S45-40–4.5; Seoul, South Korea). Airflow inputs were intermittent, with 45 min of aeration followed by 15 min without aeration, as controlled by a timer and solenoid valve (Auto Sigma HPW2120; Gyeonggi-do, South Korea). The airflow rates in the reactors were set at 0.3, 0.6, and 0.9 L min⁻¹ kg⁻¹ DM. Before the start of the main experiment, we performed two tests in two reactors with aeration rates of 0.2 and 0.6 L min/kg DM for 1 week. The first test was performed with 30 min/hr aeration rate, and the second test was performed with 15 min/hr aeration rate. The temperatures with 0.2 L min⁻¹ kg⁻¹ DM reached maximum of up to 40 and 45 °C in the two tests, respectively. Therefore, aeration rates for the main experiment were selected to be 0.3, 0.6, and 0.9 L min/kg DM, with aeration time of 15 min/hr. An airflow meter (Valve Acrylic Flowmeter, Cole-Parmer) and pressure regulators were used to control the airflow rates. The compost temperature was monitored daily (10 a.m., 2 p.m., and 6 p.m.) at three different locations in the substrate using digital thermometers (Testo 905 T1). The ambient temperature was also recorded daily.

Before entering the reactors, the air was passed over a solution of sodium hydroxide in order to remove traces of carbon dioxide. The air was then moisturized through a gas washing bottle, i.e., a 500-mL bottle of distilled water, prior to entering the reactor, so as to maintain the humidity in the reactor.

Compost materials

Fresh poultry manure and sawdust were used as experimental materials. Poultry manure was collected from poultry farms in Sacheon, South Korea, and sawdust was purchased from a wood chipping mill. The initial moisture content, organic matter content, pH, EC, total organic carbon, total nitrogen (TN), and C/N ratio of the manure and sawdust were analyzed (Table 1). The manure and sawdust were mixed using a concrete mixer to achieve effective homogenization of the materials. The initial moisture content and C/N ratio of the homogenized material were adjusted to approximately 65% and 25, respectively, prior to placing the material into the reactors. The initial physicochemical properties of compost mixtures for all reactors are shown in Table 1.

Table 1. Initial characteristics of poultry manure, sawdust, and all composts in reactors.

Parameter	Poultry manure	Sawdust	R1	R2	R3	R4	R5	R6
Moisture content, % ww	57.4 ± 0.76	16.7 ± 0.18	65.1 ± 3	65 ± 5	64.9 ± 7	65 ± 1	65.1 ± 6	64.9 ± 4
Organic matter, % dw	64.6 ± 0.80	99.1 ± 0.07	85 ± 7	85 ± 7	85 ± 7	85 ± 7	85 ± 7	85 ± 7
pH	8.14 ± 0.01	6.26 ± 0.05	7.81 ± 2	7.8 ± 4	7.81 ± 2	7.79 ± 2	7.8 ± 1	7.79 ± 3
Electrical conductivity, mS/cm	4.31 ± 0.05	1.79 ± 0.01	2.12 ± 1	2.13 ± 6	2.12 ± 4	2.11 ± 2	2.12 ± 2	2.13 ± 4
Total carbon, % dw	34.43	49.25
Total nitrogen, % dw	5.094	0.176
C/N ratio	6.76	280	25	25	25	25	25	25

Notes. ww = wet weight; dw = dry weight. For the adjustment of C/N = 25, 10.6 kg of fresh sawdust were added to 10 kg of fresh poultry manure and all the parameters were measured after mixing. For the adjustment of 65% moisture content, 17 L deionized water was added to each mixture. Values represent the mean ± standard error of three replicates.

Process monitoring and parameter analysis

For carbon dioxide determination, a Lutron MCH-383SD (Germany; accuracy ±40 ppm) electrochemical sensor was set above the composting material in each reactor. The Lutron MCH-383SD was able to record the date and time interval of the measurements, and it contained a memory card for storing these data. Carbon dioxide was measured at 10-min intervals, and the recorded data were averaged for further analysis and interpretation.

A ZDL800 ammonia meter (Environmental Sensors accuracy ±4 ppm) was used as an electrochemical sensor for ammonia determination above the compost material in each reactor. As the exposure of an electrochemical sensor to high concentrations of ammonia over long periods leads to errors in the readings, the ZDL800 sensor was used for 1-hr measurements (at the same time, daily) and the data were transferred to a computer (Qasim et al. 2017). This sensor was able to store the date, time, and the interval between each sampling point and actual exposure point.

After mixing by the bladed vertical rotating axis, the samples of compost material were taken from five different depths of compost and mixed in plastic bags. Approximately 50 g of the material was taken from plastic bags for further analysis. The moisture content of the experimental material was determined by oven drying at 105 °C for 24 hr (American Public Health Association [APHA] 1995). The C and N contents were determined using an automated LECO True Mac CNS elemental analyzer. The LECO True Mac CNS has overall measuring range from 20 ppm to 30% with high precision. EC and pH were measured in an aqueous extract, which was obtained by mechanically shaking the samples with distilled water at a solid/water ratio of 1:10 (w/v) for 1 hr. The suspension was centrifuged and filtered through Whatman no. 42 filter paper. The pH and EC measurements were carried out using digital meters (Oakton handheld pH meter, accuracy ±2 mV; CON 450 Eutech conductivity meter, accuracy ±1%). The organic matter content (volatile solids) of the material was measured by burning in an oven at 550 °C for 6 hr (APHA 1995).

The loss of organic matter (k) from the compost material was calculated from the initial and final organic matter contents, according to eq 1 (Diaz et al. 2002; Külcu and Yaldiz, 2004):

(1) $k=\frac{\left[\mathrm{O}{\mathrm{M}}_{\mathrm{i}}\left(\mathrm{\%}\right)-\mathrm{O}{\mathrm{M}}_{\mathrm{f}}\left(\mathrm{\%}\right)\right]\times 100}{\mathrm{O}{\mathrm{M}}_{\mathrm{i}}\left(\mathrm{\%}\right).\left[100-\mathrm{O}{\mathrm{M}}_{\mathrm{f}}\left(\mathrm{\%}\right)\right]}$(1)

where OM_i is the organic matter content at the beginning of the process, and OM_f is the organic matter content at the end of the process.

Mesophilic and thermophilic bacterial populations

Colony-forming units (CFU) of mesophilic and thermophilic bacteria were determined by the spread plate technique (Prescot, Harley, and Kleon 1996). Nutrient agar was used as a growth medium. One gram of each sample was suspended in 99 mL of physiological solution and then homogenized using a magnetic stirrer for 30 min. Dilution series (from 10⁻³ to 10⁻⁸) were made from the prepared suspensions. One milliliter of each sample from the above-mentioned dilution was placed into a Petri dish and spread up and down using a “hockey stick.” The Petri dishes were incubated at 37 °C for 48 hr for the growth of mesophilic bacteria and at 55 °C for 48 hr for the growth of thermophilic bacteria. The CFU of mesophilic and thermophilic bacteria were counted using a digital colony meter, and the results were expressed as CFU of mesophilic and thermophilic bacteria per gram of compost.

Radish seed germination index (GI) test

Seed germination and root length tests were conducted on water extracts obtained by mechanically shaking fresh samples for 1 hr at a solid/double-distilled water ratio of 1:10 (w/v, dry weight basis). Approximately 5.0 mL of each extract was pipetted into a sterilized plastic Petri dish lined with a Whatman no. 2 filter paper. Ten radish seeds were evenly placed on the filter paper and incubated at 25 °C in the dark for 72 hr. Triplicate measurements were made for each composting mixture. Treatments were evaluated by counting the number of germinated seeds and by measuring the length of roots. The responses were calculated as the GI, which was determined according to eq 2 (Zucconi et al. 1981):

(2) $\mathrm{G}\mathrm{I}\mathrm{\%}=\frac{\mathrm{S}\mathrm{e}\mathrm{e}\mathrm{d}\mathrm{g}\mathrm{e}\mathrm{r}\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{\%}\left(\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\right)\times \mathrm{R}\mathrm{o}\mathrm{o}\mathrm{t}\mathrm{l}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{t}\mathrm{h}\left(\mathrm{t}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\right)}{\mathrm{S}\mathrm{e}\mathrm{e}\mathrm{d}\mathrm{g}\mathrm{e}\mathrm{r}\mathrm{m}\mathrm{i}\mathrm{n}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\mathrm{\%}\left(\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\right)\times \mathrm{R}\mathrm{o}\mathrm{o}\mathrm{t}\mathrm{l}\mathrm{e}\mathrm{n}\mathrm{g}\mathrm{t}\mathrm{h}\left(\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\right)}\times 100$(2)

Statistical analysis

Statistical analysis was performed using the Prism 5 Graph Pad program. Data are presented as means (SEM) and analyzed by one-way analysis of variance (ANOVA), followed by Bonferroni’s multiple comparison test with a significance level of P < 0.05.

Results and discussion

Temperature profiles

Temperature variation during the composting process has been widely recognized as an important factor that directly correlates with microbial activity. According to the evolution of temperature, the composting process can be divided into three phases: mesophilic, thermophilic, and curing (Nada 2015). Figure 2 shows the variations in the composting temperature and ambient temperature of the cylindrical and rectangular reactors. The temperature of the middle layer of the compost was always the highest, and the temperature in the lower layer was always the lowest owing to the cooling effect of incoming air. The initial temperature in all reactors was 25 °C, but after a few hours this started to increase. In the reactors with aeration rates of 0.3 and 0.6 L min⁻¹ kg⁻¹ DM, temperatures of above 55 °C were recorded. In contrast, the reactors with aeration rates of 0.9 L min⁻¹ kg⁻¹ DM, temperatures were below 50 °C throughout the experiment due to the high aeration rate. The increase in temperature may have been due to the high availability of C, which provides favorable conditions for the growth of microorganisms (Novinscak, Surette, and Filion 2007). Troy et al. (2012) reported that temperatures rise quickly in all reactors during the heating phase, indicating a rapid establishment of microbial activity. At day 1, the temperatures at the 0.3, 0.6, and 0.9 L min⁻¹ kg⁻¹ DM aeration rates were 40, 40, and 39 °C, respectively, for the cylindrical reactors, and 39, 39, and 38 °C, respectively, for the rectangular reactors. The thermophilic phase began in all reactors on day 2, with this phase lasting longer at the 0.3 and 0.6 L min⁻¹ kg⁻¹ DM aeration rates than it did at the highest aeration rate. The highest temperatures were recorded in all reactors on days 2–4. On day 3, the temperatures at the 0.3, 0.6, and 0.9 L min⁻¹ kg⁻¹ DM aeration rates were 55, 54, and 47 °C, respectively, for the cylindrical reactors, and 53, 52, and 47 °C, respectively, for the rectangular reactors. Due to microbial activity, smaller variation in temperature may occur throughout the composting process. The temperatures at the low aeration rate (0.3 L min⁻¹ kg⁻¹ DM) were above 55 °C for 2 days, which is considered sufficient to destroy pathogens (Strauch and Ballarini 1994). At the start, due to the rapid breakdown of the available OM and nitrogenous compounds by microbial activities, the temperature of all the compost mixtures rapidly increased. In both the cylindrical and rectangular reactors, the high temperatures lasted longer at the 0.3 L min⁻¹ kg⁻¹ DM aeration rate than at the 0.6 and 0.9 L min⁻¹ kg⁻¹ DM rates. The aeration system controls the temperature of a composting material, with a high aeration rate reducing thermophilic microbial activity in the substrate (Petric, Šestan, and Šestan 2009). The aeration rate might be considered an important factor affecting temperature patterns in the substrate. The temperature starts to decrease after the thermophilic phase due to a loss of substrate and a reduction in microbial activity (Ogunwande et al. 2008). The rise and fall in temperature have been found to strongly correlate with the microbial activity. Temperature is an excellent indicator of the microbial activity in a composting pile (Bernal, Alburquerque, and Moral 2009).

Figure 2. Variations in temperature of cylindrical and rectangular reactors during the composting process.

Comparing the cylindrical and rectangular reactors at an aeration rate of 0.3 L min⁻¹ kg⁻¹ DM, the temperature in reactors R1 and R4 reached in excess of 55 °C on day 3; however, the temperature in the rectangular reactor decreased more rapidly than that in the cylindrical reactor, perhaps due to the higher SA/V ratio. Conductive and radiative heat losses depend on the external surface of the reactor; as the SA/V increases, the potential for heat loss from the reactor walls increases (VanderGheynst, Gossett, and Walker 1997). The same phenomenon occurred for R2 and R5 with an aeration rate of 0.6 L min⁻¹ kg⁻¹ DM; however, there was little difference observed between R3 and R6 with an aeration rate of 0.9 L min⁻¹ kg⁻¹ DM.

Evaluation of ammonia and carbon dioxide

Ammonia volatilization increased in all reactors at the beginning of the experiment and then gradually decreased (Figure 3a, b). At the start of composting, ammonia volatilization in all reactors increased with increasing temperature, which comports with findings reported by Dewes (1996). In all reactors, the highest ammonia volatilization rates were observed during the thermophilic phase. The highest ammonia volatilization rates in R1 were observed on days 2–4, but little ammonia was lost after the thermophilic phase, probably due to the presence of little digestible organic N. The lowest ammonia volatilization rates were observed in R3 and R6, although during the later stages of the process, the rates in R3 and R6 were higher than those of the other reactors, likely owing to the transformation of digestible organic N to ammonium. In the first week of composting, there were no significant differences in ammonia volatilization across the reactors (P > 0.05); however, in the second week, significant differences were observed between R1 and R3, R1 and R5, R1 and R6, and R2 and R6 (P < 0.05). In the third week, significant differences were observed between all reactors (except for R1 vs. R2, R3 vs. R4, R3 vs. R5, and R4 vs. R5; P < 0.05). This study confirms that increase in temperature and increase in aeration rate increase ammonia volatilization. Previous studies also reported that an increased aeration rate is responsible for an increase in ammonia (NH₃) emissions (de Guardia et al. 2008). Zhang et al. (2016) concluded after comparing three treatments in his study that a higher aeration rate resulted in more NH₃ emission.

Figure 3. Changes in ammonia and carbon dioxide volatilization in the reactors during composting.

Volatilization of carbon dioxide closely followed the temperature profile of the substrate. The volatilization of carbon dioxide increased in all reactors at the beginning of the process and then started to decrease gradually (Figure 3c, d). Guo et al. (2012) reported that CO₂-C concentrations in the outlet air during composting were significantly correlated to their temperatures. Carbon dioxide volatilization was higher during the thermophilic stage, mainly due to the degradation of easily degradable C under healthy bacterial and fungal conditions (Nada 2015). During the maturation stage, the bacterial activity decreased; therefore, carbon dioxide volatilization was reduced. The mass of carbon dioxide was directly proportional to microorganism activity during composting. There were no significant differences in carbon dioxide volatilization across the reactors during the first week of composting. However, significant differences were observed during the maturation or curing stage (P < 0.05). The stability of compost can be assessed by the evaluation of carbon dioxide production. In the maturation or curing stage, the lowest carbon dioxide values were observed in R1 and R4 due to the higher compost stability in these reactors compared with the others. The highest carbon dioxide values were observed in R3, R5, and R6, which suggests the presence of unstable and immature compost in need of further decomposition.

Influence of temperature on ammonia volatilization

Temperature and pH are the main factors that contribute to ammonia volatilization. Ammonia emissions commence when both thermophilic temperatures (>45 °C) and high pH (approximately 9) are present in the compost environment (Beck, Smars, and Jonsson 2001). In Figure 4, the ammonia volatilization observed in all reactors is presented in relation to temperature. It can be observed that the trend lines for all reactors are more likely exponential than linear. For all reactors, an exponential fit showed the positive correlation between temperature and ammonia volatilization. For all reactors, the R² values of the exponential trend lines were higher than those of the linear trend lines. This demonstrates that the correlation between ammonia and temperature was indicative of exponential growth. Ammonia emission has been shown to be exponentially correlated with temperature during the first stage of composting (Pagans et al. 2006); however, the influence of compost temperature on ammonia emissions is not clearly understood (Beck, Smars, and Jonsson 2001).

Figure 4. Correlation between ammonia volatilization and temperature in each compost reactor.

Compost settlement and volume reduction

Settlement of compost can also affect the process to achieve successful composting; it starts as soon as the compost is put in the reactors. Settlement can lead to reduction in porosity and thus volume of compost reduces after few days; such occurrence can considerably affect the efficiency of oxygen supply and water evaporation and heat ventilation rates in a material (Barrington et al. 2003; Chang 2004).

In this study, although we didn’t focus on the influence of settlement on different parameters, we observed a reduction in volume of compost up to 10 cm, comparing the initial and final volumes of the compost. It might be due to taking samples from the compost every second day. Due to mixing of compost before sampling, the porosity remained the same during the whole experiment.

Changes in physiochemical characteristics during composting

pH

The pH level is an important characteristic of compost material. The optimum range for pH during composting is 5.5–8.0 (Venglovsky et al. 2005). The change in pH is also an indicator of the maturity of the compost. At the end of our study, the pH values in all reactors were found to be lower than the initial values. The initial pH in each reactor was 7.8, and it decreased to 6.89, 6.98, 6.83, 6.95, 7.12, and 7.07 in R1, R2, R3, R4, R5, and R6, respectively, as shown in (Table 2). The increase and decrease in pH are directly related to the ammonia content in the compost. This decrease is due to an increase in the production of organic acids and/or an increase in nitrification as a result of H⁺ released during microbial nitrification (Chen et al. 2015). Eklind and Kirchmann (2000) reported that the decrease in pH with composting time is caused by a reduction in the volatilization of ammonical N and the release of carbon dioxide during the composting process. Some of this decrease may also have been caused by the production of organic acids in the compost (Sweeten and Auvermann 2008).

Table 2. Initial and final physiochemical properties of compost in all reactors.

	R1		R2		R3		R4		R5		R6
Parameter	Initial	Final	Initial	Final	Initial	Final	Initial	Final	Initial	Final	Initial	Final
pH	7.81	6.89	7.80	6.98	7.81	6.83	7.79	6.95	7.8	7.12	7.79	7.07
EC (mS/cm)	2.12	1.0	2.13	1.2	2.12	1.5	2.11	1.3	2.12	1.5	2.13	1.6
MC (% ww)	65.1	63.7	65.0	62.3	64.9	63.2	65	62.2	65	63.4	65	62.3
C/N ratio	25	20.2	25	22.5	25	22.8	25	21.5	25	22.7	25	23.4

Notes. MC = moisture content; ww = wet weight.

Electrical conductivity (EC)

The EC of compost illustrates its degree of salinity, which is a potential indicator of its phytotoxic effect when used as a fertilizer (Lin 2008). An EC higher than 4 dS/m (Chen 1999) adversely influences plant growth, e.g., by producing low germination rates and withering. At the end of our study, the EC values in all reactors were found to be lower than initial values. The initial EC value in all reactors was 2.12 mS/cm, and this decreased to 1.0, 1.2, 1.5, 1.3, 1.5, and 1.6 mS/cm in R1, R2, R3, R4, R5, and R6, respectively, as shown in (Table 2). The initial EC may increase owing to the release of mineral salts such as phosphate and ammonium ions; however, during the maturation or curing stage, the EC starts to decrease owing to the evaporation of ammonium ions and the reduction of other basic groups (Wong, Li, and Wong 1995).

Moisture content

Moisture content should be maintained within an optimum range during composting to permit microbial growth and activity (Sesay, Lasaridi, and Stentiford 1998). At the end of our study, slight decreases in moisture content were observed in all reactors. The initial moisture content in all reactors was set at 65%, and the final moisture contents were recorded as 63.7%, 62.3%, 63.2%, 62.2%, 63.4%, and 62.3% in R1, R2, R3, R4, R5, and R6, respectively, as shown in Table 2. This decrease may have been caused by aeration and heat formation in the compost; at high aeration rates, the moisture content reductions were greater than those at low aeration rates. Arslan, Ünlü, and Topal (2011) and Walker et al. (1999) also reported that drying of the solid matrix increased with increases in aeration.

C/N ratio

The C/N ratio usually indicates the degree of maturity of compost (Bernal et al. 1998). The initial C/N ratio in all reactors was set to 25, and the final C/N ratios were 20.2, 22.5, 22.8, 21.5, 22.7, and 23.4 in R1, R2, R3, R4, R5, and R6, respectively, as shown in Table 2. The greatest reductions were observed in R1 and R4, which had an airflow rate of 0.3 L min⁻¹ kg⁻¹ DM. In the literature, the lower limit C/N ratio for mature compost is accepted to be >12 (Benito et al. 2003). In various previous studies, C/N ratios lower than those that we observed were reported; this is due to the longer periods of composting in those studies.

Organic matter degradation

The loss of organic matter from compost is related to microbial degradation/respiration (Paredes et al. 2002). The rate of organic matter loss is an indicator of the overall composting rate. The organic matter contents decreased in all reactors during the composting process due to microbial activity. The initial organic matter content in all reactors was 85%, and this decreased to 73.9%, 77.9%, 78.3%, 74.8%, 79.6%, and 78.8% in R1, R2, R3, R4, R5, and R6, respectively, as shown in Table 3. The greatest losses of organic matter were observed in R1 and R4, which had an airflow rate of 0.3 L min⁻¹ kg⁻¹ DM. This may be because R1 and R4 provided an adequate oxygen concentration for microorganisms, which enhanced compost material biodegradation. There were significant differences between R1 and the other reactors (except R4; P < 0.05). At the end of experiment, the volume and mass of composting materials were considerably reduced.

Table 3. Degradation of organic matter (dw) in all compost reactors.

Organic matter	R1	R2	R3	R4	R5	R6
OMi (%)	85 ± 0.7	85 ± 0.7	85 ± 0.7	85 ± 0.7	85 ± 0.7	85 ± 0.7
OMf (%)	73.9 ± 0.5	77.9 ± 0.2	78.3 ± 0.1	74.8 ± 0.4	79.6 ± 0.3	78.8 ± 0.6
k	0.5^a	0.37^b	0.36^b	0.47^a	0.31^b	0.34^bc

Notes. k = loss of organic matter; OM_i = initial organic matter content; OM_f = final organic matter content. Different superscript letters denote the significant differences in treatments.

Mesophilic and thermophilic bacterial populations

Composting can be an aerobic as well as anaerobic process, with organic waste degradation occurring due to a biological activity of microorganisms (Partanen et al. 2010). Both bacteria and fungi produce biological activity during the composting process (Gray, Sherman, and Biddlestone 1971). According to temperature, mesophilic and thermophilic bacteria play an important role during the composting process. Figure 5 shows the growth of mesophilic and thermophilic bacteria in all reactors at different stages of the process. In all reactors, as the temperature started to increase, mesophilic bacteria also increased until the start of the thermophilic phase. Thermophilic bacteria increased as the condition shifted from mesophilic to thermophilic. All the microbial populations depend on the early stages of the process. Ryckeboer et al. (2003) also reported that one of the factors affecting microbial communities is the aerobic condition at start of the composting process, which directly affects the temperature of the substrate.

Figure 5. Growth of mesophilic and thermophilic bacteria in all reactors during the composting process.

Phytotoxicity assay

The GI is an indicator of compost phytotoxicity (Gao et al. 2010). A low GI suggests high toxicity, which is also related to excessive salinity (Tiquia 2010). At end of the composting process, the GI% values were 82.3%, 71.4%, 61.7%, 68.8%, 63.6%, and 56.3% in R1, R2, R3, R4, R5, and R6, respectively (Figure 6). This shows that R6 had the highest level of toxicity; this reactor also had the highest EC value compared with the other reactors. R1 had the GI% above 80%, which indicates the low level of toxicity. Cunha et al. (2007) reported that a GI of more than 80% is an indication of a phytotoxic-free medium for plants. Significant differences were observed between R1 and all other reactors (P < 0.05). Moreover, significant differences were observed between R6 and all other reactors (P < 0.05). However, there were no significant differences between R2 and R5 and between R3 and R5 (P > 0.05).

Figure 6. Germination index (GI%) values of all compost reactors at the end of the composting process.

The differences in GI between the composts were due to microbial activity, which is affected by aeration rate (Chen et al. 2015). Owing to the longer periods of composting used, the GI values reported in several previous studies are higher than our results.

Conclusion

It can be concluded from these experiments that combination of rectangular or cylindrical reactors with high aeration rates is not suitable for the composting due to rapid heat loss. The heat loss from rectangular reactors was higher than that of cylindrical reactors owing to the higher SA/V ratio of the rectangular reactors. Aeration rate of 0.3 L min⁻¹ kg⁻¹ DM in a cylindrical reactor is recommended to maintain longest phase of thermophilic temperatures ≥55 °C. However, it is advised to keep the compost in reactors a bit longer than in this study to get more mature compost.

Additional information

Funding

This research was supported by Korea Institute of Planning and Evaluation for Technology in Food, Agriculture, Forestry and Fisheries (IPET) through Research Centre Support Program (project no. 717001-7), Ministry of Agriculture, Food and Rural affairs (MAFRA).

Notes on contributors

Waqas Qasim

Waqas Qasim was a master's student of biosystems engineering in the College of Agriculture and Life Sciences, Gyeongsang National University, Jinju 52828, Korea.

Byeong Eun Moon

Byeong Eun Moon is a Ph.D. student of biosystems engineering in the College of Agriculture and Life Sciences, Gyeongsang National University, Jinju 52828, Korea.

Frank Gyan Okyere

Frank Gyan Okyere is a master's student of biosystems engineering in the College of Agriculture and Life Sciences, Gyeongsang National University, Jinju 52828, Korea. 

Fawad Khan

Fawad Khan is a master's student of biosystems engineering in the College of Agriculture and Life Sciences, Gyeongsang National University, Jinju 52828, Korea. 

Mohammad Nafees

Mohammad Nafees is a professor in the environmental science department, University of Peshawer, KPK, Pakistan.

Hyeon Tae Kim

Hyeon Tae Kim is a professor in the College of Agriculture and Life Sciences, Gyeongsang National University (Institute of Agriculture & Life Science), Jinju 52828, Korea.