Fermentation characteristics of bedded pack barn dairy cattle manure on methane yield, carbon, and nitrogen content in solid-state anaerobic digestion

Dairy cattle manure

The BDCM was collected at a facility with 0.02 head/m² density of dairy cattle and 3.03 kg/m² bedded sawdust in a cow farm in Yangpyung province (Latitude, 37.4614204; longitude, 127.5285745). The moisture content and volatile solids in sawdust were recorded as 36.1 ± 0.2% and 98.6 ± 0.1%, respectively. The BDCM characteristics considered in this study are listed in Table 1. Manure samples were randomly collected from at least five or more points and cow barns. Manure was collected eight times or more from each sampling point (Dou et al., 2001). The collected samples were classified using the VDI4630 method (Standard, 2006) and stored at −20 °C until further analysis.

Batch SSAD

The SSAD tests were carried out in triplicate using an approximately 1,400 mL polypropylene batch digester equipped with a 2 L gas bag at a constant temperature of 39 °C. The BDCM stored at 4 °C was pre-incubated at a temperature up to 39 °C before AD. The anaerobic state was established using N₂ gas inside the bottles which were kept in a shaking incubator (IS-971R; Jeiotech Co., Korea) until less than 1% CH₄ was produced. The substrate was supplied in the digester at 462.3 ± 4.9 g (wet basis) weight and moisture content adjusted to 82% using a mixture of distilled water, inoculum, and medium. Additionally, inoculum and medium were added at 0.88 and 0.44 g/g of total solids, respectively. The inoculum used in the study could digest cattle manure fed various substrates, and was collected from the batch type mesophilic anaerobic digester at 39 °C, which produced about 500 N mL CH₄/g volatile solids (VSs)/d of biogas with 60% methane concentration The medium used for AD was prepared using the method described by Angelidaki & Sanders (2004) (Table 2). The final pH of the medium was adjusted to 7.1 using CO₂ gas, and stored at 39 °C before AD. All assays were performed once every three days to measure gas production during the experimental period. Sampling was performed using a 50 mL gas-tight glass syringe (1001 SL SYR; Hamilton Co., Reno, USA) to measure gas production and composition every three days for 72 days. The sampling for recording pH, ammonia nitrogen, volatile fatty acids (VFA), and chemical composition was performed on the 0, 3, 6, 12, 18, 24, 30, 36, 48, 60, and 72nd days.

Analytical method

Total and volatile solids were determined using the method described by APHA APH (2018). Elemental analyses of C, H, N, S, and O were performed using an elemental analyzer (EA 1110, CE Instruments, Italy), and pH was determined using a pH meter (Orion 420A+; Thermo Electron Co., Waltham, MA, USA). Ammonia nitrogen content was determined using the method described by (Rice, Baird & Eaton, 2017) and an ELISA reader (Synergy 2, Biotek Co., USA) according to Eq. (1) (Drosg, 2013). (1)${\displaystyle {\mathrm{NH}}_3\left(\mathrm{mg}/\mathrm{L}\right)=\frac{{\mathrm{NH}}_4-\mathrm{N}}{1+10^{\left(0.0925+2728.795/\mathrm{T}-\mathrm{pH}\right)}}}$ T: 273.15 K + gas temperature (° C)

Gas production was determined daily using a 50 ml glass syringe. The amount of gas produced was calibrated to standard temperature and pressure (STP 0 °C, 1 atm) considering the temperature-dependent volume using the following calibration (Eq. (2)): (2)${\displaystyle \mathrm{V}={\mathrm{V}}_{\text{at T}°C}\times \frac{273}{\left(273+\mathrm{T}\right)}\times \frac{\left(\mathrm{P}-{\mathrm{P}}_{\mathrm{W}}\right)}{760}}$ where, V is the gas production at 0 °C and 1 atm, V_at T^∘C isthe gas production at T^∘C, T is the temperature at the time of the volume measurement, P is the pressure at the time of the volume measurement, and P_w is the saturated water vapor pressure at T^∘C. The gas composition was determined by using gas chromatography (HP 6890; Hewlett-Packard Co., Palo Alto, CA, USA) with a thermal conductivity detector. The sample (0.2 ml) containing helium as the carrier gas was injected into the column at a temperature of 60 °C and flow rate of 1.5 ml/min. The sample gas concentration was calibrated using a standard gas mixture consisting of 40% CH₄–60% CO₂ and 60% CH₄–40% CO₂.

The amount of methane produced from the serum bottle was calibrated using the calibration equation (Eq. (3)) presented by Shin (2002): (3)${\displaystyle {\mathrm{C}}_{{\mathrm{CH}}_4\left(\mathrm{R}\right)}={\mathrm{C}}_{{\mathrm{CH}}_4}\times \frac{100}{\left({\mathrm{C}}_{{\mathrm{CH}}_4}+{\mathrm{C}}_{{\mathrm{CO}}_2}\right)}}$ where, C_CH4(R) is the calibrated methane concentration (%), C_CH4 is the measured methane concentration (%), and C_CO2 is the measured carbon dioxide concentration (%). The volatile solid reduction rate was determined by the equation described by Kim et al. (2018).

Ultimate biodegradability and biomass removal

Total volatile solids (TVSs) consisted of biodegradable volatile solids (BVSs) and non-BVSs, and the TVS residue (TVS_e) after degradation was calculated using Eqs. (4)–(6). The ultimate biodegradability was calculated by plotting the ratio of TVS_e and initial TVS before degradation (TVS₀) on the Y-axis and reciprocal of the operating time (1/time) on the X-axis according to the method described by Kim et al. (2018). (4)${\displaystyle \text{Biomass removal (BMR)}={\text{CH}}_4\text{weight}+{\text{CO}}_2\text{weight}}$ (5)${\displaystyle \mathrm{BMR}=\frac{{\mathrm{V}}_0\times \left(\frac{16\mathrm{g}}{1\text{mole}}\times \frac{{\mathrm{CH}}_4}{100}+\frac{44\mathrm{g}}{1\text{mole}}\times \frac{{\mathrm{CO}}_2}{100}\right)}{\frac{22.413\mathrm{l}}{\text{mole}}}}$ (6)${\displaystyle {\text{TVS}}_{\mathrm{e}}={\text{TVS}}_0-\text{BMR}}$

Kinetic modeling

The daily methane yield rate from BDCM during SSAD was simulated using the modified Gaussian and Gomperz equations (Kim et al., 2018). This model describes the daily methane yield in batch AD assuming that the kinetic growth of microbes including methanogens follows a normal distribution during the process. The Gaussian equation is presented as Eq. (7). (7)${\displaystyle y=a\times exp\left[-0.5{\left(\frac{t-t_0}{b}\right)}^2\right]}$

y = Methane production rate (N mL/g VS) at any time t

t = digestion time (Day)

a = Constant (N mL/g VS/day)

b = Constant (Day)

t₀ = Maximal methane production rate day (Day)

The methane yield potential (P), maximum methane production (Rm; N mL/g VS), and lag phase (λ) were based on the volume and concentration of the biogas measured under the conditions of this experiment using the modified Gompertz equation (Kafle & Chen, 2016). (8)${\displaystyle \mathrm{M}=\mathrm{P}\times exp\left\{-exp\left[\frac{R_m\times \mathrm{e}}{\mathrm{P}}\left(\lambda -\mathrm{t}\right)+1\right]\right\}}$ M: Cumulative methane yield (N mL/g VS)

P: Methane yield potential (N mL/g VS)

R_m: Maximum methane production (N mL/g VS day)

λ: Lag phase

t: Day

Statistical analysis

Data were analyzed using PROC MIXED of the SAS package program (SAS Inst. Inc., Cary, NC, USA) using retention time as a fixed model. Orthogonal polynomial contrasts were performed using the CONTRAST option to verify significance and polynomials were determined using SAS PROC IML. Statistical differences were considered significant at p < 0.05, and all averages were presented as least square means. Figures were created using GraphPad Prism 7.0 (GraphPad Software, San Diego, CA, USA).

Methane production, kinetic modeling, and methane content

The methane production on daily and cumulative basis during batch SSAD, and from the kinetic model using a modified Gompertz curve are shown in Fig. 1 and Table 3, respectively. The methane production from BDCM during batch SSAD increased rapidly from day 12 onwards with maximum yield of 19.6 ± 0.8 N mL/g VS/day, whereas maximum methane production was observed at 24.0 ± 1.6 days. The R² value of the Gaussian parameter using the modified Gompertz equation was 0.5992. Cumulative methane of BDCM was 142.5 ± 6.0 N mL/g VS at the end of SSAD. The maximum methane potential of BDCM was recorded as 134.7 ± 9.9 N mL/g VS according to the Gomperz curve and maximum methane production rate was 6.1 ± 0.4 N mL/g VS/day. The lag phase was 12.9 ± 1.1 days. The R² value of the Gompertz parameter using the modified Gompertz equation was 0.9373. The methane content of BDCM in the batch SSAD is shown in Fig. 2 with a maximum value of 76.1 ± 7.3%. Results demonstrated that methane content increased rapidly until the 18th day, and it was maintained at 62.4 ± 4.8% until the end of the experiment. The R² value of the trend line obtained using the sigmoidal curve fitting was 0.9054.

Ultimate biodegradability (UB) and volatile solid removal rate

The UB and VS removal rates of BDCM in the batch SSAD were found to be 23.1 ± 3.4% and 19.9 ± 2.3, respectively (Table 4). Moreover, the BVS removal rate was 86.4 ± 1.4%.

Fiber content removal

The results of cellulose, hemicellulose, and lignin contents of BDCM in SSAD are shown in Fig. 3. The cellulose and hemicellulose contents decreased linearly during SSAD (p = 0.001 and p < 0.001, respectively). In contrast, the lignin content increased (p = 0.001). The linear regression equations for cellulose, hemicellulose, and lignin content were as follows: cellulose (%, v. b.) = −0. 282 × fermentation time + 17.135, hemicellulose (%, v. b.) = −0. 629 × fermentation time + 16.126, and lignin (%, v. b.) = 0. 485 × fermentation time + 14.733.

Carbon and nitrogen removal

The carbon content of BDCM in the batch SSAD decreased quadratically during SSAD (p < 0.001; Fig. 4). The regression equations for non-fibrous carbon fraction (NFCF), hemicellulose carbon fraction (HCF), and acid detergent insoluble carbon (ADIC) were as follows: NFCF (%, v. b.) = 0.001 × Fermentation time² - 0.1066 × fermentation time + 15.268, HCF (%, v. b.) = y = 0.0019 × Fermentation time² - 0.2172 × fermentation time + 16.125, and ADIC (%, v. b.) = y = 0.0001 × fermentation time²- 0.0417 × fermentation time + 10.76.

The nitrogen content of BDCM in the batch SSAD is shown in Fig. 5. Results revealed that the non-fibrous nitrogen fraction (NFNF) and hemicellulose nitrogen fraction (HNF) contents decreased quadratically (p = 0.002 and p = 0.038, respectively). In contrast, no significant decrease in acid detergent insoluble nitrogen (ADIN) content was observed (p = 0.836). NFNF and HNF rapidly decreased until the day 12 and 6 of digestion, respectively (p < 0.001 and p = 0.008) but did not significantly decrease after day 15 until the end of the experiment. The regression equations of the NFNF, HNF, and ADIN were as follows: NFNF (%, v. b.) = 1.119 + (0.378/(1 + (Fermentation time/4.747)^2.686), HNF (%, v. b.) = 0.268 + (0.431/(1 + (Fermentation time/2.737)^19.421), and ADIN (%, v. b.) = 0.319 + (−0.046 / (1 + (Fermentation time/20.489)^122.57)).

pH, NH₃, NH₄-N, and volatile fatty acids

Results demonstrated a change in the pH of BDCM ranging from 7.5–8.0 in batch SSAD until day 12 of fermentation. However, it increased to 8.5 and was maintained at 8.0 thereafter (Fig. 6).

The ammonia nitrogen (NH₄-N) content and maximum free ammonia (NH₃) concentration of BDCM was maintained at 1,200 mg/L and 147.7 ± 13.8 mg/L during the entire batch SSAD (Fig. 7). However, the NH₃ concentration decreased to <100 mg/L after 18th day.

Results of VFA concentration of BDCM in the SSAD batch are shown in Fig. 8, and revealed that the maximum acetic, propionic, and VFA concentrations were 223.0 ± 10.8, 110.5 ± 11.9, and 357.3 ± 21.5 mg/L, respectively. Moreover, VFA concentrations were mostly maintained after 36 days of batch SSAD.