1. Introduction
Poultry production is one of the fastest growing industries worldwide. Global poultry production exceeds 130 million tons per year, and increases annually [
1]. Every 1000 birds produce approximately 1.2 tons of manure. With around 7 batches per year, 20,000 tons of manure are produced every year [
2]. Chicken manure comprises nitrogen (70% uric acid and 30% undigested protein) along with considerable amounts of phosphorus and potassium [
3]. The main method of chicken manure disposal is its use as a fertilizer. However, the annual amounts produced exceed the amounts required for the fertilizers needed for agricultural land in many countries [
3]. Thus, it is imperative to manage CM properly, lest it cause serious environmental problems, including but not limited to foul smell, pest problems, eutrophication of surface water with a huge amount of nutrients [
4], contamination of groundwater with pathogens [
5] and heavy metals, and increasing global warming [
6].
Using microbial communities for the anaerobic digestion (AD) of chicken manure has been approved as an efficient alternative for disposing of CM via biogas production Methane produced by the co-digestion of CM is a feasible and long-term renewable energy source that can be replicated using kinetic and mathematical models [
7]. A variety of feedstocks have been utilized to produce biogas, including energy crops, agricultural and food industry waste products, and municipal waste. However, numerous countries have recommended limiting energy crops due to sustainability and agricultural land issues, and instead focusing on the use of waste and residues for biogas production [
8]. AD has been widely used to treat several types of manure, including livestock and swine manure; however, its use for chicken manure treatment has been set back by its high ammonia concentration [
9]. Many attempts have been made to mitigate the inhibitory effect of ammonia on AD, including ammonia removal [
10] and co-digestion of CM with agriculture waste [
11]. Inoculum type, activity, and inoculum-to-substrate ratio are very important factors affecting this technology [
12].
Marine sediment (MS), which constitutes 70% of the Earth’s surface, is considered a promising microbial source for AD [
13,
14]. The collective number of microbial cells in MS is estimated to be between 3 and 6 × 10
29 cells per gram, decreasing as the depth increases. MS consists of bacteria and archaea, with bacteria being more predominant.
Proteobacteria—both
Alphaproteobacteria and
Gammaproteobacteria—prevail together with
Firmicutes [
15]. Endospores from the families
Bacillaceae,
Lachnospiraceae,
Clostridiaceae, and
Ruminococcaceae, along with their vegetative forms, heavily populate MS [
16]. Miura et al [
13] assessed MS from different sources as auspicious microbial sources for the AD of
Saccharina japonica—a brown alga—at seawater salinity. They concluded that MS could be used as a microbial source for methane production from algae under high-salt conditions, without dilution [
12].
Chemometrics originated as a branch of science concerned with using mathematics and statistics to interpret the results of chemical experiments, hence the name chemometrics [
17]. Recently, Bystrzanowska [
18] referred to it as “an interdisciplinary field that uses mathematical and statistical methods to design or select optimal measurement procedures and experiments and to provide maximum chemical information by analyzing chemical data”. Response surface methodology is one of the most commonly used chemometric tools; it is now preferred for optimization studies, as they have many advantages, including cutting down the number of experiments, therefore saving time, effort, and chemicals, and providing mathematical models for evaluating the statistical significance of factors, both independently and interactively [
19]. This is especially important when factors interact significantly; in this case, univariate tools will not provide accurate maxima, and the more significant the interactions, the less precise the univariate optimization results will be; hence, the importance of using statistical models that change variables simultaneously [
18].
To the best of the authors’ knowledge, no previous studies have used MS as a source of inoculum for the AD of CM to mitigate ammonia inhibition. Accordingly, the objectives of this study were investigating the use of MS as a source of methanogenic bacteria for the AD of CM, comparing production using MS to that of OS, and applying response surface methodology to optimize methane production using MS, in addition to investigating the tolerance of methanogenic bacteria to the accumulation of ammonia.
4. Discussion
The source of anaerobic microorganisms and the substrate-to-inoculum ratio are critical factors affecting the quality of the AD process [
12,
22]. Our results revealed the biogas production at different CM–inoculum ratios (OS or MS)—from 1:30 to 1:2.5, as described in conditions 1–5—as shown in
Table 4 and
Figure 1, the former of which also shows changes in pH and the production of volatile fatty acids (VFAs; acetic and propionic acids) and ammonia over the fermentation period. Variable levels of biomethane production were recorded when culture media containing MS were tested in the presence and absence of 3% NaCl. In addition, using MS as an inoculum resulted in the successful creation of methane from CM as a sole substrate under mesophilic (35 °C) conditions. High methane yield (305 mL/g VS-CM) and final ammonia accumulation as low as 0.78 g NH
3-N/kg CM were obtained at a CM:MS ratio of 1:10. The addition of NaCl had a constructive effect on biomethane production only at the inoculum size of 1:10. However, the optimal biomethane production conditions as devised by the RSM models did not include NaCl supplementation.
Comparing methane production from CM under semi-solid conditions using either MS or OS as the inoculum source, it was noticed that OS resulted in higher methane production. However, the use of MS resulted in increased ammonia tolerance, and succeeded in producing methane (301 mL/gVS-CM) despite the accumulation of ammonia (8.58 g NH3-N/kg CM)—much higher than the reported inhibitory level. While increasing the inoculum size (CM:OS ratio = 1:30) reduced methane production when using OS, it increased methane production when using MS.
In the present study, the use of MS as an inoculum resulted in the successful production of methane from CM as a sole substrate under mesophilic (35 °C) conditions. High methane yield (305 mL/g VS-CM) and final ammonia accumulation as low as 0.78 g NH
3-N/kg CM were obtained at a CM:MS ratio of 1:10. Marine sediment (MS) was first used by Aspé et al. [
23] as a source of anaerobic bacteria to purify fish waste containing high salt content by mixing it with fresh pig manure [
23]. Tor et al. [
24] stated that the microorganisms living in marine sediments have significant potential for the anaerobic digestion of a variety of organic substrates, and found that sulfate-reducing microorganisms present in MS appear to be involved in the metabolism of both acetate and hydrogen produced by fermentative microorganisms; thus, both fermentative and sulfate-reducing microorganisms cooperate to oxidize glucose.
Miura et al. [
13] assessed MS from different sources as auspicious microbial sources for the AD of Saccharina japonica—a brown alga—at seawater salinity; they concluded that MS could be used as a microbial source for the production of methane from algae under high-salt conditions, without dilution [
12]. Accordingly, in the present study, NaCl at 3%—the concentration commonly used to simulate seawater salinity [
25]—was tested for its effect on the AD process using MS. Our results show that the addition of NaCl had a positive effect on biomethane production only at the inoculum ratio of 1:10. However, the optimal biomethane production conditions as determined by the RSM models did not include NaCl supplementation. This may be because one of the cons of using CM as a substrate is its high concentration of salts (electrical conductivity (EC) = 20 dS/m) and high pH (pH 8.0), which may be expected to inhibit methane production—especially under dry conditions [
26,
27]. Anwar et al. [
28] studied the impact of Na salts on the anaerobic digestion of kitchen waste; they found that as sodium salt concentration increased, the methane yield and the maximal methane production rate decreased, along with the lag-phase time and accumulation of VFAs. Accordingly, they recommended that sodium salt concentration in the anaerobic digesters should be maintained below 8 g/L.
Comparing methane production from CM under semi-solid conditions using either MS or OS as the inoculum source, it was found that OS resulted in higher methane production. Nevertheless, MS use resulted in increased ammonia tolerance, as it succeeded in producing methane (301 mL/GVS CM) despite the accumulation of ammonia (8.58 g NH
3-N/kg CM)—much higher than the reported inhibitory level [
29]. This may be attributed to the buffering action resulting from better degradation of VFAs (acetate), thus preventing the drop in pH (
Table 4) [
11]. While increasing inoculum size (CM:OS ratio = 1:30) decreased methane production when using OS, it increased methane production when using MS. This result may support the conclusion that MS has superior tolerance to higher ammonia levels compared to OS.
Elasri et al. [
30] stated that the energy recovery of CM depends on the waste–inoculum ratio. The maximal biogas production was obtained from a high waste–inoculum ratio (1:7), while the ratio that reached the highest methane percentage was 1:1. Marchioro et al. studied three substrate–inoculum ratios (1:1; 1:1.66, and 1:3) of poultry litter dry AD at 37 °C; they obtained the highest biogas and methane yields of 183 LN biogas.kg−1 VSadd and 74 LN methane.kg−1 VSadd, respectively, obtained at the substrate–inoculum ratio of 1:3 [
31]. Abouelenien et al. [
32] reported the highest methane yield of 49 mL/g VS when the CM–inoculum ratio was 1:2 at 35 °C, and 103.5 mL/g VS when the CM–inoculum ratio was 1:1 at 55 °C.
As shown in
Table 9, the obtained amounts of methane in this study were higher than those reported in previous studies. Cheong et al. [
7] produced 396–540 mLg
−1 VS of methane by co-digesting CM with process effluent from a bioethanol plant (10%;
v/
v), and with free ammonia (NH
3-N; less than 20 mg/L), throughout the experiment. Abouelenien et al. [
11] reported the production of 506 CH
4 mL/g VS from the co-digestion of CM with agricultural waste, with ammonia accumulation of 1.3 g NH
3-N/kg CM. The solid content of the fermentation medium is one critical factor affecting the overall performance and efficiency of the AD process [
5].
It was previously reported that dry AD needs longer lag phases and larger amounts of inoculum than wet digestion systems. Dry AD often results in poor start-up performance due to incomplete mixing and accumulation of VFAs [
33,
34]. In this study, under dry conditions (20% TS), the highest biogas production (320 mL/g VS-CM) and methane production (169.3 mL/g VS-CM) were much higher than those obtained by the authors of [
32], who obtained 31 mL/g VS-CM after an acclimatization period of 254 days. Bujoczek et al. [
35] fermented CM at different TS percentages, and they failed to produce methane from CM containing 21.7% TS, even after 120 days of fermentation at 35 °C. Magbanua et al. [
36] obtained very low amounts of methane that did not exceed 0.9 mL g
−1 VS, even after 99 days of batch testing using CM containing TS (17.4%) and VS (14.6%). It can be suggested that MS methanogenic consortia are highly tolerant to the accumulation of ammonia, regardless of the total solid percentage. This is the highest methane production from raw CM under dry conditions (20% TS) without inhibition or VFA accumulation, to the best of our knowledge. This may be attributed to the buffering action generated by the degradation of acetate (0 mM/gVS), which maintains a pH level (8) suitable for AD, despite the ammonia concentration of 4.2 g NH
3-N/kg CM. Accumulated acetate may have combined with ammonia to form ammonium acetate, resulting in a drop in pH [
11,
37].
Multivariate statistical tools have recently become the tools of choice for optimization purposes due to their multiple advantages, including reducing the number of performed experiments, thus saving time, effort, and reagents, in addition to providing mathematical models for evaluating the statistical significance of factors, both independently and interactively [
19]. This is especially important when factors interact significantly; in this case, univariate tools will not provide accurate maxima, and the more significant the interactions are, the less precise the univariate optimization results will be and, hence, the more important it is to use statistical models that change variables simultaneously [
18].
Evaluation of the statistical significance of the model was done through ANOVA (
Table 6). The designed model is highly significant, with model F-values as high as 932.94, and (P model > F) values < 0.0001; this means that there is a less than 0.01% chance that the results of this model would occur due to noise. The model shows a perfect fit, with an R2 value of 0.9908, indicating that it can account for 99% of variability [
38]; it also possesses a predicted R2 value of 0.9898, indicating the model’s excellent ability to predict the responses of non-experimented values for the test parameters [
39]. Adjusted R2 measures the ability of the model to describe variation around the mean, taking into consideration the number of terms included in the model. Accordingly, this value generally decreases as the number of factors that do not add value to the model increases. The model has an adjusted R2 value of 0.9882, indicating that the model would have the ability to explain ~98% of the variability if it were derived from different samples [
40].
A model is said to be adequately precise when the value of the parameter “adequate precision” is higher than 4.0. Adequate precision measures signal-to-noise ratio, and reflects the ability of the model to navigate the design space [
39]. The model has adequate precision of 48.68, indicating excellent precision. Nevertheless, the ability of a model to produce congruent results when tests are redone under the same conditions is assessed through its “coefficient of variation” (CV), which is calculated by dividing the standard deviation by the mean. The lower the CV value, the more consistent the results are expected to be [
41]. The CV values recorded for this model were as low as 10.9%, indicating a reliable model.
F-values and
p-values were used to assign the significant model terms, as shown in
Table 7. The 3D surface and contour plot are shown in
Figure 3. The highest
p-value was recorded for the square term of incubation time, indicating that this variable maximizes methane production; any slight change in its value will be strongly reflected in the amount of methane produced [
42].
Under the optimal conditions, a methane production of 321.73 mL/g VS was obtained—close to the predicted value. Li et al. [
33] produced ~205 mL/g VS from the semi-solid co-digestion of CM:CS (10.1–11.2% TS) under mesophilic conditions. Additionally, [
43] reported the production of 107.25 mL g
−1 TS (76.92% methane) using a mixture of CM with
Spartina alterniflora residues (SAR) at 35 °C, with an initial TS of 8%. Miura et al. [
21] used acclimatized MS for the biomethanation of raw brown algae, and they obtained methane production of 300 mL/g VS. In all of the aforementioned studies, CM was treated either by co-digestion or by ammonia stripping to enhance methane production, but in the present study CM was used as a sole substrate, without any pretreatment. Additionally, using OS as an inoculum maintains the accumulation of ammonia at all ratios below the inhibitory level of methanogenic bacteria [
10,
32,
33].
Furthermore, Elsayed et al. [
44] used three types of inocula—fresh cow manure, activated sludge, and excess sludge—in search of maximal biomethane production from primary sludge (PS) co-digested with fruit and vegetable waste, and recorded that activated sludge was the inoculum source supporting the highest methane yields (141 mL/g VS) [
44]. Likewise, Saad et al. [
45] tested the effects of using aeration tank sludge, return-activated sludge, and palm oil mill effluent sludge as microbial sources on the anaerobic co-digestion of food waste and CM; they found that the highest biogas production (120.97 N mL/g COD) was obtained when aeration tank sludge was used as the inoculum. In all of the above studies, CM was treated either by co-digestion or ammonia stripping to enhance methane production. Still, CM was used as a sole substrate without any pretreatment in the present study.
Using OS as an inoculum maintains the accumulation of ammonia at all ratios below the inhibitory level of methanogenic bacteria [
10,
11,
32]. This experiment was conducted to investigate the effect of increasing ammonia concentration on biomethane production by methanogenic consortia in MS, using an experimental glucose-containing medium. Biomethane production, expressed in mL/g glucose consumed by the microorganisms, decreased as ammonia concentration increased, as shown in
Table 8, along with the changes in pH, ammonia levels, and VFA levels over the fermentation period. Our results confirm the absence of any complete inhibition of methane production from glucose when using MS—even at an ammonia concentration of 4.2 g NH
3-N/kg CM (methane production = 30.5 mL/g glucose).
Nevertheless, methane production decreased with the increase in ammonia concentration [
10]. The highest methane production when using a glucose medium (130.6 mL/g glucose) was at an ammonia concentration of 0.8 g NH
3-N/kg CM, with the least amount of final acetate (3.2 mM/g glucose) and a pH of 7.5. The most exciting finding in this experiment was that pH and acetate accumulation were not profoundly affected by the increase in ammonia concentration (from 1.1 to 4.2 g NH
3-N/kg CM). This result supports some buffering action that inhibits the drop in pH and helps to continue methane production.
Practically, the AD of CM is a perfect challenge in the pursuit of renewable sources of energy through biogas production; at the same time, it acts as a way to preserve the environment from contamination. On the other hand, ammonia inhibition is an obstacle to this process, so seeking ammonia-tolerant methanogenic consortia was a practical aim. MS methanogenic consortia, according to this study, seem to be ammonia-tolerant. Additionally, MS is reasonably available, as it constitutes 70% of the Earth’s surface.
Despite the biogas yield from a CM–MS mixture being greater in 20% TS than in any other previous study, the amount of biogas produced under lower TS% (10% and 5% TS) was still lower than that obtained by using Ozouh sludge as the inoculum. Therefore, further improvements in the overall conditions, along with other studies on methanogenic consortia in MS, are needed.
Table 9.
Summaries of the previous trails for the co-digestion of CM with various organic wastes.
Table 9.
Summaries of the previous trails for the co-digestion of CM with various organic wastes.
Co- Substrate and/or Inoculum Source | Experimental Conditions |
---|
RetentionTime (d) | Temperature (°C) | TS% | Ratios of Substrates | Methane (CH4) or Biogas Yields | (NH3) or (NH4+) Values | References |
---|
Poultry (P) and hogwastes (H) | - -
99
| - -
35
| - -
17.4%
| - -
H:P; 0:100
| CH4 = 0.9 mL g−1 VS | 1.66 g NH3 L−1 | [36] |
Inoculum obtained from sludge(Ozouh) obtained after thermophilic anaerobic digestion of excess activated sludge | 254 d254 d | 35 °C55 °C | 25%25% | - -
CM–inoculum ratio was 1:2 - -
CM–inoculum ratio was 1:1
| - -
49 mLg−1 VS methane - -
103.5 mLg−1 VS methane
| - -
8 g-N kg−1 CM. - -
14 g-N kg−1 CM
| [32] |
Agriculture wastes (AWS) | Batch I, 2, 3 and 4 = 40 d, 35, 39 and 62 d | 35 and 55 °C | 10% | CM:AWS is 7:3 (v/v) | 506 CH4 mL/gVS | 1.3 g NH3-N/kg−1 CM | [32] |
Inoculum obtained from cattle manure under mesophilic conditions | 50 d | 35 °C | 14–32% | (1:1) to (1:7) ratio of CM and inoculum | 73.3 NmL g−1 VS biogas | - | [7] |
Mesophilic acclimated inoculum from a lab-scale CSTR (continuous stirred-tank reactor) with swine and dairy waste water | 30 d | 37 °C | 75% | Poultry litter–inoculum ratios of 1:1; 1:1.66, and 1:3 | 183 LNbiogas.kg−1 VS and 74 L Nmethane.kg−1 VS | Below 60 mgNH3.L−1 | [30] |
Process effluent from a bioethanol plant (BPE) | 165 d | - | 13.2–28.9% | 0%, 10%, and 20% (v/v) of CM: | 396–540 mLg−1 vs. methane | Less than 20 mgL−1 | [7] |
Marine sediment (MS) or Ozouh sludge(OS) (thermophilic anaerobic digestion of excess activated sludge) | - -
45 d
| - -
35
| - -
10%
| - -
CM:MS/OS 1:30; 1:20; 1:10; 1:5; 1:2.5, and 1:0
| - -
Using MS resulted in 305 CH4 mL/g VS-CM at CM: MS (1:10). using OS resulted in 633.76 CH4 mL/g VS-CM at CM: MS (1:30). | - -
For MS: 0.78 g NH3-N/kg−1 CM. For OS: 1.02 g NH3-N/kg−1 CM.
| [46] |
Marine sediment (MS) | - -
34 d
| - -
35
| - -
5% - -
10% - -
20%
| 1:2.5 | - -
Biogas = 320.46 mL/g VS CM - -
Methane 169.28 mL/g VS CM
| 4.178 g NH3-N/kg−1 CM | |