Evaluation of Biogas Production from Swine Manure Using a UASB Reactor (Upflow Anaerobic Sludge Blanket) with Long-Term Operation

: To meet Colombia’s energy needs by 2050, a total installed capacity of 42 MW across its power generation infrastructure is required. To achieve this, transitioning to cleaner energy sources, such as biomass—a non-conventional renewable energy—is necessary. Biomass is a promising renewable source for thermal and electrical energy production. This study researched the production of biogas from swine manure using a UASB reactor to valorize this waste. Swine manure was collected every 20 days from a pig farm with a capacity of 200 sows, located in Santa Rosa de Osos, Antioquia. The flow rate was increased three times (1.30 L d − 1 , 1.62 L d − 1 , and 2.08 L d − 1 ) to reduce the hydraulic retention time (HRT) and enhance biogas production. The volatile and total solids, chemical oxygen demand (COD), alkalinity, and biogas composition were measured over one year. The proposed system achieved 87.40% COD remotion from the feed stream and generated a yield of 507 mLCH 4 g VS − 1 , with an HRT of 19 days and an OLR of 4.27 gCOD L − 1 d − 1 . The reactor produced biogas with a CH 4 content of 67.7%, CO 2 content of 18.1%, and H 2 S content of 1413 ppm. This study highlights the effectiveness of the UASB reactor for biogas production using swine manure as a substrate.


Introduction
The worldwide installed renewable energy capacity has increased by 37% over the past 12 years, reaching 3.37 TW in 2022, and it is projected to reach 7.30 TW by 2028.This growth is attributed to the implementation of renewable energy policies in some countries and the decreasing costs of solar and wind technologies for electricity production.Countries like Colombia, which have abundant water resources, have historically relied heavily on hydropower, which has met over 70% of Colombia's energy demand in the last ten years [1,2].Nevertheless, the contribution of non-conventional renewable energy sources (NCRES) to the country's energy generation remains limited.In 2022, solar photovoltaic and wind power accounted for 0.6% and 0.1%, respectively, of the total energy generated [3].This suggests a high reliance on hydroelectric power plants for the supply of electricity in the country.
According to the Mining-Energy Planning Unit (UPME, its acronym in Spanish), NCRES have the potential to fulfill up to 46% of Colombia's energy needs.Biomass, small hydroelectric power plants, geothermal power, solar power, and wind power plants are classified by the UPME as NCRES in Colombia [4].To expand and diversify its energy matrix, the Colombian Government has set ambitious goals in the National Energy Plan 2020-2050 (PEN, its acronym in Spanish).The PEN proposes that by 2050, Colombia will reach an installed Energies 2024, 17, 2723 2 of 14 capacity of 42,709 MW, distributed as follows: 43.2% from NCRES, 37.4% from hydroelectric plants, and 19.2% from thermoelectric plants [3].This is to prevent energy rationing due to the reduced water storage in reservoirs during droughts (e.g., the El Niño phenomenon) and to avoid increases in energy costs.
The use of biomass to generate biogas for energy purposes has been steadily increasing worldwide.Biogas is generated through anaerobic digestion (AD), which relies on a consortium of bacteria (hydrolytic, acidogenic, acetogenic, and methanogenic) that decompose the organic matter of biomasses under anaerobic conditions to generate biogas.Biogas is a gaseous mixture composed mainly of methane, carbon dioxide, nitrogen, and traces of hydrogen sulfide, ammonia, and oxygen, among other components [5].In addition, a by-product of the process called digestate that can be used as a biofertilizer on agricultural crops, given its nitrogen, phosphorus, and magnesium content, is generated [6].China leads among the IEA Bioenergy Task 37 member countries, with over 100,000 biogas plants generating 72,000 TWh per year, followed by Germany, with more than 10,000 plants that generate 120 TWh per year.The primary uses of these biogas plants are for heat and electricity generation [7].In the South American context, Brazil leads in the implementation of biogas technologies, with 755 plants in operation and generating 12TWh per year, corresponding to a biogas generation of 2.82 billion Nm 3 year −1 [8].
In Colombia, an increasing number of studies are being conducted to produce biogas from biomass sources.A study conducted by the Colombian National University (UNAL) in 2018, established that the theoretical potential for biogas production from biomass was 149,436 TJ year −1 and the estimated technical potential was 53,554 TJ year −1 , which is equivalent to approximately 25% of the demand for natural gas in 2016 in Colombia [9].
The pig sector is important in the Colombian livestock development.According to reports from the Colombian Association of Pig Producers (Porkcolombia), a growth rate of 7.1% was seen during the last decade, with production reaching 5.2 million pigs per year.The department of Antioquia represents 43% of the national pig production, becoming one of the most important suppliers of this sector [10].According to research conducted by Pessuto et al. [11], a pig produces approximately 3.13 kg of manure per day.This is equivalent to 2.7 million tons of excrement annually, based on the country's current production.The considerable amount of waste produced has become an environmental problem due to the challenges associated with its disposal.This includes the production of greenhouse gases (GHGs), such as carbon dioxide (CO 2 ) and methane (CH 4 ), from the degradation of organic matter in the open air [12].On the other hand, the unregulated irrigation of feces on soils, as a form of biofertilization, due to its nitrogen, phosphorus, and potassium content [13], can cause an over-accumulation of nutrients in soils, resulting in the contamination of groundwater [14].
In Colombia, biogas production using swine manure has been studied mostly in systems with high HRTs and high solid load capacities, such as Taiwan-type biodigesters or stirred tanks and the choice between them depends on the farm's size [15,16].On the other hand, there are other anaerobic reactors with more efficient designs, such as the UASB.This reactor offers several advantages: simple design, small space requirement, low operating costs, minimal sludge production, and a high COD removal efficiency suited to short hydraulic retention times and low levels of total solids [17].They should be tested for this kind of process in local conditions to assess their applicability in the pork industry and to improve their biogas productivity.
Lee et al. [18] conducted a study to determine the theoretical production of biogas from the elemental composition of pig feed in Korean specialized farms; they reported 0.39 Sm 3 CH 4 kg VS −1 and 30.96Sm 3 CH 4 ton −1 of swine manure (sm 3 refers to the volume of gases at standard conditions, at a pressure and temperature of 1 atm and 0 • C, respectively).In 2021, Porkcolombia also carried out a study to establish the potential for biogas generation from swine manure; they evaluated four specialized farms located in different regions of the country and established, through theoretical models based on the elemental composition of the feed, that the maximum methane production or theoretical methane biochemical potential (TMBP) in the anaerobic fermentation process of Colombian pigs ranges between 0.58 and 0.45 m 3 CH 4 kg SV −1 [16].Additionally, they found, through experimental methods under laboratory conditions, that methane production was in a range between 0.16 m 3 CH 4 kg SV −1 and 0.38 m 3 CH 4 kg SV −1 , values that were close to those reported by the Korean study.A Brazilian study reported that swine manure from a nursery house presented the highest biogas and methane yield capacity, at 0.97 Nm 3 kg VS −1 , followed by the farrowing sows at 0.86 Nm 3 kg VS −1 , finishing pigs at 0.47 Nm 3 kg VS −1 , and gestation sows at 0.32 Nm 3 kg VS −1 [19].Swine manure can exhibit a wide range of organic matter content due to variations in feed composition, which are influenced by factors such as pig age and the feeding practices employed throughout the pig farming cycle [20,21].
In this study, an Upflow Anaerobic Sludge Blanket (UASB) reactor was monitored for 360 days using swine manure as a substrate to evaluate the process of biogas production.This study did not maintain a fixed organic loading rate (OLR), but instead aimed to simulate the fluctuating conditions that occur in a pig farm.The feed flow was increased three times, seeking to reduce the HRTs, and various parameters were monitored for subsequent evaluation.The long-term operation was conducted to observe the impact of parameter changes on biogas production, with the aim of applying this knowledge to scale up the project to a pilot plant in the future.

Collecting and Handling Substrate
The pig feces used in this study were kindly donated by a pig farm located at coordinates 6 • 45 ′ 37 ′′ N, 75 • 30 ′ 10 ′′ W, owned by Agrofranpabel SAS, Colombia.Approximately 40 kg of fresh feces were collected from gestation crates every two weeks, after the first feeding of the day (around 9:00 a.m.).The samples were transported in sealed containers for three hours to the University of Antioquia facilities and stored at 4 • C. A portion of feces was used every five days to prepare the feeding solution, following the proportions described later in Section 2.3.

Upflow Anaerobic Sludge Blanket (UASB) Reactor for Biogas Production
The reactor type was selected for its capacity to support COD levels in the range of 30-80 gO 2 L −1 and OLR up to 15 gCOD L −1 day −1 [22], which is crucial for this study given that pig manure typically exhibits COD values within this range.Furthermore, biogas production of 8.7 L day −1 has been reported, accompanied by high organic matter remotion due to the granular sludge formed [17].A UASB reactor requires less space per cubic meter of waste treated and has higher biogas productivity compared to the anaerobic digestion systems commonly used in swine farming.This technology is interesting for our study as it serves as an alternative for the generation of energy and waste treatment from an environmental perspective, particularly in future works.
The reactor was manufactured in the Bioprocess laboratory located at Universidad de Antioquia, using glass fiber and polyvinyl chloride (PVC) pipe accessories.The reactor had a total height of 174 cm, a wall thickness of 0.4 cm, and a volumetric capacity of 40 L. The reactor consisted of 2 sections: 1 reaction column and 1 liquid gas separator, with 26 L and 14 L capacity, respectively.The reaction column featured a cylindrical geometry with an internal diameter of 0.17 m and a height of 1.3 m, equipped with 3 sampling ports.The liquid gas separation section consisted of an internal bell with a height of 0.34 m and an internal diameter of 0.2 m (Figure 1).All inlets and outputs of the UASB reactor have an internal diameter of 0.02 m and are coupled to a PVC ball valve.

Performance Evaluation of UASB Reactor for Biogas Production Using Swine Manure
The swine manure solution used as a feeding substrate was prepared by mixing part of fresh swine manure with 4 parts of tap water based on preliminary results.Th mixture was prepared every 5 days using a propeller agitator (IKA RW20, Staufen, Ge many) equipped with a Rushton-type impeller.It was stirred for 10 min, left for 20 m for solids sedimentation, and only the aqueous phase was moved to the feed tank to pr vent hose obstruction by solids present in the swine manure.To evaluate the biogas pr duction process, anaerobic sludge, kindly donated by Planta de Tratamiento de Agu Residuales San Fernando, a wastewater treatment plant in Medellin, was used.For th inoculation process, 20 L of clean water, 9.6 L of sludge, and 2.4 L of swine manure we used.The mixture was then pumped using a peristaltic pump (Stenner PUMPs 45MHP1 Jacksonville, FL, USA) through a neoprene hose of 77 mm gauge with high chemical r sistance.After this, the reactor remained static for two weeks to allow for an adaptatio period to the substrate.
To increase biogas yield, a strategy of gradually increasing the feed flow rate was im plemented each time a stable trend was reached in the reactor (quasi steady state).As a r sult, the flow rate was increased three times (1.3 L d −1 , 1.6 L d −1 , and 2.1 L d −1 ), obtaining a HRT of 31, 25, and 19 days, respectively.Figure 2 depicts the process diagram.The avera local atmospheric pressure was 100.7 kPa and the ambient temperature was 24 °C.
Throughout 360 days of monitoring, samples were collected weekly from afflue and effluent, and biogas.During the monitoring process, each stream underwent chara terization including the measurement of various parameters, detailed in Sections 2.3 an

Performance Evaluation of UASB Reactor for Biogas Production Using Swine Manure
The swine manure solution used as a feeding substrate was prepared by mixing 1 part of fresh swine manure with 4 parts of tap water based on preliminary results.The mixture was prepared every 5 days using a propeller agitator (IKA RW20, Staufen, Germany) equipped with a Rushton-type impeller.It was stirred for 10 min, left for 20 min for solids sedimentation, and only the aqueous phase was moved to the feed tank to prevent hose obstruction by solids present in the swine manure.To evaluate the biogas production process, anaerobic sludge, kindly donated by Planta de Tratamiento de Aguas Residuales San Fernando, a wastewater treatment plant in Medellin, was used.For the inoculation process, 20 L of clean water, 9.6 L of sludge, and 2.4 L of swine manure were used.The mixture was then pumped using a peristaltic pump (Stenner PUMPs 45MHP10, Jacksonville, FL, USA) through a neoprene hose of 77 mm gauge with high chemical resistance.After this, the reactor remained static for two weeks to allow for an adaptation period to the substrate.
To increase biogas yield, a strategy of gradually increasing the feed flow rate was implemented each time a stable trend was reached in the reactor (quasi steady state).As a result, the flow rate was increased three times Throughout 360 days of monitoring, samples were collected weekly from affluent and effluent, and biogas.During the monitoring process, each stream underwent characterization including the measurement of various parameters, detailed in Sections 2.3 and 2.4.During experimental development, the organic load rate (OLR) was not controlled, as the aim was to simulate the daily activity of a pig farm where parameters are constantly changing.

Determination of Total Solids
Total solids (TS) were measured according to method 2540B from Standard M for Examination [23].Briefly, samples were collected in triplicate, with each sam ing a volume of 15 mL.Briefly, 15 mL of swine manure was dried in porcelain c for 24 h at 105 °C in a convection oven (Binder FD-115, Tuttlingen, Germany).Fo this, the samples were placed in a desiccator until they reached room temperatu calculation for quantifying the total solids in the sample is described below in E ( 1).An analytical balance (Shimadzu ATX-224, Kyoto, Japan) was used for weigh samples.
g total solids g sample

C
Where A is the weight of the dried sample in the capsule, B is the weight of th capsule and C is the weight of the sample.

Determination of Volatile Solids
Volatile solids (VS) were measured according to method 2540E from the S Methods for Examination [23].Samples were analyzed in triplicate and initially remove moisture content.After the drying process, the samples were heated to 55 2 h using a muffle furnace (Thermo Scientific FB1410M, Waltham, MA, USA).Fo this, the samples were placed in a desiccator until they reached room temperatu calculation for quantifying the volatile solids in the sample is described below in E (2) and is based on the change in weight, measured with an analytical balance (Sh ATX-224, Kyoto, Japan).
where D is the weight of capsule, A residue after leaving the muffle and C is the  Total solids (TS) were measured according to method 2540B from Standard Methods for Examination [23].Briefly, samples were collected in triplicate, with each sample having a volume of 15 mL.Briefly, 15 mL of swine manure was dried in porcelain capsules for 24 h at 105 • C in a convection oven (Binder FD-115, Tuttlingen, Germany).Following this, the samples were placed in a desiccator until they reached room temperature.The calculation for quantifying the total solids in the sample is described below in Equation ( 1).An analytical balance (Shimadzu ATX-224, Kyoto, Japan) was used for weighing the samples.
where A is the weight of the dried sample in the capsule, B is the weight of the empty capsule and C is the weight of the sample.

Determination of Volatile Solids
Volatile solids (VS) were measured according to method 2540E from the Standard Methods for Examination [23].Samples were analyzed in triplicate and initially dried to remove moisture content.After the drying process, the samples were heated to 550 • C for 2 h using a muffle furnace (Thermo Scientific FB1410M, Waltham, MA, USA).Following this, the samples were placed in a desiccator until they reached room temperature.The calculation for quantifying the volatile solids in the sample is described below in Equation ( 2) and is based on the change in weight, measured with an analytical balance (Shimadzu ATX-224, Kyoto, Japan).
where D is the weight of capsule, A residue after leaving the muffle and C is the weight of the sample.Volatile fatty acids (VFAs) and alkalinity (ALK) were measured using the Kapp Method (1984), developed for mesophilic anaerobic digesters.To implement this method, the effluent samples were collected in triplicate, with each sample having a volume of 40 mL, then these were centrifuged at 5000 rpm for 10 min using a centrifuge (Sigma 2-16PK, Osterode am Harz, Germany) to avoid any solid particles and then, the liquid supernatant was titrated using sulfuric acid 0.1 N (ITW reagents, Milan, Italy), at three final pH points: 5.0, 4.3, and 4.0.Titration was conducted using an automatic titrator (SI Analytics TL-7000, Mainz, Germany), ensuring precise and accurate measurements of the VFA and ALK levels in the samples.

Quantification of Chemical Oxygen Demand (COD)
Chemical oxygen demand (COD) was measured using the 8000 method from Hach instruments, employing COD TNTPLUS Vials Test by Hach (Product Number: TNT82206).For this process, 200 µL of homogenized sample was introduced into a vial and placed in a thermoreactor (Hach DRB-200, Ames, IA, USA) at 150 • C for 2 h for the reaction to take place.After completing the reaction time, the samples were allowed to cool.Then, their concentration of COD was measured using a colorimeter (HACH DR-900, Ames, IA, USA).

Calculation of Organic Load Remotion Using COD as a Parameter
The calculation of organic load remotion can be performed using a variety of reference parameters, including COD and VS.It indicates the efficiency of a treatment process in removing organic pollutants from the influent.In this study, COD is employed as the reference parameter, and the remotion is calculated in accordance with Equation (3).
where COD in and COD out are the values of the chemical oxygen demand (gO 2 L −1 ) corresponding to the feed and effluent streams of the reactor.

. Measurement of Biogas Production
A High-Density Polyethylene (HDPE) bag with a capacity of 150 L and a thickness of 0.5 mm was used to collect the biogas produced by the UASB reactor.Biogas production was measured using a conventional gas meter of the diaphragm type (Koreabi Shin Han Precision G-1.6, Seoul, South Korea).Biogas production in each period was calculated using Equations ( 4) and (5). V where V biogas is the volume of biogas measured in liters, V f and V i are the final and initial reading of the gas meter in liters, respectively, and time is the period in which the biogas sample was collected in days.

Biogas Composition Analysis
The composition of biogas was determined using a biogas analyzer (MRU AIR Op-tima7, Rostock, Germany) attached to the biogas HDPE bag.The analyzer quantified the content of methane, carbon dioxide, nitrogen, and oxygen in volumetric percentage (%V/v).Hydrogen sulfide content was quantified in volumetric parts per million (ppm).

Swine Manure Characterization
In Table 1, the characterization of the swine manure solution used for this study is presented.Despite the swine manure collected always being from breeding females, the standard deviation values were high, likely due to uncontrolled factors such as changes in feeding, weather conditions, and the natural degradation of the organic matter present in the swine manure due to environmental conditions.The swine manure solution presented a COD value of 39.154 ± 13.992 g O 2 L −1 , a TS value of 3.35 ± 1.33%, and a VS value of 2.39 ± 0.98% (Table 1).The high variation in these parameters (35-40%) indicates that, although the ratio of pig feces/water used in the feed was constant, this parameter varied significantly due to the productive conditions of the pig farm.High temperature is one of the most important climatic factors influencing pig performance [24]; in a meta-analysis, Renaudeau et al. [25] showed that high temperatures have an undesirable effect on the voluntary feed intake and body weight gain, with this effect being more pronounced in heavier pigs.Collin et al. [26] found a significant decrease in the voluntary feed intake (45 g d −1 ) and body weight gain (37 g d −1 ) in group-housed young pigs between temperatures of 23 and 33 • C. Climatic variations resulting from current weather conditions cause irregular feeding patterns, which are reflected in the composition of swine manure.In addition, the growth of fungi and other microorganisms, which reduce the available organic matter prior to entering the reactor, is often caused by the sub-optimal storage conditions of agro-industrial waste.However, this reflects the genuine state of a pig farm and any system seeking to harness the benefits of pig waste must exhibit tolerance for such substrate variability.This is why we designed the experiment to simulate laboratory conditions closely resembling the actual operational environment of the proposed system by intentionally avoiding initial organic load control.Despite this variation, the VS/TS ratio tends to remain stable, with a variation of less than 4.7% with an average value of 0.72.The variability in the composition is permanent and difficult to control, since it is also strongly affected by humans, who carry out cleaning tasks and perform the final disposal of the pig manure.These situations never interrupted the biogas generation process, providing evidence of the adaptation capacity of microorganisms within the UASB reactor.
Wu et al. [27] reported that the pig manure they used for biogas production had a TS value of 23.1%, a VS value of 18.0%, and a COD value 243 gO 2 L −1 .These values were higher than those reported in this study because it was not filtrated or diluted with water; however, the VS/TS ratio of 0.78 is very close to ours, showing that despite the differences in the production processes of the pig farming cycle and the proportion of water in the swine manure, the VS/TS ratio is only slightly affected by these changes.Lansing et al. [28] reported that swine manure diluted with wash water in a 4:1 ratio had a pH of 7.5, a COD of 20 gO 2 L −1 and a VS value of 1.37%.This characterization shows that pig manure has a pH that tends to be alkaline and is consistent with the values reported in this study.On the other hand, due to its dilution with water, the values of the VS and COD were lower compared to those reported by Wu et al. [27] and closer to those reported in this study.

Reactor UASB Performance for Biogas Production
The stability trend was identified twice over the course of a year of monitoring and evaluation.The proposed feeding system allowed the established flow values to be met; however, it is not the most suitable for this substrate.A monthly cleaning and maintenance routine was necessary for the neoprene hoses to prevent clogging and breakage.In addition, undigested grain was detected, which did not settle in the feed tank and entered the hoses.To address this issue, starting from the second month of operation, the aqueous phase of the substrate was filtered through a stainless steel strainer with a pore size of 1 mm 2 .The presence of numerous unusual elements, such as stones, wires, packing fibers, and even nuts, which are often found in the substrate, can eventually cause interruption in the pumping of the substrate.These elements are a consequence of cleaning procedures used in pig fattening pens.This factor must be considered when designing pumping systems for mixtures of pig manure or other animal feces.We suggest using a slurry pump that is non-clogging and can tolerate particles bigger than 7 mm.
For the swine manure solution feed flows of 1.3 L d −1 (Q1), 1.6 L d −1 (Q2), and 2.1 L d −1 (Q3), the average biogas productions were 12.1 L d −1 , 19.2 L d −1 , and 37.9 L d −1 , respectively (Figure 3).Likewise, the methane content has an average volume percentage of 67.69% with a standard deviation of 2.81%.This suggests that despite fluctuating flow and increasing the biogas production, the content remained stable within a reasonable range (standard deviation of 4.10%).
The stability trend was identified twice over the course of a year of monitoring and evaluation.The proposed feeding system allowed the established flow values to be met; however, it is not the most suitable for this substrate.A monthly cleaning and maintenance routine was necessary for the neoprene hoses to prevent clogging and breakage.In addition, undigested grain was detected, which did not settle in the feed tank and entered the hoses.To address this issue, starting from the second month of operation, the aqueous phase of the substrate was filtered through a stainless steel strainer with a pore size of 1 mm 2 .The presence of numerous unusual elements, such as stones, wires, packing fibers, and even nuts, which are often found in the substrate, can eventually cause interruption in the pumping of the substrate.These elements are a consequence of cleaning procedures used in pig fattening pens.This factor must be considered when designing pumping systems for mixtures of pig manure or other animal feces.We suggest using a slurry pump that is non-clogging and can tolerate particles bigger than 7 mm.
For the swine manure solution feed flows of 1.3 L d −1 (Q1), 1.6 L d −1 (Q2), and 2.1 L d −1 (Q3), the average biogas productions were 12.1 L d −1 , 19.2 L d −1 , and 37.9 L d −1 , respectively (Figure 3).Likewise, the methane content has an average volume percentage of 67.69% with a standard deviation of 2.81%.This suggests that despite fluctuating flow and increasing the biogas production, the content remained stable within a reasonable range (standard deviation of 4.10%).
On the other hand, the rest of the components in the biogas were CO2, O2, N2, and H2S at 18.06%, 8.99%, 3.91%, and 1413 ppm, respectively, and according to the results obtained, the mean methane production yield was 507 mL CH4 gVS −1 when the reactor operated with an HRT of 19 days and an OLR of 2.68 gVS L −1 d −1 .The H2S concentration in raw biogas ranges from 50 to 10,000 ppm, depending upon the feedstocks [29].H2S is formed during the microbiological reduction of compounds that contain sulfur, and it can lead to corrosion problems, damage to biogas equipment, generators, combustion engines, and other metal components [30,31].A concentration higher On the other hand, the rest of the components in the biogas were CO 2 , O 2 , N 2 , and H 2 S at 18.06%, 8.99%, 3.91%, and 1413 ppm, respectively, and according to the results obtained, the mean methane production yield was 507 mL CH 4 g VS −1 when the reactor operated with an HRT of 19 days and an OLR of 2.68 g VS L −1 d −1 .
The H 2 S concentration in raw biogas ranges from 50 to 10,000 ppm, depending upon the feedstocks [29].H 2 S is formed during the microbiological reduction of compounds that contain sulfur, and it can lead to corrosion problems, damage to biogas equipment, generators, combustion engines, and other metal components [30,31].A concentration higher than 500 ppm of H 2 S can cause unconsciousness and death in people; H 2 S content Energies 2024, 17, 2723 9 of 14 can be almost fully removed by several methods [32].Wang et al. [29] evaluated the addition of biochar (BC), steam-treated wood, and untreated poplar wood chips as additives to reduce the H 2 S content in dairy manure; they found that adding 3 g of biochar per 500 g of manure during AD in a plug flow reactor can remove up to 90% of the H 2 S during the first week of treatment, without impacting CH 4 production, thus making it a viable option for various end-use technologies.
On the other hand, biological methods can also be employed to reduce the H 2 S content in biogas.Kumdhitiahutsawakul et al. [32] investigated the efficiency of vertical and horizontal pilot-scale biofilters for H 2 S remotion in swine waste biogas, utilizing bacterial community analysis in both laboratory and pilot-scale biofiltration systems.This study revealed that a porous glass (PG) biofilter immobilized with Paracoccus versutus bacteria eliminated H 2 S when inlet concentrations ranged from 2005 to 3644 ppm H 2 S. The system operated consistently with a remotion efficiency of 99-100%, showing no fluctuations throughout the experiments.On the other hand, Agudelo et al. [33] evaluated H 2 S reduction using three materials: steel wool, iron rods, and compost (biofilter) in biogas produced from Palm Oil Mill Effluent (POME).The results showed that steel wool had a high potential for H 2 S remotion compared to the other materials, achieving a 99.7% remotion of H 2 S in a biogas with 1327 ppm H 2 S.

COD, VS, Organic Load Rate (OLR), and Hydraulic Retention Time (HRT) Results
The feed stream showed significant variations in the COD values, ranging from 15.95 g O 2 L -1 to 67.30 g O 2 L -1 (Figure 4).This variability is attributed to the fact that the COD is a parameter dependent on the organic load of the swine manure, and as previously explained, the composition of this substrate is influenced by factors such as pig metabolism and pig feed composition and its natural degradation.tent in biogas.Kumdhitiahutsawakul et al. [32] investigated the efficiency of vertical and horizontal pilot-scale biofilters for H2S remotion in swine waste biogas, utilizing bacterial community analysis in both laboratory and pilot-scale biofiltration systems.This study revealed that a porous glass (PG) biofilter immobilized with Paracoccus versutus bacteria eliminated H2S when inlet concentrations ranged from 2005 to 3644 ppm H2S.The system operated consistently with a remotion efficiency of 99-100%, showing no fluctuations throughout the experiments.On the other hand, Agudelo et al. [33] evaluated H2S reduction using three materials: steel wool, iron rods, and compost (biofilter) in biogas produced from Palm Oil Mill Effluent (POME).The results showed that steel wool had a high potential for H2S remotion compared to the other materials, achieving a 99.7% remotion of H2S in a biogas with 1327 ppm H2S.

COD, VS, Organic Load Rate (OLR), and Hydraulic Retention Time (HRT) Results
The feed stream showed significant variations in the COD values, ranging from 15.95 g O2 L -1 to 67.30 g O2 L -1 (Figure 4).This variability is attributed to the fact that the COD is a parameter dependent on the organic load of the swine manure, and as previously explained, the composition of this substrate is influenced by factors such as pig metabolism and pig feed composition and its natural degradation.
Despite high variations in the feed stream, the biomass of the reactor responded well and showed stability, which is reflected in the fact that the COD of the effluent has a more constant behavior with an average value of 4.86 ± 1.66 g O2 L −1 and a remotion percentage of about 87.38%.Figure 4 shows two notable decreases in the percentage of COD remotion (days 128 and 320), which occurred immediately after the increase in the feed flow that was made in the system.These decreases are due to the adaptation time of the microorganisms to process more organic matter, which saturates their processing capacity and is reflected in the decrease of this parameter in the process.However, the remotion rate increases after a few days, indicating that the system can stabilize with increases in the OLR.Despite high variations in the feed stream, the biomass of the reactor responded well and showed stability, which is reflected in the fact that the COD of the effluent has a more constant behavior with an average value of 4.86 ± 1.66 g O 2 L −1 and a remotion percentage of about 87.38%.Figure 4 shows two notable decreases in the percentage of COD remotion (days 128 and 320), which occurred immediately after the increase in the feed flow that was made in the system.These decreases are due to the adaptation time of the microorganisms to process more organic matter, which saturates their processing capacity and is reflected Energies 2024, 17, 2723 10 of 14 in the decrease of this parameter in the process.However, the remotion rate increases after a few days, indicating that the system can stabilize with increases in the OLR.
Considering these values (Figure 4) and the biogas production presented in Section 3.2.(Figure 3), it is evident that an increase in the feed flow leads to a rise in biogas production; this is due to the direct correlation between the OLR and the flow rate.On the other hand, the HRT of the process decreases as the feed flow rate increases while the working volume remains constant.The HRT values for Q1, Q2, and Q3 were 31 days, 25 days, and 19 days, respectively, which enabled the system to digest a greater amount of organic matter in less time, enhancing its efficiency, reducing the HRT by 38%, and maintaining high remotion efficiencies in the VS and COD, with values of 85.12% and 87.38%, respectively.These values exceed those reported by Duan et al. [34], who achieved remotion efficiencies in the VS and COD of 66.22% and 76.18%, respectively, while operating a CSTR with an OLR of 1.89 g VS L −1 d −1 .Table 2 shows a summary of the different measured parameters and calculations for the three feed flows used.The values for the COD reduction ranged from 61% to 90% (Figure 4).Comparing the higher or lower OLR, the stage with the highest OLR exhibited the higher average biogas production and an organic matter reduction of around 89%.Although a heightened OLR denotes an increased treatment capacity and methane yield, it may also precipitate overloading, thereby engendering process instability and even system failure [34], emphasizing the need for exercising caution with OLR increases.Table 3 presents additional findings on methane production from various authors, as reported in the literature.The study explored how varying OLR impacted the efficiency of AD of pig manure on a CSTR system.Duan et al. [34] In this paper, the methane productivity of manure in terms of volatile solids (VS), volume, and livestock production were determined.Moller et al. [35] This study employs a life cycle assessment and cost-benefit analysis to assess and contrast the environmental impacts and cost benefits associated with the implementation of a wastewater treatment system on a swine farm.The results presented in Table 3 show that the production of biogas from swine manure in a continuous reactor-type UASB is superior to those reported by other authors and shows great potential for its implementation in pig farms for the energy use of this type of waste.It is important to note that the results are comparable despite the operation temperatures being different, as this can be justified by the fact that the production of biogas in anaerobic digestion is enhanced when the system operates at higher temperatures in the mesophilic range (24-40 • C) [37].Therefore, in the context of the experiment, it could be stated that there is no advantage in operating at a lower temperature over the mentioned studies.
3.4.Alkalinity (ALK), Volatile Fatty Acids (VFAs), and ALK/VFA Ratio Monitoring alkalinity (ALK) and volatile fatty acids (VFAs) in the process effluent stream is important because the stability of an anaerobic digestion process can be qualified by the VFA/ALK ratio.A study reported that a favorable VFA/ALK ratio to prevent acidification of the system is in the range of 0.1-0.25 and showed that if it is in the range of 0.3-0.4,there may be disturbances in the reactor, requiring corrective measures [38].On the other hand, it shows that the optimum pH range to avoid inhibition processes of methanogenic bacteria is 6.8-8.0.Finally, Kim et al. [39] researched the condition of lag-phase reduction during AD and they found that to reduce the lag phase to achieve a good performance, the VFA/ALK ratio should be less than 0.4.

Alkalinity (ALK), Volatile Fatty Acids (VFAs), and ALK/VFA Ratio
Monitoring alkalinity (ALK) and volatile fatty acids (VFAs) in the process effluent stream is important because the stability of an anaerobic digestion process can be qualified by the VFA/ALK ratio.A study reported that a favorable VFA/ALK ratio to prevent acidification of the system is in the range of 0.1-0.25 and showed that if it is in the range of 0.3-0.4,there may be disturbances in the reactor, requiring corrective measures [38].On the other hand, it shows that the optimum pH range to avoid inhibition processes of methanogenic bacteria is 6.8-8.0.Finally, Kim et al. [39] researched the condition of lag-phase reduction during AD and they found that to reduce the lag phase to achieve a good performance, the VFA/ALK ratio should be less than 0.4.
From Figure 5, it is noted that at the beginning of the operation (until day 29), the value of the VFA/ALK ratio is approximately 0.49 and the pH is 6.68.This deviation from the values observed during the rest of the monitoring period is attributed to the inability of microorganisms to digest all the organic matter, resulting in an accumulation of VFAs and temporary acidification during the start-up and adaptation of the process.However, in the following days, it exhibited remarkable stability, with the ratio consistently staying below the maximum limit and the pH within the recommended range.From Figure 5, it is noted that at the beginning of the operation (until day 29), the value of the VFA/ALK ratio is approximately 0.49 and the pH is 6.68.This deviation from the values observed during the rest of the monitoring period is attributed to the inability of microorganisms to digest all the organic matter, resulting in an accumulation of VFAs and temporary acidification during the start-up and adaptation of the process.However, in the following days, it exhibited remarkable stability, with the ratio consistently staying below the maximum limit and the pH within the recommended range.

Conclusions
To meet Colombia's 2050 energy goals of 42 MW, transitioning to cleaner energy sources, specifically biomass, is imperative.This study, which researched biogas production from swine manure using a UASB reactor with long-term operation (360 days), demonstrated the viability of converting pig waste into a valuable energy resource.The UASB reactor effectively achieved remotion efficiency with VS and COD values of 85.12% and 87.38%, respectively, and a methane production yield of 507 mLCH 4 g VS −1 with an HRT of 25 days and an OLR of 4.27 gCOD L −1 d −1 .The produced biogas has a composition of 67.69% CH 4 and 18.06% CO 2 .However, to ensure good stability in the process, pumping systems with tolerance to suspended solids are required.These findings underscore the potential of swine manure as a sustainable substrate for biogas production, with the UASB reactor proving to be efficient technology.
The utilization of biogas technology in the Colombian agricultural sector, particularly that which harnesses swine manure as an energy source, has the potential to significantly enhance the profitability of agribusinesses.This sustainable approach not only addresses environmental concerns but also offers a promising avenue for diversifying income streams and reducing waste-related costs within the agricultural landscape.Future research endeavors should focus on further enhancing biogas composition for electricity generation.

Figure 1 .
Figure 1.Diagram of the UASB reactor used in this study.

Figure 1 .
Figure 1.Diagram of the UASB reactor used in this study.
(1.3 L d −1 , 1.6 L d −1 , and 2.1 L d −1 ), obtaining an HRT of 31, 25, and 19 days, respectively.Figure 2 depicts the process diagram.The average local atmospheric pressure was 100.7 kPa and the ambient temperature was 24 • C.

Figure 2 .
Figure 2. Process diagram for biogas production using anaerobic digestion on UASB re with swine manure solution.

Figure 2 .
Figure 2. Process diagram for biogas production using anaerobic digestion on UASB reactor fed with swine manure solution.

2. 5 .
Determination of Performance of Reactor UASB fed with Swine Manure 2.5.1

Figure 4 .
Figure 4. COD in influent and effluent streams and its remotion in the system.The dashed lines represent the points where the system's feed flow was increased.

Figure 4 .
Figure 4. COD in influent and effluent streams and its remotion in the system.The dashed lines represent the points where the system's feed flow was increased.

Figure 5 .
Figure 5. Monitoring of the VFA/ALK ratio and pH during the study period.The dashed red line represents the maximum value of the VFA/ALK ratio necessary for good performance without the need for corrective action, according to the literature.

Figure 5 .
Figure 5. Monitoring of the VFA/ALK ratio and pH during the study period.The dashed red line represents the maximum value of the VFA/ALK ratio necessary for good performance without the need for corrective action, according to the literature.

Table 1 .
Characterization of aqueous phase of swine manure-water mixture used for this study.COD: chemical oxygen demand; TS: total solids; VS: volatile solids; ALK: alkalinity.

Table 2 .
Average of VS, COD, OLR, and HRT for each flow rate.

Table 3 .
Comparison of methane yield (mL CH 4 g VS −1 ) obtained from swine manure in standard conditions.