Evaluation of Biogas Production from the Co-Digestion of Municipal Food Waste and Wastewater Sludge at Refugee Camps Using an Automated Methane Potential Test System

: The potential beneﬁts of the application of a circular economy—converting biomass at Za’atari Syrian refugee camps into energy—was investigated in this study. Representative organic waste and sludge samples were collected from the camp, mixed in different ratios, and analyzed in triplicate for potential biogas yield. Numerous caloriﬁc tests were also carried out. The tangential beneﬁt of the co-digestion that was noticed was that it lowered the value of the total solid content in the mixture to the recommended values for wet digestion without the need for freshwater. To test the potential methane production, the automated methane potential test system (AMPTS) and the graduated tubes in the temperature-controlled climate room GB21 were utilized. Also, caloriﬁc values were determined for the organic waste and sludge on both a dry and a wet basis. The maximum biogas production from 100% organic waste and 100% sludge using AMPTS was 153 m 3 ton − 1 and 5.6 m 3 ton − 1 , respectively. Methane yield reached its maximum at a Vs sub/Vs inoculum range of 0.25–0.3. In contrast, the methane yield decreased when the Vs sub/Vs inoculum exceeded 0.46. The optimum ratio of mixing of municipal food waste to sludge must be carefully selected to satisfy the demands of an energy production pilot plant and avoid the environmental issues associated with the sludge amount at wastewater treatment plants (WWTPs). A possible ratio to start with is 60–80% organic waste, which can produce 21–65 m 3 · biogas ton − 1 fresh matter (FM). The co-digestion of organic waste and sludge can generate 38 Nm 3 /day of methane, which, in theory, can generate about 4 MW in remote refugee camps.


Introduction
Jordan is a semi-arid country with scarce fresh water resources and relatively limited agricultural activities [1,2]. Currently, Jordan is ranked second in the world in water scarcity [3][4][5]. Therefore, the biomass availability in arid and semi-arid regions can be considered to be relatively low compared with non-arid regions [6]. Biomass residues available in Jordan may be derived from localized agriculture-related activities, the organic fraction of municipal solid waste (OFMSW), animal manure,

Study Area
For the purpose of this research, the Za'atari Syrian Refugee Camp (ZSRC) was selected as the study area, which is located approximately 8 miles from the Syrian border (32 • 17 44.4"N 36 • 19 25.5"E). ZSRC is the largest refugee camp in the Middle East, and the second-largest refugee camp in the world [32]. The ZSRC has on average 80,000 residents, and regularly generates 28-50 tons of unsegregated municipal waste per day (40 tons/day on average). The ZSRC's waste management is fully described in the work undertaken by Saidan et al in 2017. However, it is now considered to be the fourth-largest city in Jordan and is gradually evolving into a permanent settlement [33]. Running the ZSRC tends to be expensive due to the high costs of waste treatment and energy supply [34]. In this context, the introduction of a biogas production and energy-producing plant would help to minimize Energies 2019, 12, 32 3 of 11 the cost of running the refugee camp. As a result of that, progress towards sustainability and better living conditions in the camp can be ensured [35].

Waste Sampling and Preparation in the ZSRC
Due to restrictions on, and control of, access to the ZSRC, the ZSRC's managing personnel provided representative MSW samples over a period of one week together with sludge from the Za'atari wastewater treatment plant (WWTP). However, in order to achieve representative samples, guidelines were provided to the ZSRC's technical personnel. The weight of the sample was 15 kg, and it was delivered as a composite sample.
The MSW sample was harvested for its organic fraction, then the organic fraction was milled down to achieve a particle size of less than 10 mm. This size distribution ensures the homogeneity and repeatability of the test.

Preliminary Testing of Waste Samples
All experiments were conducted in the biogas lab located at the German Jordanian University. The room temperature and humidity in the lab were controlled using an air conditioning system with a set value of 24 • C and 60% relative humidity. Within the lab, a climate-insulated room is installed where the temperature is kept at 36 • C.
As a preliminary and an important step in setting up the experiment, detailed information on the volatile solids (VS) and the total solids (TS) in all the experimental content was estimated. This importance arises from the impact of the ratio between the VS of the inoculum and the VS of the substrate on the bio-methane production rates. All experiments were carried out at least in triplicate for each tested parameter, and the corresponding statistical variation is shown.
TS and VS were calculated for the OFMSW, the sludge from the Za'atari WWTP, and the inoculum used in the experiment. The inoculum used in the experiment was received from an anaerobic digester located at the Shallalah WWTP; its TS and VS values were (1.08 ± 0.02)% and (72.22 ± 11.45)%, respectively. The measurement of the VS and TS for all constituents in the experiments was carried out in accordance with DIN 38 414-S2 [36]. For the determination of TS, three crucibles were heated at 105 • C (the tests were done in triplicate) for an hour, then they were cooled down in a desiccator to room temperature. After that, the crucibles were filled with a sample and weighed. Subsequently, it was placed in the oven overnight at 105 • C, then left to cool down to the ambient temperature. Eventually, the sample's corresponding weight was taken, and the TS in the crucible were calculated. Similarly, after that, VS measurement was conducted using the dried sample from the TS measurement. The dried sample was ground to a particle size of less than 0.25 mm according to DIN 38 409 [37]. Then, the sample was weighed and placed in crucibles, and the sample was heated at 200 • C, 300 • C, and 500 • C for 20 min, 30 min, and 5 h, respectively.
After acquiring the TS and VS values for all constituents, the ratio of volatile solids in the substrate to the volatile solid in the inoculum was determined. Then, weight calculations for the different percentages of mixing between the OFMSW and the WWTP sludge were performed. Setting up the reactors for both the GB21 experiment and the automated methane potential test system (AMPTS) is performed in a similar manner; however, the tests operate differently. While the AMPTS operates automatically for both mixing and measurements, the climate room GB21 test requires the readings to be taken, and the mixing to be performed, manually. The experiments were conducted in controlled conditions regarding temperature and environment. The AMPTS has a water bath that self-regulates temperature, and each reaction chamber is submerged in the water bath and closed off to the surrounding environment. Moreover, the climate room is tightly sealed with self-regulating heaters.
Based on the VS values, different co-digestion ratios were selected between the OFMSW and the WWTP sludge and tested accordingly. These ratios were then analyzed by GB21 [38] using the standardized graduated glass tubes in the temperature-controlled climate room and in the AMPTS as well.

Experimental Setup for Aerobic Digestion
AMPTS equipment (Bioprocess Control, Sweden), due to its accuracy and high throughput, was employed for the on-line measurements of biogas and biomethane flows, which were then used in the determination of the biogas potential of various substrates on a laboratory scale [39]. AMPTS consists of three main components: a digester, a carbon dioxide (CO 2 ) fixing unit, and a gas enumerator. The digester is a batch system with automated mixing motors, and each digester is connected through plastic lines to the CO 2 fixing unit then to the gas flow enumerator [40,41]. Each of these batch reactors (digesters) were kept at a constant 37 • C throughout the experiment's duration. The volume readings are automatically logged, standardized, and tabulated wirelessly by the system. AMPTS is further described in McEniry et al. (2014) [42] and Bioprocess Control Sweden AB (2014) [43].
The GB21 method was implemented in the present study in the climate room using the eudiometer system. The system consists of 500-ml reactors, where each of them is connected to a eudiometer tube and filled with a buffer solution (The buffer solution consisted of 200 g NaCl/L and 5 g/L of citric acid), which traps all biogas produced in the reactors. The climate room was kept at a constant temperature of 36 • C, and the pressure measurements were recorded on a reading basis. The total volume of the gas was measured by measuring the displacement of the buffer solution due to the gas production. Then, the total volume reading was standardized based on the climate room's pressure reading and temperature. The generated biogas volume was collected and eventually analyzed for its composition.

Waste Analysis and Characterization
The TS and VS parameters of the OFMSW, the sludge, and the inoculum were characterized, and the results were considered as a preliminary step to both the AMPTS and the climate room GB21 method test. The TS and VS for the OFMSW samples collected from the ZSRC are shown in Table 1  To investigate biogas generation potential, it is unnecessary to measure the calorific values of an OFMSW and WWTP sludge, since the calorific value should not be equal to the anaerobic fermentation ability. However, this was done in the present work in order to assess different waste-to-energy options, such as incineration and anaerobic co-digestion. The calorific values of both the OFMSW and the WWTP sludge on a dry basis are shown in Table 2. These values were acquired in accordance with DIN 51900 [44]. It is clear that the calorific values of the OFMSW and the WWTP sludge are relatively low even on a dry basis. The situation is worsened due to the fact that both the WWTP sludge and the OFMSW have a high water content; hence, incineration is not a feasible option in comparison with the biogas production option.   Table 3 and Figure 1 show the yields of bio-methane production from the OFMSW mixed with the Za'atari WWTP's sludge at different ratios. The production volumes are presented in Normal milliliter per gram of volatile solids. The production volume showed a strong correlation with the amount of organic fraction: the greater the percent of OFMSW, the greater the bio-methane yield.  Table 3 and Figure 1 show the yields of bio-methane production from the OFMSW mixed with the Za'atari WWTP's sludge at different ratios. The production volumes are presented in Normal milliliter per gram of volatile solids. The production volume showed a strong correlation with the amount of organic fraction: the greater the percent of OFMSW, the greater the bio-methane yield.

Evaluation of the Biogas Production and Performance of AMPTS
Further, it is clear that most of the biogas production occurred in the first 10 days. Thereafter, a small amount of bio-methane was cumulatively produced (i.e., 4-12% of the total production).   Further, it is clear that most of the biogas production occurred in the first 10 days. Thereafter, a small amount of bio-methane was cumulatively produced (i.e., 4-12% of the total production).
The AMPTS contains a CO 2 fixation unit (where the gas produced in the climate room is trapped in the void between the reactor and the buffer solution). Therefore, it only provides bio-methane production values. Table 4 and Figure 2 show the production volume of biogas. As stated before, the production volume is directly proportional to the amount of OFMSW used. The AMPTS contains a CO2 fixation unit (where the gas produced in the climate room is trapped in the void between the reactor and the buffer solution). Therefore, it only provides bio-methane production values. Table 4 and Figure 2 show the production volume of biogas. As stated before, the production volume is directly proportional to the amount of OFMSW used.
Based on the highest methane yield produced as shown in Table 4, it could be considered that a Vs sub/ Vs inoculum range of 0.25-0.3 is the ideal ratio for methane yield. Some offsets can be noticed in Table 4, where two values of Vs sub/ Vs inoculum are higher than 0.3: 0.457 at a lower mixing ratio (20-80%) and 0.462 at a higher mixing ratio (80-20%). In Figure 2, the cumulative methane trend showed no delay (the lag phase) in all experiments. However, for 0.457 at the lower mixing ratio (20-80%) and 0.462 at the higher mixing ratio (80-20%), the biogas ceased after 23 days, while the volume of methane production was slightly lower than that of the 0.25-0.3 Vs sub/ Vs inoculum range. This indicates that the amount of micro-organism required for an anaerobic reaction to take place is not enough for the biodegradation process [45,46].  Based on the highest methane yield produced as shown in Table 4, it could be considered that a Vs sub/Vs inoculum range of 0.25-0.3 is the ideal ratio for methane yield. Some offsets can be noticed in Table 4, where two values of Vs sub/Vs inoculum are higher than 0.3: 0.457 at a lower mixing ratio (20-80%) and 0.462 at a higher mixing ratio (80-20%). In Figure 2, the cumulative methane trend showed no delay (the lag phase) in all experiments. However, for 0.457 at the lower mixing ratio (20-80%) and 0.462 at the higher mixing ratio (80-20%), the biogas ceased after 23 days, while the volume of methane production was slightly lower than that of the 0.25-0.3 Vs sub/Vs inoculum range. This indicates that the amount of micro-organism required for an anaerobic reaction to take place is not enough for the biodegradation process [45,46].
Unlike the AMPTS, the biogas production from the climate room system includes other gases in addition to methane, which is CO 2 . Therefore, the biogas produced from the climate room system was analyzed for its composition as shown in Table 5, which shows the percentages of CH 4 and CO 2 in the biogas. Multiplying the percentage of CH 4 with the total produced amount of biogas yields CH 4 volumes that are relatively close to those found in the AMPTS test, despite the fact that both tests were performed under different (VS substrate/VS inoculum) ratios. The obtained biogas production reported in this work reflects previously reported values for similar substrates. Elbeshbishy et al. investigated in 2012 the bio-methane potential from food wastes and primary sludge, and reported that 400-1400 L methane /kg VS can be produced depending on the used inoculum [47]. Pavi et al. investigated in 2017 the biogas potential of the co-digestion of an OFMSW and fruit and vegetable wastes at different ratios. They found that the best mixing ratio was 1:3 OFMSW to the fruit and vegetables, with an approximate production of 396 Nml/g VS methane [48]. Fiore et.al studied in 2016 the biogas potential from industrial food-processing waste [49]. In their work, they reviewed several potential types of industrial waste. The results vary from 180 L methane /kg VS for apple pulp to 300 L methane /kg VS for Glycerol [49,50]. Moreover, fruit and vegetable waste has been reported to produce 320-630 L biogas /kg VS [51].

Biogas Plant Potential and Analysis
Based on the experimental results, it is recommended to undertake co-digestion of the feed for the pilot biogas plant. The feed should comprise sludge from the ZSRC's WWTP and the organic waste portion of the municipal waste generated at the ZSRC.
For the calculations, and as stated before, the ZSRC has an average population of 80,000, and generates an average of 0.8 kg/capita per day of daily waste. The MSW composition generated at the ZSRC contains 50% organic waste. Moreover, the WWTP produces approximately 2000 m 3 of sludge on a daily basis. In summary, the conducted biogas potential test indicates the following: (1): the organic portion of the municipal waste was about 40-50%, with the TS and VS being 20% and 15% of the total amount of solid, respectively; and (2) the total solid in the sludge was about 2-6% (depending on the sampling location), and the volatile solid was 0.3-3% of the fresh matter.
The maximum production of biogas from the APMTS system was found for the 100% organic waste, with an approximate value of 81 m 3 of bio-methane per ton of fresh organic waste, while the smallest quantity was for sludge from the WWTP only, with an approximate value of 1.6 m 3 of bio-methane per ton of fresh sludge.
One possible scenario to operate a large-scale biogas plant is to use 11% of the residues as feed to the biogas plant. For that, 500 kg of the sludge should be mixed with about 280 kg of organic municipal waste per day. The biogas potential test indicates that about 38 Nm 3 /day of biogas can be generated. However, the potential percentage of methane that can be generated in the biogas is about 62%. Using a combustion engine to generate heat and electricity with an efficiency of 40%, and taking into consideration the calorific value of methane, 38 Nm 3 /day can generate about 98,000 kWh/day as electricity, 98,000 kWh/day as thermal energy, or about 4 MW in theory. For that, four digesters are needed, each with a capacity of about 20,000 m 3 .

Conclusions
In the present study, it was demonstrated that laboratory-scale testing provides a good basis for the anaerobic co-digestion of the OFMSW and wastewater treatment sludge generated at ZSRC. In addition, the results of AMPTS and climate room testing could be valuable in optimizing critical parameters for the further implementation of anaerobic digestion/co-digestion under continuous operation. Co-digestion of organic waste and sludge increases the dissolved nutrients and water content in the wet fermentation process (88-95%). The maximum production of biogas from the APMTS system was from 100% organic waste, with an approximate value of 81 m 3 bio-methane·ton −1 fresh organic waste. The smallest quantity was for sludge from the WWTP only, with an approximate value of 1.6 m 3 bio-methane ton −1 fresh sludge. The results from the climate room system experiments were found to be in good agreement with those of the APMTS system experiments, with slight deviations. The maximum production was found to be from 100% organic waste, with an approximate production of 153 m 3 bio-gas ton −1 fresh organic waste and only 5.6 m 3 bio-gas ton −1 fresh sludge. Optimization of the Vs sub/Vs inoculum ratios plays a vital role in a batch digester's operation. The results indicated that Vs sub/Vs inoculum ratios ranging from 0.25 to 0.3 produce the maximum methane yields. The average composition of the bio-methane in the biogas was 65%, and the remainder was mainly carbon dioxide. The ratio of mixing between the municipal organic waste and the sludge from the WWTP should be chosen carefully to satisfy the demands of an energy production pilot plant and the environmental problems associated with the sludge amount at WWTPs. A possible ratio to start with is 60-80% organic waste, which can produce 21-65 m 3 ·ton −1 fresh matter (FM). Finally, having a biogas plant at the ZSRC to co-digest organic waste and sludge could generate 38 Nm 3 /day of methane, which could potentially produce about 4 MW in theory. This represents a potential application of a circular economy at the ZSRC and at other similar camps and small communities. Funding: The authors are thankful for the seed grant fund provided by the deanship of graduate studies and research at the German Jordanian University. The funding for all used equipment was from a project called "Biogas Production in Local Communities in Jordan" that was funded by the European Union.

Conflicts of Interest:
The authors declare no conflict of interest.