Semicontinuous Temperature-Phased Anaerobic Digestion (TPAD) of Organic Fraction of Municipal Solid Waste (OFMSW). Comparison with single-stage processes

Chemical Engineering Journal 285 (2016) 409–416

Contents lists available at ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Semicontinuous Temperature-Phased Anaerobic Digestion (TPAD) of Organic Fraction of Municipal Solid Waste (OFMSW). Comparison with single-stage processes J. Fernández-Rodríguez a,b,⇑, M. Pérez a,c, L.I. Romero a,b a b c

Agrifood Campus of Excellence (ceiA3), University of Cádiz, Spain Department of Chemical Engineering and Food Technology, Faculty of Science, University of Cádiz, Spain Department of Environmental Technology, Faculty of Sea and Environment Sciences, University of Cádiz, Spain

h i g h l i g h t s  Semicontinuous temperature-phased anaerobic digestion has been studied.  Industrial OFMSW from non-selective collection of MBT plant has been digested.  Thermophilic–mesophilic temperatures and dry condition have been selected.  Different thermophilic–mesophilic combination of times were analyzed.  In each condition, the methane yield and the organic material removal were studied.

a r t i c l e

i n f o

Article history: Received 18 August 2015 Received in revised form 5 October 2015 Accepted 7 October 2015

Keywords: Temperature Phased Anaerobic Digestion (TPAD) OFMSW Semicontinuous Thermophilic Mesophilic

a b s t r a c t The optimization of anaerobic digestion of organic wastes is a challenge to maximize energy production through biogas production process. In this study, semicontinuous TPAD (Temperature Phased Anaerobic Digestion) process has been used for the treatment of the Organic Fraction of Municipal Solid Waste (OFMSW) coming from a mechanical–biological-treatment (MBT) plant. TPAD combines the advantages of operating in different temperature ranges getting better efficiencies of organic matter removal and higher methane productivities than single-stage anaerobic digestion. In this study, the configuration used in the overall process was a thermophilic reactor (55–57 °C) for the first phase followed by a mesophilic one (35–37 °C) for the second phase. Two TPAD conditions have been tested in this paper: 4:10 and 3:6. The first digit means the SRT used in the first thermophilic phase while the second digit is related to the SRT used in the second mesophilic phase. Moreover, the performance of TPAD processes was compared with those from single-stage digesters operating at similar SRT (i.e., mesophilic and thermophilic reactors operating at 15 days SRT and thermophilic reactor operating at 10 days). The results showed that achievement reached in TPAD 4:10 was better than the corresponding one in TPAD 3:6, obtaining higher productivity of methane (35–45%) and removal of organic matter (6–19%). Moreover, the results indicate that TPAD processes reach higher efficiencies for organic matter removal (16%, 10% and 30% for DOC, CODsoluble and VS, respectively) and higher methane yields (26–60%) than single-stage systems operating at similar SRT. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction Anaerobic digestion (AD) has been considered as the most viable option from an economic point of view for the treatment ⇑ Corresponding author at: Department of Chemistry and Soil Science, School of Sciences, University of Navarra, Spain. Tel.: +34 948 425 600x806271. E-mail addresses: [email protected], [email protected] (J. Fernández-Rodríguez). http://dx.doi.org/10.1016/j.cej.2015.10.027 1385-8947/Ó 2015 Elsevier B.V. All rights reserved.

and recycling of Municipal Solid Wastes (MSWs), because produces methane and generates a digested waste similar to compost produced aerobically [1]. The AD process is strongly influenced by certain variables such as the solids content of the feed waste, the solids retention time and the temperature. According to their growth temperature range, microorganisms are classified into three groups: psychrophilic (T < 20 °C), mesophilic (20 < T < 45°C) and thermophilic (T > 45 °C). Mesophilic processes (33–37 °C) are the most common facilities for anaerobic

410

J. Fernández-Rodríguez et al. / Chemical Engineering Journal 285 (2016) 409–416

digestion at industrial scale because they are more stable [2], lower concentration of organic matter in the effluent are reached [3] and require less energy expenditure. Besides, other advantage of the mesophilic process is the lower risk of inhibition by ammonium [4–5] and long chain fatty acids [6]. Moreover, thermophilic anaerobic digestion (50–55 °C) is characterized by higher processing rate, enabling use of smaller digestion units. Indeed, it can be highlighted the increment in the hydrolysis rate and the VFA concentration reached in thermophilic range [7], especially when working with wastes with high solid content such as the OFMSW [8]. The higher and faster solubilization of organic substrates with the increase of VFA concentration, justify the thermophilic operation in the first phase in a combined temperature system. TPAD (Temperature Phased Anaerobic Digestion) processes, which is based on a combination of two temperature ranges with the aim to obtain the advantages of both ranges, is considered as a promising option for the treatment of organic wastes. The first studies were conducted by Welper et al. [9] and Han et al. [10], in systems with low solids content such as whey or activated sludge. In this type of systems, the process stability is increased with a higher efficiency in the specific production of biogas [11]. Several studies have been developed with diverse organic wastes and different combinations of temperature: hyperthermophilic (70 °C)–thermophilic (55 °C) [12–14] or thermophilic (55 °C)–mesophilic (35 °C), which is the most common combination and the selected one for this work. In fact, some authors recommend the thermophilic regime for the starting of the anaerobic digestion process as it is more effective [15–16]. The combination of both temperature regimes in TPAD can lead to a more efficient process [17– 18] which allows to take advantage of the thermophilic range to promote the hydrolysis of the waste and, then, to ensure higher stability by means of the subsequent mesophilic stage. Schmit and Ellis [19] studied the efficiency of the thermophilic phase compared with a TPAD process in codigestion sewage sludge and MSW, finding higher volatile solids reduction and production of methane in the TPAD system. The assumptions for selecting TAD configuration are that the application of a thermophilic stage can increase the rate of the rate-limiting step of anaerobic process of OFMSW – i.e., hydrolysis – whereas the use of subsequent mesophilic stage allows working with a stable process, with less risk of inhibition by ammonia and a lower concentration of volatile fatty acids. Based on that, the concatenation of the two stages allows the process to be faster and, at the same time, to achieve a higher degree of degradation of the organic matter contained in the OFMSW, maximizing the amount of biogas produced and reducing the organic content in the waste. The most common TPAD process reported in literature is the treatment of sewage sludge. In this case, the thermophilic phase not only accelerates the rate limiting step of anaerobic digestion, but also achieves hygienization by sterilization of pathogenic microorganisms based on the temperature effect [18,20–21] allowing further use of the effluent of the process for agronomic applications. Moreover, the system in phases of temperature can also improve the dewaterability of the sludge [22]. Demirer and Othman [23] studied the digestion of activated sludge in a thermophilic acidogenic reactor and the effluent was fed on a mesophilic methanogenic reactor. The thermophilic reactor was operated at SRT of 2–4 days and the subsequent degradation was conducted in a mesophilic batch reactor. Phase separation (thermophilic acidogenic and mesophilic methanogenic) allowed reaching higher removal percentages of organic matter, in the range of 26% to 49% as COD. Riau et al. [24] recently have been testing the influence of ultrasonic pretreatment on the temperature-phased anaerobic digestion of waste-activated sludge (WAS) before the thermophilic stage. The results showed that total methane production was

enhanced by more than 50% and the volatile solid removal increased by 13% in comparison to the TPAD control process. Furthermore, several studies with other wastes using TPAD codigestion system have been published. Thus, Kim et al. [25] using food waste in co-digestion with sewage sludge in a temperature phased system and Lee et al. [26] developed a mathematical model for a temperature phased system treating dog food mixed with corn meal, showing a good fit with the experimental data. Recent studies show new trends for waste treatment in TPAD processes. Kobayashi et al. [27] have published a study on the treatment of activated sludge in a temperature phased system with intermediate ozonation. They compared two systems, a thermophilic–mesophilic system and a thermophilic-ozonation–meso philic system. The intermediate ozonation improved hydrolysis for the subsequent anaerobic treatment process of the sludge, with highest methane yield and COD reduction in comparison with the classic TPAD process, thermophilic–mesophilic [28]. The comparison of the system in phases and the system with intermediate ozonation shows that the last one is more effective than the former system, achieving higher methane production and improving the final sludge dewaterability. The aim of this paper is to determine the feasibility of TPAD (t hermophilic–mesophilic) process for the Treatment of OFMSW and to compare the system performance with single-stage reactors operating at thermophilic and mesophilic temperatures. In TPAD process the initial hypothesis is that the thermophilic reactor generates a partially degraded effluent, which was used as the feed of the mesophilic reactor. Thus, the hydrolysis time (the rate-limiting step for particulate wastes as OFMSW) decreased by using the thermophilic stage while the overall process stability increased due to the further organic matter degradation in the mesophilic digestion. Indeed, two different configurations of TPAD process have been tested: TPAD 4:10 and TPAD 3:6 in order to compare with single-stage processes performed to 15 and 10 days SRT. Despite several papers about TPAD reactors for the treatment of organic wastes have been published, only a few of them are dealing with wastes as heterogeneous as the industrial OFMSW. Therefore, this paper aims to characterize and optimize the operational parameters of TPAD of industrial OFMSW.

2. Material and methods 2.1. Waste characterization The different wastes used in this study were:  Industrial OFMSW. It was used as the feed of the process and source of organic matter. The OFMSW was obtained from the recycling and composting plant ‘‘Las Calandrias”, coming from a mechanical–biological-treatment (MBT) plant, operating in semicontinuous regime, located in Jerez de la Frontera, CádizSpain.  Effluent from a lab-scale thermophilic reactor (55 °C). The effluent was collected from a semicontinuous stirred tank reactor, operating at 15 days SRT and adapted to degradation of OFMSW, because was feeding with this waste for months in stable conditions. This effluent was used as inoculum of the thermophilic phase.  Mesophilic digested sludge (35 °C), coming from the sludge reactor of a Waste Water Treatment Plant (WWTP), which is also located in Jerez de la Frontera, Cádiz-Spain. This sludge was used as inoculum of the mesophilic phase. Both effluents were used as inocula of the two reactors operating in Temperature Phased Anaerobic Digestion process. It can be

J. Fernández-Rodríguez et al. / Chemical Engineering Journal 285 (2016) 409–416

pointed out that the good acclimation of the inocula to the substrate has been a key factor to carry out the experiments. The physico-chemical characterization of the feeds of both the thermophilic and mesophilic reactors in TPAD process (OFMSW and the effluents of thermophilic reactor, respectively) is shown in Table 1. The Organic Loading Rate (OLR) added to the TPAD reactors is also shown in the table.

411

measured with a total organic carbon analyzer Analytic Jena. The CODs was measured by colorimetric method and the solid quantification by gravimetric methods. The production and the composition of the biogas was determined by gas chromatography using a gas chromatograph (Shimadzu Model 2014, Japan) equipped with a thermal conductivity detector (TCD) and a Carbosieve S-II packed column 3 m  1/8 inch Internal Diameter (Supelco, USA). The quantity and the composition of biogas were measured daily.

2.2. Equipment used 2.4. Operational conditions Two lab-scale semi-continuously stirred tank reactors (SSTR) were used (as described previously by Fernández-Rodríguez et al. [8]. The TPAD process was composed by two 5 L (4.5 L useful volume) stainless steel reactors provided with devices for mixing and heating. The system had a stainless steel cover with different inlet/outlet ports (biogas output, pH and temperature probes, and several inputs for the addition of chemical agents for pH control, feeding inlet and stirring system). The bottom of the reactor had a valve for sampling and discharging. The homogenization of the reactor was carried out by mechanical stirring. To keep the temperature constant, both reactors were equipped with a thermostatic jacket and water coming from two thermostatic circulating baths (35 °C and 55 °C, respectively) was used for temperature control. Tedlar bags were used to collect the biogas produced. To determine the biogas composition, the content was sampled using a gas-syringe. The first reactor was operated at thermophilic temperature (55–57 °C) and the other reactor at mesophilic temperature (35– 37 °C). The Fig. 1 shows a scheme of the process. As can be seen, the OFMSW is added to the thermophilic reactor and the effluent of this unit is the feeding of the mesophilic reactor.

 Thermophilic reactor, operating at 4 days SRT, followed by mesophilic reactor, operating at 10 days SRT, namely TPAD 4:10.  Thermophilic reactor, operating at 3 days SRT, followed by mesophilic reactor, operating at 6 days SRT, namely TPAD 3:6. In both cases, the output effluent of the thermophilic reactor was used as the feed of the mesophilic digester. The operating systems were maintained under these conditions for more than 3 times SRT to ensure that steady state conditions were achieved. 3. Results and discussion

2.3. Analytical methods The following parameters have been determined for system monitoring and characterization of the wastes and the inoculum thermophilic and mesophilic: pH, volatile fatty acids (VFA), dissolved organic carbon (DOC), soluble chemical oxygen demand (CODs), alkalinity, total solids (TS) and volatile solids (VS). All the parameters were determined according to Standard Methods [29]. However, the methods were adapted to high solids samples by leaching in water according to the procedure proposed by Alvarez-Gallego [30]. The procedure consisted in leaching 1 g of OFMSW in 10 mL distilled water, for 2 h. After that, the mixture was filtered to separate liquid fraction, which contains the soluble organic matter. The samples from the reactors were taken as minimum three times per week. The VFA were determined by gas chromatography methods using a Shimadzu Model 2014 with a flame ionization detector (FID) and a Nukol capillary column 30 m  0.25 mm Internal Diameter (Supelco, USA). The DOC was Table 1 Characterization of industrial OFMSW coming from MBT plant, the inoculum both T and M, and the feeds used in for the two configurations of TPAD process.

*

The study was conducted in dry anaerobic digestion conditions (20–30% TS) and thus, initially, the reactors were loaded with 30% (w/w) of inoculum and 20% (w/w) of OFMSW and then completed with distilled water, as previously reported by FernandezRodriguez et al. [8,31]. Considering previous results [8,31] operating with a similar waste, it was decided to study the following combinations of Solid Retention Times (SRT) in the sequence of reactors in TPAD:

Parameter

OFMSW

Inoculum T

Inoculum M

TPAD 4:10

TPAD 3:6

TS (%) VS (%)* DOC (ppm) COD (ppm) Vadded (L/day) OLRadded (mgDOC/ Lreactor
day) OLRadded (gSV/Lreactor day)

82.19 30.69 538.84 1591.17 — —

4.42 3.69 581.26 779.95 — —

3.89 1.93 652.84 1172.12 — —

24.01 8.97 108.89 312.61 1.125 134.71

24.01 8.97 108.89 312.61 1.500 179.61

—

—

—

22.41

29.88

VS % calculated over the total weight of the sample.

3.1. Analysis of TPAD reactors In this section the evolutions of the most representative parameters of the two tested configurations for the TPAD processes are shown in Figs. 2–4. In each Figure, the results obtained in both configurations, TPAD 4:10 and TPAD 3:6, have been splitted by a vertical dotted line. In addition, in each Figure two data sets can be seen: one of them, represented by circles, shows the evolution of different parameters measured at the thermophilic effluent and the other one, represented by squares, shows the characteristic parameters of the effluent from the mesophilic reactor, which was added with the thermophilic effluent. 3.1.1. Characteristics of semisolid effluents from the reactors The results show a more marked decrease in pH trend for the thermophilic reactor, reaching values below 7 due to the hydrolysis of the waste. However, pH values for the mesophilic reactor lie in the range 7.4–7.6, in both configurations (TPAD 4:10 and TPAD 3:6), which is optimal for the development of the methanogenic microbiota. In Fig. 2(a) the evolution of the total acidity, expressed as acetic acid concentration, can be observed. In the TPAD 4:10, acid generation takes place during the thermophilic phase, with values close to 600 ppm. It is also noted that the mesophilic reactor reaches a high VFA degradation efficiency, since the VFA concentrations in mesophilic effluent were below 50 ppm. The TPAD 3:6 shows a higher accumulation of acids during the thermophilic stage reaching values close to 800 ppm. However, the mesophilic reactor consumes most of VFAs produced in the thermophilic stage, reaching a final average concentration of 100 ppm in the effluent. Other studies carried out using temperature phased systems [32] also showed a significant increase of VFA in the effluent of

412

J. Fernández-Rodríguez et al. / Chemical Engineering Journal 285 (2016) 409–416

Fig. 1. Scheme and photograph of the TPAD process.

Fig. 2. Evolution of total acidity (a) and individual VFAs (b) (acetic, propionic and butyric) in TPAD systems: thermophilic–mesophilic.

Fig. 3. Evolution of DOC (a) and CODs (b) in TPAD systems (thermophilic–mesophilic).

the thermophilic process with a decrease in methane production. Those effects were subsequently compensated in the mesophilic reactor without jeopardize the stability of the system, due to the mesophilic reactor can break down the VFAs. Results obtained in this study indicate that the degradation of OFMSW in a TPAD system leads to a sharp generation of VFAs in the thermophilic stage and a high rate of removal of VFAs in the mesophilic one. Thus, the elimination of acids in the mesophilic reactor reaches 98.03% after a SRT of 10 days, being slightly smaller, 87.46%, after a SRT of 6 days. The high rate of VFAs removal in

the TPAD 4:10 endorses the application of that SRT combination for the TPAD process. Furthermore, the TPAD system allows the degradation of VFAs, such as propionic acid, that can cause inhibition of anaerobic processes. As shown in the Fig. 2(b), the propionic acid concentration is negligible in the mesophilic effluent of TPAD 4:10 after 10 days of operation, and it is not measured again. In that case, TPAD process allowed the removing of propionic acid. Moreover, propionic acid was not detected in TPAD 3:6 as a consequence of the adaptation of mesophilic reactor to its degradation. Kim et al. [25] have

413

J. Fernández-Rodríguez et al. / Chemical Engineering Journal 285 (2016) 409–416

Fig. 4. Evolution of methane production daily (a) and cumulative (b) in TPAD systems: thermophilic–mesophilic.

reported for wastewater treatment that the thermophilic effluent of the process shows a range of concentration of propionic acid which lies between 1500 and 2500 ppm, whereas this concentration is reduced to around 100 ppm at the effluent of the mesophilic reactor if the system is operated in TPAD. The acidity/alkalinity ratio followed a similar trend as that observed for acidity. The thermophilic effluent shows the highest values of this ratio, due to the release of VFAs during this stage. Effluent from TPAD 4:10 shows lower values of the ratio than those of the effluent of TPAD 3:6. In both cases, data are below 0.1 indicating a high stability of the system. Fig. 3(a) shows the temporal evolution of the organic matter, measured as DOC, in TPAD systems. The mesophilic reactors degraded 70.73% and 59.46% of the DOC in the corresponding thermophilic effluents for TPAD 4:10 and 3:6 systems, respectively. The average values at the output of TPAD systems (i.e., the mesophilic reactors) were 169.15 and 251.20 ppm for SRT of 10 and 6 days, respectively. Trends followed by COD in the two TPAD systems (Fig. 3b) do not show significant differences. The solubilization of organic matter, calculated from CODs, of the OFMSW in the thermophilic phase was 22.32% and 16.13% for SRT of 4 days and 3 days, respectively; data which are in agreement with the contact time between waste and microorganisms. In a previous study dealing with TPAD system with a hyperthermophilic phase (70°C), 40% solubilization of organic matter was reached operating at SRT between 1 and 5 days and working with fresh sewage sludge [12]. The high percentage of solubilization reported by these authors is due to the lower solids concentration in the waste and, to lesser extent, to the high temperature [33]. In the TPAD 4:10 process, the mesophilic reactor reached 63.51% removal of soluble COD of the thermophilic effluent reactor while in TPAD 3:6 the percentage was 59.79% (Table 2). 3.1.2. Biogas production The daily productions of methane for TPAD 4:10 and TPAD 3:6 systems are shown in Fig. 4(a). Thus, in TPAD 4:10, the average value of the biogas produced in the thermophilic reactor was 6.03 Lbiogas/(Lreactor
day) whereas in the mesophilic reactor the

production was 2.44 Lbiogas/(Lreactor
day). However, in the TPAD 3:6, the biogas production was higher with 7.98 Lbiogas/(Lreactor
day) in the thermophilic reactor and 2.79 Lbiogas/(Lreactor
day) in the mesophilic reactor. Therefore, the overall biogas production in TPAD 3:6 was 10.77 Lbiogas/(Lreactor
day) which is a 27% higher than the TPAD 4:10 (8.47 Lbiogas/(Lreactor
day)). This results show that the decrease of the SRT from 4 to 3 days is not disturbing so much the system, since OLR is 1/3 higher in 3 days SRT in front of 4 days SRT. Initially, methane production decreased in the thermophilic reactor in response to change of SRT in TPAD 3:6. However, at the end of the assay, productions previously reached in TPAD 4:10 system were achieved again. Specifically, methane production ranged from 2.97 to 3.88 LCH4/(Lreactor
day) as shown in Fig. 4(a). Moreover, the methane production in mesophilic reactors increased slightly when SRT was decreased, averaging 1.58 LCH4/(Lreactor
day) and 1.78 LCH4/(Lreactor
day) for STRs of 10 and 6 days, respectively. CO2 temporal production follows similar trend than that of CH4. The average rates at the thermophilic reactor for SRT of 4 and 3 days were 2.77 and 3.39 LCO2/(Lreactor
day), respectively, and 0.66 and 0.80 LCO2/(Lreactor
day) for STRs of 10 and 6 days at mesophilic reactor, respectively. The temporal cumulative production trend of methane, per liter of reactor, is close to a straight line whose slope can report information about the rate of the process, as can be seen in the Fig. 4 (b). Slopes were calculated using the data obtained once the system was stabilized at the solids retention time considered (i.e., the last period of the test using this SRT). Additionally, slopes were calculated expressed as LCH4/(Lreactor
day), to standardize the results. In this regard, mesophilic reactor operating at SRT of 10 days showed a slope of 1.6062 and mesophilic reactor at SRT of 6 days had a slope of 1.6933 LCH4/(Lreactor
day). In thermophilic reactor, the slope values expressed as LCH4/(Lreactor
day) were 3.2948 and 3.8736 for SRT of 4 and 3 days, respectively (Table 3). These data show that methane generation is higher for the lowest SRT. The theoretical values are in concordance with the experimental productivity of methane obtained in the stable time of operation for each condition.

Table 2 Organic matter removal in the mesophilic reactor added with pre-treated waste from the thermophilic reactor. DOC (ppm)

SRT

TPAD 10:4 TPAD 6:3

COD soluble (ppm)

Added

Effluent

Removal (%)

Added

Effluent

Removal (%)

577.9 ± 2.9 619.8 ± 3.6

169.2 ± 3.3 251.2 ± 5.4

70.7 59.5

1236.0 ± 4.7 1423.7 ± 8.3

451.0 ± 2.1 572.4 ± 5.4

63.5 59.8

414

J. Fernández-Rodríguez et al. / Chemical Engineering Journal 285 (2016) 409–416

Table 3 Results of the regression fitting of accumulated methane versus time and data of methane productivity in thermophilic and mesophilic reactors of TPAD process. SRT (days)

Regression fitting

Real productivity of methane Methane yields

TPAD 4:10

TPAD 3:6

4T

10 M

3T

6M

Slope (LCH4/ day) R2 Slope expressed as LCH4/(Lreactor
day) LCH4/(Lreactor
day)

14.81 0.9658 3.29

7.25 0.9992 1.61

17.42 0.9979 3.87

7.61 0.9992 1.69

3.47

1.58

3.88

1.78

LCH4/g LCH4/g LCH4/g LCH4/g LCH4/g

8.73 33.38 22.01 – 0.15

12.81 22.51 27.39 59.63 –

7.28 66.98 21.61 – 0.13

7.51 12.01 17.26 34.81 –

CODadded CODcons DOCadded DOCcons SVadded

3.2. Study of the TPAD system as overall system (black box analysis) The two reactors in the temperature phased process can be considered as one entity in a black box analysis. In that case, only the inputs and outputs to the global system are considered. Thus, it is considered that the feed of the process is the inlet of the thermophilic reactor and the effluent of the process is the outlet of the mesophilic one. Additionally, the biogas production, of both reactors has been considered jointly as the overall biogas production of the process. Fig. 5 shows the percentage of removal for DOC, COD and VS in TPAD 4:10 and TPAD 3:6 (as a black box). TPAD 3:6 showed differences with regard to TPAD 4:10 system. It can be observed that all removal percentages in TPAD 4:10 system were around 68–73%, measured once SRT of the system was stabilised, except in VS removal. In case of VS, the elimination rate in TPAD 3:6 is around 81.3% VS, which is slightly higher than observed in TPAD 4:10 (75.1%). The soluble COD removal reached a value of 65.6% and DOC removal was 53.2%, which are lower than those obtained in the TPAD 4:10. The comparison between the two TPAD assayed, shows that TPAD 4:10 showed 9.32% higher removal percentage than TPAD 3:6. On the other hand, higher DOC removal percentages can be obtained operating at TPAD 4:10, being 29.48% higher than that in TPAD 3:6. Studies performed by other authors [9] using temperature phased process for the treatment of non-fat dry residue from the dairy industry showed a 90% removal operating at total SRT of 18 h (6 h in thermophilic plus 12 h in mesophilic). In the present study, the high concentration of solids (dry anaerobic digestion)

hinders to work at such low SRT and, furthermore the complexity and heterogeneity of the waste avoid achieving high rates of removal compared with the cited study. The yield of methane calculated as an overall process is shown in Table 4. The TPAD 4:10 is more effective than TPAD 3:6 system because the specific methane production (SMP) – expressed as COD and DOC both added and consumed – is higher. Thus, the yield of methane for TPAD 4:10 is approximately 40% higher than that for TPAD 3:6. The methane productivities obtained in this study were 2.17 LCH4/(Lreactor
day) and 2.45 LCH4/(Lreactor
day) in the TPAD 4:10 and TPAD 3:6, respectively. These productivities are higher than those obtained by other authors such as Kim et al. [25] in a study for the co-digestion of organic kitchen waste and sewage sludge in a TPAD system with 10 days HRT in total, who obtained 0.41 LCH4/(Lreactor
day). One possible reason for these higher rates of methane production is that OFMSW used in this study has an appropriate relationship between the organic matter and alkalinity (low organic matter content and high alkalinity). Thus, a lower concentration of organic matter reduces risk of inhibition by acidification and allows working with lower SRT, 4 days and 3 days in thermophilic. The comparison of both processes TPAD shows a higher methane productivity in the TPAD 3:6 than in TPAD 4:10. However, the specific productivity of methane referred to organic matter added or consumed is higher in the TPAD 4:10. Specifically, the specific productivity of methane is 35–45% higher in the TPAD 4:10 versus TPAD 3:6. The performance of TPAD 3:6 is only slightly worse than TPAD 4:10. However, it can be pointed out that TPAD 3:6 achieves high rates of organic matter removal despite of operating in an overall SRT of 9 days, which is lower than the SRT commonly used at industrial scale for the anaerobic treatment of OFMSW. In general, it was found that TPAD 4:10 process is favoured with respect to TPAD 3:6, since both a higher SMP as a higher percentage of organic matter removal were reached. However, the production rate of methane expressed as LCH4/(Lreactor
day) is higher for TPAD 3:6. 3.3. Comparison of TPAD systems with the single-systems operating at mesophilic and thermophilic ranges TPAD systems have shown better performance, in terms of elimination of organic matter, than single-stage systems operating at mesophilic or thermophilic temperature ranges. In this regards, Shantha Sung et al. [34] have reported very good results for organic matter removal and gas productivity at a temperature phased sys-

Fig. 5. Removal percentages of organic matter (referred to DOC, COD and VS) in TPAD 4:10 and TPAD 3:6 systems, temporal evolution (a) and histogram of comparative average efficiencies (b).

415

J. Fernández-Rodríguez et al. / Chemical Engineering Journal 285 (2016) 409–416 Table 4 Comparison of the productivity of methane and the methane yields in TPAD and single-reactor systems. Reference T 15 T 10 M 20 M 15 TPAD 4:10 TPAD 3:6

Fernández-Rodríguez Fernández-Rodríguez Fernández-Rodríguez Fernández-Rodríguez This study This study

et et et et

al. al. al. al.

[8] [8] [31] [31]

LCH4/(Lreactor
day)

LCH4/gDOCadded

LCH4/gDOCcons

LCH4/gCODadded

LCH4/gCODcons

1.40 1.95 1.03 1.07 2.17 2.45

30.70 35.73 35.56 29.75 56.29 41.00

59.62 65.94 51.62 51.73 81.73 76.80

– 8.79 9.24 10.07 19.06 13.01

– 25.61 15.01 15.15 26.60 19.85

tem treating cattle manure. These authors have obtained similar removal percentages of VFA, ammonium and pathogens when operated at SRT of 14 days in a TPAD system thermophilic–mesophilic and when used single-stage process operating at SRT of 20 days in mesophilic range. For the treatment of OFMSW, the application of TPAD process is a very interesting option because the thermophilic first phase in the system enhances the hydrolysis rate of solid waste and the mesophilic second phase allows a high organic matter removal and a deeper stabilization of the waste. Hence, a comparison between TPAD systems (TPAD 4:10 and TPAD 3:6), and the individual mesophilic and thermophilic processes for the treatment of OFMSW can be performed on the basis of DOC, COD and VS removals and methane productions. In order to compare TPAD and single systems, similar SRT should be selected. Therefore, taking into account that the overall SRT in TPAD systems tested have been 14 and 9 days, the following data published by Fernández-R odríguez et al. [8,31] have been considered:  Thermophilic anaerobic digestion operating at SRT of 15 days (T15) and 10 days (T10).  Mesophilic anaerobic digestion operating at SRT of 20 days (M20) and 15 days (M15). The average of DOC removal is higher in the TPAD 4:10 compared to single-stage M15 and T15, 68.87% versus 57.51% and 58.95%, respectively. TPAD 3:6 and single T10 show a similar DOC removal percentage, around 50–55%. As regards the COD, TPAD 4:10 and 3:6 attain a higher removal of COD compared to those of single-stage mesophilic and thermophilic systems. Thus, the TPAD 4:10 reaches a COD removal of 71.66% which is 15% higher than that obtained in the mesophilic system M15 (66.51%). In the case of TPAD 3:6, the removal percentage was 65.55% while in the single-stage thermophilic system T10 the removal was 59.83%. Concerning to the removal of VS, TPAD 3:6 presents a value 34% higher than that of the one-stage T10. Similar results have been obtained by Ge et al. [35] who observed a higher reduction of VS in TPAD systems (between 20% and 25%) compared to a mesophilic single-stage, in a study about degradation of sewage sludge, working at SRT of 2 days in the thermophilic phase and 13 days in mesophilic one. Reusser and Zelinka [36] also observed a higher VS destruction in TPAD thermophilic–mesophilic at HRT 10 and 12.5 days compared to single-stage mesophilic or thermophilic digestion at 15 to 20 days. In this study, the average VS removal was 73.05% in TPAD 4:10, 73.67% in M15 and 80.40% in T15 and, hence, no significant differences were observed for that SRT. Table 4 shows the specific productivity of methane per gram of organic matter (CODsoluble and DOC) added and consumed into the system. The temperature phased process has shown to be feasible for the treatment of an extremely heterogeneous waste with a high solid content such as the industrial OFMSW. Additionally, data presented in this study show that systems in temperature phased have a specific productivity of methane per gram of organic matter higher than the equivalent single-reactor systems. Specifically, the TPAD 4:10 showed higher productivity of methane than that of the

T15 and M15 reactors. Additionally, the TPAD 3:6 also reaches a higher SMP than that of T10. The methane yield in single-reactor systems ranged from 35 to 40 LCH4/gDOCadded whereas the SMP in TPAD systems reaches 56.29 and 41.00 LCH4/gDOCadded in TPAD 4:10 and 3:6, respectively. The values of the parameters listed are higher in the temperature system except when productivity refers to COD consumed, the values in the TPAD 3:6 and T10 are in the same range: 20 to 25 LCH4/(Lreactor
day). 4. Conclusions Based on the results presented in this study, it can be concluded that:  The TPAD process (thermophilic–mesophilic) is feasible in both configurations tested: TPAD 4:10 (4 days SRT in the thermophilic reactor followed by 10 days SRT in the mesophilic reactor), and TPAD 3:6 (3 days SRT at thermophilic stage and 6 days SRT at mesophilic stage).  The TPAD systems have shown higher removal efficiencies of organic matter (measured as DOC, CODsoluble and VS) as well as higher methane productivity than those of the singlereactor systems, for similar retention times.  The best results were obtained for the TPAD 4:10 system as higher specific productivities of methane (35–45%) and higher removal efficiencies of organic matter (6–19%) were reached compared to the TPAD 3:6 system. However, the TPAD 3:6 reaches a high productivity of methane, 2.45 LCH4/(Lreactor
day), which together with the significant decreasing in the overall SRT, also makes it an interesting industrial option. In short, it can be concluded that TPAD systems are feasible to the treatment of OFMSW, reaching higher rates of removal of organic matter and higher methane productivities than singlereactor systems at similar solids retention times. Acknowledgements This work has been developed in the research group of Environmental Technologies of the University of Cádiz (TEP 181), Group of Excellence of Andalusian Plan of Research, Development and Innovation TEP-181. It has been funding thanks to the Ministry of Science and Innovation (Project CTM2010-17654); the funding of the Fellowship José Castillejo CAS14/00003 Ministry of Education, Culture and Sport of the Government of Spain; and the European Regional Development Funds. References [1] G. Tchobanoglous, in: Integral Management of Solid Wastes, McGraw-Hill, D.L. Madrid, 1998. [2] K.F. Fannin, R. Biljetina, Reactor Design, in: D.P. Chynoweth, R. Isaacson (Eds.), Anaerobic Digestion of Biomass, Elsevier Applied Science, London, 1987, pp. 109–128. [3] J. Fernández-Rodríguez, M. Perez, L.I. Romero, Comparison of mesophilic and thermophilic dry anaerobic digestion of OFMSW: kinetic analysis, Chem. Eng. J. 232 (2013) 59–64.

416

J. Fernández-Rodríguez et al. / Chemical Engineering Journal 285 (2016) 409–416

[4] I. Angelidaki, B.K. Ahring, Thermophilic anaerobic digestion of livestock waste: effect of ammonia, Appl. Microbiol. Biotechnol. 38 (1994) 560–564. [5] K.H. Hansen, I. Angelidaki, B.K. Ahring, Anaerobic digestion of swine manure: inhibition by ammonia, Water Res. 32 (1) (1998) 5–12. [6] J. Palatsi, B. Fernández, V.A. Vavilin, X. Flotats, Anaerobic biodegradability of fresh slaughterhouse waste: interpretation of results by a simplified model, in:11th World Congress on Anaerobic Digestion (AD11) Bioenergy for our future, Australia, 2007. [7] Q. Li, W. Qiao, X.C. Wang, K. Takayanagi, M. Shofie, Y. Li, Kinetic characterization of thermophilic and mesophilic anaerobic digestion for coffee grounds and waste activated sludge, Waste Manage. 36 (2015) 77–85, http://dx.doi.org/10.1016/j.wasman.2014.11.016. [8] J. Fernández-Rodríguez, M. Perez, L.I. Romero, Dry thermophilic anaerobic digestion of the organic fraction of municipal solid wastes: solid retention time optimization, Chem. Eng. J. 251 (1) (2014) 435–440. [9] L.L. Welper, S. Sung, R.R. Dague, Laboratory studies on the temperature-phased ASBR system, Water Sci. Technol. 36 (2–3) (1997) 295–302. [10] Y. Han, R. Dague, Laboratory studies on the temperature-phased anaerobic digestion of domestic primary sludge, Water Environ. Res. 69 (6) (1997) 1139– 1143. [11] S. Ke, Z. Shi, H.H.P. Fang, Applications of two-phase anaerobic degradation in industrial wastewater treatment, Int. J. Environ. Pollut. 23 (1) (2005) 65–80. [12] D. Bolzonella, P. Pavan, M. Zanette, F. Cecchi, two-phase anaerobic digestion of waste activated sludge: effect of an extreme thermophilic prefermentation, Ind. Eng. Chem. Res. 46 (21) (2007) 6650–6655. [13] M.Y. Lee, J.H. Cheon, T. Hidaka, H. Tsuno, The performance and microbial diversity of temperature-phased hyperthermophilic and thermophilic anaerobic digestion system fed with organic waste, Water Sci. Technol. 57 (2) (2008) 283–289. [14] F. Wang, T. Hidaka, H. Tsuno, J. Tsubota, Co-digestion of polylactide and kitchen garbage in hyperthermophilic and thermophilic continuous anaerobic process, Bioresour. Technol. 112 (2012) 67–74. [15] L. Shu-Guang, I. Tsuyoshi, U. Masao, S. Masahiko, Start-up performances of dry anaerobic mesophilic and thermophilic digestions of organic solid wastes, J. Environ. Sci. 19 (4) (2007) 416–420. [16] J. Fernández-Rodríguez, M. Perez, L.I. Romero, Temperature-phased anaerobic digestion of industrial organic fraction of municipal solid waste: a batch study, Chem. Eng. J. 270 (15) (2015) 597–604, http://dx.doi.org/10.1016/j. cej.2015.02.060. [17] M. Duran, R.E. Speece, Temperature-staged anaerobic processes, Environ. Technol. 18 (7) (1997) 747. [18] V. Riau, M.A. de la Rubia, M. Pérez, Temperature-phased anaerobic digestion (TPAD) to obtain class A biosolids: a semi-continuous study, Bioresour. Technol. 101 (8) (2010) 2706–2712. [19] K.H. Schmit, T.G. Ellis, Comparison of temperature-phased and two-phase anaerobic co-digestion of primary sludge and municipal solid waste, Water Environ. Res. 73 (3) (2001) 314–321. [20] M.D. Aitken, G.W. Walters, P.L. Crunk, J.L. Willis, J.B. Farrell, P.L. Schafer, C. Arnett, B.G. Turner, Laboratory evaluation of thermophilic-anaerobic digestion to produce class A biosolids. 1. Stabilization performance of a continuous-flow reactor at low residence time, Water Environ. Res. 77 (6) (2005) 3019–3027. [21] E. Viau, J. Peccia, Survey of wastewater indicators and human pathogen genomes in biosolids produced by class A and class B stabilization treatments, Appl. Environ. Microbiol. 75 (1) (2009) 164–174. [22] J.L. Bivins, J.T. Novak, Changes in dewatering properties between the thermophilic and mesophilic stages in temperature-phased anaerobic digestion systems, Water Environ. Res. 73 (4) (2001) 444–449. [23] G. Demirer, M. Othman, Two-phase thermophilic acidification and mesophilic methanogenesis anaerobic digestion of waste-activated sludge, Environ. Eng. Sci. 25 (9) (2008) 1291–1300. [24] V. Riau, M.A. De la Rubia, M. Pérez, Upgrading the temperature-phased anaerobic digestion of waste activated sludge by ultrasonic pretreatment, Chem. Eng. J. 259 (1) (2008) 672–681.

[25] H.-W. Kim, S.-K. Han, H.-S. Shin, Anaerobic co-digestion of sewage sludge and food waste using temperature-phased anaerobic digestion process, Water Sci. Technol. 50 (9) (2004) 107–114. [26] M.-Y. Lee, C.-W. Suh, Y.-T. Ahn, H.-S. Shin, Variation of ADM1 by using temperature-phased anaerobic digestion (TPAD) operation, Bioresour. Technol. 100 (11) (2009) 2816–2822. [27] T. Kobayashi, Y.Y. Li, H. Harada, H. Yasui, T. Noike, Upgrading of the anaerobic digestion of waste activated sludge by combining temperature-phased anaerobic digestion and intermediate ozonation, Water Sci. Technol. 59 (1) (2009) 185–193. [28] M. Weemaes, H. Grootaerd, F. Simoens, A. Huysmans, W. Verstraete, Ozonation of sewage sludge prior to anaerobic digestion, Water Sci. Technol. 42 (9) (2000) 175–178. [29] APHA, AWWA, WPCF, in: Métodos normalizados. Para análisis de aguas potables y residuales, Díaz de Santo, S.A., 1992. [30] C.J. Álvarez Gallego, Different test procedures for starting a process of anaerobic co-digestion of OFMSW and dried sewage sludge in thermophilic range (PhD thesis), University of Cadiz, 2005. [31] J. Fernández-Rodríguez, M. Perez, L.I. Romero, Mesophilic anaerobic digestion of the organic fraction of municipal solid waste: Optimisation of the semicontinuous process, Chem. Eng. J. 193–194 (2012) 10–15. [32] S.K. Kaiser, R.R. Dague, W.L. Harris, Initial studies on the temperature-phased anaerobic biofilter process, Water Environ. Res. 67 (7) (1995) 1095–1103. [33] K.H. Schmit, Comparison of temperature phased and other state of the art processes for anaerobic digestion of municipal solid waste (Retrospective Theses and Dissertations), Digital Repository Iowa State University, Paper 12522, 1998. [34] S. Sung, H. Santha, Performance of temperature-phased anaerobic digestion (TPAD) system treating dairy cattle wastes, Water Res. 37 (7) (2003) 1628– 1636. [35] H. Ge, D. Jensen Paul, J. Batstone Damien, Pre-treatment mechanisms during thermophilic–mesophilic temperature phased anaerobic digestion of primary sludge, Water Res. 44 (1) (2010) 123–130. [36] S. Reusser, G. Zelinka, Laboratory-scale comparison of anaerobic-digestion alternatives, Water Environ. Res. 76 (4) (2004) 360–379.

Glossary and notations AD: anaerobic digestion COD: chemical oxygen demand, expressed as mg/L CODs: soluble chemical oxygen demand, expressed as mg/L CTSRs: continuous stirred tank reactors DOC: dissolved organic carbon, expressed as mg/L FID: flame ionization detector HRT: SRT: Hydraulic Retention Time, expressed as days M 15: 15 days SRT mesophilic reactor M 20: 20 days SRT mesophilic reactor MBT: mechanical–biological-treatment OFMSW: Organic Fraction of Municipal Solid Waste SRT: Solid Retention Time, expressed as days T 10: 10 days SRT thermophilic reactor T 15: 15 days SRT thermophilic reactor TCD: thermal conductivity detector TPAD 3:6: Temperature Phased Anaerobic Digestion reactors, 3 days SRT in thermophilic followed by 6 days SRT in mesophilic TPAD 4:10: Temperature Phased Anaerobic Digestion reactors, 4 days SRT in thermophilic followed by 10 days SRT in mesophilic TPAD: Temperature Phased Anaerobic Digestion TS: total solids concentration, expressed as % VFA: volatile fatty acids, expressed as mg/L VS: volatile solids concentration, expressed as % WAS: waste activated sludge