Abstract

Anaerobic mono-digestion and co-digestion are nowadays widely used in wastewater treatment plants (WWTP). However, the data processing of the conventional biochemical potential test (BMP) carried out to assess potential substrates should be enhanced to reduce the uncertainty of the results. In this study, two methodologies aiming to improve the data processing in anaerobic digestion studies were proposed. The methodologies aimed at the estimation of synergy in anaerobic co-digestion of organic waste and the standardization of the BMP test results by considering the activity of the inoculums under mono-digestion conditions. Both methodologies comprise the application of the Gompertz equation. For the first methodology, four cosubstrates and two types of substrates were used. Regarding synergy estimation, the cosubstrates dairy whey and grease sludge had an impact on the degradation kinetic. In regard to the second methodology, the results indicate that the activity of the inoculums exerts an influence on the BMP analysis, and it should be considered. This can be meaningful when comparing results among studies when different inoculums are used or even for studies where the same inoculum is used but it is taken at different reactor operational moments.

NOTATION

     
  • WWTP

    Wastewater treatment plant

  •  
  • AcoD

    Anaerobic co-digestion

  •  
  • MSS

    Mixed sewage sludge

  •  
  • TPS

    Thermally pretreated sludge

  •  
  • Gr

    Grease

  •  
  • DW

    Dairy whey

  •  
  • GrS

    Fatty sludge

  •  
  • WAS

    Biological activated sludge

  •  
  • InC

    Inoculum from a digester treating MSS

  •  
  • InA

    Inoculum from a digester treating TPS

  •  
  • B

    Biogas production on time

  •  
  • P

    Maximum biogas production

  •  
  • Rm

    Maximum biogas production rate

  •  
  • λ

    Lag-phase parameter

  •  
  • EV

    Expected value

  •  
  • OV

    Observed value

  •  
  • MP

    Model parameter

  •  
  • NMP

    Normalized model parameter

  •  
  • CODt

    Chemical oxygen demand (total)

  •  
  • CODs

    Chemical oxygen demand (soluble)

  •  
  • VFA

    Volatile fatty acids

  •  
  • TAN

    Total ammonium nitrogen

  •  
  • TKN

    Total Kjeldahl nitrogen

  •  
  • VS

    Volatile solid

INTRODUCTION

Anaerobic co-digestion (AcoD) of waste is defined as the feeding of two or more substrates of different characteristics and/or origin to the same anaerobic digester. According to the data collected by the Water Environment Federation (WEF), out of the 1,191 wastewater treatment plants (WWTP) that have anaerobic digesters in the USA, 159 are importing substrates to be co-digested in their systems (WEF 2018). The application of AcoD leads to important benefits, particularly in terms of biogas productivity and yield. This performance enhancement has been associated, compared to conventional mono-digestion, with better nutrient availability, a major presence of trace elements, the dilution of potential inhibitory compounds, and changes of the media rheology that improve the mass transfer and mixing (Mata-Alvarez et al. 2011).

Synergy is the preferred word used to describe the positive outcome, especially in terms of biogas production, driven by the mix of substrates for AcoD (Aichinger et al. 2015). However, the use or even abuse of this concept has led to some controversy (Insam & Markt 2016). It is easy to overestimate or misuse the term synergy of waste interactions (especially when positive results are obtained) since the sole increase in the biogas production in terms of the total amount, or standardized per mass of organic matter added, is not enough. For instance, a lipid-rich waste will yield more biogas than a carbohydrate-rich waste per gram of volatile material. Different authors have estimated the synergy due to the AcoD in different ways. The approaches have been based on the proportional summation of the experimental methane yields obtained from single substrate digestion (mono-digestion) versus those obtained from the AcoD of such substrates (Labatut et al. 2011; Ara et al. 2015), a COD balance (Aichinger et al. 2015; Xie et al. 2017), dynamic or steady state modeling approaches (Astals et al. 2014; Aichinger et al. 2015; Kim et al. 2017) or statistical modeling (Kashi et al. 2017).

In regard to the experimental set-up, the so-called biochemical potential test (BMP) is carried out to study the anaerobic biodegradability of a certain substrate.The BMP assay provides a platform to test the anaerobic degradability of different organic wastes as well as a proper comparison criterion to assess their degradability properties. However, the reported BMP results show significant variability, even for the same substrates. This is due to the large quantity of variables that can affect the results and that cannot be easily standardized, such as the substrate heterogeneity given by its size, composition or bioavailability, the inoculum's origin and quality, and the biogas measurement method (Raposo et al. 2011; Strömberg et al. 2014). Amid these variables, the quality of the inoculum exerts an important influence upon the results of the BMP test and there is no simple solution to overcome this situation (Angelidaki et al. 2009; Holliger et al. 2016; Steinmetz et al. 2016). For instance, Angelidaki et al. (2009) and Holliger et al. (2016) suggested several preparation methodologies for the inoculum selection and the digester seeding as well as inoculum activity estimation using acetate and cellulose as the substrate. Steinmetz et al. (2016) proposed a method for inoculum enrichment and maintenance in order to improve the reliability of the BMP results.

Therefore, the first aim of this study is to propose a methodology to estimate the absence or presence of synergy in anaerobic co-digestion. The second aim is to propose a methodology to discard the activity of the inoculum as a variable in a conventional mono-digestion batch test. Both proposals are based on a kinetic analysis of the biogas production curves.

MATERIALS AND METHODS

Synergy determination in AcoD test

Substrates, cosubstrates and inoculum

Two substrates, mixed sewage sludge (MSS) and thermally pretreated sludge (TPS) were used. MSS was composed of 60% primary sludge and 40% secondary sludge from the activated sludge system of the wastewater treatment plant (WWTP) Mapocho-Trebal of Santiago, Chile. TPS is a mixture of raw primary sludge (60%) and thermally pretreated secondary sludge (40%) that was hydrolyzed in a pilot system for 20 min at 6 bars/170 °C. The specific details of the mixtures are presented in Table 1. As cosubstrates, four organic wastes were evaluated: grease (Gr) from a food processing industry, dairy whey (DW) from a cheese factory, fatty sludge (GrS) coming from the primary settler of the WWTP of a food processing industry and waste activated sludge (WAS) from an extended aeration-based WWTP. For the tests where MSS was used as substrate, an inoculum (named InC) coming from a full-scale digester treating MSS was seeded. Likewise, for the tests where TPS was used as substrate, an inoculum (named InA) coming from a full-scale digester treating TPS was seeded. Both inoculums are expected to have different microbial diversity in terms of methanogens that is driven by the tolerance of the presence of ammonia and VFAs that are present in the substrates (Wett et al. 2014).

Table 1

Characterization of the substrates, cosubstrates and inoculums

Substrates
Co-substrates
Inoculums
MSSTPSGrDWWASGrSInCInA
pH 5.7–6.1 5.5–6.1 4.0 3.5 7.9 6.8 7.6–8.3 7.7–8.2 
TS (g/kg) 37.6–44.6 39.8–58.9 660 101.5 139.0 131.3 21.7–26.5 25.5–32.2 
VS (g/kg) 28.6–34.2 30.1–47.2 647.5 83.2 102.8 111.8 14.4–17.2 17.0–21.0 
CODt (g/L) 56.7–71.8 61.5–93.7 – 100.9 152.6 203.5 26.5–32.9 35.5–38.0 
CODs (g/L) 0.8–1.6 5.7–8.0 30.2 6.0 6.0 3.4 0.5–1.0 1.3–1.6 
VFA (mg/L) 1460–4500 1976–4988 – 2392 840 4800 204–348 404–520 
TAN (mg/L) 312–1044 432–1870 39 <1 980 1622 960–1718 1760–3080 
Proteins (%) 31 45 70 26   
Lipids (%) 41 35 100 11 53   
Carbohydrates (%) 28 20 89 24 21   
Substrates
Co-substrates
Inoculums
MSSTPSGrDWWASGrSInCInA
pH 5.7–6.1 5.5–6.1 4.0 3.5 7.9 6.8 7.6–8.3 7.7–8.2 
TS (g/kg) 37.6–44.6 39.8–58.9 660 101.5 139.0 131.3 21.7–26.5 25.5–32.2 
VS (g/kg) 28.6–34.2 30.1–47.2 647.5 83.2 102.8 111.8 14.4–17.2 17.0–21.0 
CODt (g/L) 56.7–71.8 61.5–93.7 – 100.9 152.6 203.5 26.5–32.9 35.5–38.0 
CODs (g/L) 0.8–1.6 5.7–8.0 30.2 6.0 6.0 3.4 0.5–1.0 1.3–1.6 
VFA (mg/L) 1460–4500 1976–4988 – 2392 840 4800 204–348 404–520 
TAN (mg/L) 312–1044 432–1870 39 <1 980 1622 960–1718 1760–3080 
Proteins (%) 31 45 70 26   
Lipids (%) 41 35 100 11 53   
Carbohydrates (%) 28 20 89 24 21   

Experimental setup

The experimental evaluation was based on the BMP test, following the guidelines given in Angelidaki et al. (2009). For the experimental set-up, glass bottles of 120 mL with 100 mL of working volume were used. A substrate inoculum (S/I) ratio of 0.5 gVS gVS−1 and an inoculum concentration of 10 gVS L−1 were established. Sodium bicarbonate was added in a concentration of 0.5 g NaHCO3 gVS−1 of substrate and co-substrate to ensure neutral pH values during the assay. Blank trials with only inoculum were used to quantify the amount of biogas produced by endogenous respiration. Each BMP test was performed in duplicate. The temperature of the assay was set at mesophilic temperature (35 °C) and the bottles were stirred at 40 rpm in a thermoregulated shaker system. The biogas production was measured by an interchangeable 50 ml glass plug syringe (Luer lock). The net value of the biogas yield was obtained by subtracting the endogenous production of the blank bottle and converted to normal conditions of temperature (0 °C) and pressure (1 atm). The inoculum samples were degasified for at least three days before the test started.

Each experimental AcoD run (one for each cosubstrate tested with both substrates, separately) was carried out approximately a month apart from each other. Six proportions, namely: 0, 10, 25, 50, 75 and 100% of cosubstrate as percentage of total VS, were used. The fact that the inoculums were taken at different operational moments could affect the observed results and add an additional uncertainty when comparing different co-substrates to assess AcoD performance. However, the full-scale digesters, where the inoculums were taken from, are working in continuous mode so that, at least a priori, one could expect to have stable properties of the inoculums in the evaluation period. In any case, the influence of the inoculum's activity will be discussed in the Inoculum activity assessment section.

Data processing

From the accumulative biogas production, three kinetic parameters (P, Rm and λ) were estimated by fitting the Gompertz equation (Equation (1)). 
formula
(1)
where B: biogas production at time t (d), P: Maximum biogas production (ml gVS−1), Rm: Maximum biogas production rate (ml gVS−1 d−1) λ: Lag-phase time (d). To estimate the parameters, single least squares criteria was used as the minimization procedure. The standard deviation was estimated from the covariance matrix of the parameters obtained from the inverse of the Fisher Information Matrix (FIM), which gives a lower bound on the achievable parameter error covariance matrix. The estimated values of these parameters were defined as the observed values (OV) from the experiments.
The proposed methodology defines synergy (Si) as the standardized difference between the observed value (OV) of a certain parameter (P, Rm or λ) obtained from the Gompertz equation and the expected value (EV) of the parameter, as is shown in Equation (2). 
formula
(2)
The expected value of the parameter is determined from Equation (3). The approach chosen was based on the proportional effect of the cosubstrate and the substrate fractions in the AcoD test (Ara et al. 2015). 
formula
(3)
where S_F and CoS_F represent the decimal fraction (dimensionless) related to the proportion of the substrate and cosubstrate added to test, respectively. MPs and MPCoS correspond to the model parameter values (P, Rm or λ) obtained during the mono-digestion of the substrate and cosubstrate, respectively.

Discarding inoculum activity in mono-digestion batch test

Experimental set-up, substrate and inoculum

BMP tests following the same guidelines described in the Substrates, cosubstrates and inoculum section were set up. As substrates, MSS and TPS were used for the BMP test and chopped cellulose-based paper (5 mm square) was used for the control BMP test. Chopped cellulose-based paper was chosen so that the activity of the main metabolic pathways of the anaerobic digestion process to produce biogas are tested, from the disintegration to the methanogenesis. InC and InA were used as inoculums for MSS and TPS, respectively. In contrast, both inoculums were used with chopped cellulose-based paper as substrate. The two inoculums (InC and InA) were taken from their respective digesters in four different moments, defined as M1, M2, M3 and M4, approximately a month apart from each other. It is worth pointing out that M1, M2, M3 and M4 also match with each of the four AcoD runs described in the Experimental setup section.

Data processing

The proposed methodology is based on the normalization of the kinetic parameters (OV) from the BMP test by the same parameter obtained from the control test. Therefore, first, the Gompertz equation fitting processing has to be carried out as described in the Data processing section, in order to get the parameters for all the substrates and the control test. Afterwards, and in order to discard the activity inoculum effect on the result and make a better comparison of the results, the parameters obtained for the BMP test of MSS and TPS were normalized by the kinetic parameters from the control test according to Equation (4). 
formula
(4)
where, NMP represents the normalized model parameter (P, Rm or λ), MPBMP is the model parameter obtained from the BMP test for the studied substrate and MP_control is the model parameter value (P, Rm or λ) for the control test.

Analytical methods

Total and volatile solids were measured using the gravimetric method according to APHA (2012). The chemical oxygen demand (COD) was measured by spectrophotometry and volatile fatty acids (VFA) were measured using the titration method according to APHA (2012). The pH was measured using a laboratory meter (WTW inoLab® pH7110) and the total ammonium nitrogen (TAN) was measured by an ion selective electrode (Dual Star Orion model). The total Kjeldahl nitrogen (TKN) was measured by digestion and oxidation according to APHA (2012). The protein content (Pc) was calculated based on the assumption that the protein contains 16% (w/w) nitrogen, therefore, the difference between the TKN and TAN was multiplied by a factor of 6.25 to estimate the protein content (Donoso-Bravo et al. 2011). Lipids were measured by Soxhlet extraction according to APHA (2012). Carbohydrates were estimated using a balance (Donoso-Bravo et al. 2011); that is, the total COD of the cosubstrates minus the COD of the proteins and the COD of the lipids.

RESULTS AND DISCUSSION

Characterization

The physicochemical characterization of the substrates, cosubstrates and inoculums used in the experiments is shown in Table 1. Regarding the substrates, the results show similarities in regard to the total organic content, although a wider range of values was obtained for MSS. The effect of the thermal hydrolysis process on WAS is shown in the higher content of CODs and TAN for TPS compared to MSS. With respect to the cosubstrates, they all have a high organic matter content; however, their macromolecular content is quite different, which allows us to categorize them as: pure-lipid (Gr), carbohydrate-rich (DW), protein-rich (WAS) and lipid-rich (GrS). According to this content, it can be estimated that the highest methane potential is expected for Gr, GrS, WAS and DW in decreasing order (Raposo et al. 2011). With respect to the inoculum, the physicochemical parameters were higher for InA than InC. The digester, where InA comes from, is fed with TPS which has an impact on the final organic matter content of the digestate that is related to the presence of more recalcitrant soluble organic material formed during the hydrolysis process.

Parameter estimation and synergy in AcoD

The cumulative biogas production curves of the cosubstrates are shown in Figure 1. The model simulations for all conditions can be consulted in the Supplementary Material (Figures S1 to S8, available with the online version of this paper). In general, the determination coefficient (r2) for all the curves was always above 0.98. Values of r2 as low as 0.93 were obtained when WAS was used as cosubstrate in a proportion of the feeding above 50% as VS. This was caused because, as it is shown in Figures S7 and S8 from the Supplementary Material (available online), there was an underestimation of the initial biogas production rate, therefore the interpretation of the results must be done carefully. The estimated parameters of the Gompertz equation, for all the studied conditions, are presented in Table 2. The standard deviation of the parameters was also estimated, and it is presented in the Supplementary Material (Table S1, available online). Overall, all the values were always below 10% of the average value (coefficient of variation) except for WAS where the values of Rm showed higher error values, which reflects the underestimation of the initial biogas production rate.

Table 2

Kinetic parameter values from the BMP test of the AcoD experiments

Model parameters
P (mlbiogas gVS−1)
Rm (mlbiogas gVS−1 d−1)
λ (d)
Cosubstrate proportion (%)
Cosubstrate proportion (%)
Cosubstrate proportion (%)
010255075100010255075100010255075100
 MSS (substrate) - InC (inoculum) 
Gr 545 641 792 1083 1322 1428 77 86 97 116 124 107 0.8 1.0 1.2 1.8 2.4 2.9 
DW 597 595 604 616 683 692 108 110 113 101 122 129 0.8 0.6 0.5 0.1 0.0 0.0 
SGr 543 563 614 662 735 753 104 105 114 121 124 129 0.5 0.5 0.6 0.7 0.8 0.8 
WAS 556 538 485 425 346 276 116 101 109 86 67 43 0.3 0.2 0.2 
 TPS (substrate) – InA (inoculum) 
Gr 540 649 782 1089 1341 1521 92 92 88 100 95 64 0.3 0.4 0.4 0.9 1.7 2.8 
DW 607 629 655 681 703 674 106 110 111 111 118 130 0.5 0.4 0.3 0.0 0.0 0.0 
SGr 547 570 614 713 817 888 92 106 114 124 140 147 0.4 0.3 0.4 0.5 0.6 0.6 
WAS 593 572 537 492 459 397 75 72 66 55 42 27 
Model parameters
P (mlbiogas gVS−1)
Rm (mlbiogas gVS−1 d−1)
λ (d)
Cosubstrate proportion (%)
Cosubstrate proportion (%)
Cosubstrate proportion (%)
010255075100010255075100010255075100
 MSS (substrate) - InC (inoculum) 
Gr 545 641 792 1083 1322 1428 77 86 97 116 124 107 0.8 1.0 1.2 1.8 2.4 2.9 
DW 597 595 604 616 683 692 108 110 113 101 122 129 0.8 0.6 0.5 0.1 0.0 0.0 
SGr 543 563 614 662 735 753 104 105 114 121 124 129 0.5 0.5 0.6 0.7 0.8 0.8 
WAS 556 538 485 425 346 276 116 101 109 86 67 43 0.3 0.2 0.2 
 TPS (substrate) – InA (inoculum) 
Gr 540 649 782 1089 1341 1521 92 92 88 100 95 64 0.3 0.4 0.4 0.9 1.7 2.8 
DW 607 629 655 681 703 674 106 110 111 111 118 130 0.5 0.4 0.3 0.0 0.0 0.0 
SGr 547 570 614 713 817 888 92 106 114 124 140 147 0.4 0.3 0.4 0.5 0.6 0.6 
WAS 593 572 537 492 459 397 75 72 66 55 42 27 
Figure 1

Accumulative biogas production of the BMP test using MSS (for In C) and TPS (for inA) as the main substrates. Co-substrates: GR (a and b) DW (c and d) GrS (e and f) and WAS (g and h). Co-substrate proportion: 0% red, 10% green, 25% purple, 50% light blue, 75% orange, 100% dark blue. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/wst.2019.292.

Figure 1

Accumulative biogas production of the BMP test using MSS (for In C) and TPS (for inA) as the main substrates. Co-substrates: GR (a and b) DW (c and d) GrS (e and f) and WAS (g and h). Co-substrate proportion: 0% red, 10% green, 25% purple, 50% light blue, 75% orange, 100% dark blue. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/wst.2019.292.

As expected, Gr showed the highest P value among the evaluated cosubstrates due to its lipid content, such that if the goal of implementing co-digestion is to increase the biogas production, this would be the best cosubstrate. However, the feeding of Gr may lead to some operational problems due to mass transfer problems, inhibition and foaming, so pilot studies should be first carried out (Long et al. 2012). On the other hand, DW and SGr showed the highest values of Rm, which means that both cosubstrates may be treated with the same or even lower (depending on the presence of synergy) residence time. The highest values for the lag phase were found for Gr for both types of inoculums, which demonstrated the difficulties encountered by the microorganism to access this cosubstrate and start the biodegradation process. The literature comparison of the results was carried out only with studies where the same model was applied to fit the data. For DW, Zahan et al. (2018), who used yoghurt whey, obtained similar values for P, but far lower values of Rm probably due to the pH of the tests. No lag phase was observed. In regard to the parameter for SGr, the values of P are similar to those obtained for similar substrates such as grease trap waste in Yalcinkaya & Malina (2015). In that study, the maximum productivity almost doubled the value obtained in this study. The lag phase, in turn, was four times greater than the one obtained in this study. For other lipid-rich residues, such as nut residues, the parameter values obtained here were higher (Shen et al. 2018).

The synergy determination results of the AcoD test are shown in Table 3. The major impact of the cosubstrate addition lies in the biogas production kinetic (Rm) over the biogas potential (P). Only Gr at 50% of cosubstrate fraction yielded some significant positive synergy for the biogas potential when InC was used. It is worth pointing out that synergy percentage values (parameter Si in Equation (2)) of at least 10% are considered as significant. Regarding Rm, there were significant positive synergies with Gr with both inoculums. In general, the addition of cosubstrates had a positive impact in reducing the lag phase for all cosubstrates with both inoculums, although it is worth stressing that the importance of this parameter is limited to a batch operation and enzymatic reactions, which does not apply to continuous reactors where the cosubstrate is applied on a regular basis. It may become relevant when using unacclimated biomass or to the start-up of a continuous reactor. The proposed methodology includes the Gompertz equation fitting process in the estimation of the synergy of the AcoD. The main differences to other approaches are, first, that the effect of cosubstrate addition in terms of synergy can be assessed using three parameters instead of only the experimental observation of the biogas yield and, second, the direct use of the Gompertz equation to draw the parameter values. Ebner et al. (2016) used the first order equation to draw the hydrolytic coefficient and the extent of degradation from the AcoD BMP test. To estimate the presence of synergy, they defined two parameters, the co-digestion performance index (CPI) and co-digestion rate index (CRI); however, their determination required a mathematical analysis or numerical estimation, which is not as straightforward as the one presented in this study.

Table 3

Synergy percentage of the AcoD test

P
Rm
λ
Cosubstrate proportion (%)
102550751025507510255075
 MSS (substrate) - InC (inoculum) 
Gr +1 +3 +10 +9 +7 +15 +26 +25 −1 +12 +4 
DW −2 −2 −4 −14 −2 +10 +29 +215 >1000 
SGr +3 +2 +5 −2 +3 +3 +1 +6 −1 −2 −4 
WAS +2 +2 −7 +12 +8 +9 +9 −8 >1000 >1000 
 TPS (substrate) – InA (inoculum) 
Gr +2 +6 +5 +2 +3 +28 +33 +32 +151 +80 +26 
DW +3 +5 +6 +7 +1 −5 −4 +61 +100 >1000 >1000 
SGr +2 −1 −3 −2 +8 +8 +4 +5 +34 +22 +3 +1 
WAS −1 −1 +3 −11 12 +24 +32 +9 +42 +90 >1000 
P
Rm
λ
Cosubstrate proportion (%)
102550751025507510255075
 MSS (substrate) - InC (inoculum) 
Gr +1 +3 +10 +9 +7 +15 +26 +25 −1 +12 +4 
DW −2 −2 −4 −14 −2 +10 +29 +215 >1000 
SGr +3 +2 +5 −2 +3 +3 +1 +6 −1 −2 −4 
WAS +2 +2 −7 +12 +8 +9 +9 −8 >1000 >1000 
 TPS (substrate) – InA (inoculum) 
Gr +2 +6 +5 +2 +3 +28 +33 +32 +151 +80 +26 
DW +3 +5 +6 +7 +1 −5 −4 +61 +100 >1000 >1000 
SGr +2 −1 −3 −2 +8 +8 +4 +5 +34 +22 +3 +1 
WAS −1 −1 +3 −11 12 +24 +32 +9 +42 +90 >1000 

Inoculum activity assessment

BMP test of the control test

The cumulative biogas production curves of the control tests are shown in Figure 2 and the kinetic parameters are shown in Table 4. The standard deviation of the parameters is presented in Table S2 in the Supplementary Material (available online). It can be seen that the P values were relatively constant throughout the experimental runs for both inoculums, with similar averages for both: 667 ± 34 ml gVS−1 and 655 ± 36 ml gVS−1, respectively. On the contrary, the results are quite different in regard to Rm; from the initial value (M1) there was a significant increase (double) for both inoculums. After that, the values remained fairly constant. With respect to the values themselves, Rm for InC is between 75% and 90% greater that the values for InA, which shows the hydrolytic capacity of the inoculum that comes from a digester fed with MSS in comparison to the inoculum that deals with a substrate that has a higher fraction of soluble organic matter. Likewise, the 30% difference between the values of the lag phase for the inoculum confirmed what is abovementioned regarding the hydrolytic capacity. This control test allows us not only to ensure the activity of the inoculum of the BMP test analysis but also to analyze the current operation of the digester from which the anaerobic biomass is taken.

Table 4

Kinetic parameter values from the control test (chopped paper used as substrate)

InC
InA
BMP testP (ml gVS−1)Rm (ml gVS−1 d−1)λ(d)P (ml gVS−1)Rm (ml gVS−1 d−1)λ(d)
M1 690 69 2.3 626 34 2.6 
M2 673 115 1.8 656 72 2.1 
M3 618 110 2.0 632 68 2.2 
M4 688 123 1.5 704 64 3.0 
InC
InA
BMP testP (ml gVS−1)Rm (ml gVS−1 d−1)λ(d)P (ml gVS−1)Rm (ml gVS−1 d−1)λ(d)
M1 690 69 2.3 626 34 2.6 
M2 673 115 1.8 656 72 2.1 
M3 618 110 2.0 632 68 2.2 
M4 688 123 1.5 704 64 3.0 
Figure 2

Accumulative biogas production of the control BMP test. M1, green, M2, blue, M3, red, M4, black. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/wst.2019.292.

Figure 2

Accumulative biogas production of the control BMP test. M1, green, M2, blue, M3, red, M4, black. The full color version of this figure is available in the online version of this paper, at http://dx.doi.org/10.2166/wst.2019.292.

Standardization of the BMP test by inoculum activity

The incorporation of the inoculum activity results into the BMP analysis by parameter estimation is presented in Table 5 and the values of the kinetic parameters (P, Rm, λ) are normalized by the values obtained for the inoculum activity test (Table 4). Moreover, the values of the parameters from a traditional analysis are also included in Table 5.

Table 5

Conventional and normalized parameters taking the results of the control test into account

BMP testPa (ml gVS−1)NMPPRma (ml gVS−1 d−1)NMPRmλa (d)NMPλ
 MSS-InC 
M1 545 0.79 ± 0.02 77 1.11 ± 0.05 0.8 0.33 ± 0.14 
M2 597 0.88 ± 0.02 108 0.94 ± 0.09 0.8 0.44 ± 0.18 
M3 543 0.88 ± 0.01 104 0.94 ± 0.07 0.5 0.26 ± 0.26 
M4 556 0.81 ± 0.02 116 0.94 ± 0.13 0.3 0.18 ± 0.68 
 TPS-InA 
M1 540 0.86 ± 0.04 92 2.73 ± 0.07 0.3 0.12 ± 0.29 
M2 607 0.92 ± 0.02 106 1.46 ± 0.06 0.5 0.25 ± 0.18 
M3 547 0.86 ± 0.02 92 1.35 ± 0.02 0.4 0.19 ± 0.24 
M4 593 0.84 ± 0.01 75 1.17 ± 0.08 
BMP testPa (ml gVS−1)NMPPRma (ml gVS−1 d−1)NMPRmλa (d)NMPλ
 MSS-InC 
M1 545 0.79 ± 0.02 77 1.11 ± 0.05 0.8 0.33 ± 0.14 
M2 597 0.88 ± 0.02 108 0.94 ± 0.09 0.8 0.44 ± 0.18 
M3 543 0.88 ± 0.01 104 0.94 ± 0.07 0.5 0.26 ± 0.26 
M4 556 0.81 ± 0.02 116 0.94 ± 0.13 0.3 0.18 ± 0.68 
 TPS-InA 
M1 540 0.86 ± 0.04 92 2.73 ± 0.07 0.3 0.12 ± 0.29 
M2 607 0.92 ± 0.02 106 1.46 ± 0.06 0.5 0.25 ± 0.18 
M3 547 0.86 ± 0.02 92 1.35 ± 0.02 0.4 0.19 ± 0.24 
M4 593 0.84 ± 0.01 75 1.17 ± 0.08 

aValues taken from Table 2 where no cosubstrates were added to the test.

Overall, the P values follow the same trend for both inoculums under a traditional analysis (Table 2). The most staggering result is the difference for Rm. In the traditional analysis, a significant increase from M1 to M2 is observed, which would indicate a possible change in the sludge properties with the presence of a more soluble fraction, which is also indicated in the CODs analysis. However, when the values are normalized by the control test parameter, on the contrary, in the case of InC, a slight decrease in Rm is observed, which agrees with the slight decrease in the total COD measured. In the case of InA, with the traditional comparison from M1 to M2, a small increase in Rm is observed, which is totally the opposite to the large drop observed in the parameter after normalization. The result from the conventional analysis indicates a significant change in the soluble fraction of the TPS, which is known by the presence of soluble recalcitrant compounds formed during the thermo-hydrolysis reaction (Carrère et al. 2010). However, after normalizing those values for the inoculum activity parameters, the situation is the opposite since a decrease from M1 to M2 (not as sharp as the increase though) was observed. In regard to λ, the large variability of the NMP makes it difficult to draw any reliable conclusion from the analysis. This methodology allows us to better discard the actual influence of the inoculum activity and isolate the influence of the substrate alone.

It is worth pointing out that other control substrates such as microcrystalline cellulose can also be employed for standardization of the inoculum activity. In fact, the biogas production curves are quite similar between that substrate and the one presented in this study (Díaz et al. 2011).

CONCLUSIONS

In this study, two methodologies aiming to improve the data processing in anaerobic digestion studies were proposed. The first one addresses the estimation of synergy in anaerobic co-digestion of organic waste and sewage sludge by using an integrated methodology based on the BMP test, Gompertz equation fitting and comparison of parameter values. The impact of four cosubstrates and two types of sewage sludge (substrates) were evaluated. In regard to the synergy estimation, dairy whey and grease sludge had an impact on the degradation kinetic, although only dairy whey decreased the lag phase compared to mono-digestion. The other methodology comprises the standardization of the BMP test results by taking into account the activity of the inoculums. This is carried out by normalizing the parameters obtained from the Gompertz equation fit from the BMP of the substrates with the parameter values obtained from the control test, where chopped paper was used as the substrate. The results indicate that the assessment can be significantly altered by taking this into account and gives a more realistic picture of the evaluation. This can be meaningful to compare results among studies when different inoculums are used or even for studies where the same inoculum is used but it is taken at different reactor operational moments.

REFERENCES

REFERENCES
Angelidaki
I.
,
Alves
M.
,
Bolzonella
D.
,
Borzacconi
L.
,
Campos
J. L.
,
Guwy
a. J.
,
Kalyuzhnyi
S.
,
Jenicek
P.
&
van Lier
J. B.
2009
Defining the biomethane potential (BMP) of solid organic wastes and energy crops: a proposed protocol for batch assays
.
Water Sci. Technol.
59
,
927
934
.
APHA (American Public Health Association)
2012
Standard Methods for the Examination of Water and Wastewater
, 22nd edn.
American Water Works Association, American Public Heath Association, Water Environment Federation
,
Washington, DC, USA
.
Astals
S.
,
Batstone
D. J.
,
Mata-Alvarez
J.
&
Jensen
P. D.
2014
Identification of synergistic impacts during anaerobic co-digestion of organic wastes
.
Bioresour. Technol.
169
,
421
427
.
Carrère
H.
,
Dumas
C.
,
Battimelli
A.
,
Batstone
D. J.
,
Delgenès
J. P.
,
Steyer
J. P.
&
Ferrer
I.
2010
Pretreatment methods to improve sludge anaerobic degradability: a review
.
J. Hazard. Mater.
183
(
1–3
),
1
15
.
Díaz
I.
,
Donoso-Bravo
A.
&
Fdz-Polanco
M.
2011
Effect of microaerobic conditions on the degradation kinetics of cellulose
.
Bioresour Technol.
102
,
10139
10142
.
Ebner
J. H.
,
Labatut
R. A.
,
Lodge
J. S.
,
Williamson
A. A.
&
Trabold
T. A.
2016
Anaerobic co-digestion of commercial food waste and dairy manure: characterizing biochemical parameters and synergistic effects
.
Waste Manag.
52
,
286
294
.
Holliger
C.
,
Alves
M.
,
Andrade
D.
,
Angelidaki
I.
,
Astals
S.
,
Baier
U.
,
Bougrier
C.
,
Buffière
P.
,
Carballa
M.
,
De Wilde
V.
,
Ebertseder
F.
,
Fernández
B.
,
Ficara
E.
,
Fotidis
I.
,
Frigon
J. C.
,
De Laclos
H. F.
,
Ghasimi
D. S. M.
,
Hack
G.
,
Hartel
M.
,
Heerenklage
J.
,
Horvath
I. S.
,
Jenicek
P.
,
Koch
K.
,
Krautwald
J.
,
Lizasoain
J.
,
Liu
J.
,
Mosberger
L.
,
Nistor
M.
,
Oechsner
H.
,
Oliveira
J. V.
,
Paterson
M.
,
Pauss
A.
,
Pommier
S.
,
Porqueddu
I.
,
Raposo
F.
,
Ribeiro
T.
,
Pfund
F. R.
,
Strömberg
S.
,
Torrijos
M.
,
Van Eekert
M.
,
Van Lier
J.
,
Wedwitschka
H.
&
Wierinck
I.
2016
Towards a standardization of biomethane potential tests
.
Water Sci. Technol.
74
,
2515
2522
.
Kim
M.
,
Chowdhury
M. M. I.
,
Nakhla
G.
&
Keleman
M.
2017
Synergism of co-digestion of food wastes with municipal wastewater treatment biosolids
.
Waste Manag.
61
,
473
483
.
Labatut
R. a.
,
Angenent
L. T.
&
Scott
N. R.
2011
Biochemical methane potential and biodegradability of complex organic substrates
.
Bioresour. Technol.
102
,
2255
2264
.
Long
J. H.
,
Aziz
T. N.
,
Reyes
F. L. D. L.
&
Ducoste
J. J.
2012
Anaerobic co-digestion of fat, oil, and grease (FOG): a review of gas production and process limitations
.
Process Saf. Environ. Prot.
90
,
231
245
.
Mata-Alvarez
J.
,
Dosta
J.
,
Macé
S.
&
Astals
S.
2011
Codigestion of solid wastes: a review of its uses and perspectives including modeling
.
Crit. Rev. Biotechnol.
31
,
99
111
.
Raposo
F.
,
Fernández-Cegrí
V.
,
de la Rubia
M. a.
,
Borja
R.
,
Béline
F.
,
Cavinato
C.
,
Demirer
G.
,
Fernández
B.
,
Fernández-Polanco
M.
,
Frigon
J. C.
,
Ganesh
R.
,
Kaparaju
P.
,
Koubova
J.
,
Méndez
R.
,
Menin
G.
,
Peene
A.
,
Scherer
P.
,
Torrijos
M.
,
Uellendahl
H.
,
Wierinck
I.
&
de Wilde
V.
2011
Biochemical methane potential (BMP) of solid organic substrates: evaluation of anaerobic biodegradability using data from an international interlaboratory study
.
J. Chem. Technol. Biotechnol.
86
,
1088
1098
.
Steinmetz
R. L. R.
,
Mezzari
M. P.
,
da Silva
M. L. B.
,
Kunz
A.
,
do Amaral
A. C.
,
Tápparo
D. C.
&
Soares
H. M.
2016
Enrichment and acclimation of an anaerobic mesophilic microorganism's inoculum for standardization of BMP assays
.
Bioresour. Technol.
219
,
21
28
.
WEF
2018
Water Environment Federation. Resource recovery data. Retrieved on 9/6/2018 from http://www.resourcerecoverydata.org/biogasdata.php
.
Wett
B.
,
Takács
I.
,
Batstone
D.
,
Wilson
C.
&
Murthy
S.
2014
Anaerobic model for high-solids or high-temperature digestion – additional pathway of acetate oxidation
.
Water Sci. Technol.
69
,
1634
1640
.
Xie
S.
,
Wickham
R.
&
Nghiem
L. D.
2017
Synergistic effect from anaerobic co-digestion of sewage sludge and organic wastes
.
Int. Biodeterior. Biodegradation
116
,
191
197
.