Many digesters in Germany are not operated at full capacity; this offers the opportunity for co-digestion. Within this research the potentials and limits of a flexible and adapted sludge treatment are examined with a focus on the digestion process with added food waste as co-substrate. In parallel, energy data from a municipal wastewater treatment plant (WWTP) are analysed and lab-scale semi-continuous and batch digestion tests are conducted. Within the digestion tests, the ratio of sewage sludge to co-substrate was varied. The final methane yields show the high potential of food waste: the higher the amount of food waste the higher the final yield. However, the conversion rates directly after charging demonstrate better results by charging 10% food waste instead of 20%. Finally, these results are merged with the energy data from the WWTP. As an illustration, the load required to cover base loads as well as peak loads for typical daily variations of the plant's energy demand are calculated. It was found that 735 m³ raw sludge and 73 m³ of a mixture of raw sludge and food waste is required to cover 100% of the base load and 95% of the peak load.

INTRODUCTION

Anaerobic digestion of sewage sludge continuously provides biogas that is typically converted to heat and electricity via a combined heat and power plant (CHP) at wastewater treatment plants (WWTPs). In general, the plant's own heat requirements can be covered; however, the electricity demand, especially the peak in demand, cannot be covered without the extension of system boundaries and/or the use of external (often renewable) energy sources (Nowak et al. 2011; Chae & Kang 2013). Considering that many digesters in Germany are currently not operated at full capacity (Krupp et al. 2005; Roos 2008), easily degradable material is increasingly used as additional input material to obtain more biogas, i.e. electricity and heat (Weiland 2000).

General guidelines for implementing co-digestion in the operation of WWTPs recommend steady operation of the digester, for example, constant feeding of co-substrate (DWA 2009). By integrating co-digestion in a more flexible manner with regard to the actual energy demand, the overall efficiency of the WWTP's energy management might be improved (Hahn et al. 2014). Therefore, knowledge about composition, degradability and gas production rates of the used material (raw sludge vs. co-substrate), as well as limits of shock loads and inhibitory influences, is indispensable.

In general food waste is high in organics, therefore suitable for anaerobic digestion (Chen et al. 2008; Nayono 2009). The composition of food waste can vary regionally and seasonally and depends strongly on its origin (Zhang et al. 2007; Iacovidou et al. 2012). Food waste from industrial sources varies significantly according to the processed food (e.g. dairy industry). Mixed food waste from canteens, restaurants, and kitchens is generally high in carbohydrates (55–78% of total solids (TS)), followed by proteins (15–21%), and fats (5–22%) (Iacovidou et al. 2012).

The main benefit of food waste is its high degradability and methane yield, compared to sewage sludge (Iacovidou et al. 2012; Mata-Alvarez et al. 2014). Many studies have been performed to determine these values for various food wastes and substrate ratios (e.g., Cecchi et al. 1988; Stroot et al. 2001; Heo et al. 2003; Kim et al. 2003; Sosnowski et al. 2008; Koch et al. 2015). Specific values for the degree of degradability and amount of gas production depend on its specific composition/origin as well as the operational design (batch vs. continuous experiments, hydraulic retention time (HRT), temperature). Additionally the reference value of the methane yield is often unspecified; therefore, it is difficult to compare the results directly. Nevertheless, by increasing the ratio of food waste to sewage sludge, the overall degradability as well as the methane yield is likely to increase as long as no inhibition occurs, as shown in Table 1.

Table 1

Methane yields and degradation rates for several mixtures of sewage sludge and food waste under mesophilic conditions

Substrate Ratio
 
Methane yield
 
Degradability Reference 
(base) l/kg TVS Degraded/added % TVS 
WAS/FW 90/10 (TVS) 194 added 39 Heo et al. (2003)  
70/30 230 44 
50/50 339 56 
RS/FW 100/0 (mass) 310 unspecified not determined Koch et al. (2015)  
90/10 350 
0/100 450 
RS/FW 100/0 (TVS) 116 unspecified not determined Kim et al. (2003)  
80/20 157 
50/50 215 
20/80 257 
RS/OFMSW 100/0 (Vol.) 318 degraded not determined Sosnowski et al. (2008)  
75/25 439 
0/100 234 
Substrate Ratio
 
Methane yield
 
Degradability Reference 
(base) l/kg TVS Degraded/added % TVS 
WAS/FW 90/10 (TVS) 194 added 39 Heo et al. (2003)  
70/30 230 44 
50/50 339 56 
RS/FW 100/0 (mass) 310 unspecified not determined Koch et al. (2015)  
90/10 350 
0/100 450 
RS/FW 100/0 (TVS) 116 unspecified not determined Kim et al. (2003)  
80/20 157 
50/50 215 
20/80 257 
RS/OFMSW 100/0 (Vol.) 318 degraded not determined Sosnowski et al. (2008)  
75/25 439 
0/100 234 

FW: food waste.

OFMSW: organic fraction of municipal solid waste.

RS: raw sludge.

WAS: waste-activated sludge.

Implementing co-digestion in a flexible manner requires the correct timing of methane production; therefore, the conversion rates of sludge and food waste are of interest, i.e., the rate-limiting step of their digestion. In general, the rate-limiting step of the digestion process is determined decisively by the composition: when large amounts of dissolved/hydrolysed organic material are present, the acetogenic, as well as the acetate utilizing methanogenic phase is limiting. During digestion of more complex material with a high solids content, the hydrolysis is likely to be the rate-limiting step (Noike et al. 1985; Roediger et al. 1990; Bischofsberger et al. 2005). If material with a fast hydrolysis rate is used, the level of volatile fatty acids (VFAs) can increase rapidly (Zoetemeyer et al. 1982; Noike et al. 1985; Sosnowski et al. 2008; Yang et al. 2015); under stable conditions, VFAs subsequently decrease, due to the degradation process (Bischofsberger et al. 2005). However, when VFAs accumulate too fast (high organic loading rate/large amount of degradable material), the uptake rate of the microorganisms might be reduced; in the worst case, the process failed completely due to acidification (Bischofsberger et al. 2005; Iacovidou et al. 2012).

Besides process instability caused by overloading, inhibitory substances present in the food waste can negatively influence the digestion process. Possible inhibitory substances in the anaerobic digestion process are, for example, ammonia, sulphides, light metal ions, heavy metals, and organic compounds (Chen et al. 2008). Nevertheless, the inhibition is a question of concentration, adaptation and acclimation, as well as complex mechanisms such as synergism and antagonism (Kugelman & Chin 1971). Considering the addition of food waste from the food processing industry (no mixed food waste), the inhibition due to lipids, proteins, and salts are of particular interest: during co-digestion of lipid-rich material, technical problems (e.g., clogging, flotation of biomass) might occur; additionally long-chain fatty acids may accumulate and inhibit the anaerobic digestion process (Cirne et al. 2007; Chen et al. 2008). Protein-rich material is degraded to ammonia/ammonium (depending on the pH value); free ammonia, in particular, has a toxic effect (Chen et al. 2008). Low concentrations of salts can be stimulating, but at high concentrations the effect can be toxic: the cells dehydrate, due to osmotic pressure (McCarty & McKinney 1961; de Baere et al. 1984). On the other hand, inhibitors present in sewage sludge (e.g., heavy metals, pharmaceuticals, pathogens) can be diluted by the addition of co-substrate (Mata-Alvarez et al. 2014).

MATERIALS AND METHODS

In order to determine the options (or the limits and potential) of a flexible and target-oriented digester operation, a municipal WWTP is examined. The focus of the examination is on dosage strategies, with and without co-substrate loads, which are adapted to the power load of the municipal WWTP. The examined municipal WWTP has a total capacity of 240,000 population equivalents and operates three digesters with a total volume of 12,000 m³ at mesophilic conditions.

The examination is subdivided into two parts:

  • In order to examine the influence of co-substrate and different loading rates, laboratory-scale semi-continuous as well as batch digestion tests are conducted at the Technische Universität Darmstadt.

  • Energy consumption data from the municipal WWTP are analysed to determine typical diurnal variations of power loads.

The municipal WWTP provides input material for digestion tests as well as data for the theoretical analysis. Experimental data and energy data from the WWTP have been merged to discuss the opportunities of flexible operation (Figure 1).
Figure 1

Scheme of the examination procedure.

Figure 1

Scheme of the examination procedure.

Experimental study

The reactors used for semi-continuous digestion tests have a volume of 15 litres, are stirred and equipped with manometer, gas meter and overflow. For the batch test, 1-litre bottles with pressure heads (Oxitop, WTW GmbH) are used. During the examination, the reactors and bottles are located in a climatic chamber to ensure mesophilic operation at 37°C. Raw sewage sludge (RS: mixture of primary sludge and waste-activated sludge) from the municipal WWTP and food waste (FW: collected and thermally processed organic waste from gastronomy, industry, and trade) are used as input material. For the batch tests, digested sewage sludge (DS) from the reference digester of the semi-continuous tests is additionally used. The input materials are analysed in terms of TS, total volatile solids (TVS) and chemical oxygen demand (COD) (Table 2). Furthermore, organic acids of dissolved solids are measured regularly with cuvette tests from the Hach Company. One dried sample of FW and RS is additionally analysed according to the Weende analysis (see Figure 2).
Table 2

Composition of input material

  Method   Semi-continuous tests*
 
Batch tests
 
  RS FW DS RS FW 
TS DIN EN 12880 (2001) 4.2 19.0 3.0 5.2 18.5 
TVS DIN EN 15935 (2012) % TS 72.3 89.1 57.7 63.9 86.7 
COD DIN 38414 – part 9 (1986) via potassium dichromate mg/g TS 1,130 1,706 870 1,041 1,757 
mg/g TVS 1,563 1,914 1,508 1,629 2,027 
  Method   Semi-continuous tests*
 
Batch tests
 
  RS FW DS RS FW 
TS DIN EN 12880 (2001) 4.2 19.0 3.0 5.2 18.5 
TVS DIN EN 15935 (2012) % TS 72.3 89.1 57.7 63.9 86.7 
COD DIN 38414 – part 9 (1986) via potassium dichromate mg/g TS 1,130 1,706 870 1,041 1,757 
mg/g TVS 1,563 1,914 1,508 1,629 2,027 

*Average values during total testing period.

Figure 2

Composition of RS and FW according to the Weende analysis (Fischer & Glomb 2015).

Figure 2

Composition of RS and FW according to the Weende analysis (Fischer & Glomb 2015).

For the semi-continuous digestion tests, the proportion of RS to FW has been varied within three test series with ratios (Vol.-% RS/Vol.-% FW) of 100/0 (#C1), 90/10 (#C2) and 80/20 (#C3) for 6–8 weeks (Table 3). Throughout the test series, one digester is used as reference digester and charged only with RS. Both reactors are charged once a day (‘shock loaded’) with an average HRT of 20 days. The gas production has been measured by drum-type gas meters and converted to standard conditions (273 K and 101,325 Pa) and the gas quality (Vol.-% of CH4 and CO2) by gas chromatography. The digested sludge has been analysed in accordance with the input material. For the detailed examination of shock loads, one random day within each testing period has been chosen; gas production and quality has been recorded and measured every 30 minutes, starting at loading time.

Table 3

Characteristics of test series

    #C1 #C2 #C3   
Duration 56 49 42  
HRT 19.8 20.1 20.0  
Amount of RS Vol.-% 100 90 80  
FW Vol.-% – 10 20  
Organic loading rate (OLR) kg TVS/(m³·d) 1.6 1.9 2.9  
kg COD/(m³·d) 2.4 3.4 5.2  
 #B1 #B2 #B3 #B4 
Duration 22 22 22 22 
Amount of RS Vol.-% 100 90 80 – 
FW Vol.-% – 10 20 100 
    #C1 #C2 #C3   
Duration 56 49 42  
HRT 19.8 20.1 20.0  
Amount of RS Vol.-% 100 90 80  
FW Vol.-% – 10 20  
Organic loading rate (OLR) kg TVS/(m³·d) 1.6 1.9 2.9  
kg COD/(m³·d) 2.4 3.4 5.2  
 #B1 #B2 #B3 #B4 
Duration 22 22 22 22 
Amount of RS Vol.-% 100 90 80 – 
FW Vol.-% – 10 20 100 

The batch tests are conducted in accordance with the guideline VDI 4360 (Verein Deutscher Ingenieure 2006). The ratio of RS to FW is varied from 100 Vol.-% of RS to 100 Vol.-% of FW (see Table 3) in four series running in parallel and in triplicate. #B1, #B2 and #B3 comply with the test series of the semi-continuous digestion tests. In accordance with the semi-continuous digestion tests, gas production is recorded regularly and the final gas quality is analysed by gas chromatography.

Data analysis

For the analysis of the energy consumption of the municipal WWTP, data from the year 2013 are examined to define a typical diurnal variation of the power load. The total electrical energy consumption of the WWTP is calculated via the electricity purchase from the provider and the generated electricity of the CHP; from time to time, the plant feeds electricity into the public grid and this has to be deducted: 
formula
These data are only available in monthly intervals. However, data on the electricity purchase from the provider are available in 15 min intervals and therefore used for the determination of diurnal variations, under the assumption that the amount of generated electricity is constant (due to a constant/steady operation of the digester); data are converted to 2 h data; weekends and wet weather conditions (>0 mm on the same day and >1 mm on the previous day) are excluded.

RESULTS AND DISCUSSION

Results of experimental study

Within the three test series of the semi-continuous digestion tests, the shock loads show similar effects, although to varying degrees. In general, examination of the gas quality after intermittent charge reveals a visible drop in methane concentration and peak of carbon dioxide concentration, respectively, within the first 9 hours (Figure 3). Without co-substrate addition (#C1), the methane concentration decreases slightly, by one percentage point, within the first two hours. When charging with co-substrate, this drop is more significant and, the larger the co-substrate addition, the stronger the drop.
Figure 3

Development of the CH4 and CO2 concentrations after charging (semi-continuous test).

Figure 3

Development of the CH4 and CO2 concentrations after charging (semi-continuous test).

The observed drop in gas quality immediately after charging (especially #C2 and #C3) can be explained by the rapid acidification of easily degradable material within the co-substrate: during hydrolysis and acidogenesis predominantly H2 and CO2, as well as acetic acids, are produced; due to high acetate values, acetate-utilizing methanogenesis is favoured, whereby CO2 and CH4 are produced in equal amounts (Bischofsberger et al. 2005). Sun et al. (2013) observed a similar drop of gas quality accompanying increasing VFA and soluble COD levels in the first two hours after charging with a mixture of WAS and FW.

The basic trend of the average gas composition of the semi-continuous digestion tests complies with the values of the batch digestion tests (see Table 4): the average methane concentrations are significantly lower than the concentrations after 24 hours (semi-continuous tests) and 21 days (batch tests), respectively. In the case of the average values, it is relevant that the values with FW are higher than without; however, the ratio of FW is not decisive. In terms of the values 24 hours after charging (semi-continuous tests) and the values after 21 days (batch tests), the methane concentration increases with increasing FW addition.

Table 4

Average and final methane concentrations and yields of semi-continuous and batch digestion tests

RS/FW ratio   100/0 90/10 80/20 0/100 
  #C1 #C2 #C3  
Average CH4 concentration Vol.-% 61.1 64.1 64.8  
CH4 concentration after 24 h Vol.-% 63.2 67.0 71.0  
Average CH4 yield l/kg TVSadded 225 387 404  
CH4 yield after 24 h l/kg TVSadded 165 322 353  
  #B1 #B2 #B3 #B4 
Average CH4 concentration Vol.-% 60.8 62.6 62.6 62.2 
Final CH4 concentration Vol.-% 65.0 66.5 68.0 68.9 
Final CH4 yield l/kg TVSadded 234 355 424 534 
RS/FW ratio   100/0 90/10 80/20 0/100 
  #C1 #C2 #C3  
Average CH4 concentration Vol.-% 61.1 64.1 64.8  
CH4 concentration after 24 h Vol.-% 63.2 67.0 71.0  
Average CH4 yield l/kg TVSadded 225 387 404  
CH4 yield after 24 h l/kg TVSadded 165 322 353  
  #B1 #B2 #B3 #B4 
Average CH4 concentration Vol.-% 60.8 62.6 62.6 62.2 
Final CH4 concentration Vol.-% 65.0 66.5 68.0 68.9 
Final CH4 yield l/kg TVSadded 234 355 424 534 

The used FW contains about 20% of ether extracts, i.e., lipids/fats (see Figure 2). This is not unusual in terms of the general composition of FW, although relatively high (Iacovidou et al. (2012): 5–22%). The theoretical stoichiometric conversion of lipids produces biogas with a high methane content (Lensch et al. 2013). Furthermore, lipids are highly degradable; however, their hydrolysis takes time (Bischofsberger et al. 2005). For long retention times or during waiting after charging, a higher potential can be utilized; this means that higher values can be obtained.

In contrast to the methane concentration, the gas production increases immediately after charging (Figure 4): the cumulated specific gas production of #C2 and #C3 (with FW addition) is significantly higher than without (#C1). Within the first 10 hours there is no significant difference between #C2 and #C3 but, subsequently, the slope of #C3 exceeds that of #C2, resulting in a final gas production of approximately 600 l/(kg TVSadded). The values from the batch tests (Figure 5) match these observations: final gas productions are higher as more FW is added. Approximately 380 l/(kg TVSadded) is produced without FW (#B1) and 850 l/(kg TVSadded) with (#B4).
Figure 4

Development of the cumulated specific gas production after charging (semi-continuous test).

Figure 4

Development of the cumulated specific gas production after charging (semi-continuous test).

Figure 5

Development of the cumulated specific gas production (batch test).

Figure 5

Development of the cumulated specific gas production (batch test).

As stated above, FW contains, on the one hand, easily degradable material, resulting in high gas (CO2 and CH4) productions directly after charging, although of low quality (see above). On the other hand, the lipid content takes effect after about 10 hours (semi-continuous tests) and 3 days (batch tests) after charging, respectively.

When calculating the resulting specific methane production (superposition of both effects), the drop in gas quality becomes less significant, but still visible (Figure 6): not until 17 to 18 hours after charging is the specific methane production of #C2 higher than that of #C3. Twenty-four hours after charging, the specific methane production reaches values of 353 l/kg TVSadded (#C3) and 322 l/kg TVSadded (#C2). The reference digester, on the other hand, produces approximately 165 l/kg TVSadded within the same time. The basic trend follows the methane yields obtained in the batch tests, as well as the average values of the total operating period of the semi-continuous tests; however, the yields after 24 hours are lower (see Table 4).
Figure 6

Development of the specific methane production after charging (semi-continuous test).

Figure 6

Development of the specific methane production after charging (semi-continuous test).

In general, these values are consistent with values from the literature (see Table 1): the larger the FW addition, the higher the methane yields. The variations between batch and semi-continuous tests are due to varying FW and RS composition (see Table 2). Furthermore, it has to be considered that the value 24 hours after charging is a random sample of one day within the testing period, whereas the average value is the overall value of the testing period. Therefore, the latter value is more representative.

In Figure 7, the energy conversion rates in watt hours per hour (or watt) per kg TVSadded are plotted. These values are calculated in terms of the produced methane over the time (or the slope/change of methane production) and converted to kWh by the calorific value of 9.968 kWh/m³ methane (Brandt 1999). The conversion rate curves can be described as logarithmic functions starting with a high conversion rate within the first hour that then decreases rapidly. As expected, the conversion rates of the semi-continuous test series with co-substrate (#C2 and #C3) show the highest values. One hour after charging, almost 300 to 400 Wh/(kg TVSadded · h) can be provided (neglecting the efficiency factor of the CHP plant). Nevertheless, charging 10% FW (#C2) provides higher values than charging 20% (#C3).
Figure 7

Development of the conversion rate after charging (semi-continuous test).

Figure 7

Development of the conversion rate after charging (semi-continuous test).

The slower start of the methane production of #C3, compared to #C2, is the result of the two effects explained above: rapid hydrolysis of easily degradable material in the FW results in higher VFA levels that decrease the methanogenic activity and increase the risk of acidification. The hydrolysis of lipids also requires more time before its high energy potential can be used. Due to the RS, there is buffer capacity to stabilize the process.

However, observing the total testing period reveals the risk of acidification more clearly: as shown in Figure 8, the organic acid concentration increases rapidly at the end of #C3, to 2,000 mg/l. According to WEF & ASCE (2010), values below 300 mg/l are not critical for the digestion process; DWA (2009) states that, by applying co-substrate, this value can be raised to 500 mg/l. Above 1,000 mg/l, the process stability is at risk and, above 2,000 mg/l, the digester is likely to acidify (ATV 1996; DWA 2009).
Figure 8

Development of the organic acids (semi-continuous test).

Figure 8

Development of the organic acids (semi-continuous test).

Without resting time or without a decrease in the OLR, the digester faces the risk of acidifying. The OLR of the semi-continuous digestion tests, as shown in Table 3, is still within the general range of recommended values (2–5 kgTVS/(m³·d) and 2.5–7.5 kgCOD/(m³·d), depending on plant size, according to Bischofsberger et al. (2005) and Urban & Scheer (2011)). This shows that, when applying co-substrate, recommended values have to be questioned. In particular, when applying lipid-rich material, the process stability has to be observed carefully. Mata-Alvarez et al. (2014) list several studies of sewage sludge co-digestion with fats, oils and greases in which inhibition occurs.

Results of the data analysis of the municipal WWTP

Over the year, the electrical self-consumption of the WWTP is steady at around 630,000 to 720,000 kWhel (Figure 9). The electricity purchases, on the other hand, fluctuate slightly between 170,000 and 315,000 kWhel, depending on the gas production. During the year, about 54% to 76% can be covered by digester gas; surpluses (∼feed-in) are minor (900 to 21,000 kWhel).
Figure 9

Electrical consumption and production of the WWTP during the course of the year 2013.

Figure 9

Electrical consumption and production of the WWTP during the course of the year 2013.

The analysis of the diurnal variations of the electricity purchase shows varying forms during the year. There is no distinct relationship to seasons or weather conditions. However, a direct comparison to the precipitation values reveals (Figure 10) a trend towards higher electricity purchases with higher precipitation and low daily electricity purchases (below 5,000 kWh) can only occur if the precipitation is below 10 mm. In this case, it can be assumed that the higher demand might be due to the electricity consumption of the pumping station and not only to the consumption of the wastewater treatment itself.
Figure 10

Relation of daily precipitation and electricity purchase (reference year 2013).

Figure 10

Relation of daily precipitation and electricity purchase (reference year 2013).

Despite variations in its form, the daily electricity purchase curve of the WWTP shows the following typical diurnal variations: in the morning, the consumption is low, then increases from around 4 a.m. until noon; subsequently the consumption decreases again. Throughout the day, an average of approximately 200 kWh/h is purchased; at peak times, this value increases up to 530 kWh/h. Taking into account maximum values, the peak load can increase up to 980 kWh/h. Considering a steady operation of the CHP the overall electricity consumption (electricity purchase plus generated electricity) shifts 570 kW (in 2013, generated electricity of the CHP ∼ 5 GWh/a) in relation to the electricity purchase. The base load during the day is approximately 770 kWh/h (Figure 11).
Figure 11

Diurnal variation of the overall energy consumption of a municipal WWTP.

Figure 11

Diurnal variation of the overall energy consumption of a municipal WWTP.

Outlook: a target-oriented dosage strategy

Taking into account the specific methane production of RS (225 L/kg TVSadded), the calorific value of methane (9.968 kWh/m³) and the electrical efficiency factor of a CHP plant (ηel = 35%) 785 Whel/kg TVSadded can be generated. In order to cover the base load of the overall consumption 735 m³/d or 31 m³/h is required.

To cover the required energy during peak load the mixture #C2 (10% FW) is chosen due to better/faster conversion rates directly after charging (see Figure 7). With the determined regression curve the required substrate amount is calculated as shown in Figure 12. In total 73 m³ of the mixture can cover the required amount (of the peak load) over the day by 95%. However, this is a theoretical calculation. It has to be proved that the short-term high loading rates do not provoke any inhibition.
Figure 12

Exemplary dosage strategy to manage the peak load by charging a mixture of RS and FW (electrical yield is calculated with ηel = 35%).

Figure 12

Exemplary dosage strategy to manage the peak load by charging a mixture of RS and FW (electrical yield is calculated with ηel = 35%).

CONCLUSIONS

The conducted digestion tests show – as expected – higher values for energy production and energy conversion rates with co-substrate addition than without. The results of the final methane yields (batch digestion tests) and the yields 24 hours after charging (semi-continuous tests) match this theory in particular. However, within the first 10 hours after charging, the addition of more co-substrate (20% FW instead of 10%) does not improve the methane yield. Quite the contrary, the conversion rates of #C2 (10% FW) at the beginning are higher. A possible cause for the delayed methane production is a high lipid content within the FW. An increase in organic acids at the end of #C3 underlines this theory.

By observing the diurnal variation of the energy consumption of a municipal WWTP, base loads and peak loads can be determined. Whereas the base load can be covered by continuously charging sewage sludge (31 m³/h), peak loads offer the opportunity of adding co-substrates. One dosage strategy for peak load requirement is shown. Here, 95% of required electrical energy can be covered.

In general, the examination of the energy conversion rate by charging sewage sludge with the addition of co-substrate shows the potential of a flexible and target-oriented energy production of WWTPs due to the increase in energy efficiency. However, before implementing such flexible dosage strategies the process stability during short-term high loading rates needs to be examined further. It has to be considered that the co-digestion might influence the overall treatment process (dewatering step, back charge due to process water). This includes guaranteeing that the requested discharge quality has to be the primary objective of the WWTP at any time.

REFERENCES

REFERENCES
ATV
1996
ATV-Handbuch ‘Klärschlamm’
.
Ernst & Sohn Verlag
,
Berlin
.
Bischofsberger
W.
Dichtl
N.
Rosenwinkel
K.-H.
Seyfried
C.-F.
Böhnke
B.
2005
Anaerobtechnik
.
Springer Verlag
,
Heidelberg, Berlin
.
Brandt
F.
1999
Brennstoffe und Verbrennungsrechnung
.
Vulkan-Verlag
,
Essen
.
Chen
Y.
Cheng
J. J.
Creamer
K. S.
2008
Inhibition of anaerobic digestion process: a review
.
Bioresource Technology
99
(
10
),
4044
4064
.
Cirne
D. G.
Paloumet
X.
Bjornsson
L.
Alves
M. M.
Mattiasson
B.
2007
Anaerobic digestion of lipid-rich waste – effects of lipid concentration
.
Renewable Energy
32
(
6
),
965
975
.
de Baere
L. A.
Devocht
M.
Van Assche
P.
Verstraete
W.
1984
Influence of high NaCl and NH4Cl salt levels on methanogenic associations
.
Water Research
18
(
5
),
543
548
.
DWA
2009
Co-Vergärung in kommunalen Klärschlammfaulbehältern, Abfallvergärungsanlagen und landwirtschaftlichen Biogasanlagen: Merkblatt DWA
.
DWA
,
Hennef
.
Fischer
M.
Glomb
M. A.
2015
Moderne Lebensmittelchemie
.
Behr's Verlag
,
Hamburg
.
Hahn
H.
Krautkremer
B.
Hartmann
K.
Wachendorf
M.
2014
Review of concepts for a demand-driven biogas supply for flexible power generation
.
Renewable & Sustainable Energy Reviews
29
,
383
393
.
Heo
N.
Park
S.
Lee
J.
Kang
H.
Park
D.
2003
Single-stage anaerobic codigestion for mixture wastes of simulated Korean food waste and waste activated sludge
.
Applied Biochemistry and Biotechnology
107
(
1–3
),
567
579
.
Iacovidou
E.
Ohandja
D. G.
Voulvoulis
N.
2012
Food waste co-digestion with sewage sludge – realising its potential in the UK
.
Journal of Environmental Management
112
,
267
274
.
Kim
H. W.
Han
S. K.
Shin
H. S.
2003
The optimisation of food waste addition as a co-substrate in anaerobic digestion of sewage sludge
.
Waste Management & Research
21
(
6
),
515
526
.
Kugelman
I. J.
Chin
K. K.
1971
Toxicity, synergism, and antagonism in anaerobic waste treatment processes
.
Advances in Chemistry Series
105
(
105
),
55
–90
.
Lensch
D.
Schaum
C.
Cornel
P.
2013
An integrated view of sludge treatment processes: organic matter and its influence
. In:
1st International IWA Conference on Holistic Sludge Management
,
IWA
,
Västerås
,
Sweden
.
Mata-Alvarez
J.
Dosta
J.
Romero-Guza
M. S.
Fonoll
X.
Peces
M.
Astals
S.
2014
A critical review on anaerobic co-digestion achievements between 2010 and 2013
.
Renewable & Sustainable Energy Reviews
36
,
412
427
.
McCarty
P. L.
McKinney
R. E.
1961
Salt toxicity in anaerobic digestion
.
Journal Water Pollution Control Federation
33
(
4
),
399
415
.
Nayono
S. E.
2009
Anaerobic Digestion of Organic Solid Waste for Energy Production. Bauingenieur-, Geo- und Umweltwissenschaften
.
Dissertation
,
Universität Fridericiana zu Karlsruhe
,
Karlsruhe
.
Noike
T.
Endo
G.
Chang
J. E.
Yaguchi
J. I.
Matsumoto
J. I.
1985
Characteristics of carbohydrate degradation and the rate-limiting step in anaerobic-digestion
.
Biotechnology and Bioengineering
27
(
10
),
1482
1489
.
Nowak
O.
Keil
S.
Fimml
C.
2011
Examples of energy self-sufficient municipal nutrient removal plants
.
Water Science and Technology
64
(
1
),
1
6
.
Roediger
H.
Roediger
M.
Kapp
H.
1990
Anaerobe alkalische Schlammfaulung
.
Oldenbourg
,
München, Wien
.
Roos
H.-J.
2008
Perspektiven der Co-Vergärung auf Kläranlagen
.
GWF Wasser Abwasser
149
(
4
),
364
368
.
Sosnowski
P.
Klepacz-Smolka
A.
Kaczorek
K.
Ledakowicz
S.
2008
Kinetic investigations of methane co-fermentation of sewage sludge and organic fraction of municipal solid wastes
.
Bioresource Technology
99
(
13
),
5731
5737
.
Sun
Y. F.
Wang
D.
Qiao
W.
Wang
W.
Zhu
T. L.
2013
Anaerobic co-digestion of municipal biomass wastes and waste activated sludge: dynamic model and material balances
.
Journal of Environmental Sciences – China
25
(
10
),
2112
2122
.
Urban
I.
Scheer
H.
2011
Co-Vergärung in kommunalen Faulbehältern: Ein CSB-basierter Ansatz zur Bestimmung von Freivolumen
.
KA Korrespondenz Abwasser, Abfall
58
(
1
),
56
62
.
Verein Deutscher Ingenieure
2006
Vergärung organischer Stoffe – Substratcharakterisierung, Probenahme, Stoffdatenerhebung, Gärversuche. 4630
.
Beuth Verlag GmbH
,
Berlin
.
WEF & ASCE
2010
Design of Municipal Wastewater Treatment Plants: WEF Manual of Practice No. 8 ASCE Manuals and Reports on Engineering Practice No. 76
.
WEF Press, Alexandria, VA, American Society of Civil Engineers, Environmental and Water Resources Institute, McGraw-Hill
.
Yang
G.
Zhang
P. Y.
Zhang
G. M.
Wang
Y. Y.
Yang
A. Q.
2015
Degradation properties of protein and carbohydrate during sludge anaerobic digestion
.
Bioresource Technology
192
,
126
130
.
Zhang
R. H.
El-Mashad
H. M.
Hartman
K.
Wang
F. Y.
Liu
G. Q.
Choate
C.
Gamble
P.
2007
Characterization of food waste as feedstock for anaerobic digestion
.
Bioresource Technology
98
(
4
),
929
935
.
Zoetemeyer
R. J.
Matthijsen
A. J. C. M.
Cohen
A.
Boelhouwer
C.
1982
Product inhibition in the acid forming stage of the anaerobic-digestion process
.
Water Research
16
(
5
),
633
639
.