- Email: [email protected]
Monitoring foaming potential in anaerobic digesters
Waste Management xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Monitoring foaming potential in anaerobic digesters Chenjing Jiang a,b, Rong Qi c, Liping Hao a, Simon Jon McIlroy a, Per Halkjær Nielsen a,⇑ a
Center for Microbial Communities, Department of Chemistry and Bioscience, Aalborg University, 9220 Aalborg, Denmark Key Laboratory of Engineering Oceanography, Second Institute of Oceanography, SOA, Hangzhou 310012, China c State Key Laboratory of Environmental Aquatic Chemistry, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 10085, China b
a r t i c l e
i n f o
Article history: Received 15 October 2017 Revised 6 February 2018 Accepted 12 February 2018 Available online xxxx Keywords: Foaming Anaerobic digestion Foaming potential aeration test Foam height/propensity Foam stability
a b s t r a c t Foaming in anaerobic digestion (AD) systems for biogas generation can give serious operational problems. The cause of such foaming events is often unclear, and it is therefore not an easy task to predict and subsequently apply preventative measures. Methods for the measurement of the foaming potential of digester sludge are often implemented, but no standardized method is available. In this study, we investigated parameters influencing the foam formation during experimental aeration tests of fullscale digester sludge, including air flow, time, and total solids concentration, and proposed an optimized method for standard use. In a survey of 16 full-scale AD systems located at wastewater treatment plants in Denmark, all sludge samples were classified into three groups (non-foaming, pre-foaming, and actually foaming) according to their foam height/propensity and stability. Extensive surveillance of plants with the proposed classification system will enable the determination of cut-off values to help to identify foaming or pre-foaming sludge, and to associate these with operational conditions leading to foaming episodes. Ó 2018 Published by Elsevier Ltd.
1. Introduction Foaming is a serious operational problem in full-scale biogas plants at wastewater treatment plants (WWTPs), where it causes inefficient gas recovery, blockages of gas mixing, fouling of gas collection pipes and covering the digester wall with foam solid (Ganidi et al., 2009; Pagilla et al., 1997). Besides reduced gas production, it is very costly in repairing and cleaning (Ganidi et al., 2009; Moeller and Görsch, 2015; Westlund et al., 1998). A combination of factors are thought to lead to stable foam formation in anaerobic digestion (AD) systems, including gas bubbles surrounded by liquid film, and hydrophobic particles in the form of microorganisms or suspended solids (Boe et al., 2012; Ganidi et al., 2009). Although this phenomenon has been known for decades, and numerous studies have addressed it in AD, the exact mechanism of how foaming is initiated and stabilized is still not fully understood (Barjenbruch et al., 2000; Ganidi et al., 2009, 2011; Pagilla et al., 1997; Ross and Ellis, 1992; Westlund et al., 1998). Several methods have been implemented to measure the foaming propensity of both activated sludge wastewater treatment plants and their associated anaerobic digester systems. Foaming is also a well described operational problem in the activated sludge
⇑ Corresponding author. E-mail address: [email protected] (P.H. Nielsen).
part of the WWTPs, where it is easy to observe on top of the process or settling tanks. In the closed AD tanks, it is very difficult to observe the foam layer, hence therefore it is necessary to develop simple methods that can monitor the general foaming state of the sludge and detect the onset of foaming events for early intervention measures, such as reducing loading or addition of antifoaming chemicals. The foaming potential aeration test, also known as the bubble test, is one of most popular methods applied for monitoring or evaluating the foaming propensity of sludge. It is a simple analysis attempting to generate and measure foam to determine foaming tendency. It was first applied in activated sludge systems (Blackall and Marshall, 1989), but may not represent the fullscale plants due to different wall effect and hydraulic conditions (Fryer and Gray, 2012; Hug, 2006). However, there are some studies that found a correlation between foaming potential and foam event in the plant (Blackall et al., 1991; Hladikova et al., 2002; Hug, 2006; Torregrossa et al., 2005). In AD systems, aeration test were initially applied to determine foaming tendency by measuring foaming height and surface tension, but it was unsuccessful as no significant difference was observed in foam height among foaming and non-foaming samples (Ross and Ellis, 1992). However, there seems to be a number of issues that make the aeration tests more difficult in AD, such as high and varying total solids (TS), and a wide range of salinity values that may affect the surface properties of the particles.
https://doi.org/10.1016/j.wasman.2018.02.021 0956-053X/Ó 2018 Published by Elsevier Ltd.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
2
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
In recent years, the foaming potential has been defined by two new terms: foaming propensity and foam stability describing how easily foam forms and how stable it is. Foaming propensity was found to be able to evaluate foaming potential. However, it is doubted whether the foam stability is sensitive enough to predict foaming in digesters (Boe et al., 2012; Ganidi et al., 2011; Kougias et al., 2013). Additionally, foaming potential was calculated in terms of an unstable and stable foam ratio, and analyzed in two full-scale digesters (Subramanian and Pagilla, 2014; Subramanian et al., 2015). However, the results of these two studies showed that the foam potential test may not be well-suited to the prediction of foaming in full-scale digesters due to the complexity of the mechanisms that contribute to foaming. The aeration tests applied in the past 10 years on ADs are very diverse in terms of apparatus, test parameters, and calculation methods (Table 1), and in many cases, key information is not given in the original papers, such as apparatus size, aeration time, flow rate, etc. Moreover, the effect of TS concentration on foaming potential aeration test was only investigated in one study, where it was shown to be of key importance to the outcome of the test (Boe et al., 2012). Consequently, the outcome of these tests is questionable and hard to compare, thus there is a strong need for an easy, fast, and reliable procedure and evaluation scale of the foaming potential of digester sludge. The aim of this study is to establish and test a protocol for monitoring foaming potential in anaerobic digesters. We investigated parameters influencing the foaming propensity during the foaming potential aeration test in digester sludge and optimized the experimental conditions by using an orthogonal design experiment. This is a statistical design where an orthogonal table is used to mix and analyze the multifactors in order to obtain optimal combinations of factors and levels. The proposed protocol was used in a survey into sludge samples from sixteen full-scale mesophilic and thermophilic digesters located at 14 WWTPs in Denmark. The foaming
properties of these digester sludges were characterized and classified, some of which with serious foaming. 2. Materials and methods 2.1. Sample collection and physico-chemical analyses Digester sludge samples were obtained from 16 anaerobic digesters at 14 Danish WWTPs. Fresh samples were investigated immediately when they arrived in the laboratory with a transportation time less than 24 h. Most digesters were mesophilic (14), one of which had thermal hydrolysis process pre-treatment (THP), and two were thermophilic. Primary and secondary sludge were fed to all the digesters, except Fredericia, which only had secondary sludge. The primary sludge fraction was 50–70% of the organics. The TS, volatile solids (VS) values, and foam status of digester sludge samples are listed in Table 2, and other characteristics of the anaerobic digester plants and sludge are presented in Table S1 in supplementary materials. TS, VS, pH, total nitrogen, ammonia nitrogen, and orthophosphate were measured according to APHA (2005). Short chain organic acids (C1-C6) were analyzed by High Pressure Ion Chromatography (HPIC, Dionex ICS-5000 system), equipped with an IonPac AS11-HC capillary column and a Conductivity Detector (CD). Potassium hydroxide (60 mM) was used as eluent for a multi-step gradient elution. All measurements were performed in triplicate. 2.2. Foaming potential aeration test apparatus and calculation The apparatus used to measure foaming potential was constructed from a plexiglass tube (inner diameter of 3 cm, height of 85 cm) containing a diffuser stone attached to the bottom (Fig. 1). Air, which was used in most studies (Table 1), was supplied
Table 1 Literature review of methods to characterize the foaming potential foaming potential in ADs. NO.
Digester
Parameter
Apparatus
Aeration Time
Flow Rate
Sludge Volume
Reference
1
WWTPs digester
Foam height (while bubbling and 1 min after the bubbling was stopped)
NA
Air, NA
NA
Ross and Ellis (1992)
2
WWTPs digester WWTPs digester
Foam height
A graduated plexiglass tube 130 cm long and 4.5 cm diameter A vertical glass tube 33 cm long and 2.2 cm diameter Sintered glass with porosity S1 in 2 L cylinder
1 min
Air, 40 L/ h Nitrogen, 1 L/min
3 mL
Westlund et al. (1998)
1L
Zábranská et al. (2002)
A column 1 m long and 5.2 cm diameter
10 min
Air, 0.5 L/ min
NA
Ganidi et al. (2011)
NA
10 min
Air, 60 mL/min
50 mL
Boe et al. (2012); Kougias et al. (2013)
NA
1 min
Air, 1 L/ min
100 mL
Suhartini et al. (2014)
2 L graduated cylinder
NA
Air, 1.5 L/ min
200 mL
Subramanian and Pagilla (2014), Subramanian et al. (2015)
3
4
WWTPs digester
5
Manure digester
6
Sugar beet pulp digester WWTPs digester
7
Foaming potential (calculated based on the level of foamy sludge after 5 min of the aeration over the volume of digester sludge) Foam stability (percentage of foaming remaining in the cylinder at 5 min after aeration compared with the volume of foam right after aeration) Foaming propensity (calculated based on the amount of foam generated from a sample after aeration normalized over the solids content of the sample) Foam stability (monitored indirectly by measuring the foam height 1 h after aeration ceased) Foaming tendency (the volume of foam right after aeration divided by air flow rate) Foaming stability (percentage of foam remaining in the settling cone at 1 h after aeration compared with the volume of foam right after aeration) Foaming potential, Foaming Tendency, Foaming Stability, Foaming Propensity Foaming potential (calculated in terms of an unstable foam and stable ratio)
5 min
NA: not available.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
3
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx Table 2 Characteristics of anaerobic digester plants and sludge samples. Sample location
Process
Sample foam status
Total solids (TS, g/L)
Volatile solids (VS, g/L)
VS/TS (%)
AAOE RT2 Avedoere RT1 Bjergmarken RT1 Damhusaaen RT1 Ejby Moelle Esbjerg Vest RT1A Esbjerg Vest RT2B Fornaes Fredericia RT2 Hjoerring Mariagerfjord Slagelse RT1 Soeholt Randers RT1 Randers RT2 Viborg RT1
Thermophilic Mesophilic Thermophilic Mesophilic Mesophilic Mesophilic Mesophilic Mesophilic THP-Mesophilic Mesophilic Mesophilic Mesophilic Mesophilic Mesophilic Mesophilic Mesophilic
No No Noa No No Noa Yes No Noa No Yes No No No No No
35.1 ± 0.6 15.0 ± 0.4 45.8 ± 1.1 34.2 ± 2.0 35.3 ± 1.4 32.4 ± 0.6 35.5 ± 1.9 36.0 ± 3.0 30.2 ± 0.5 25.1 ± 0.8 29.1 ± 0.2 31.9 ± 0.7 45.1 ± 0.2 35.3 ± 0.8 35.5 ± 0.2 19.7 ± 0.2
21.0 ± 0.5 6.64 ± 1.3 26.9 ± 0.8 17.7 ± 0.03 20.6 ± 1.0 19.8 ± 0.3 23.6 ± 1.1 23.9 ± 2.1 19.1 ± 0.3 12.0 ± 0.2 19.5 ± 0.1 20.5 ± 0.4 30.4 ± 0.9 19.8 ± 0.3 20.9 ± 0.9 12.0 ± 0.2
59.9 44.1 58.7 51.9 58.5 61.1 66.4 66.3 63.2 47.6 66.9 64.2 67.4 56.1 58.9 60.8
NA: not available. a Frequent foaming events were registered at this WWTP.
at the bottom to pass through the stone by an air pump (AM-TOP, Denmark). Air flow rate was measured and controlled using a flowmeter (Porter, USA). Maximum foam height during aeration (Hmax), and foam height immediately after the aeration (Hafter) were recorded. The maximum foaming propensity (Pmax) and foaming propensity immediately after the aeration (Pafter) were calculated based on Hmax or Hafter as well as the solids content of the sample (Eq. (1)). These parameters were applied to evaluate the foaming potential in AD sludge according to the literature shown in Table 1.
Pmax or after ¼
mm of Hmax or Hafter gram total solids
ð1Þ
2.3. Optimization of foaming potential aeration test For the method, appropriate foam height should be obtained in this specific apparatus, neither too low nor out of the tube. Three
parameters, including air flow rate (R), sludge volume (V), and aeration time (T), were examined at three different levels according to the preliminary experiments and the limitation by the tube length. The air flow rates were 125, 250, and 500 mL/min, the sludge volumes were 25, 50, and 100 mL, and the aeration times were 5, 10, and 15 min, nine different schemes in total, based on the orthogonal table L9(33). All the measurements were performed in triplicate. The sludge tested was from an AD at Mariagerfjord WWTP. 2.4. Effect of TS concentration Our previous survey showed a broad range of TS concentration in Danish digesters. Therefore, various TS concentrations (10, 20, 30, 40, 50, 60, and 70 g/L) were used on sludge samples from 16 digesters to optimize the foaming potential aeration test. The sludges were diluted to appropriate concentrations with supernatant from the same sample, which were obtained using centrifugation (Sigma 3-18k, UK) at 4200g for 10 min. Hmax and Hafter were measured during the test procedure. All measurements were performed in triplicate. 2.5. Statistical analysis The orthogonal experiment data were analyzed using univariate analysis of variance (ANOVA) in IBM SPSS Statistics 24 (IBM Corp., Armonk, NY, USA). Principal component analysis (PCA) was performed in R version 3.4 (R Development Core Team, 2017) through RStudio version 1.0.143 (http://www.rstudio.com), using R CRAN packages: ampvis (v1.13) (Albertsen et al., 2015), ggplot2 (Wickham, 2016). Other statistical analyses and visualizations were performed in OriginPro 2017 (OriginLab, Northampton, MA, USA). 3. Results and discussion 3.1. Optimization of the foaming potential aeration test
Fig. 1. Foaming potential aeration test apparatus.
An orthogonal experiment was performed to measure the foaming propensity of the digester sludge from a full-scale AD (Mariagerfjord) as shown in Table 3. It revealed that the highest foam propensity was generated at 15 min aeration time, 25 mL sludge volume, and 500 mL/min flow rate, while the lowest was made by 10 min aeration time, 100 mL sludge volume, and 125 mL/min flow rate. Fig. 2 demonstrated that flow rate had much stronger effect on the foaming propensity (largest range and F val-
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
4
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
Table 3 The result of foaming propensity test by orthogonal design experiment. Experiment number
1 2 3 4 5 6 7 8 9 a b
Variable Aeration time (T, min)
Sludge volume (V, mL)
Flow rate (R, mL/min)
1 1 1 2 2 2 3 3 3
1 2 3 1 2 3 1 2 3
1 2 3 2 3 1 3 1 2
(5) (5) (5) (10) (10) (10) (15) (15) (15)
(25) (50) (100) (25) (50) (100) (25) (50) (100)
(125) (250) (500) (250) (500) (125) (500) (125) (250)
Pmax (mm/g)a
Pafter (mm/g)b
190 ± 5 331 ± 8 295 ± 3 359 ± 31 538 ± 5 182 ± 38 872 ± 51 313 ± 36 264 ± 13
113 ± 31 231 ± 26 249 ± 23 72 ± 10 421 ± 41 164 ± 36 810 ± 51 174 ± 31 213 ± 44
Pmax: the maximum foaming propensity during aeration. Pafter: foaming propensity immediately.
ues), both for Pmax and Pafter, compared to sludge volume and aeration time. Higher flow rate resulted in higher foaming propensity values, which was also observed by Fryer et al., (2011), who found that, in activated sludge, flow rate caused significant differences in measured foam potential. As for the other two factors, sludge volume had a bigger range of Pmax than aeration time, while aeration time had a bigger range of Pafter than sludge volume. According to the principles of orthogonal design which define the optimal conditions to be the highest propensity value for each factor, 15 min aeration time, 25 mL sludge volume, and 500 mL/min flow rate (T3V1R3) were the optimized conditions. However, these results were limited by the height of the tube and influenced by TS concentration (see below). Therefore, we chose to use 500 mL/min flow rate, 10 min aeration time, and 50 mL sludge volume (T2V2R3) to obtain an appropriate foam height.
3.2. The effect of TS concentration We investigated the influence of TS concentration on foam height/propensity in mesophilic digester samples with the optimized protocol on sludge from Viborg RT1, Fredericia RT2, and Mariagerfjord. At the time of sampling, Viborg RT1 had no foaming; Mariagerfjord had foaming; but the foaming status of Fredericia RT2 was unknown, however, frequent foaming events were registered at this WWTP. The results showed that all measurements (Hmax, Hafter, Pmax, Pafter) varied widely with TS (Fig. 3), and for each sample, the foam height changed with increased TS concentration. Both Fredericia RT2 and Mariagerfjord had much higher foam height than the non-foaming Viborg RT1. Moreover, the tendency of the changes of foam height along with TS was different between the three samples with different foaming status. For
Fig. 2. Range and variance analysis of each factor of orthogonal experiment. Comparison of K1, K2, K3, and range values of each factor on Pmax (A) and Pafter (B) of the orthogonal experiment. Ki (i = 1, 2, 3) is the mean of corresponding propensity at different levels, while the range value of each factor is the difference between maximum and minimal values of Ki. The larger the range of one factor, the stronger the effect of the factor on propensity. The variance value represented by F-value was indicated on the top of each factor. The results of variance analysis are consistent with the results of the range analysis, i.e. FR > FV > FT and RangeR > RangeV > RangeT for Pmax.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
5
Fig. 3. Effect of TS concentration on foaming potential aeration test in three digester sludges. Comparison of Hmax and Hafter (A), Pmax and Pafter (B) at different TS concentrations in three digester sludges using foaming potential aeration test.
Viborg RT1 sludge, the foaming height exhibited a slight increase when TS increased from 10 to 50 g/L, and then decreased when TS exceeded 50 g/L. For Fredericia RT2 sludge, foam height fluctuated in the TS range tested. For Mariagerfjord sludge, the foaming sample foam height had the biggest value at TS = 10 g/L, and decreased strongly with the rising TS. In all samples, Hafter showed the same tendency as Hmax, indicating there was a correlation between Hafter and Hmax. Different from the foam height, foaming propensity values showed the same overall pattern in all three samples: the higher TS, the less propensity (Fig. 3B). However, when comparing different samples at the same TS level, it could be seen that the samples that showed bigger foam height also demonstrated higher propensity. This could be due to the fact that propensity was determined by both foam height and TS (Eq. (1)). The foaming propensity after aeration Pafter showed the same tendency as Pmax. It is seen that TS had a great effect on foam height as well as propensity, not only between different samples, but also within the same sample. The foaming tests in previous studies do not consider TS, except for a few investigations of activated sludge, where the foaming potential was shown to increase with the suspension solid (SS) concentration (Blackall et al., 1991; Hug, 2006; Oerther et al., 2001). However, the performance of the test depended on the type of activated sludge samples, and can hardly be compared to AD samples because normal TS concentrations in activated sludge systems (2–6 g/L) are much lower than in digesters (10– 50 g/L). In systems with AD sludge, only one study with manure sludge showed that TS was important, and the authors found, in
contrast to our study, that increase of TS from 10 to 60 g/L resulted in an increase in foam volume by foaming potential aeration test (Boe et al., 2012). Moeller et al. (2015) also found that the digester with highest foam-generation had highest viscosity and TS (1170 mPa s and 82.8 g/L). In the current study, we observed that when TS was higher than 50 g/L, the foam height and propensity of all three ADs decreased and stabilized as the high viscosity prevented the bubbles from rising (Hug, 2006; Moeller et al., 2015). When TS was in the range of 10–50 g/L, it was difficult to find a correlation between foam height/propensity and TS concentrations. In order to see whether any overall pattern could be identified for nonfoaming and foaming plants, a larger number of different ADs were tested using the same method. 3.3. Classification of digester sludge We analyzed Hmax, Hafter, and Hmax-Hafter (DHmax-after) in 16 digester samples from 14 WWTPs. According to the results, the digester sludge could be classified into three groups, each including several digesters with very similar height trends (Fig. 4). Group 1 had Hmax, Hafter, and DHmax-after ranging between 13 and 163 mm, 4–114 mm, and 3–80 mm, respectively. All the samples in Group 1 had foam height lower than 200 mm (green lines in Fig. 4). Group 2 samples (blue lines) had much larger foam height (57–770 mm for Hmax), and DHmax-after fluctuated strongly within the range of 13– 513 mm, whereas Group 3 samples (red lines) had a foam height range similar to that of Group 2, but with a much smoother tendency of DHmax-after within the range of 7–143 mm.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
6
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
Fig. 4. Classification of 16 digester sludges by foaming potential aeration test on foam height. Hmax, Hafter, and DHmax-after of Group 1 (A1, A2, A3), Group 2 (B1, B2, B3), Group 3 (C1, C2, C3). Hmax = maximum foam height during aeration, Hafter = foam height immediately after the aeration, and DHmax-after = Hmax Hafter.
The classification of these digester samples was investigated by principal component analysis (PCA) (Fig. 5). Overall, samples in Group 1 clustered together and were separated from the other two groups along the primary axis, indicating Group 1 was very different from the other groups. It was consistent with our survey that digesters in Group 1 did never experience foaming. Groups 2 and 3 had a certain similarity along the axes. Hence, they were difficult to separate. However, for samples in Group 3, except the thermophilic digester AAOE, all the other 4 digesters clustered together, including the foaming samples from Mariagerfjord and Esbjerg Vest RT2B. Interestingly, Esbjerg Vest RT1A, Fredericia RT2, and Bjergmarken were classified into Group 2 (prefoaming), and had no foaming during sampling (Table 2). However, they had frequently foaming events so these plants may continuously be close to the critical level giving problems. Furthermore, the clustering of Groups 2 and 3 among the mesophilic digesters may imply that other parameters such as microbial communities or chemical parameters also affect the foam height. However, no study, to our knowledge, compares foam height between mesophilic and thermophilic digesters. This should be investigated in future studies of AD systems. Additionally, there was a significant correlation between Hmax and Hafter of all the samples (p < 0.05), except Slagelse RT1 and Fredericia RT2 (Table S2 in supplementary materials). Also, according to the observation during the tests, we found that DHmax-after is a sensitive parameter that indicates the foam stability. The smaller the value, the more stable the foam. The foaming potential aeration test is actually a kind of adsorptive bubble separation technique
which can be used to separate microorganisms and small particles by flotation (Blackall and Marshall, 1989). Surfactants produced by microorganisms are attached to the surfaces of bubbles rising through the liquid phase in the digester. The surfactant was concentrated on the top of the digester, which can cause foaming. Both surfactants and microorganisms are necessary to contribute to stable formation of foam since it has been demonstrated that foams were unstable in the absence of microorganisms and will not be formed without surfactant (Blackall and Marshall, 1989). Group 1 had smallest Hmax, Hafter, and DHmax-after values. The small amount of foam produced during the aeration process was probably caused by the aeration bubbles. In contrast, Group 2 generated much more foam, which disappeared quickly after aeration, as is indicated by the higher DHmax-after values (Fig. 4-B3). According to the theories of Blackall and Marshall (1989), there might be surfactants in Group 2, but not enough foam-forming microorganisms to stabilize the foam. We consider this group as pre-foaming sludge. Group 3 exhibited stable foam as the values of DHmax-after were stable, indicating that both foam-forming microorganisms and surfactant were present. This group represents actual foaming or high risk of foaming sludge. Interestingly, it was found that the average VS/TS ratio of three group were 55.3%, 60.5%, and 63.6%, respectively (see Table 2 for each plant). It indicates a possible correlation between organic content and foaming potential of digester sludge, which may suggest that a high level of microorganism contributes to the foam height.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
7
Fig. 5. Principal component analysis of 16 digester samples on the Hmax, Hafter, DHmax-after. THP = thermal hydrolysis process pre-treatment.
Fig. 6. Classification of 16 digester sludges by foaming potential aeration test on foam propensity. Pmax, Pafter, and DPmax-after of Group 1 (A1, A2, A3), Group 2 (B1, B2, B3), Group 3 (C1, C2, C3). Pmax = maximum foam propensity during aeration, Pafter = foam propensity immediately after the aeration, and DPmax-after = Pmax Pafter.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
8
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
3.4. The recommended foaming potential aeration test in AD systems The proposed set-up is an apparatus-specific method which was optimized according to a survey of 16 AD samples. These represent various types of ADs installed at WWTPs. Manure-based AD sludges and other types have not been tested, and modifications of the method may be needed. However, the general principles of foam formation and stabilization are similar and can be investigated by such tests. Generally, it seems that measurement of Hmax and Hafter is a direct way of analyzing the foam properties. The values of Hmax and Hafter can be used to classify digester samples roughly as (a) Group 1: Hmax < 200 mm at all TS concentrations, no abundant surfactants or foam-forming microorganisms to cause foam; (b) Group 2: Hmax > 200 mm, and DHmax-after > 200 mm at some of TS concentrations, absence of foam-forming microorganisms; (c) Group 3: Hmax > 200 mm at some of TS concentrations, and DHmax-after < 200 mm at all TS concentrations, abundant surfactant, and foam-forming. Group 1 represents non-foaming status, Groups 2 and 3 are most likely pre-foaming and actually foaming, respectively, but difficult to distinguish. Extensive surveillance of plants with the proposed classification system will enable the determination of more precise cut-off values. The cut-off values can also be used to correlate the foaming status with operational conditions to identify key factors leading to foaming episodes. As Fig. 4 shows, Hmax and Hafter at TS higher than 50 g/L are similar for the three groups, especially for Groups 2 and 3. This could be due to the high viscosity at high TS. On the other hand, 10 g/L is also problematic because the foam floated easily out of the tube, which occurred for two ADs in Group 2 (Soeholt and Bjergmarken RT2). This was the reason for the flat curves at the beginning of these two ADs (Fig. 4-B1). Thus, we recommend that the TS is kept in the range of 20–30 g/L, which is also the normal TS concentration in most ADs. Dilution with sludge water or up-concentration by settling and removal of supernatant is needed when TS is outside this interval. Fig. 6 showed that the foam propensity varied with TS in the three groups. Pmax and Pafter of nearly all the samples decreased with increasing TS, except Pafter of Damhusaaen RT1 at TS = 20 g/ L. The trends on foam propensity of these three groups showed the same as foam height, indicating that foam propensity can also be used as a parameter to classify the foam property. Foam stability is another parameter sometimes applied to describe/define the foaming potential. To measure foam stability, typically, the foam height should be recorded at a certain time point/after aeration for a fixed time period, often 1 h after aeration stops (Boe et al., 2012; Ganidi et al., 2011; Kougias et al., 2013). However, according to our results, several samples had zero foam height 1 h after the aeration stopped (results not shown). This agrees with Ganidi et al. (2011) and Kougias et al. (2013), who found that this value was not sensitive enough to predict or link to foaming in digesters. Moreover, this parameter is not relevant for full-scale operation since a state without gas production for 1 h does not take place in full-scale digester tanks. Attempts to use the ratio of Hafter to Hmax to define the foam stability were unsuccessful as Group 1 and 2 were undistinguishable (Fig. S1 in supplementary materials). Instead, we suggest that the parameter of DHmax-after or Pmax-Pafter (DPmax-after) could be used to evaluate the foam stability.
4. Conclusion A simple protocol for foaming potential aeration test for monitoring foam properties in digester sludge is proposed: a simple apparatus (tube, inner diameter 3 cm, 85 cm height), flow rate of 500 mL/min, digester sludge volume of 50 mL, and 10 min aeration
time with the TS range of 20–30 g/L. The digester sludge can be classified into three groups, based on the foam height and propensity. It is a simple, fast, and inexpensive method that produces comparative results in a short time suitable for digester surveillance. Acknowledgements We thank the plant operators for providing digester sludge samples. This research was funded by the Innovation Fund Denmark through the project NomiGas [grant number 1305-00018B]. The Chinese Scientific Research Council is acknowledged for providing financial support for C. Jiang and R. Qi. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.wasman.2018.02. 021. References Albertsen, M., Karst, S.M., Ziegler, A.S., Kirkegaard, R.H., Nielsen, P.H., 2015. Back to basics – the influence of DNA extraction and primer choice on phylogenetic analysis of activated sludge communities. PLoS One 10, e0132783. APHA, 2005. Standard Methods for the Examination of Water and Wastewater. American Public Health Association, Washington, DC, USA. Barjenbruch, M., Hoffmann, H., Kopplow, O., Trnckner, J., 2000. Minimizing of foaming in digesters by pre-treatment of the surplus-sludge. Water Sci. Technol. 42, 235–241. Blackall, L.L., Marshall, K.C., 1989. The mechanism of stabilization of actinomycete foams and the prevention of foaming under laboratory conditions. J. Ind. Microbiol. 4, 181–187. Blackall, L.L., Harbers, A.E., Greenfield, P.F., Hayward, A.C., 1991. Activated sludge foams: effects of environmental variables on organism growth and foam formation. Environ. Technol. 12, 241–248. Boe, K., Kougias, P.G., Pacheco, F., O-Thong, S., Angelidaki, I., 2012. Effect of substrates and intermediate compounds on foaming in manure digestion systems. Water Sci. Technol. 66, 2146–2154. Fryer, M., Gray, N.F., 2012. Foaming scum index (FSI) – a new tool for the assessment and characterisation of biological mediated activated sludge foams. J. Environ. Manage. 110, 8–19. Fryer, M., O’Flaherty, E., Gray, N.F., 2011. Evaluating the measurement of activated sludge foam potential. Water 3, 424–444. Ganidi, N., Tyrrel, S., Cartmell, E., 2009. Anaerobic digestion foaming causes – A review. Bioresour. Technol. 100, 5546–5554. Ganidi, N., Tyrrel, S., Cartmell, E., 2011. The effect of organic loading rate on foam initiation during mesophilic anaerobic digestion of municipal wastewater sludge. Bioresour. Technol. 102, 6637–6643. Hladikova, K., Ruzickova, I., Klucova, P., Wanner, J., 2002. An investigation into studying of the activated sludge foaming potential by using physicochemical parameters. Water Sci. Technol. 46, 525–528. Hug, T., 2006. Characterization and controlling of foam and scum in activated sludge systems. Swiss Federal Institute of Technology Zurich. Kougias, P.G., Boe, K., Angelidaki, I., 2013. Effect of organic loading rate and feedstock composition on foaming in manure-based biogas reactors. Bioresour. Technol. 144, 1–7. Moeller, L., Görsch, K., 2015. Foam formation in full-scale biogas plants processing biogenic waste. Energy Sustain. Soc. 5, 1. Moeller, L., Eismann, F., Wißmann, D., Nägele, H.-J., Zielonka, S., Müller, R.A., Zehnsdorf, A., 2015. Innovative test method for the estimation of the foaming tendency of substrates for biogas plants. Waste Manage. 41, 39–49. Oerther, D.B., de los Reyes, F.L., de los Reyes, M.F., Raskin, L., 2001. Quantifying filamentous microorganisms in activated sludge before, during, and after an incident of foaming by oligonucleotide probe hybridizations and antibody staining. Water Res. 35, 3325–3336. Pagilla, K.R., Craney, K.C., Kido, W.H., 1997. Causes and effects of foaming in anaerobic sludge digesters. Water Sci. Technol. 36, 463–470. Ross, R.D., Ellis, L.M., 1992. Laboratory-scale investigation of foaming in anaerobic digesters. Water Environ. Res. 64, 154–162. Subramanian, B., Pagilla, K.R., 2014. Anaerobic digester foaming in full-scale cylindrical digesters – effects of organic loading rate, feed characteristics, and mixing. Bioresour. Technol. 159, 182–192. Subramanian, B., Miot, A., Jones, B., Klibert, C., Pagilla, K.R., 2015. A full-scale study of mixing and foaming in egg-shaped anaerobic digesters. Bioresour. Technol. 192, 461–470. Suhartini, S., Heaven, S., Banks, C.J., 2014. Comparison of mesophilic and thermophilic anaerobic digestion of sugar beet pulp: Performance, dewaterability and foam control. Bioresour. Technol. 152, 202–211.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx Torregrossa, M., Viviani, G., Vinci, V., 2005. Foaming estimation tests in activated sludge systems. Acta Hydrochim. Hydrobiol. 33, 240–246. Westlund, Å.D., Hagland, E., Rothman, M., 1998. Operational aspects on foaming in digesters caused by Microthrix parvicella. Water Sci. Technol. 38, 29–34.
Wickham, H., 2016. ggplot2: Elegant Graphics for Data York. Zábranská, J., Dohányos, M., Jenícˇek, P., Zaplatílková, contribution of thermophilic anaerobic digestion to wastewater sludge treatment. Water Sci. Technol. 46,
9 Analysis. Springer, New P., Kutil, J., 2002. The the stable operation of 447–453.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Monitoring foaming potential in anaerobic digesters Chenjing Jiang a,b, Rong Qi c, Liping Hao a, Simon Jon McIlroy a, Per Halkjær Nielsen a,⇑ a
Center for Microbial Communities, Department of Chemistry and Bioscience, Aalborg University, 9220 Aalborg, Denmark Key Laboratory of Engineering Oceanography, Second Institute of Oceanography, SOA, Hangzhou 310012, China c State Key Laboratory of Environmental Aquatic Chemistry, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 10085, China b
a r t i c l e
i n f o
Article history: Received 15 October 2017 Revised 6 February 2018 Accepted 12 February 2018 Available online xxxx Keywords: Foaming Anaerobic digestion Foaming potential aeration test Foam height/propensity Foam stability
a b s t r a c t Foaming in anaerobic digestion (AD) systems for biogas generation can give serious operational problems. The cause of such foaming events is often unclear, and it is therefore not an easy task to predict and subsequently apply preventative measures. Methods for the measurement of the foaming potential of digester sludge are often implemented, but no standardized method is available. In this study, we investigated parameters influencing the foam formation during experimental aeration tests of fullscale digester sludge, including air flow, time, and total solids concentration, and proposed an optimized method for standard use. In a survey of 16 full-scale AD systems located at wastewater treatment plants in Denmark, all sludge samples were classified into three groups (non-foaming, pre-foaming, and actually foaming) according to their foam height/propensity and stability. Extensive surveillance of plants with the proposed classification system will enable the determination of cut-off values to help to identify foaming or pre-foaming sludge, and to associate these with operational conditions leading to foaming episodes. Ó 2018 Published by Elsevier Ltd.
1. Introduction Foaming is a serious operational problem in full-scale biogas plants at wastewater treatment plants (WWTPs), where it causes inefficient gas recovery, blockages of gas mixing, fouling of gas collection pipes and covering the digester wall with foam solid (Ganidi et al., 2009; Pagilla et al., 1997). Besides reduced gas production, it is very costly in repairing and cleaning (Ganidi et al., 2009; Moeller and Görsch, 2015; Westlund et al., 1998). A combination of factors are thought to lead to stable foam formation in anaerobic digestion (AD) systems, including gas bubbles surrounded by liquid film, and hydrophobic particles in the form of microorganisms or suspended solids (Boe et al., 2012; Ganidi et al., 2009). Although this phenomenon has been known for decades, and numerous studies have addressed it in AD, the exact mechanism of how foaming is initiated and stabilized is still not fully understood (Barjenbruch et al., 2000; Ganidi et al., 2009, 2011; Pagilla et al., 1997; Ross and Ellis, 1992; Westlund et al., 1998). Several methods have been implemented to measure the foaming propensity of both activated sludge wastewater treatment plants and their associated anaerobic digester systems. Foaming is also a well described operational problem in the activated sludge
⇑ Corresponding author. E-mail address: [email protected] (P.H. Nielsen).
part of the WWTPs, where it is easy to observe on top of the process or settling tanks. In the closed AD tanks, it is very difficult to observe the foam layer, hence therefore it is necessary to develop simple methods that can monitor the general foaming state of the sludge and detect the onset of foaming events for early intervention measures, such as reducing loading or addition of antifoaming chemicals. The foaming potential aeration test, also known as the bubble test, is one of most popular methods applied for monitoring or evaluating the foaming propensity of sludge. It is a simple analysis attempting to generate and measure foam to determine foaming tendency. It was first applied in activated sludge systems (Blackall and Marshall, 1989), but may not represent the fullscale plants due to different wall effect and hydraulic conditions (Fryer and Gray, 2012; Hug, 2006). However, there are some studies that found a correlation between foaming potential and foam event in the plant (Blackall et al., 1991; Hladikova et al., 2002; Hug, 2006; Torregrossa et al., 2005). In AD systems, aeration test were initially applied to determine foaming tendency by measuring foaming height and surface tension, but it was unsuccessful as no significant difference was observed in foam height among foaming and non-foaming samples (Ross and Ellis, 1992). However, there seems to be a number of issues that make the aeration tests more difficult in AD, such as high and varying total solids (TS), and a wide range of salinity values that may affect the surface properties of the particles.
https://doi.org/10.1016/j.wasman.2018.02.021 0956-053X/Ó 2018 Published by Elsevier Ltd.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
2
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
In recent years, the foaming potential has been defined by two new terms: foaming propensity and foam stability describing how easily foam forms and how stable it is. Foaming propensity was found to be able to evaluate foaming potential. However, it is doubted whether the foam stability is sensitive enough to predict foaming in digesters (Boe et al., 2012; Ganidi et al., 2011; Kougias et al., 2013). Additionally, foaming potential was calculated in terms of an unstable and stable foam ratio, and analyzed in two full-scale digesters (Subramanian and Pagilla, 2014; Subramanian et al., 2015). However, the results of these two studies showed that the foam potential test may not be well-suited to the prediction of foaming in full-scale digesters due to the complexity of the mechanisms that contribute to foaming. The aeration tests applied in the past 10 years on ADs are very diverse in terms of apparatus, test parameters, and calculation methods (Table 1), and in many cases, key information is not given in the original papers, such as apparatus size, aeration time, flow rate, etc. Moreover, the effect of TS concentration on foaming potential aeration test was only investigated in one study, where it was shown to be of key importance to the outcome of the test (Boe et al., 2012). Consequently, the outcome of these tests is questionable and hard to compare, thus there is a strong need for an easy, fast, and reliable procedure and evaluation scale of the foaming potential of digester sludge. The aim of this study is to establish and test a protocol for monitoring foaming potential in anaerobic digesters. We investigated parameters influencing the foaming propensity during the foaming potential aeration test in digester sludge and optimized the experimental conditions by using an orthogonal design experiment. This is a statistical design where an orthogonal table is used to mix and analyze the multifactors in order to obtain optimal combinations of factors and levels. The proposed protocol was used in a survey into sludge samples from sixteen full-scale mesophilic and thermophilic digesters located at 14 WWTPs in Denmark. The foaming
properties of these digester sludges were characterized and classified, some of which with serious foaming. 2. Materials and methods 2.1. Sample collection and physico-chemical analyses Digester sludge samples were obtained from 16 anaerobic digesters at 14 Danish WWTPs. Fresh samples were investigated immediately when they arrived in the laboratory with a transportation time less than 24 h. Most digesters were mesophilic (14), one of which had thermal hydrolysis process pre-treatment (THP), and two were thermophilic. Primary and secondary sludge were fed to all the digesters, except Fredericia, which only had secondary sludge. The primary sludge fraction was 50–70% of the organics. The TS, volatile solids (VS) values, and foam status of digester sludge samples are listed in Table 2, and other characteristics of the anaerobic digester plants and sludge are presented in Table S1 in supplementary materials. TS, VS, pH, total nitrogen, ammonia nitrogen, and orthophosphate were measured according to APHA (2005). Short chain organic acids (C1-C6) were analyzed by High Pressure Ion Chromatography (HPIC, Dionex ICS-5000 system), equipped with an IonPac AS11-HC capillary column and a Conductivity Detector (CD). Potassium hydroxide (60 mM) was used as eluent for a multi-step gradient elution. All measurements were performed in triplicate. 2.2. Foaming potential aeration test apparatus and calculation The apparatus used to measure foaming potential was constructed from a plexiglass tube (inner diameter of 3 cm, height of 85 cm) containing a diffuser stone attached to the bottom (Fig. 1). Air, which was used in most studies (Table 1), was supplied
Table 1 Literature review of methods to characterize the foaming potential foaming potential in ADs. NO.
Digester
Parameter
Apparatus
Aeration Time
Flow Rate
Sludge Volume
Reference
1
WWTPs digester
Foam height (while bubbling and 1 min after the bubbling was stopped)
NA
Air, NA
NA
Ross and Ellis (1992)
2
WWTPs digester WWTPs digester
Foam height
A graduated plexiglass tube 130 cm long and 4.5 cm diameter A vertical glass tube 33 cm long and 2.2 cm diameter Sintered glass with porosity S1 in 2 L cylinder
1 min
Air, 40 L/ h Nitrogen, 1 L/min
3 mL
Westlund et al. (1998)
1L
Zábranská et al. (2002)
A column 1 m long and 5.2 cm diameter
10 min
Air, 0.5 L/ min
NA
Ganidi et al. (2011)
NA
10 min
Air, 60 mL/min
50 mL
Boe et al. (2012); Kougias et al. (2013)
NA
1 min
Air, 1 L/ min
100 mL
Suhartini et al. (2014)
2 L graduated cylinder
NA
Air, 1.5 L/ min
200 mL
Subramanian and Pagilla (2014), Subramanian et al. (2015)
3
4
WWTPs digester
5
Manure digester
6
Sugar beet pulp digester WWTPs digester
7
Foaming potential (calculated based on the level of foamy sludge after 5 min of the aeration over the volume of digester sludge) Foam stability (percentage of foaming remaining in the cylinder at 5 min after aeration compared with the volume of foam right after aeration) Foaming propensity (calculated based on the amount of foam generated from a sample after aeration normalized over the solids content of the sample) Foam stability (monitored indirectly by measuring the foam height 1 h after aeration ceased) Foaming tendency (the volume of foam right after aeration divided by air flow rate) Foaming stability (percentage of foam remaining in the settling cone at 1 h after aeration compared with the volume of foam right after aeration) Foaming potential, Foaming Tendency, Foaming Stability, Foaming Propensity Foaming potential (calculated in terms of an unstable foam and stable ratio)
5 min
NA: not available.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
3
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx Table 2 Characteristics of anaerobic digester plants and sludge samples. Sample location
Process
Sample foam status
Total solids (TS, g/L)
Volatile solids (VS, g/L)
VS/TS (%)
AAOE RT2 Avedoere RT1 Bjergmarken RT1 Damhusaaen RT1 Ejby Moelle Esbjerg Vest RT1A Esbjerg Vest RT2B Fornaes Fredericia RT2 Hjoerring Mariagerfjord Slagelse RT1 Soeholt Randers RT1 Randers RT2 Viborg RT1
Thermophilic Mesophilic Thermophilic Mesophilic Mesophilic Mesophilic Mesophilic Mesophilic THP-Mesophilic Mesophilic Mesophilic Mesophilic Mesophilic Mesophilic Mesophilic Mesophilic
No No Noa No No Noa Yes No Noa No Yes No No No No No
35.1 ± 0.6 15.0 ± 0.4 45.8 ± 1.1 34.2 ± 2.0 35.3 ± 1.4 32.4 ± 0.6 35.5 ± 1.9 36.0 ± 3.0 30.2 ± 0.5 25.1 ± 0.8 29.1 ± 0.2 31.9 ± 0.7 45.1 ± 0.2 35.3 ± 0.8 35.5 ± 0.2 19.7 ± 0.2
21.0 ± 0.5 6.64 ± 1.3 26.9 ± 0.8 17.7 ± 0.03 20.6 ± 1.0 19.8 ± 0.3 23.6 ± 1.1 23.9 ± 2.1 19.1 ± 0.3 12.0 ± 0.2 19.5 ± 0.1 20.5 ± 0.4 30.4 ± 0.9 19.8 ± 0.3 20.9 ± 0.9 12.0 ± 0.2
59.9 44.1 58.7 51.9 58.5 61.1 66.4 66.3 63.2 47.6 66.9 64.2 67.4 56.1 58.9 60.8
NA: not available. a Frequent foaming events were registered at this WWTP.
at the bottom to pass through the stone by an air pump (AM-TOP, Denmark). Air flow rate was measured and controlled using a flowmeter (Porter, USA). Maximum foam height during aeration (Hmax), and foam height immediately after the aeration (Hafter) were recorded. The maximum foaming propensity (Pmax) and foaming propensity immediately after the aeration (Pafter) were calculated based on Hmax or Hafter as well as the solids content of the sample (Eq. (1)). These parameters were applied to evaluate the foaming potential in AD sludge according to the literature shown in Table 1.
Pmax or after ¼
mm of Hmax or Hafter gram total solids
ð1Þ
2.3. Optimization of foaming potential aeration test For the method, appropriate foam height should be obtained in this specific apparatus, neither too low nor out of the tube. Three
parameters, including air flow rate (R), sludge volume (V), and aeration time (T), were examined at three different levels according to the preliminary experiments and the limitation by the tube length. The air flow rates were 125, 250, and 500 mL/min, the sludge volumes were 25, 50, and 100 mL, and the aeration times were 5, 10, and 15 min, nine different schemes in total, based on the orthogonal table L9(33). All the measurements were performed in triplicate. The sludge tested was from an AD at Mariagerfjord WWTP. 2.4. Effect of TS concentration Our previous survey showed a broad range of TS concentration in Danish digesters. Therefore, various TS concentrations (10, 20, 30, 40, 50, 60, and 70 g/L) were used on sludge samples from 16 digesters to optimize the foaming potential aeration test. The sludges were diluted to appropriate concentrations with supernatant from the same sample, which were obtained using centrifugation (Sigma 3-18k, UK) at 4200g for 10 min. Hmax and Hafter were measured during the test procedure. All measurements were performed in triplicate. 2.5. Statistical analysis The orthogonal experiment data were analyzed using univariate analysis of variance (ANOVA) in IBM SPSS Statistics 24 (IBM Corp., Armonk, NY, USA). Principal component analysis (PCA) was performed in R version 3.4 (R Development Core Team, 2017) through RStudio version 1.0.143 (http://www.rstudio.com), using R CRAN packages: ampvis (v1.13) (Albertsen et al., 2015), ggplot2 (Wickham, 2016). Other statistical analyses and visualizations were performed in OriginPro 2017 (OriginLab, Northampton, MA, USA). 3. Results and discussion 3.1. Optimization of the foaming potential aeration test
Fig. 1. Foaming potential aeration test apparatus.
An orthogonal experiment was performed to measure the foaming propensity of the digester sludge from a full-scale AD (Mariagerfjord) as shown in Table 3. It revealed that the highest foam propensity was generated at 15 min aeration time, 25 mL sludge volume, and 500 mL/min flow rate, while the lowest was made by 10 min aeration time, 100 mL sludge volume, and 125 mL/min flow rate. Fig. 2 demonstrated that flow rate had much stronger effect on the foaming propensity (largest range and F val-
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
4
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
Table 3 The result of foaming propensity test by orthogonal design experiment. Experiment number
1 2 3 4 5 6 7 8 9 a b
Variable Aeration time (T, min)
Sludge volume (V, mL)
Flow rate (R, mL/min)
1 1 1 2 2 2 3 3 3
1 2 3 1 2 3 1 2 3
1 2 3 2 3 1 3 1 2
(5) (5) (5) (10) (10) (10) (15) (15) (15)
(25) (50) (100) (25) (50) (100) (25) (50) (100)
(125) (250) (500) (250) (500) (125) (500) (125) (250)
Pmax (mm/g)a
Pafter (mm/g)b
190 ± 5 331 ± 8 295 ± 3 359 ± 31 538 ± 5 182 ± 38 872 ± 51 313 ± 36 264 ± 13
113 ± 31 231 ± 26 249 ± 23 72 ± 10 421 ± 41 164 ± 36 810 ± 51 174 ± 31 213 ± 44
Pmax: the maximum foaming propensity during aeration. Pafter: foaming propensity immediately.
ues), both for Pmax and Pafter, compared to sludge volume and aeration time. Higher flow rate resulted in higher foaming propensity values, which was also observed by Fryer et al., (2011), who found that, in activated sludge, flow rate caused significant differences in measured foam potential. As for the other two factors, sludge volume had a bigger range of Pmax than aeration time, while aeration time had a bigger range of Pafter than sludge volume. According to the principles of orthogonal design which define the optimal conditions to be the highest propensity value for each factor, 15 min aeration time, 25 mL sludge volume, and 500 mL/min flow rate (T3V1R3) were the optimized conditions. However, these results were limited by the height of the tube and influenced by TS concentration (see below). Therefore, we chose to use 500 mL/min flow rate, 10 min aeration time, and 50 mL sludge volume (T2V2R3) to obtain an appropriate foam height.
3.2. The effect of TS concentration We investigated the influence of TS concentration on foam height/propensity in mesophilic digester samples with the optimized protocol on sludge from Viborg RT1, Fredericia RT2, and Mariagerfjord. At the time of sampling, Viborg RT1 had no foaming; Mariagerfjord had foaming; but the foaming status of Fredericia RT2 was unknown, however, frequent foaming events were registered at this WWTP. The results showed that all measurements (Hmax, Hafter, Pmax, Pafter) varied widely with TS (Fig. 3), and for each sample, the foam height changed with increased TS concentration. Both Fredericia RT2 and Mariagerfjord had much higher foam height than the non-foaming Viborg RT1. Moreover, the tendency of the changes of foam height along with TS was different between the three samples with different foaming status. For
Fig. 2. Range and variance analysis of each factor of orthogonal experiment. Comparison of K1, K2, K3, and range values of each factor on Pmax (A) and Pafter (B) of the orthogonal experiment. Ki (i = 1, 2, 3) is the mean of corresponding propensity at different levels, while the range value of each factor is the difference between maximum and minimal values of Ki. The larger the range of one factor, the stronger the effect of the factor on propensity. The variance value represented by F-value was indicated on the top of each factor. The results of variance analysis are consistent with the results of the range analysis, i.e. FR > FV > FT and RangeR > RangeV > RangeT for Pmax.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
5
Fig. 3. Effect of TS concentration on foaming potential aeration test in three digester sludges. Comparison of Hmax and Hafter (A), Pmax and Pafter (B) at different TS concentrations in three digester sludges using foaming potential aeration test.
Viborg RT1 sludge, the foaming height exhibited a slight increase when TS increased from 10 to 50 g/L, and then decreased when TS exceeded 50 g/L. For Fredericia RT2 sludge, foam height fluctuated in the TS range tested. For Mariagerfjord sludge, the foaming sample foam height had the biggest value at TS = 10 g/L, and decreased strongly with the rising TS. In all samples, Hafter showed the same tendency as Hmax, indicating there was a correlation between Hafter and Hmax. Different from the foam height, foaming propensity values showed the same overall pattern in all three samples: the higher TS, the less propensity (Fig. 3B). However, when comparing different samples at the same TS level, it could be seen that the samples that showed bigger foam height also demonstrated higher propensity. This could be due to the fact that propensity was determined by both foam height and TS (Eq. (1)). The foaming propensity after aeration Pafter showed the same tendency as Pmax. It is seen that TS had a great effect on foam height as well as propensity, not only between different samples, but also within the same sample. The foaming tests in previous studies do not consider TS, except for a few investigations of activated sludge, where the foaming potential was shown to increase with the suspension solid (SS) concentration (Blackall et al., 1991; Hug, 2006; Oerther et al., 2001). However, the performance of the test depended on the type of activated sludge samples, and can hardly be compared to AD samples because normal TS concentrations in activated sludge systems (2–6 g/L) are much lower than in digesters (10– 50 g/L). In systems with AD sludge, only one study with manure sludge showed that TS was important, and the authors found, in
contrast to our study, that increase of TS from 10 to 60 g/L resulted in an increase in foam volume by foaming potential aeration test (Boe et al., 2012). Moeller et al. (2015) also found that the digester with highest foam-generation had highest viscosity and TS (1170 mPa s and 82.8 g/L). In the current study, we observed that when TS was higher than 50 g/L, the foam height and propensity of all three ADs decreased and stabilized as the high viscosity prevented the bubbles from rising (Hug, 2006; Moeller et al., 2015). When TS was in the range of 10–50 g/L, it was difficult to find a correlation between foam height/propensity and TS concentrations. In order to see whether any overall pattern could be identified for nonfoaming and foaming plants, a larger number of different ADs were tested using the same method. 3.3. Classification of digester sludge We analyzed Hmax, Hafter, and Hmax-Hafter (DHmax-after) in 16 digester samples from 14 WWTPs. According to the results, the digester sludge could be classified into three groups, each including several digesters with very similar height trends (Fig. 4). Group 1 had Hmax, Hafter, and DHmax-after ranging between 13 and 163 mm, 4–114 mm, and 3–80 mm, respectively. All the samples in Group 1 had foam height lower than 200 mm (green lines in Fig. 4). Group 2 samples (blue lines) had much larger foam height (57–770 mm for Hmax), and DHmax-after fluctuated strongly within the range of 13– 513 mm, whereas Group 3 samples (red lines) had a foam height range similar to that of Group 2, but with a much smoother tendency of DHmax-after within the range of 7–143 mm.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
6
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
Fig. 4. Classification of 16 digester sludges by foaming potential aeration test on foam height. Hmax, Hafter, and DHmax-after of Group 1 (A1, A2, A3), Group 2 (B1, B2, B3), Group 3 (C1, C2, C3). Hmax = maximum foam height during aeration, Hafter = foam height immediately after the aeration, and DHmax-after = Hmax Hafter.
The classification of these digester samples was investigated by principal component analysis (PCA) (Fig. 5). Overall, samples in Group 1 clustered together and were separated from the other two groups along the primary axis, indicating Group 1 was very different from the other groups. It was consistent with our survey that digesters in Group 1 did never experience foaming. Groups 2 and 3 had a certain similarity along the axes. Hence, they were difficult to separate. However, for samples in Group 3, except the thermophilic digester AAOE, all the other 4 digesters clustered together, including the foaming samples from Mariagerfjord and Esbjerg Vest RT2B. Interestingly, Esbjerg Vest RT1A, Fredericia RT2, and Bjergmarken were classified into Group 2 (prefoaming), and had no foaming during sampling (Table 2). However, they had frequently foaming events so these plants may continuously be close to the critical level giving problems. Furthermore, the clustering of Groups 2 and 3 among the mesophilic digesters may imply that other parameters such as microbial communities or chemical parameters also affect the foam height. However, no study, to our knowledge, compares foam height between mesophilic and thermophilic digesters. This should be investigated in future studies of AD systems. Additionally, there was a significant correlation between Hmax and Hafter of all the samples (p < 0.05), except Slagelse RT1 and Fredericia RT2 (Table S2 in supplementary materials). Also, according to the observation during the tests, we found that DHmax-after is a sensitive parameter that indicates the foam stability. The smaller the value, the more stable the foam. The foaming potential aeration test is actually a kind of adsorptive bubble separation technique
which can be used to separate microorganisms and small particles by flotation (Blackall and Marshall, 1989). Surfactants produced by microorganisms are attached to the surfaces of bubbles rising through the liquid phase in the digester. The surfactant was concentrated on the top of the digester, which can cause foaming. Both surfactants and microorganisms are necessary to contribute to stable formation of foam since it has been demonstrated that foams were unstable in the absence of microorganisms and will not be formed without surfactant (Blackall and Marshall, 1989). Group 1 had smallest Hmax, Hafter, and DHmax-after values. The small amount of foam produced during the aeration process was probably caused by the aeration bubbles. In contrast, Group 2 generated much more foam, which disappeared quickly after aeration, as is indicated by the higher DHmax-after values (Fig. 4-B3). According to the theories of Blackall and Marshall (1989), there might be surfactants in Group 2, but not enough foam-forming microorganisms to stabilize the foam. We consider this group as pre-foaming sludge. Group 3 exhibited stable foam as the values of DHmax-after were stable, indicating that both foam-forming microorganisms and surfactant were present. This group represents actual foaming or high risk of foaming sludge. Interestingly, it was found that the average VS/TS ratio of three group were 55.3%, 60.5%, and 63.6%, respectively (see Table 2 for each plant). It indicates a possible correlation between organic content and foaming potential of digester sludge, which may suggest that a high level of microorganism contributes to the foam height.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
7
Fig. 5. Principal component analysis of 16 digester samples on the Hmax, Hafter, DHmax-after. THP = thermal hydrolysis process pre-treatment.
Fig. 6. Classification of 16 digester sludges by foaming potential aeration test on foam propensity. Pmax, Pafter, and DPmax-after of Group 1 (A1, A2, A3), Group 2 (B1, B2, B3), Group 3 (C1, C2, C3). Pmax = maximum foam propensity during aeration, Pafter = foam propensity immediately after the aeration, and DPmax-after = Pmax Pafter.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
8
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx
3.4. The recommended foaming potential aeration test in AD systems The proposed set-up is an apparatus-specific method which was optimized according to a survey of 16 AD samples. These represent various types of ADs installed at WWTPs. Manure-based AD sludges and other types have not been tested, and modifications of the method may be needed. However, the general principles of foam formation and stabilization are similar and can be investigated by such tests. Generally, it seems that measurement of Hmax and Hafter is a direct way of analyzing the foam properties. The values of Hmax and Hafter can be used to classify digester samples roughly as (a) Group 1: Hmax < 200 mm at all TS concentrations, no abundant surfactants or foam-forming microorganisms to cause foam; (b) Group 2: Hmax > 200 mm, and DHmax-after > 200 mm at some of TS concentrations, absence of foam-forming microorganisms; (c) Group 3: Hmax > 200 mm at some of TS concentrations, and DHmax-after < 200 mm at all TS concentrations, abundant surfactant, and foam-forming. Group 1 represents non-foaming status, Groups 2 and 3 are most likely pre-foaming and actually foaming, respectively, but difficult to distinguish. Extensive surveillance of plants with the proposed classification system will enable the determination of more precise cut-off values. The cut-off values can also be used to correlate the foaming status with operational conditions to identify key factors leading to foaming episodes. As Fig. 4 shows, Hmax and Hafter at TS higher than 50 g/L are similar for the three groups, especially for Groups 2 and 3. This could be due to the high viscosity at high TS. On the other hand, 10 g/L is also problematic because the foam floated easily out of the tube, which occurred for two ADs in Group 2 (Soeholt and Bjergmarken RT2). This was the reason for the flat curves at the beginning of these two ADs (Fig. 4-B1). Thus, we recommend that the TS is kept in the range of 20–30 g/L, which is also the normal TS concentration in most ADs. Dilution with sludge water or up-concentration by settling and removal of supernatant is needed when TS is outside this interval. Fig. 6 showed that the foam propensity varied with TS in the three groups. Pmax and Pafter of nearly all the samples decreased with increasing TS, except Pafter of Damhusaaen RT1 at TS = 20 g/ L. The trends on foam propensity of these three groups showed the same as foam height, indicating that foam propensity can also be used as a parameter to classify the foam property. Foam stability is another parameter sometimes applied to describe/define the foaming potential. To measure foam stability, typically, the foam height should be recorded at a certain time point/after aeration for a fixed time period, often 1 h after aeration stops (Boe et al., 2012; Ganidi et al., 2011; Kougias et al., 2013). However, according to our results, several samples had zero foam height 1 h after the aeration stopped (results not shown). This agrees with Ganidi et al. (2011) and Kougias et al. (2013), who found that this value was not sensitive enough to predict or link to foaming in digesters. Moreover, this parameter is not relevant for full-scale operation since a state without gas production for 1 h does not take place in full-scale digester tanks. Attempts to use the ratio of Hafter to Hmax to define the foam stability were unsuccessful as Group 1 and 2 were undistinguishable (Fig. S1 in supplementary materials). Instead, we suggest that the parameter of DHmax-after or Pmax-Pafter (DPmax-after) could be used to evaluate the foam stability.
4. Conclusion A simple protocol for foaming potential aeration test for monitoring foam properties in digester sludge is proposed: a simple apparatus (tube, inner diameter 3 cm, 85 cm height), flow rate of 500 mL/min, digester sludge volume of 50 mL, and 10 min aeration
time with the TS range of 20–30 g/L. The digester sludge can be classified into three groups, based on the foam height and propensity. It is a simple, fast, and inexpensive method that produces comparative results in a short time suitable for digester surveillance. Acknowledgements We thank the plant operators for providing digester sludge samples. This research was funded by the Innovation Fund Denmark through the project NomiGas [grant number 1305-00018B]. The Chinese Scientific Research Council is acknowledged for providing financial support for C. Jiang and R. Qi. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.wasman.2018.02. 021. References Albertsen, M., Karst, S.M., Ziegler, A.S., Kirkegaard, R.H., Nielsen, P.H., 2015. Back to basics – the influence of DNA extraction and primer choice on phylogenetic analysis of activated sludge communities. PLoS One 10, e0132783. APHA, 2005. Standard Methods for the Examination of Water and Wastewater. American Public Health Association, Washington, DC, USA. Barjenbruch, M., Hoffmann, H., Kopplow, O., Trnckner, J., 2000. Minimizing of foaming in digesters by pre-treatment of the surplus-sludge. Water Sci. Technol. 42, 235–241. Blackall, L.L., Marshall, K.C., 1989. The mechanism of stabilization of actinomycete foams and the prevention of foaming under laboratory conditions. J. Ind. Microbiol. 4, 181–187. Blackall, L.L., Harbers, A.E., Greenfield, P.F., Hayward, A.C., 1991. Activated sludge foams: effects of environmental variables on organism growth and foam formation. Environ. Technol. 12, 241–248. Boe, K., Kougias, P.G., Pacheco, F., O-Thong, S., Angelidaki, I., 2012. Effect of substrates and intermediate compounds on foaming in manure digestion systems. Water Sci. Technol. 66, 2146–2154. Fryer, M., Gray, N.F., 2012. Foaming scum index (FSI) – a new tool for the assessment and characterisation of biological mediated activated sludge foams. J. Environ. Manage. 110, 8–19. Fryer, M., O’Flaherty, E., Gray, N.F., 2011. Evaluating the measurement of activated sludge foam potential. Water 3, 424–444. Ganidi, N., Tyrrel, S., Cartmell, E., 2009. Anaerobic digestion foaming causes – A review. Bioresour. Technol. 100, 5546–5554. Ganidi, N., Tyrrel, S., Cartmell, E., 2011. The effect of organic loading rate on foam initiation during mesophilic anaerobic digestion of municipal wastewater sludge. Bioresour. Technol. 102, 6637–6643. Hladikova, K., Ruzickova, I., Klucova, P., Wanner, J., 2002. An investigation into studying of the activated sludge foaming potential by using physicochemical parameters. Water Sci. Technol. 46, 525–528. Hug, T., 2006. Characterization and controlling of foam and scum in activated sludge systems. Swiss Federal Institute of Technology Zurich. Kougias, P.G., Boe, K., Angelidaki, I., 2013. Effect of organic loading rate and feedstock composition on foaming in manure-based biogas reactors. Bioresour. Technol. 144, 1–7. Moeller, L., Görsch, K., 2015. Foam formation in full-scale biogas plants processing biogenic waste. Energy Sustain. Soc. 5, 1. Moeller, L., Eismann, F., Wißmann, D., Nägele, H.-J., Zielonka, S., Müller, R.A., Zehnsdorf, A., 2015. Innovative test method for the estimation of the foaming tendency of substrates for biogas plants. Waste Manage. 41, 39–49. Oerther, D.B., de los Reyes, F.L., de los Reyes, M.F., Raskin, L., 2001. Quantifying filamentous microorganisms in activated sludge before, during, and after an incident of foaming by oligonucleotide probe hybridizations and antibody staining. Water Res. 35, 3325–3336. Pagilla, K.R., Craney, K.C., Kido, W.H., 1997. Causes and effects of foaming in anaerobic sludge digesters. Water Sci. Technol. 36, 463–470. Ross, R.D., Ellis, L.M., 1992. Laboratory-scale investigation of foaming in anaerobic digesters. Water Environ. Res. 64, 154–162. Subramanian, B., Pagilla, K.R., 2014. Anaerobic digester foaming in full-scale cylindrical digesters – effects of organic loading rate, feed characteristics, and mixing. Bioresour. Technol. 159, 182–192. Subramanian, B., Miot, A., Jones, B., Klibert, C., Pagilla, K.R., 2015. A full-scale study of mixing and foaming in egg-shaped anaerobic digesters. Bioresour. Technol. 192, 461–470. Suhartini, S., Heaven, S., Banks, C.J., 2014. Comparison of mesophilic and thermophilic anaerobic digestion of sugar beet pulp: Performance, dewaterability and foam control. Bioresour. Technol. 152, 202–211.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021
C. Jiang et al. / Waste Management xxx (2018) xxx–xxx Torregrossa, M., Viviani, G., Vinci, V., 2005. Foaming estimation tests in activated sludge systems. Acta Hydrochim. Hydrobiol. 33, 240–246. Westlund, Å.D., Hagland, E., Rothman, M., 1998. Operational aspects on foaming in digesters caused by Microthrix parvicella. Water Sci. Technol. 38, 29–34.
Wickham, H., 2016. ggplot2: Elegant Graphics for Data York. Zábranská, J., Dohányos, M., Jenícˇek, P., Zaplatílková, contribution of thermophilic anaerobic digestion to wastewater sludge treatment. Water Sci. Technol. 46,
9 Analysis. Springer, New P., Kutil, J., 2002. The the stable operation of 447–453.
Please cite this article in press as: Jiang, C., et al. Monitoring foaming potential in anaerobic digesters. Waste Management (2018), https://doi.org/10.1016/ j.wasman.2018.02.021