The efficacy of sonication as a pre-treatment to anaerobic digestion (AD) was assessed using thickened waste activated sludge (TWAS). Efficiency was measured in relation to solubilisation, dewaterability, and AD performance. Eighteen experimental conditions were evaluated at low frequency (20 kHz), duration (2–10 s), amplitude (∼8–12 μm) and applied pressure (0.5–3.0 barg), using a sonix™ patented titanium sonoprobe capable of delivering an instantaneous power of ∼6 kW provided by Doosan Enpure Ltd (DEL). An optimised experimental protocol was used as a pre-treatment for biochemical methane potential (BMP) testing and semi-continuous trials. Four digesters, with a 2-L working volume were operated mesophilically (37 ± 0.5 °C) over 22 days. The results showed that the sonix™ technology delivers effective sonication at very short retention times compared to conventional system. Results demonstrate that the technology effectively disrupts the floc structures and filaments within the TWAS, causing an increase in solubilisation and fine readily digestible material. Both BMP tests and semi-continuous trials demonstrated that sonicated TWAS gave higher biodegradability and methane potential compared to untreated TWAS. Partial-stream sonication (30:70 sonicated to untreated TWAS) resulted in a proportionate increase in biogas production illustrating the benefits of full-stream sonication.

INTRODUCTION

The efficiency of anaerobic digestion (AD) is often limited by the rate at which anaerobic bacteria perform cell lysis, i.e. degradation of the primary organic material within the digesters (Hogan et al. 2004; Appels et al. 2008). From the previous research experiences, Doosan Enpure Ltd (DEL) found that within the municipal wastewater sector, an AD plant operating within mesophilic ranges 35 °C–45 °C will have an average hydraulic retention time (HRT) of 15–18 days, destroying about 45% of its organic matter content during that time and as a consequence producing 1 m3 of biogas per kg of volatile solids (VS) destroyed with a volumetric methane content of ∼60% (Rooksby 2001; Hogan et al. 2004; Collett et al. 2005).

It is generally accepted that operational issues facing AD plants treating municipal wastes include the ever increasing amount of sludge from municipal water to be treated as regulatory and other pressures require greater treated standards before discharge. This can lead to shorter retention times within the digester and as a consequence poorer solids destruction rates. The second issue relates to the quality and nature of the sludge (Rooksby 2001; Collett et al. 2005). This sludge, a by-product of secondary treatment known as surplus activated sludge (SAS), is mainly composed of cellular material which, by its very nature, is difficult to digest anaerobically because the cell lysis by the AD consortia is relatively slow and inefficient (Rooksby 2001; Appels et al. 2008). Typically, the SAS is pre-thickened before digestion, referred to as thickened or waste SAS (TWAS or WAS) cannot be fed into the AD process at proportions much greater than 25% of the feed without compromising the stability or performance of the digester within the HRTs applied (Rooksby 2001; Hogan et al. 2004; Collett et al. 2005). Pre-treatment, therefore, is often required to improve TWAS characteristics prior to AD (Rooksby 2001; Collett et al. 2005; Appels et al. 2008).

Sonication, as a pre-treatment method, has been widely applied to wastewater treatment for the destruction of organic pollutants into less recalcitrant and therefore more readily digestible compounds (Pilli et al. 2011). This treatment results in a reduction of sludge particle size, the release of soluble carbohydrates and organic substances, and a reduction of bacterial population (Chu et al. 2001). The basic principle of sonication is the formation and collapse of gas bubbles resulting from rapid changes in pressure (known as cavitation). When cavitation occurs and the bubbles collapse at the cell wall, extreme pressure and temperature conditions can disrupt the cell structure and floc matrix, inducing cell lysis. Simultaneously as the cells are lysed, the cavitation process acts upon the released intracellular material and other non-cellular particles which break down into smaller and more readily degradable fractions. In addition to the mechanical disruption, chemical reactions also take place as a result of the formation of OH, HO2, H radicals, which also have a beneficial impact on sludge disintegration (Pilli et al. 2011). Several factors affecting the cavitation phenomena during sonication pre-treatment include ultrasonic frequency and intensity, ultrasonic density, solvent characteristics (e.g. viscosity, surface tension, vapour pressure), external pressure, the presence of gas, particulate matter and temperature (Pilli et al. 2011), sludge type, total solids (TS) content, particle size, contact time, pH and power input (Show et al. 2007).

Low power ultrasound technologies are not new and have been used commercially in non-destructive applications such as echography or cleaning baths (Rooksby 2001; Pilli et al. 2011). There is a large library of published material which has reported the use of ultrasound in sludge treatment, however much of this work has involved small reaction chambers and long retention times (up to 2.5 h) (Braguglia et al. 2008; Salsabil et al. 2009; Ruiz-Hernando et al. 2014). In addition, these studies have applied sonication in the range of ∼20 to 3,250 kHz, so far, a review of more recently presented results considering the typically used frequency of ∼20 kHz attributed to a strong hydrodynamic shear resulting from cavitation effects and better sludge disintegration. Such condition enhances the rate of degradation and overall methane yield during the AD process.

Many studies have reported the positive effects of the application of sonication for treating sludges (TWAS or WAS) prior to AD for many decades in a variety of small scale laboratory-based studies under mesophilic or thermophilic condition (Pérez-Elvira et al. 2009; Salsabil et al. 2009; Erden & Filibeli 2010; Ruiz-Hernando et al. 2014). Whilst these studies have demonstrated an increase in soluble chemical oxygen demand (sCOD), destruction of VS and increased biogas production at laboratory scale, the application at full scale has, until now, not been proven to be commercially viable.

The focus of this work was to determine optimal operating condition for sonication pre-treatment to mimic a pilot scale or full scale application operating continuously or in-line mode with frequencies above the audible range, starting at about 20 kHz, taking into account the efficacy of floc or cell disruption, organic materials disintegration and full scale techno-economic considerations such as retention time and energy input which would influence process viability. This project also investigated in more detail the impact of sonication pre-treatment on the TWAS characteristics, biodegradability and AD performance under batch and semi-continuous condition. The results obtained in this study is used as the foundation to on-going pilot scale sonication pre-treatment under recirculation mode to process complex organic material such as agricultural crop residues.

MATERIALS AND METHODS

Feedstock and inoculum

TWAS was collected from full scale activated sludge plants at Minworth Sewage Treatment Works, Severn Trent Water, Birmingham, UK, following polymer addition (at a dose of 3.611 ml l−1 or 2.96 kg tonne−1 dry SAS) and belt thickening. This site is one of the largest sites in the EU treating domestic and industrial discharges (as well as additional sludge from smaller works around the region) from a population equivalent of 1.7 million people. TWAS samples were collected immediately after a thickening process. Characterisation and pre-treatment of TWAS was conducted on the same day as the sample collection to avoid any deterioration and the subsequent treated (or untreated) sample was then fed to the digesters for evaluation.

Inoculum for biochemical methane potential (BMP) test trials was prepared using digestate collected from a full scale mesophilic AD plant treating (Minworth Sewage Treatment Works, Severn Trent Water, Birmingham, UK). Inoculum was sieved through a 1 mm screen to remove larger particles and degassed for 48 h at a temperature of 37 °C prior to the BMP test. The characteristics of both the inoculum and untreated TWAS are shown in Table 1.

Table 1

Characteristics of inoculums and untreated TWAS

ParametersUnitsInoculumsUntreated TWASa
pH – 7.54 5.96–6.73 
Temperature °C 18.4 10.3–19.3 
EC mS cm−1 1.16 0.65–1.71 
TSb %ww 3.09 ± 0.001 2.12–6.60 
VSb %TS 59.67 ± 0.001 77.42–78.67 
Ashb %ww 1.25 ± 0.005 0.62–1.42 
CST 289.7 35.0–1535.7 
tCODb g l−1 30.63 ± 0.06 29.56–82.21 
sCODb mg l−1 1098.0 ± 4.24 106.3–2801.0 
ParametersUnitsInoculumsUntreated TWASa
pH – 7.54 5.96–6.73 
Temperature °C 18.4 10.3–19.3 
EC mS cm−1 1.16 0.65–1.71 
TSb %ww 3.09 ± 0.001 2.12–6.60 
VSb %TS 59.67 ± 0.001 77.42–78.67 
Ashb %ww 1.25 ± 0.005 0.62–1.42 
CST 289.7 35.0–1535.7 
tCODb g l−1 30.63 ± 0.06 29.56–82.21 
sCODb mg l−1 1098.0 ± 4.24 106.3–2801.0 

aUntreated TWAS used in optimisation trials.

bData are expressed as an average of triplicate samples.

Experimental procedures

Sonication pre-treatment

Experiments were carried out using approximately 55% of the available instantaneous power from a 6 kW sonix™ patented titanium sonoprobe and stack device (provided by Doosan Enpure Ltd (DEL), UK) at an ultrasonic frequency of 20 ± 0.5 kHz. The operational parameters were adjusted to establish an optimised protocol. These parameters included duration (2–10 s), amplitude (70–100% or ∼8–12 μm) and applied pressure (0.5–3.0 barg). The selected operation parameters were based on historical empirical data from previous research experiences carried out by DEL. For example, Rooksby (2001) reported that sonication exposure using sonix™ only takes place in a matter of few seconds (i.e. 1.5 s), which is short and quick compared to other equipment used in other studies (Bougrier et al. 2006; Appels et al. 2008; Braguglia et al. 2008; Muller et al. 2009; Pérez-Elvira et al. 2009; Salsabil et al. 2009; Erden & Filibeli 2010; Ruiz-Hernando et al. 2014). There are two phases in the process optimisation of sonication.

The key development in the 6 kW sonix™ device, used in this study, is the focusing of the ultrasound by means of a radial horn forged from a single piece of virgin 6/4 titanium and heat treated before use (Hogan et al. 2004). The intense energy focused by the horn is sufficient to cause significant cavitation within the fluid. This cavitation generates extremely high temperatures and pressures of many thousands of degrees and atmospheres at the foci of the collapsing bubbles. As the energy is so intensely focused retention times within the reactor can be reduced to 1–2 s, previously unachievable with other systems using the more conventional block horns. The effect of this is that sonix™ power requirements of 4–5 kJ l−1 are at the very least between 25% and an order of magnitude lower than competing systems. Also, its design allows for continuous in-line applications rather than in batch mode and offers greater efficiency for use in full scale applications (personal communication of Tony Amato).

In each test run, approximately 3 L of TWAS was added to the test rig through the inlet (Figure 1). The pre-determined operating parameters (i.e. amplitude and duration) were adjusted via the personal computer (PC) and software. The reaction chamber is pressurised using an external compressor and actual pressure measured using the systems own integrated gauge. The cavitation process was initiated and, upon completion, any change in pressure recorded. Excess pressure was released from the reaction chamber by gently opening the pressure valve. The first 60 mL of sample (within the static region of the chamber below the horn) was removed from the chamber via the outlet valve. Subsequently, a 750 mL aliquot of sample was removed (this volume corresponds to the volume in direct contact with the horn forming the ultrasonic reaction zone). Sample pH and temperature were recorded immediately following treatment.
Figure 1

Bench scale test rig.

Figure 1

Bench scale test rig.

Specific energy or Es (in kJ kg−1 TS) is calculated based on formula as shown in Equation (1) (Bougrier et al. 2005): 
formula
1
where P is ultrasonic power in kW, t is ultrasonic time in seconds (s), V is volume of sonicated sludge in litres, and TS0 is initial TS concentration in kg TS l−1. The power figures used in the calculation are based on what is assumed to be drawn by the equipment and, therefore, any loss between the input and the end of the radial horn/sonoprobe have been ignored.

BMP test procedures

The BMP tests were performed in an Automatic Methane Potential Test System (AMPTS II) (Bioprocess Control AB, Sweden) for 34 days to comply with the typical length of BMP test (>30 day) and allow for accurate determination of cumulative methane production, as well as to continue and investigate whether methane production is ceased due to the inhibitory conditions or the complete degradation of the organic materials (Erden & Filibeli 2010; Lim & Fox 2013). Three TWAS samples were evaluated in the BMP test including untreated TWAS, sonicated TWAS, and mixed TWAS (sonicated:untreated TWAS with ratio of 30:70 (w/w)). A ratio 30:70 of sonicated:untreated TWAS was selected based on the historical experimental data, previously reported by Hogan et al. (2004). Blank samples were included in the test to measure the indigenous methane production from the inoculums. The positive controls (α-cellulose from Sigma-Aldrich, Germany) were used to test the activity of the inoculum. Mixed TWAS was tested to investigate the possibility of the application of partial-stream sonication prior to the AD process. Partial-stream sonication may enhance methane potential and provide more economic benefits (e.g. energy and cost reduction) compared to that of full-stream sonication.

The test was carried out at 37 °C in triplicates with an inoculum to substrate ratio (I/S ratio) of 5.2:1 (untreated TWAS) and 5.1:1 (sonicated and mixed TWAS) on a VS basis. The total volume of each reactor was 400 mL and the following settings were applied: CH4 concentration assumption (60%), CO2 content in flushed gas (0%), and intermittent stirring, set at 1.5 Hz with 60 s on and 60 s off to ensure an adequate mixing in the reactors. CO2 is removed from the biogas via an alkaline (3 M NaOH) fixing unit and methane production is recorded using real-time data logging.

Semi-continuous AD trials

Experiments were carried out in duplicate using four mesophilic (37 ± 0.5 °C), continuously-stirred tank reactor digesters, each with a 2-L working volume. Digester temperature was maintained through an external indirect heating element wrapped around the digester. Semi-continuous operation was achieved by daily feeding of TWAS (either untreated or sonicated), via a hole in the top plate; digestate was removed through an outlet port in the base plate immediately before feed addition. The digesters were connected to a gas collection bag (Tedlar bag). Biogas productivity was measured continuously apart from a brief interval during feeding and digestate removal. Daily biogas volume (measured and reported) are corrected to standard temperature and pressure of 0 °C and 101.325 kPa as described by Walker et al. (2009).

For the first day of the trial, the mesophilic digesters (M1–M4) were initially fed with untreated and sonicated TWAS at an organic loading rate (OLR) of 1.4 g VS L−1 day−1 which then, on day 2, started to feed at target OLR of 2.4–3.1 g VS L−1 day−1, depended on the initial VS value of TWAS samples on a daily basis. Digesters were then operated for 22 days HRT. There was no trace element addition to any of the digesters during these trials.

Analytical methods

TS and VS contents were determined according to Standard Method 2540 G (APHA 2005). pH was measured using a Hanna HI-208 meter (Hanna Instruments, Italy) with a combination glass electrode and temperature-compensating device (APHA 2005). The tCOD and sCOD were analysed using a HACH Reactor Digestion Method and then measured with DR 2800 Spectrophotometer (HACH Lange, UK). Total alkalinity measurement was based on Standard Method 2320B (APHA 2005).

TS and VS destruction was calculated on a weekly basis by mass balance. For this purpose, it was assumed that the wet weight (ww) of digestate removed was equal to that of feedstock added, minus the weight of biogas removed. The weight of biogas removed was estimated from the weekly average volume and gas composition in terms of % CH4 and CO2 (ignoring water vapour and other gases).

Samples for elemental composition (C, H, O, N, S) content determination were sent to an external laboratory for analysis by inductively coupled plasma mass spectrometry (ICP-MS) (Aston University Birmingham, UK). The values were used to calculate theoretical methane concentration using Buswell equation, which then further used in semi-continuous AD trials.

Dewaterability was measured by capillary suction time (CST) test using a Triton-WRPL type 304 M single CST apparatus and paper (Triton Electronics Ltd, UK).

To evaluate the efficiency of sonication pre-treatments, the solubilisation degree (SD) (%) of COD was calculated according to Equation (2) (Donoso-Bravo et al. 2011): 
formula
2
where CODs is the soluble COD after pre-treatment, CODSo is the soluble COD in the untreated sample and CODT the total COD of untreated sample.

RESULTS AND DISCUSSION

Process optimisation

Untreated TWAS characteristics

From Table 1, it can be seen that the TWAS had a TS and VS in the range of 2–7%ww and 1.9–6%ww, respectively, giving a VS content within the range of 96–99% of TS. The TS concentration of TWAS samples was within the range of TS concentration considered for these trials as suitable for sonication pre-treatment (≤8%) (Collett et al. 2005). The pH values were in the range of 5.96 to 6.73. The CST value varied from 35 to 1,536 s, indicating that TWAS samples have poor dewaterability characteristics. The sCOD concentration varied from 110.7 to 2,235 mg L−1. The variation in all parameters measured was possibly due to operational changes within the activated sludge plant (ASP) at Minworth.

Phase I

Effects of sonication pre-treatment on COD solubilisation
There were no significant changes in pH of samples prior to and after sonication. There was also minimal change in temperature or concentrations of TS and VS between the untreated and sonicated TWAS. The results, however, showed that sonication pre-treatment increased sCOD in the liquid fraction of the TWAS sample by 1.2- to 9-fold, indicating an enhanced disintegration and solubility of TWAS (Table 2 and Figure 2(a)). This is in agreement with other studies reported in the literature (Bougrier et al. 2006; Pérez-Elvira et al. 2009; Salsabil et al. 2009; Erden & Filibeli 2010; Ruiz-Hernando et al. 2014).
Table 2

Characteristics of TWAS before and after sonication treatment

TrialTreatmentSonication condition
Es (kJ kg−1 TS)sCOD (mg l−1)CST (s)
Amp. (%)sHRT (sec)P (barg)
Phase I – Optimisation 
Untreated (TS of TWAS = 2.62%ww) 275.3 35.0 
S3 70 1.5 774.25 754.3 196.3 
Untreated (TS of TWAS = 6.59%ww) 517.7 1,009.8 
S4 70 1.5 341.91 1,142.7 1,365.4 
S5 70 10 1.5 404.45 776.0 1,374.0 
S6 80 10 1.5 481.33 1,065.0 1,555.4 
Untreated (TS of TWAS = 6.60%ww)  2,235.0 1,535.7 
S7 90 10 1.5 569.88 3,561.3 2,311.9 
S9 70 10 0.5 324.44 2,932.3 2,044.3 
Untreated (TS of TWAS = 5.21%ww) 110.7 99.8 
S10 70 10 1.0 481.59 497.7 422.5 
S11 70 10 2.0 615.24 680.7 951.3 
S12 70 10 2.5 654.85 994.7 1,046.4 
Untreated (TS of TWAS = 6.20%ww) 287.3 604.2 
S1 70 1.5 88.52 325.7 617.0 
S2 70 1.5 174.36 439.0 796.1 
S8 100 10 1.5 751.17 1,095.3 1,993.8 
S13 70 10 3.0 557.81 570.3 1,413.4 
Phase II – Optimisation 
Untreated (TS of TWAS = 4.20%ww) 106.3 35.8 
S14 80 2.0 167.06 209.3 52.9 
S15 90 2.5 235.14 321.3 127.2 
S16 100 3.0 303.06 375.0 148.8 
Untreated (TS of TWAS = 6.08%ww) 233.7 295.5 
S17 90 2.0 146.63 441.7 640.5 
S18 100 2.0 176.11 556.3 763.6 
TrialTreatmentSonication condition
Es (kJ kg−1 TS)sCOD (mg l−1)CST (s)
Amp. (%)sHRT (sec)P (barg)
Phase I – Optimisation 
Untreated (TS of TWAS = 2.62%ww) 275.3 35.0 
S3 70 1.5 774.25 754.3 196.3 
Untreated (TS of TWAS = 6.59%ww) 517.7 1,009.8 
S4 70 1.5 341.91 1,142.7 1,365.4 
S5 70 10 1.5 404.45 776.0 1,374.0 
S6 80 10 1.5 481.33 1,065.0 1,555.4 
Untreated (TS of TWAS = 6.60%ww)  2,235.0 1,535.7 
S7 90 10 1.5 569.88 3,561.3 2,311.9 
S9 70 10 0.5 324.44 2,932.3 2,044.3 
Untreated (TS of TWAS = 5.21%ww) 110.7 99.8 
S10 70 10 1.0 481.59 497.7 422.5 
S11 70 10 2.0 615.24 680.7 951.3 
S12 70 10 2.5 654.85 994.7 1,046.4 
Untreated (TS of TWAS = 6.20%ww) 287.3 604.2 
S1 70 1.5 88.52 325.7 617.0 
S2 70 1.5 174.36 439.0 796.1 
S8 100 10 1.5 751.17 1,095.3 1,993.8 
S13 70 10 3.0 557.81 570.3 1,413.4 
Phase II – Optimisation 
Untreated (TS of TWAS = 4.20%ww) 106.3 35.8 
S14 80 2.0 167.06 209.3 52.9 
S15 90 2.5 235.14 321.3 127.2 
S16 100 3.0 303.06 375.0 148.8 
Untreated (TS of TWAS = 6.08%ww) 233.7 295.5 
S17 90 2.0 146.63 441.7 640.5 
S18 100 2.0 176.11 556.3 763.6 
Figure 2

Results of process optimisation: trends of sCOD and CST in Phase I (a) and (b) and Phase II (c) and (d).

Figure 2

Results of process optimisation: trends of sCOD and CST in Phase I (a) and (b) and Phase II (c) and (d).

The results, highlighted in Figure 3(a)3(c), also suggest that the degree of sludge disintegration is directly related to the Es input (expressed as the amount of energy delivered related to duration, amplitude, and applied pressure). For example, increasing duration from 2 to 10 s resulted in a cell disintegration proportional to the Es applied during the sonication process, with an R2 of 0.8785 (Figure 3(a)). This is in agreement with findings from Huan et al. (2009), who also found that sonication at higher Es resulted in an increased concentration of collapsing sonication bubbles, which in turn resulted in greater mechanical disintegration of the cell membranes. This subsequently resulted in a significant improvement in whole cell disruption and a greater concentration of intracellular material released into the solution. At Es higher than 600 kJ kg−1 TS, however, the sCOD release seems to decrease. Salsabil et al. (2009) and Erden & Filibeli (2010) observed similar trends whereby higher Es lead to a decrease in sCOD concentration. This is possibly due to exhaustion of readily disintegrable organic particles present in the sludge or exhaustion of dissolved gases that aid the sonication bubble formation (Pilli et al. 2011).
Figure 3

The degree of sludge solubilisation (SDCOD) as function of the sonication Es. Phase I: (a) 2–10 s HRT, 70% amp., 1.5 barg pressure; (b) 10 s HRT, 70–100% amp., 1.5 barg pressure; and (c) 10 s HRT, 70% amp., 0.5–3.0 barg pressure. Phase II: (d) 2 s HRT, 90% amp., 2.0–3.0 barg pressure. TS concentration of TWAS samples varied from 2.61–6.68%ww.

Figure 3

The degree of sludge solubilisation (SDCOD) as function of the sonication Es. Phase I: (a) 2–10 s HRT, 70% amp., 1.5 barg pressure; (b) 10 s HRT, 70–100% amp., 1.5 barg pressure; and (c) 10 s HRT, 70% amp., 0.5–3.0 barg pressure. Phase II: (d) 2 s HRT, 90% amp., 2.0–3.0 barg pressure. TS concentration of TWAS samples varied from 2.61–6.68%ww.

Variation in untreated TWAS characteristics (e.g. TS concentration and initial sCOD) may also contribute to the relative increase in sCOD released after sonication. This study also illustrated that higher initial TS concentrations (>than 6%ww) inhibited release of sCOD release (i.e. lower than expected increase in %sCOD). This has been confirmed by Show et al. (2007) that the TS concentration as one of the factors limiting optimum sonication and subsequent disintegration and disruption of the cell or floc structure. They found that optimal TS concentration for sonication pre-treatment was in the range of 2.3%–3.2%. Similarly, this study also indicated that when the initial sCOD of untreated TWAS is lower (<600 mg L−1), a greater percentage increase in sCOD was observed (Table 2 and Figure 2(a)). Higher initial sCOD concentrations may be an indication of the availability and nature of the organic fractions within the TWAS (e.g. sCOD, particulate COD, inert COD, readily biodegradable COD, etc.) (Dulekgurgen et al. 2006). Measuring the COD fractions within the sample prior to and after sonication may improve our understanding of the mechanism of disruption and conversion.

Effects of sonication pre-treatment on dewaterability

In terms of CST values, as expected, the sonication pre-treatment increased CST values indicating that the floc structure had been fragmented into finer particle and the cellular material had been disrupted resulting in a release of extracellular polymeric substances (EPS) (Table 2 and Figure 2(b)). This is reflected as a deterioration in dewaterability characteristics of the TWAS sample. The impact of sonication on floc structure is well documented and supports the observation that disintegration of sludges with a corresponding decrease in floc size and increase in number of fine particles leads to deterioration in sludge dewaterability (Erden & Filibeli 2010). A reduction in particle size (evidenced as an increase in CST values) confirms that even a short-period of sonication can be effective in producing more readily digestible material which, in turn, can enhance biogas and methane production.

The increase of Es was also proportional to % increase of CST (Figure 3(a)3(c)). At Es lower than 600 kJ kg−1 TS, the fragmentation of cellular material was reduced as indicated by a small increase in CST values, however, the CST values were still higher than the acceptable CST value for better dewatering processes. At Es above 600 kJ kg−1 TS, the sonicated TWAS samples exhibited higher CST values as indicated by the presence of more fine particles. Many have reported that increasing Es in sonication pre-treatment causes more production of fine particles and more release of EPS in aqueous phase, thus the separation of the liquid from the solid fraction becomes more difficult (Chu et al. 2001; Erden & Filibeli 2010). An increase in the number of fine particles' presence in sludge increases the surface area, which contributed to a great amount of water attached onto the large surfaces of fine particles (bound water) (Liming et al. 2009). In this regards, it is clear that Es of sonication pre-treatment affect the TWAS solubility and dewaterability, as have been reported in other studies (Bougrier et al. 2006; Show et al. 2007; Huan et al. 2009; Muller et al. 2009; Salsabil et al. 2009; Erden & Filibeli 2010).

Phase II

From the optimisation trials in Phase I, a process duration of 2 s was selected to further explore the impact of Es at different amplitudes (∼8–12 μm) and applied pressures (1.5–3.0 barg). The short sonication period of 2 s was selected to reduce the operational cost, as well as to enhance the efficiency and the economy and commercial viability of the sonication technology. This trial demonstrated that, although the TWAS samples were treated with a short-period sonication, the impact on sCOD released and CST were obvious (Table 2 and Figure 2(c) and 2(d)). There was also seemingly small changes in TS and VS concentration before and after sonication pre-treatment, as well as a decrease in pH values and an increase in temperature after sonication, however, these changes were not significant. This is in agreement with Erden & Filibeli (2010) who found that sonication pre-treatment has changed the physico-chemical characteristics of WAS, such as a decrease in pH value because of acidic compound formation resulted from floc disintegration and lipids hydrolysis to volatile fatty acids, or an increase in temperature due to increasing Es applied.

Effects of sonication pre-treatment on COD solubilisation

The results indicate that increasing the Es supplied during sonication pre-treatment resulted in an increase in sCOD released into liquid fraction, in agreement with Salsabil et al. (2009). In this study, the results showed that increasing amplitude has a significant increase of soluble COD. Increasing amplitude and applied pressure resulted in an increase in Es applied (in the range of ∼147 to ∼304 kJ kg−1 TS), causing a subsequent increase sCOD released by 2- to 3.5-fold, respectively. The degree of sludge solubilisation from optimisation trials also indicated a proportional increase to the logarithm of the Es input, with R2 of 0.8388 (Figure 3(d)). This finding further confirmed previous results that the degree of sludge solubilisation is directly related to the Es input consumed during sonication process, and the TS concentration of TWAS samples.

Effects of sonication pre-treatment on dewaterability

The results also demonstrated that increasing sonication energy greatly increased the CST values of TWAS sample, with R2 of 0.9984 (Figure 3(d)). Sonication was not shown to improve sludge filterability, however this was not deemed to be a problem for full scale operation as sonicated TWAS would be fed directly to the AD plant where finer particles and greater surface area is considered to be beneficial to biogas production, as stated by Erden & Filibeli (2010).

In summary, with consideration upon the historical data from DEL and the findings from the initial optimisation trials, the optimal sonication conditions suggested for pre-treatment were a HRT of 2 s, amplitude of 90% and applied pressure of 2.0–2.5 barg. These parameters were selected as the pre-treatment protocol for further batch and semi-continuous AD evaluations.

Effects of sonication pre-treatment on batch AD process (BMP test)

The specific methane production (SMP) was calculated for each sample by dividing the final (day 34) cumulative methane volume by the weight (on a VS basis) of substrate added. The values obtained for blank, α-cellulose, untreated TWAS, sonicated TWAS and mixed TWAS samples are shown in Figure 4 and Table 3. The SMP value obtained for the blank sample was 0.057 m3 CH4 kg−1 VSadded. The SMP of positive control (α-cellulose) was 0.330 m3 CH4 kg−1 VSadded. This is significantly lower than the value of the theoretical BMP of cellulose based on its molecular composition (C6H10O5) of 0.415 m3 CH4 kg−1 VSadded. The difference in the SMP of cellulose is possibly due to the activity of the inoculum used in these trials. Optimised inoculum could enhance SMP.
Table 3

Summary results of BMP test

Sample IDsCOD (mg L−1)Net methane prod. (Nml)Methane increaseSMP (m3 kg−1 VSadded)SMP increase (%)
Sonication at 2 s sHRT, 2.5 barg & 90% amp. Equivalent Es input = ∼171 kJ kg−1 TS 
 Inoculum 1,221 398 n.a. 0.057 n.a. 
α-cellulose (C6H10O5n.a. 382 n.a. 0.330 n.a. 
 Untreated TWAS 460 228 1x 0.203 – 
 Sonicated TWAS 740 272 1.19x 0.234 15.5 
 Sonicated TWAS:Untreated TWAS (30:70) 644 245 1.08 0.214 5.3 
Sample IDsCOD (mg L−1)Net methane prod. (Nml)Methane increaseSMP (m3 kg−1 VSadded)SMP increase (%)
Sonication at 2 s sHRT, 2.5 barg & 90% amp. Equivalent Es input = ∼171 kJ kg−1 TS 
 Inoculum 1,221 398 n.a. 0.057 n.a. 
α-cellulose (C6H10O5n.a. 382 n.a. 0.330 n.a. 
 Untreated TWAS 460 228 1x 0.203 – 
 Sonicated TWAS 740 272 1.19x 0.234 15.5 
 Sonicated TWAS:Untreated TWAS (30:70) 644 245 1.08 0.214 5.3 

Note: Data represented are average values of triplicate sample analysis.

Figure 4

SMP from the BMP test. Data are expressed as means of triplicate samples and bars represent range.

Figure 4

SMP from the BMP test. Data are expressed as means of triplicate samples and bars represent range.

For untreated TWAS, the SMP was 0.203 m3 kg−1 VSadded. The methane potential for untreated TWAS obtained in this study, however, was higher than reported elsewhere in the literature. For example, Lim & Fox (2013) reported that the SMP of untreated TWAS at an I/S ratio of 1:1, 1:3 and 1:8 were 0.0514, 0.0763, and 0.0219 m3 CH4 kg−1 VSadded. The differences observed in this study could be attributed to the nature of TWAS samples or inoculum collected from this particular wastewater treatment plant.

The results showed that sonication pre-treatment at Es of 101 kJ kg−1 TS enhanced the biodegradability of TWAS by 15.5%, with a SMP of 0.234 m3 kg−1 VSadded. The results are in agreement with other studies which have demonstrated enhanced methane production as a result of sonication of TWAS samples (Braguglia et al. 2008; Salsabil et al. 2009). The mixed TWAS samples with a ratio of 30:70 (sonicated TWAS:untreated TWAS) gave an SMP of 0.214 m3 kg−1 VSadded, which was 5.3% higher than that of untreated TWAS. This result confirmed that blending a proportion of sonicated sludge with the main feed into the digester will also have a beneficial impact on overall biogas production. However, the partial-stream sonication gave a proportionate increase based on the actual ratio of sonicated to untreated TWAS in the feed. This is in agreement to the findings from Pérez-Elvira et al. (2010).

The BMP findings demonstrate that the sonication pre-treatment in this study had superior performance (in terms of percentage increase in biogas relative to power input) compared to the other technologies reported in the literature (Table 4). The measured value based on an assumed methane content within the biogas of 60% resulted in a 0.0905% increase per kJ kg−1 TS, this was in a good agreement with the theoretical value based on the Buswell equation of 0.0989% increase per kJ kg−1 TS. These relatively close values indicate that both approaches are valid and support the validity of the experimental results. However, BMP tests are not dynamic and continuous; therefore, a semi-continuous trial was performed to provide a more accurate and realistic estimation of efficiency and performance, particularly on improving biogas (methane) production from TWAS.

Table 4

DEL performance compared with literature reported values

SourcesType of sludgeEs (kJ kg−1 TS)Biogas increase (%)Normalised % increase in biogas relative to power input (% increase/ kJ kg−1 TS)
Batch AD (BMP test) 
 This study (2015) TWAS 171 15.5a 0.0905 
16.9b 0.0989 
Ruiz-Hernando et al. (2014)  WAS 27,000 13.0 0.0005 
Salsabil et al. (2009)  TWAS 3,600 39.8 0.0111 
31,500 143.0 0.0045 
108,000 362.4 0.0034 
Erden & Filibeli (2010)  WAS 9,690 44.0 0.0045 
Pérez-Elvira et al. (2009)  AS 5,040 6.0 0.0012 
14,400 17.0 0.0012 
29,160 32.0 0.0011 
38,880 42.0 0.0011 
Appels et al. (2008)  WAS 168 45.0 0.2676 
456 21.3 0.0468 
768 15.4 0.0200 
1,248 18.6 0.0149 
1,560 16.5 0.0106 
5,850 10.8 0.0018 
8,180 2.3 0.0003 
Bougrier et al. (2006)  TWAS 6,250 47.1 0.0075 
9,350 51.1 0.0055 
Semi-continuous AD 
 This study (2015) TWAS 151 5.9 0.0391 
Pérez-Elvira et al. (2009)  AS 38,880 55.2 0.0014 
38,880 3.08 0.0001 
38,880 8.24 0.0002 
Braguglia et al. (2008)  WAS 2,500 21.9 0.0088 
5,000 37.9 0.0076 
5,000 14.6 0.0029 
Muller et al. (2009)  TWAS 6,409 18.8 0.0029 
4,068 12.0 0.0029 
4,205 23.0 0.0055 
SourcesType of sludgeEs (kJ kg−1 TS)Biogas increase (%)Normalised % increase in biogas relative to power input (% increase/ kJ kg−1 TS)
Batch AD (BMP test) 
 This study (2015) TWAS 171 15.5a 0.0905 
16.9b 0.0989 
Ruiz-Hernando et al. (2014)  WAS 27,000 13.0 0.0005 
Salsabil et al. (2009)  TWAS 3,600 39.8 0.0111 
31,500 143.0 0.0045 
108,000 362.4 0.0034 
Erden & Filibeli (2010)  WAS 9,690 44.0 0.0045 
Pérez-Elvira et al. (2009)  AS 5,040 6.0 0.0012 
14,400 17.0 0.0012 
29,160 32.0 0.0011 
38,880 42.0 0.0011 
Appels et al. (2008)  WAS 168 45.0 0.2676 
456 21.3 0.0468 
768 15.4 0.0200 
1,248 18.6 0.0149 
1,560 16.5 0.0106 
5,850 10.8 0.0018 
8,180 2.3 0.0003 
Bougrier et al. (2006)  TWAS 6,250 47.1 0.0075 
9,350 51.1 0.0055 
Semi-continuous AD 
 This study (2015) TWAS 151 5.9 0.0391 
Pérez-Elvira et al. (2009)  AS 38,880 55.2 0.0014 
38,880 3.08 0.0001 
38,880 8.24 0.0002 
Braguglia et al. (2008)  WAS 2,500 21.9 0.0088 
5,000 37.9 0.0076 
5,000 14.6 0.0029 
Muller et al. (2009)  TWAS 6,409 18.8 0.0029 
4,068 12.0 0.0029 
4,205 23.0 0.0055 

aCalculated using an assumption of 60% methane concentration in biogas.

bCalculated using theoretical biogas composition based on the Buswell equation.

Semi-continuous AD performance

Variability of TWAS feedstock

The characteristics of TWAS feedstock used in this trial are shown in Table 5. There was no significant difference found in terms of tCOD from untreated and treated TWAS. The sCOD of untreated TWAS was in the range of 234.5 ± 0.71–646.0 ± 0.00 mg L−1, which was suitable for the trials based on the historical data from DEL. Occasionally, operational events on site (outside of our control) resulted in samples with higher than expected sCOD values. These samples were deemed to be unrepresentative and may not, therefore, demonstrate the impact of sonication on degradation of organic material to sCOD.

Table 5

Characteristics of TWAS before and after sonication over 22 days

ParametersUntreated TWASSonicated TWAS
Specific energy (Es) (kJ kg−1 TS) 129.13–214.95 
pH 6.34–6.61 6.29–6.61 
TS (%ww) 4.41–6.51 4.40–6.52 
VS (%ww) 3.46–5.08 3.48–5.12 
VS (%TS) 77.61–79.67 78.42–79.46 
Ash (g l−19.50–14.24 9.20–13.95 
CST (s) 39.4–987.7 163.1–1,569.9 
tCOD (g l−152.4–75.8 51.7–73.3 
sCOD (mg l−1234.5–646.0 437.0–1084.5 
Elemental analysis (in %TS) 
 C 37.95 37.39 
 H 6.55 6.43 
 N 6.61 6.60 
 O (by difference) 48.2 48.9 
 S 0.69 0.64 
ParametersUntreated TWASSonicated TWAS
Specific energy (Es) (kJ kg−1 TS) 129.13–214.95 
pH 6.34–6.61 6.29–6.61 
TS (%ww) 4.41–6.51 4.40–6.52 
VS (%ww) 3.46–5.08 3.48–5.12 
VS (%TS) 77.61–79.67 78.42–79.46 
Ash (g l−19.50–14.24 9.20–13.95 
CST (s) 39.4–987.7 163.1–1,569.9 
tCOD (g l−152.4–75.8 51.7–73.3 
sCOD (mg l−1234.5–646.0 437.0–1084.5 
Elemental analysis (in %TS) 
 C 37.95 37.39 
 H 6.55 6.43 
 N 6.61 6.60 
 O (by difference) 48.2 48.9 
 S 0.69 0.64 

After sonication pre-treatment, as expected, both sCOD and CST values were increased (Table 5), indicating an improvement in the solubility and disintegration of TWAS samples. In general, the variation of CST values recorded in this study was due to the variation of initial TS and VS values of daily TWAS feedstock collected from the ASP plant.

AD performance parameters

Results obtained from each of the trial digesters are shown in Figure 5. The average values for key parameters are given in Table 6 taken over an HRT of 22 days, period. These results are for samples taken from the digester prior to fresh feed being added and therefore accurately reflect the conditions within the digester at that time. The tCOD from all digesters increased throughout the digestion period (Figure 5(a)): with average values of ∼36.14 g L−1 (digester fed with untreated TWAS), and 36.27 g L−1 (digester fed with sonicated TWAS). This was attributed to an accumulation of tCOD caused by a partial or incomplete removal during the methanogenic phase of digestion. This indicates the steady state (or stable) conditions were not achieved as typically it requires at least 3 HRT (Appels et al. 2008). Digesters fed with untreated or sonicated TWAS exhibited similar sCOD (Figure 5(b)), and the sCOD values remained constant in the range of ∼1.1–1.25 g L−1.
Table 6

Average of performance indicators for duplicate digesters over 22 day period

ParametersUntreated TWASSonicated TWAS
Specific biogas production (Nm3 kg−1 VSadded day−10.290 0.307 
SMPa(Nm3 kg−1 VSadded day−10.165 0.173 
Specific biogas production (Nm3 kg−1 VSadded day−10.551 0.578 
SMPa (Nm3 kg−1 VSadded day−10.314 0.325 
Vol. biogas production (l l−1 day−10.739 0.830 
Vol. methane production (l l−1 day−10.421 0.467 
Digestate TS (g l−136.49 36.76 
Digestate VS (g l−124.52 24.47 
VS destruction (%) 54.27 54.58 
tCOD (g l−136.14 36.27 
sCOD (g l−11.22 1.26 
pH 7.35 7.35 
Total alkalinity (g CaCO3 kg−1ww) 8.38 8.14 
Theoretical CH4 concentrationb (%) 57.0 56.3 
CST (s) ∼934 ∼1,114 
ParametersUntreated TWASSonicated TWAS
Specific biogas production (Nm3 kg−1 VSadded day−10.290 0.307 
SMPa(Nm3 kg−1 VSadded day−10.165 0.173 
Specific biogas production (Nm3 kg−1 VSadded day−10.551 0.578 
SMPa (Nm3 kg−1 VSadded day−10.314 0.325 
Vol. biogas production (l l−1 day−10.739 0.830 
Vol. methane production (l l−1 day−10.421 0.467 
Digestate TS (g l−136.49 36.76 
Digestate VS (g l−124.52 24.47 
VS destruction (%) 54.27 54.58 
tCOD (g l−136.14 36.27 
sCOD (g l−11.22 1.26 
pH 7.35 7.35 
Total alkalinity (g CaCO3 kg−1ww) 8.38 8.14 
Theoretical CH4 concentrationb (%) 57.0 56.3 
CST (s) ∼934 ∼1,114 

aCalculated using theoretical methane concentration.

bCalculated using Buswell equation.

Figure 5

Trend of tCOD, sCOD, pH, total alkalinity and CST at digesters fed untreated and sonicated TWAS. Data are expressed as the average of duplicate digesters.

Figure 5

Trend of tCOD, sCOD, pH, total alkalinity and CST at digesters fed untreated and sonicated TWAS. Data are expressed as the average of duplicate digesters.

Although minor fluctuations were observed, all digesters had a pH in the range of 7.2 to 7.5 (Figure 5(d)), and within normally acceptable optimum range for an AD process. There was also little difference between all digesters in terms of total alkalinity (Figure 5(e)). No significant differences in TS and VS were observed across pairs of digesters fed with untreated TWAS and sonicated TWAS (Figure 5(f) and 5(g)). Under mesophilic conditions, the average VS destruction was ∼54.3% (untreated TWAS), and ∼54.6% (sonicated TWAS). Although there was a slightly higher VS destruction (∼0.3%) than those without sonication, there is still some biogas production increased in sonicated TWAS. This was explained by solids retention time (SRT) applied, as reported by Appels et al. (2008) that there was a minor improvement in VS destruction when SRT applied was more than 12–13 days. Also, an increase VS destruction is linked to enhanced biogas production, as the volume of biogas or methane produced is correlated as a function of organic matter being destroyed by the microconsortia in the digester (Ahn & Forster 2002). Other factors that influence the VS destruction in this study are the TWAS composition and OLR applied which likely to limit the solids hydrolysis and the overall biochemical reactions (Braguglia et al. 2008); the sonication intensity supplied to the sludge (Show et al. 2007; Salsabil et al. 2009); and the re-flocculation phenomenon of the sonicated sludge due to the release of intracellular or extracellular material and the changes of the sludge's physico-chemical characteristics (Erden & Filibeli 2010). Indeed, the VS destruction values from this semi-continuous trial also reflected the BMP test findings, where sonicated TWAS has higher VS destruction leading to higher biogas (or methane) production than those without sonication.

Biogas and methane production

Results for biogas and methane production during the trial are shown in Figure 6 and average values are given in Table 6. Specific biogas and methane production in digesters fed with untreated TWAS increased during the experiment, giving the average of 0.290 l biogas g−1 VSadded day−1 and 0.165 l CH4 g−1 VSadded day−1 over the 22-day period. Digesters fed with sonicated TWAS exhibited an increase in specific biogas production by 5.9%, with average values of 0.307 Nm3 kg−1 VS day−1. The SMP increased by 4.8% giving the average value of 0.173 Nm3 CH4 kg−1 VS day−1. This percentage increase was lower than that of found in the BMP test (15.5% improvement).
Figure 6

Daily specific biogas and methane production from semi-continuous trials. Data are expressed as the average of duplicate digesters.

Figure 6

Daily specific biogas and methane production from semi-continuous trials. Data are expressed as the average of duplicate digesters.

There are several factors which may influence methane production and explain the difference seen. Firstly, the short digestion period of 22 days (1 HRT) applied in the chemostat trial with daily feeding caused partial degradation of organic matter, as well as an accumulation of undigested organic matter which may inhibit the AD performance. BMP tests, however, were operated for 34 days under batch conditions which allowed complete degradation of organic materials within TWAS. It has been reported that at short HRTs, the microorganism consortia in the digester are under considerable hydraulic stress which may result in microbial washout, thus reducing the efficiency of the overall digestion process (Appels et al. 2008). Secondly, the daily variability of TWAS feedstock (e.g. VS concentration) may impact upon the performance of a semi-continuous digestion process, unlike BMP tests where the feedstock used showed little variation under controlled conditions. It is widely acknowledged that biogas yield depends on the concentration of VS in the input feedstock (Braguglia et al. 2008). Finally, as previously explained, differences may be attributed to the variation in the specific energy provided during the sonication process prior to chemostat trials, with average values of 151.01 kJ kg−1 TS and in the BMP trials (Es applied was higher at 171 kJ kg−1 TS). This may impact on the amount of finer, readily digestable organic matter released after sonication.

Several studies on the semi-continuous AD of sonicated sludge using higher specific energy resulted in a similar biogas increase. For example, Braguglia et al. (2008), who studied mesophilic AD of WAS (untreated and sonicated) at OLR of 1.4 and 0.7 g VS l−1 day−1, with HRT of 10 and 20 days. The results indicated that specific biogas production from untreated biogas was 0.198 m3 kg−1 VS day−1. While for sonicated WAS, the biogas production increase by 21.89% (HRT of 10 days, Es of 2,500 kJ kg−1 TS), 14.64% (HRT of 10 days, Es of 5,000 kJ kg−1 TS), and 37.88% (HRT of 20 days, Es of 5,000 kJ kg−1 TS). Pérez-Elvira et al. (2009) studied the mesophilic AD of activated sludge (untreated and sonicated) operated for HRT of 15 and 20 days. The results demonstrated that specific biogas production for untreated AS was 0.325 m3 kg−1 VSadded day−1 and 0.425 m3 kg−1 VSadded day−1, for HRT of 15 and 20 days, respectively. Sonication pre-treatment at optimum Es of 38,880 kJ/kgTS resulted in an increase in biogas production by 3.08% (HRT of 15 days) and 8.24% (HRT of 20 days). When these results were standardised based on % increase in biogas relative to power input, the % biogas obtained from this study was 3.91%, for instance, higher than that of reported by Braguglia et al. (2008) at 0.29–0.88% and by Pérez-Elvira et al. (2009) at 0.01–0.14%. This further confirms that the sonication device provided by DEL was more effective in disintegrating the floc structure and cellular material within the TWAS samples. The comparison of methane potential from semi-continuous AD in this study with other studies can be seen in Table 4.

Dewaterability characteristics

The dewaterability characteristic of the digestates from mesophilic AD of TWAS is shown in Table 6. Daily measurement of CST showed that filterability of the mesophilic digestate deteriorated over the period of 22 days (Figure 5(h)). Digestate from digesters fed with untreated TWAS exhibited an increase in the CST from ∼214 s to ∼934 s (in average) by the end of experiment. A similar trend was observed with digesters fed with sonicated TWAS, where the average CST values increased from ∼216 s to ∼1,114 s. The presence of finer particle in the digested sonicated TWAS was contributed to its higher CST values. The composition and the variability of TWAS collected from the plant may also contributed to the deterioration of the digestate dewaterability, as stated by Liming et al. (2009). Another possible contributing factor is the accumulation of hydrophobic by-products (e.g. proteins) (Muller et al. 2009; Erden & Filibeli 2010). Neyens & Baeyens (2003) reported that the hydrophobic fraction of EPS was made up of predominantly proteins and high concentrations of hydrophilic fractions may contribute to the more hydrophilic nature of sludge. The authors reported that EPS composition (relative proportions of proteins and carbohydrates) determined the overall surface charge of the sludge flocs, whereby a high surface charge represents poor settleability and poor dewaterability characteristics. In addition, they stated that water can become entrapped when higher concentrations of EPS are present leading to higher viscosity which, in turn, can lead to a deterioration in sludge dewaterability. Visual observations made during this study supported this theory.

CONCLUSIONS

The design of the sonix™ radial horn technology is innovative and allows for the application of high intensity energy and ultrasound with relatively short retention times. This technology is effective in disrupting the floc structures and filaments within the TWAS, thus increasing sCOD concentration and producing finer particles. Sonication pre-treatment improved sludge characteristics and thus performance of the AD process, as confirmed by BMP and semi-continuous trials. The key mechanism is mechanical disruption which results in a more readily digestible organic carbon. Partial-stream sonication at a ratio of 30:70 (sonicated:untreated TWAS) produced a proportional increase in biogas and methane production. Fine particles and an accumulation of undigested organic materials present in digesters caused a moderate deterioration in sludge dewaterability.

ACKNOWLEDGEMENTS

Authors want to thank Doosan Enpure Ltd for financial supports and equipment contributions; as well as to Severn Trent Water for their access to site and samples.

REFERENCES

REFERENCES
APHA
2005
Standard Methods for the Examination of Water and Wastewater
, 21st edn.
American Public Health Association, American Water Works Association, Water Environment Federation, Washington
,
DC, USA
.
Appels
L.
Baeyens
J.
Degrève
J.
Dewil
R.
2008
Principles and potential of the anaerobic digestion of waste-activated sludge
.
Prog. Energy Combust. Sci.
34
(
6
),
755
781
.
Bougrier
C.
Carrère
H.
Delgenès
J. P.
2005
Solubilisation of waste-activated sludge by ultrasonic treatment
.
Chem. Eng. J.
106
(
2
),
163
169
.
Chu
C. P.
Chang
B.-V.
Liao
G. S.
Jean
D. S.
Lee
D. J.
2001
Observations on changes in ultrasonically treated waste-activated sludge
.
Water Res.
35
(
4
),
1038
1046
.
Collett
S.
Amato
T.
Griffiths
M.
Rooksby
F.
2005
Sonix™ has gone global
. In:
Proceedings of the 9th European Biosolids Conference
,
14–17 November 2004
,
West Yorkshire, UK, Paper 60, ref. 3
.
Dulekgurgen
E.
Doğruel
S.
Karahan
Ö.
Orhon
D.
2006
Size distribution of wastewater COD fractions as an index for biodegradability
.
Water Res.
40
(
2
),
273
282
.
Hogan
F. M.
Mormede
S.
Clark
P. B.
Crane
M. J.
2004
Ultrasonic sludge treatment for enhanced anaerobic digestion
.
Water Sci. Technol.: Water Supply
50
(
9
),
25
32
.
Huan
L.
Yiying
J.
Mahar
R. B.
Zhiyu
W.
Yongfeng
N.
2009
Effects of ultrasonic disintegration on sludge microbial activity and dewaterability
.
J. Hazard. Mater.
161
(
2
),
1421
1426
.
Liming
S.
Peipei
H. E.
Guanghui
Y. U.
Pinjing
H. E.
2009
Effect of proteins, polysaccharides, and particle sizes on sludge dewaterability
.
J. Environ. Sci.
21
,
83
88
.
Pérez-Elvira
S.
Fdz-Polanco
M.
Plaza
I.
Garralón
G.
Fdz-Polanco
F.
2009
Ultrasound pre-treatment for anaerobic digestion improvement
.
Water Sci. Technol.
60
,
1525
1532
.
Pérez-Elvira
S.
Ferreira
L.
Donoso-Bravo
A.
Fdz-Polanco
M.
Fdz-Polanco
F.
2010
Full-stream and part-stream ultrasound treatment effect on sludge anaerobic digestion
.
Water Sci. Technol.
61
(
6
),
1363
1372
.
Pilli
S.
Bhunia
P.
Yan
S.
LeBlanc
R.
Tyagi
R.
Surampalli
R.
2011
Ultrasonic pretreatment of sludge: a review
.
Ultrason. Sonochem.
18
(
1
),
1
18
.
Rooksby
F.
2001
Ultrasound for enhancement of sludge digestion
. In:
CIWEM/Aqua Enviro Conference
,
Wakefield, UK
, pp.
1
10
.
Ruiz-Hernando
M.
Martín-Díaz
J.
Labanda
J.
Mata-Alvarez
J.
Llorens
J.
Lucena
F.
Astals
S.
2014
Effect of ultrasound, low-temperature thermal and alkali pre-treatments on waste activated sludge rheology, hygienization and methane potential
.
Water Res.
61
,
119
129
.
Salsabil
M. R.
Prorot
A.
Casellas
M.
Dagot
C.
2009
Pre-treatment of activated sludge: effect of sonication on aerobic and anaerobic digestibility
.
Chem. Eng. J.
148
(
2
),
327
335
.
Show
K.-Y.
Mao
T.
Lee
D.-J.
2007
Optimisation of sludge disruption by sonication
.
Water Res.
41
(
20
),
4741
4747
.