Review of Removal of Fats, Oil and Greases from Effluents from Meat Processing Plants

Review of Removal of Fats, Oil and Greases from Effluents from Meat Processing Plants

Prepared by:

Neil McPhail CSIRO Animal, Food and Health Sciences

Date Submitted:

21 May 2014

Date Published:

March 2015

Published by:

Australian Meat Processor Corporation

Acknowledgements The project was undertaken by CSIRO and funded by the Australian Water Recycling Centre of Excellence under the Commonwealth’s National Urban Water and Desalination Plan. Further information: www.csiro.au and www.australianwaterrecycling.com.au

Disclaimer This publication is published by the Australian Meat Processor Corporation Ltd ABN 67 082 373 448. Care is taken to ensure the accuracy of the information contained in this publication. However, AMPC cannot accept responsibility for the accuracy or completeness of the information or opinions contained in this publication. No part of this work may be reproduced, copied, published, communicated or adapted in any form or by any means (electronic or otherwise) without the express written permission of Australian Meat Processor Corporation Ltd. All rights are expressly reserved. Requests for further authorisation should be directed to the Company Secretary, AMPC, Suite 1, Level 5, 110 Walker Street Sydney NSW. For further information please contact AMPC on 02 8908 5500 or [email protected]

Contents Acknowledgements........................................................................................................................ 2

Executive summary ......................................................................................................................... 5 1 2 3

Introduction.............................................................................................................................. 6 Value of fat ................................................................................................................................ 8 Primary treatment to reduce wastewater fat levels ................................................. 9

3.1

Screening ........................................................................................................................... 9

3.4

Physico-chemical treatment.....................................................................................10

3.2 3.3

4

3.5 3.6

Saveall ...............................................................................................................................10 Dissolved air flotation (DAF) ...................................................................................10 Hydrocyclones ...............................................................................................................10 Electrocoagulation .......................................................................................................11

Anaerobic treatment ...........................................................................................................11

4.1 4.2 4.3 4.4

The anaerobic digestion of FOG ..............................................................................11

Operational issues with anaerobic digestion of FOG ......................................12 Effect of FOG level on digestion ..............................................................................13 Pre-treatment technologies......................................................................................14

4.4.1

Saponification........................................................................................................15

4.4.4

Ultrasound ..............................................................................................................16

4.4.2 4.4.3

5

6

4.4.5

Thermal hydrolysis .............................................................................................15 Enzyme hydrolysis ..............................................................................................15

Combination pre-treatments ..........................................................................17

Digester configuration ........................................................................................................18

5.1 5.2 5.3

Two-phase digestion ...................................................................................................18

Temperature-phased digestion ..............................................................................19

Inverted anaerobic sludge blanket ........................................................................19

Conclusions and recommendations ..............................................................................20

Abbreviations ..................................................................................................................................22

References ........................................................................................................................................23

3

Figures Figure 1: Covered anaerobic lagoon (Courtesy A.J. Bush & Sons) ................................ 7 Figure 2: First full-scale inverted anaerobic sludge blanket reactor in Portugal (from Picavet & Alves, 2013).....................................................................................................20

Tables

Table 1: Range in concentration of contaminants in combined wastewater streams from three Australian abattoirs ................................................................................ 6 Table 2: Potential biogas production from different classes of substrates (adapted from Alves et al 2009) .................................. Error! Bookmark not defined. Table 3: Value of 1 kg of fat recovered at different stages of the process ................. 9 Table 4: Summary of treatments to reduce fat levels ......................................................11 Table 5: Summary of pre-treatments to anaerobic digestion of oily wastewaters ...............................................................................................................................................................18

4

Executive Summary The potentially high fat content of effluent from meat processing plants can lead to treatment issues in covered anaerobic lagoons such as extended digestion times, scum formation and inhibition of methane production. In most cases, keeping the fat out of the water and processing it in the existing on-site rendering plant provides the meat processor with the best return. Gross fat particles are difficult to digest and should either be prevented from entering the anaerobic system or removed by screening and flotation. Otherwise they are likely to float to the surface of the lagoon and form a scum or crust. A range of pre-treatments to anaerobic digestion to reduce the effects of fat have been investigated by researchers. These pre-treatments which include saponification, thermal and enzymatic hydrolysis and homogenisation by ultrasound appear to offer limited benefit at considerable added processing cost. Two-stage anaerobic treatment with a separate vessel or pond for the initial hydrolysis and acidogenesis may offer benefits for the treatment of high-fat wastewaters and further investigation under Australian conditions is warranted. New high-rate anaerobic systems are also being developed in Europe to specifically handle high-fat wastewaters. The long-term performance of the recently developed inverted anaerobic sludge blanket (IASB) reactor should be monitored. Laboratory trials should be undertaken to determine the effects of different fat levels in abattoir effluent on anaerobic digestion to address a lack of basic data on the performance of anaerobic treatment of wastewater from Australian meat processing plants. These should be complemented by continued investigations of the performance of existing and new covered anaerobic lagoons.

5

1

Introduction

Liquid effluent from meat processing plants is variable in nature depending on the processes taking place within the abattoir and can be highly contaminated with organic material. A survey of wastewater streams in three Australian plants processing beef and sheep by Jensen & Batstone (2012) found wide variations in levels of contaminants between different streams within the plants and different plants. Table 1 shows the ranges for the combined streams for the three plants. Table 1: Range in concentration of contaminants in combined wastewater streams from three Australian abattoirs

Parameter Total COD Soluble COD Total solids Oil & grease N P

Range (mg/L) 9,600 – 12,900 890 – 1,970 4,300 – 8,400 790 – 3,350 230 – 260 30 – 50

They found the rendering plant to be the major contributor of oil and grease (O&G) in the effluent, with the rendering stickwater containing 5,500 to 6,000 mg/L O&G which contributes up to 1,900 kg per day to fat, oil and grease (FOG) in the wastewater. This is equivalent to approximately 14.5 kg FOG per tonne hot standard carcase weight (tHSCW). Due to high levels of BOD and nutrients, abattoir effluent must be treated to varying extents prior to disposal to sewer, land irrigation or waterways. Green (1992) stated that anaerobic treatment is ideal for meatworks effluent because it has a high BOD, nutrients and a high temperature of up to 40°C. Anaerobic systems are more cost effective than aerobic treatment as there is no aeration cost, less sludge is produced and there is the potential to produce a combustible fuel, methane, as a major end product (Greenfield & Johns 1992). Anaerobic ponds or lagoons have been used as the first stage of the secondary treatment of abattoir effluent for many years in Australia and have been preferred over high-rate anaerobic digestion systems due to simplicity of operation, land availability, the Australian climate and their inherent stability. Although anaerobic ponds were initially uncovered, covered anaerobic lagoons (CALs - Figure 1) are becoming increasingly common for several reasons. Uncovered anaerobic ponds however, have one significant problem: they can emit considerable odour, particularly during commissioning (Green 1992). Historically, the meat industry has relied on the crust formed from floating solids and FOG to reduce odours (White et al 2013) but this is not generally sufficient to satisfy current environmental demands and urban 6

development that may have encroached on meat plants. For these reasons, plus the ability to capture methane and reduce greenhouse gas emissions, and the improved durability and affordability of polymeric covers, most plants are covering existing ponds or installing new CALs.

Figure 1: Covered anaerobic lagoon (Courtesy A.J. Bush & Sons)

The practice of allowing a crust to form is not desirable for efficient and effective operation of a CAL, since the floating crust poses a risk to the integrity of the cover and the biogas capture system (White et al 2013). The layering of recalcitrant fats within the digester also limits the solid-liquid contact area required for efficient conversion to biogas. For these reasons, and others further discussed in Section 4.2, effective primary treatment of the wastewater is essential to break down FOG into a dispersed and useable form or to ensure removal of excess FOG and solids. Despite the issues that FOG can lead to with anaerobic digestion, lipids are an ideal substrate for methane production, since their degradation produces more biogas with higher methane content than the degradation of proteins and carbohydrates does (Table 2).

7

Table 2: Potential biogas production from different classes of substrates (adapted from Alves et al 2009)

Component Lipids Carbohydrates Proteins

Biogas (L/g) 1.425 0.830 0.921

Methane (%) 69.5 50.0 68.8

Jensen & Batstone (2012) also measured the biochemical methane potential of the different streams from the abattoirs surveyed and found that the rendering stream with the high FOG level of 6,000 mg/L provided the highest methane potential of about 650 L/kg VS added. This review aims to present the current knowledge on the effect of FOG content of wastewater on the operation and performance of anaerobic treatment systems and techniques available to optimise operation and biogas production.

2

Value of fat

Waste fat from the meat processing industry is a potentially valuable byproduct and can be recovered at basically three stages of the process: 1. Collection from the processing areas and rendered to produce a highgrade tallow; 2. Recovery from the wastewater primary treatment and processed to produce a low-grade tallow; and 3. Conversion to methane in an anaerobic digestion system. A 1% FFA (free fatty acid) tallow produced from fresh fat was valued at $900 per tonne in January 2014 (MLA 2014). Tallow produced from fat that has had prolonged contact with water will have a much higher FFA and hence lower sale value. ‘Saveall tallow’ produced from saveall or DAF float may have a FFA of 10 – 40% and a value of about $750 per tonne (W. Spooncer pers. comm. Feb. 2014). If the fat in the effluent is allowed to flow to the anaerobic lagoon and is fully converted to biogas it could have a significant value. When anaerobically digested, one gram of fat can produce 1.425 L of biogas at 69.5% methane (Alves et al 2009). Therefore 1 kg of fat in the wastewater could produce 0.99 m3 of CH 4 with a net heating value of 35 MJ/m3. If this is used on the plant to replace natural gas priced at 1.5 to 2 c per MJ, the value of a kilogram of fat could be $0.52 to $0.69 ($520 - $690 per tonne). Table 3 shows a summary of estimated potential sale values of fat recovered at each of these stages of the process (the cost of processing must be deducted). 8

Table 3: Value of 1 kg of fat recovered at different stages of the process

Process stage Freshly rendered fat Primary effluent treatment Covered anaerobic lagoon

Form 1% FFA tallow Saveall tallow Methane

Value ($/kg fat) 0.90 0.75 0.52 – 0.69

In most cases, keeping the fat out of the water and processing it in the existing on-site rendering plant provides the meat processor with the best return. However, if the fat can be anaerobically digested without detrimental effect to the process, the value of the methane produced may be greater than the net value of a low-grade tallow.

3

Primary treatment to reduce wastewater fat levels

The main aim of primary treatment is to remove coarse suspended solids, oil and grease and recover solids and fat for treatment and sale. Fat and other recovered solids can be processed in the rendering plant to produce meat meal and tallow. Fine fat particles and emulsified fat, oil and grease will need to be reduced by other means. The performance of this first stage of wastewater treatment will have an effect on the operation of the next stage and the quality of the discharged effluent (Husband 1992). The options available for primary treatment are discussed below.

3.1 Screening Screening is normally the first stage and aims to remove solid material including fat particles from the wastewater. It is desirable to separate the fat-laden and non fat-laden streams, at least until after they have been screened (Husband 1992) as paunch (stomach contents) and manure recovered from the low-fat ‘green’ stream will have a different disposal route to the fat and protein solids. A range of screen types have been developed and some that are successfully used in the meat processing industry are (MLA 2007): •

•

Static and vibrating screens: Static screens are normally of wedge wire, while vibrating screens are of woven wire. Both of these screen types are more suited to manure-laden streams as they can be blinded by fat particles. Rotary screens: Cylindrical rotary screens are usually of wedge wire with the liquid entering in the centre and the solids discharged at the other end. They are more suited to fatty effluents than other screen types.

9

•

•

Screw press: Several designs of screw press are available which combine a screen with a screw auger to de-water solids. These are suited to fibrous material such as manure and paunch solids. Other screening arrangements such as the Baleen Filter have also been successfully applied to a fatty wastewater stream.

The above screening processes will remove most of the gross fat particles.

3.2 Saveall A saveall is the traditional settling tank which allows suspended fat to rise to the surface to be recovered whilst the heavier solids sinks. A minimum residence time of 30 minutes is recommended (MINTRAC 2012) but elevated effluent temperatures will reduce efficiency (MLA 2007). A welldesigned and managed saveall will recover 85% of the fat in the effluent stream (Husband 1992).

3.3 Dissolved air flotation (DAF) DAF normally involves pressurisation of a portion of the wastewater stream by air injection. The release of the pressurised stream in a circular or rectangular tank, results in the release of micro-bubbles of air which carry fat and lighter solid particles to the surface where they can be recovered. A well-designed DAF can recover in excess of 90% of the fat (Husband 1992) but removal efficiencies can be poor when the water temperature is above 39°C (MLA 2007). DAF units can also be fitted with parallel plates to increase the separation surface and enhance performance.

3.4 Physico-chemical treatment The effluent stream feeding a DAF can be dosed with chemicals or polymers to coagulate solids to improve removal efficiency. Oil and grease removal efficiencies of 89-98% have been reported (Johns 1995). This process results in large quantities of float which must either be further processed or disposed of and the use of chemicals greatly increases the operating cost (MINTRAC 2012).

3.5 Hydrocyclones Hydrocyclones use centrifugal force to separate solids and fat from water. Originally designed for the oil and mining industries, hydrocyclones have a small footprint and retention time and no moving parts (MLA 2007). Hydrocyclones remove 40-90% of oil and grease and are a capital costeffective method compared with other technologies (GHD 2003). 10

3.6 Electrocoagulation The wastewater is passed across iron and aluminium electrode plates supplied with a DC current which solubilises metal ions promoting coagulation and flocculation of the organic material which is subsequently separated in a downstream unit (Tetreault 2003). Oil and grease separation efficiencies of 90 – 95% were achieved when treating rendering stickwater. A summary of FOG removal efficiencies for primary treatment technologies is presented in Table 4. Table 4: Summary of treatments to reduce fat levels

4

Process Screens

FOG removal Gross fat particles

Saveall

Up to 85%

DAF

~90%

Physico-chemical

89 – 98%

Hydrocyclones

40 – 90%

Electrocoagulation

90 – 95%

Comments Careful screen selection needed for fatty wastewater to avoid blockage. High temperatures reduce efficiency. Reduced efficiency when temperature above 39°C. High chemical cost and large sludge volumes. Vulnerable to blockage from fine solids and fat. Unproven with dilute high-volume streams

Anaerobic treatment 4.1 The anaerobic digestion of FOG During anaerobic digestion, lipids are first hydrolysed to glycerol and free long chain fatty acids (LCFAs) (Cirne et al 2007, Pittaway 2011). Glycerol is converted to acetate by acidogenesis, while the LCFAs are degraded anaerobically via the β-oxidation pathway to acetate and hydrogen, which are subsequently converted to methane (methanogenesis). The following reaction expresses the degradation of LCFAs via the β-oxidation pathway: CH 3 (CH 2 ) n COOH + 2H 2 O -> CH 3 (CH 2 ) n-2 COOH + CH 3 COOH + 2H 2 (Kim et al 2004) The greatest reduction of COD occurs when the methanogens convert hydrogen and volatile fatty acids into methane and carbon dioxide (Pittaway 2011). 11

Anaerobic digestion can occur in the mesophilic temperature range of 3040°C or in the thermophilic range (55-75°C). Li et al (2013) demonstrated higher biogas production and degradation of organics when digesting FOG with primary sludge at 55°C than at the mesophilic temperature of 37°C. When Gannoun et al (2009) compared thermophilic digestion of abattoir wastewater with mesophilic digestion in an anaerobic filter they achieved higher COD removal and biogas yield under thermophilic conditions. However supplementary heating is required to maintain thermophilic conditions. A study of covered and uncovered anaerobic lagoons treating abattoir wastewater found a FOG removal rate of 94.7% and 92.8% respectively (UNSW, 1998). A mass balance found that of the 94.7% of the FOG removed in the CAL, 25.9% accumulated in the scum layer, 0.4% in the sludge and the remaining 68.4% was most likely converted to biogas. In the uncovered pond, 41.4% accumulated in the scum layer, 2.5% in the sludge and the remaining 56.1% was converted. The scum layer was a significant contributor to the FOG removal from the wastewater but as it thickens, a reduction in the effective volume of the lagoon would occur over time.

4.2 Operational issues with anaerobic digestion of FOG Alves et al (2009) reported two main issues with anaerobic treatment of wastewaters containing FOG: (i) sludge flotation and biomass washout due to the adsorption of lipids/LCFA on the biomass, and (ii) inhibition of the acetogenic bacteria and methanogenic archaea by LCFA. In a review, Long et al (2012) reported that anaerobic digestion of high-lipid wastes causes inhibition of acetoclastic and methanogenic bacteria, substrate and product transport limitation, sludge flotation, digester foaming, blockages of pipes and pumps, and clogging of gas collection and handling systems. Other researchers also report inhibition of methane generation due to lipid levels in the waste, but most of these studies relate to high-rate digestion systems, such as the upflow anaerobic sludge blanket (UASB). Sayed et al (1988) found that the protein fraction of slaughterhouse wastewater degraded more efficiently than the fats fraction. During laboratory trials using a slaughterhouse wastewater with 50-100 mg/L fat, at a process temperature of 30°C, 45.5 to 47% of the fat was degraded compared with 87% of the protein. They recommended that a good fat separator be installed to prevent excessive scum layers in the reactor. Based on tests with a model waste, Cirne et al (2007) reported inhibition of methane production but only with lipid content over about 30% of COD. They reported an initial lag phase of 6-10 days attributed to the rapid buildup of VFA and/or LCFA. However even with a high lipid content of 47% of COD, the process was able to recover and produce close to 100% methane recovery. Masse et al (2003) treated slaughterhouse waste 12

containing pork fat particles of 60 µm to 450 µm and found that larger fat particles seemed to inhibit methane production. When treating abattoir wastewater in a covered anaerobic lagoon (CAL), fat in the effluent can lead to the development of a crust which may damage the cover (MINTRAC 2012). In the evaluation of a CAL treating abattoir wastewater with an influent O&G level of 1,125 mg/L, UNSW (1998) noticed that the thickness of the scum layer fluctuated, indicating that equilibrium may have been reached. White et al (2013) also noted the formation of a scum layer under the cover within 2 months of commissioning a CAL treating abattoir wastewater with 156 mg/L FOG. After 6 months, the scum was evident under the cover for the first 9 metres of the inlet, but had thinned towards the outlet end. It was expected to disappear as the pond settled into normal operation. In a New Zealand CAL operating for 7 years treating wastewater from an abattoir and rendering plant, solid fat has accumulated under about 25% of the area of the cover at the inlet end (A. van Oostrom, pers. comm. February 2014). There was little or no pre-treatment to reduce fat other than the use of a waste heat evaporator in the rendering plant for stickwater solids recovery and the fat accumulation has raised the cover by 100-200 mm but has possibly reached a steady state.

4.3 Effect of FOG level on digestion Ambiguity remains as to the maximum loadings of FOG that can be managed within anaerobic digesters. Several factors can influence the effective levels of FOG, including temperature (mesophilic versus thermophilic) and the specific composition of the wastewater, including the method of pre-treatment. In a review of waste lipids to energy, Alves et al (2009) stated that in general, hydrolysis of fats and oils to glycerol and LCFA proceeds rapidly in anaerobic processes, resulting in accumulation of LCFA in wastewater. If the rate of hydrolysis is higher than the rate of methanogenesis, excess LCFAs may accumulate. This could lead to an increase in the lag phase in methanogenic activity (Long et al (2012). Shock loading of LCFAs may also severely retard the anaerobic digestion process. During laboratory tests with a UASB reactor, Hwu et al (1998) determined that sludge flotation occurred when the LCFA concentration reached 263 mg/L which was far below the minimum concentration for inhibition of methanogenesis of 401 mg/L reported by Yuan (1995) (cited by Hwu et al (1998)). Therefore deterioration of the UASB process by LCFA adsorption and subsequent sludge washout is likely to occur prior to inhibition of the methanogenic organisms.

13

Vidal et al (2000) carried out anaerobic biodegradability tests on dairy wastewaters with fat contents ranging from 4 mg/L to 2,890 mg/L. They concluded that although biodegradation of fat-rich wastewaters was slower than that of fat-poor wastewater, due to the slower rate of the fat hydrolysis step, this avoids the accumulation of VFA and the overall process is favoured. The intermediates of fat degradation (glycerol and LCFA) seem not to reach concentrations high enough to affect the anaerobic process. During anaerobic digestion at 25°C, Masse et al (2002) found that pork fat particle size ranging from 60 to 450 µm had no significant effect on the rate of fat hydrolysis. At an average fat level of about 725 mg/L, they concluded that hydrolysis pre-treatment should not substantially accelerate anaerobic treatment. In anaerobic digestion tests of wastewater from a dairy factory, Kim and Shin (2010) used lipid concentrations ranging from 0.7 to 6.1 g COD/L. They found that as the lipid concentration was increased, the lag-phase time for secondary methane production increased. They suggested that the long lag time was due to the diffusion limitations imposed by the lipid layer surrounding the bacterial cells. However the accumulated lipids were eventually converted to methane within 50 days, even at the highest concentration of 6.1 g COD/L. Cirne et al (2007) also found that a model waste with a high lipid content of up to 47% w/w COD basis could be digested, although a lag phase of 60 days was observed before commencement of significant methane production. In a review of anaerobic co-digestion of FOG, Long et al (2012) noted that most of the tests of inhibition of methane production by fatty wastes have been performed on samples of pure free fatty acids which may not be representative of an industrial waste. Laboratory and pilot-scale experimental tests are needed to determine the effect a mixture of oil, grease, fat, food solids and detergents on methane production. More data is also needed to identify optimum reactor conditions such as percent FOG loading, solids retention time, and reactor configuration.

4.4 Pre-treatment technologies Researchers have studied the effects of different pre-treatments and additives to the waste to enhance the digestibility of fat, oil and grease in anaerobic systems. Pre-treatment technologies which aim to improve digestion of FOG and improve biogas yield include: • •

Saponification Hydrolysis

14

• • •

Enzymatic hydrolysis Ultrasound disintegration Combination pre-treatments

4.4.1 Saponification Saponification is the hydrolysis reaction between a lipid and an alkali, resulting in LCFA salts and the release of glycerol. The conversion of the lipids into soluble soaps and free LCFA should improve the contact between the substrate and the micro-organisms, thereby enhancing their anaerobic biodegradability (Battimelli et al 2009). When a DAF float and carcass fats from a slaughterhouse were saponified with 60°C sodium hydroxide at pH 12 and then fed to an anaerobic digester in a batch process, there was little increase in the total volume of biogas (Battimelli et al 2009). Pre-treatment did provide improved initial reaction kinetics, indicating better initial bio-availability. Affes et al (2013) applied similar saponification to fatty slaughterhouse wastes which were fed to an anaerobic digestion vessel in which the digestate solid fraction could be recirculated. An improved biogas yield was obtained with pre-treatment and solids recirculation. In this case, saponification promoted emulsification and bioavailability of solid fatty residues and reactors with solids recirculation were able to rapidly adapt to an increase in organic loading rate compared with the normal reactor which exhibited a slow transition. 4.4.2 Thermal hydrolysis As discussed earlier, hydrolysis is the first stage of anaerobic digestion where the FOG is converted to LCFAs and glycerol. Bouchy et al (2012) trialled thermal hydrolysis as a pre-treatment to digestion of a mixture of DAF float (0.15-1.0% fat) and primary and secondary sewage sludge. The mixture was treated with steam at 8 bar for 30 min resulting in 51.8% increase in biogas yield with better treated sludge characteristics such as solids/liquid separation during centrifuging. There is a large body of work on the use of thermal hydrolysis as a pre-treatment for anaerobic digestion of sewage sludge demonstrating significant improvements in biogas yield. The Cambi™ thermal hydrolysis process has been widely applied to sewage sludge, but not to industrial wastewaters. 4.4.3 Enzyme hydrolysis Enzyme pre-treatment of high-fat wastewaters has been extensively researched (Masse et al 2003, Pereira et al 2006, Leal et al 2006, Valladao et al 2011, Bouchy et al 2012, Zawadzki et al 2013, DonosoBravo & Fdz-Polanco 2013). Commercially available lipase enzymes were utilised in the hydrolysis of FOG in slaughterhouse wastewater by Masse 15

et al (2003) and Pereira et al (2006). Masse et al (2003) found little improvement in methane yield when the enzyme PL-250 was added (at 250 mg/L) to slaughterhouse effluent containing 500 mg/L fat particles and hydrolysed for 5.5 h at 25°C. Cirne et al (2007) also found little benefit from the addition of lipase although it enhanced hydrolysis due to the accumulation of LCFAs. On the other hand, Pereira et al (2006) obtained about 4 times more biogas from slaughterhouse effluent after treatment with 0.4% lipase compared with anaerobic digestion of untreated wastewater. Bouchy et al (2012) also achieved an increase in methane yield when a commercial lipase enzyme was used to hydrolyse DAF fat mixed with sewage sludge. Donoso-Bravo & Fdz-Polanco (2013) assessed the effect of the addition of a commercial enzyme (Biolipase L®) to a mixture of grease trap waste and sewage sludge prior to anaerobic digestion. An enzyme dosage of 0.33% more than doubled the methane yield with 5% grease trap waste. The cost of the enzyme addition could be offset by the reduction in retention time and increased yield. Unfortunately, the use of commercial enzymes for the hydrolysis of fats is very expensive. Several investigators (Leal et al 2006) have used solidstate fermentation of the fungus Penicillium restrictum as a source of enzyme for the hydrolysis of wastewater in an attempt to reduce the cost. The solid enzymatic preparation (0.1%) was added to dairy wastewater containing up to 1,000 mg/L fat followed by hydrolysis at 35°C for 24 h prior to anaerobic digestion. There was little benefit at 600 mg/L fat, but significant improvement in performance was found when the fat level in the effluent was 1,000 mg/L. Valladao et al (2010) treated poultry abattoir wastewater at 800 mg/L fat with 0.1% of the same enzyme for 4 h at 30°C prior to digestion in a UASB digester. The hydrolysis pre-treatment resulted in slightly higher COD removal and methane yield. Zawadzki et al (2013) developed a continuous enzymatic pre-hydrolysis treatment in a packed-bed reactor for high-fat wastewaters from a meat factory. The reactor was packed with a solid-state fermentation of Rhizopus microspores on a mixture of sugarcane bagasse and sunflower seed meal and used to treat the wastewater containing approximately 600 mg/L oil and grease. After a hydraulic retention time of 24 h, the O&G content was reduced by 96% and the biodegradability of the wastewater improved such that it would be more suited to anaerobic digestion or activated sludge treatment. 4.4.4 Ultrasound Ultrasound refers to sound waves at frequency higher than the threshold of human hearing (>18 kHz). Low frequency (20 kHz to 1,000 kHz), high power ultrasound is used for improving the efficiency of physical and biochemical processes. In general, the process of cavitation is the basis for the effects of ultrasound on anaerobic digestion and other physical and 16

biochemical processes. Cavitation is the formation, growth and collapse of tiny gas bubbles during the propagation of soundwaves in a liquid media. There is a large body of work on the effects of ultrasound pre-treatment of sewage sludge for anaerobic digestion (Carrere et al 2010) but limited information on use with fat-laden industrial wastewaters. Bouchy et al (2012) reported an increased of almost 90% in methane yield from an ultrasound homogeniser cavitation pre-treatment of a DAF float and sewage sludge mixture prior to anaerobic digestion. However no details of ultrasound parameters and treatment time were provided. Peng et al (2014) treated 1 L of oily wastewater for 10 min with ultrasound apparatus operating at a frequency of 28 kHz and a supplied power of 700 W. Subsequent anaerobic digestion resulted in 23% increase in methane yield compared with the untreated control. Ultrasound was installed to treat the effluent containing in excess of 2,000 mg/L fat from a rendering plant in Queensland prior to entering a CAL. However, the use of ultrasonic treatment was discontinuted due to the additional operating costs and an already satisfactory performance of the CAL without ultrasonic treatment (S. Boyers pers comm. February 2014). 4.4.5 Combination pre-treatments The synergistic effect of combining different pre-treatments on the effectiveness of anaerobic digestion of high-fat wastewaters has also been investigated (Damasceno et al 2012, Peng et al 2014). Damasceno et al (2012) used a combination of a rhamnolipid biosurfactant produced by Pseudomonas aeruginosa and an enzyme pool produced by solid-state fermentation with Penicillium simplicissimum in the anaerobic treatment of poultry processing plant wastewater with a fat content of 2,400 mg/L. A combination of 0.19% (w/v) enzyme pool and 114 mg/L biosurfactant at 33°C yielded the highest specific methane production. The claimed advantage of this treatment was that the effluent could be anaerobically digested without using the fat flotation stage to reduce the fat level. A combination of an oil-degrading bacteria (Bacillus), ultrasound and citric acid was used by Peng et al (2014) to pre-treat an oily wastewater containing 6 g-VS waste oil per litre of water. Pre-mixing Bacillus at 9% (w/w) for 24 h, then combined with ultrasound (28 kHz, 700 W) for 10 min and a citric acid concentration of 500 mg/L provided the optimum conditions, which resulted in methane yield of 280% greater than the untreated control. The ultrasound promoted homogenisation and the citric acid maintained the homogenisation throughout the digestion. The benefits and disadvantages of the various wastewater pre-treatments are summarised in Table 5. 17

Table 5: Summary of pre-treatments to anaerobic digestion of oily wastewaters

Pre-treatment Saponification

Thermal hydrolysis

Enzyme hydrolysis

Ultrasound

Combined treatments

5

Benefits Disadvantages Improves bioavailability More suited to fatty of fat and biogas yield. wastes that can be handled separately and increase in pH may affect digester operation. Likely 50% increase in Would be expensive for biogas yield and dilute wastes such as improved sludge settling meat plant effluent. properties. Relatively easy to High cost of enzymes implement and some and variable results. researchers report improved CH 4 yield. Relatively simple to High power cost and implement in an existing sonotrode replacement system. Increased CH 4 cost. yield. Increased CH 4 yield Unproven outside from oily wastewaters by laboratory and potentially pre-treatment in one high operating costs. vessel.

Digester configuration 5.1 Two-phase digestion The complete digestion process for both lagoon and high-rate anaerobic systems normally take place in the one vessel, but a separate vessel is utilised for the acidogenesis phase in some processes. In laboratory trials, Kim & Shin (2010) utilised a continuously stirred tank reactor (CSTR) followed by a UASB reactor to treat fatty wastewater (1,000 – 2,500 mg/L total lipids) from a dairy factory. The two-phase system showed enhanced COD removal, lipid removal and methane production as the organic loading rate increased above 2.5 g COD/L/d compared to a single UASB. Gannoun et al (2009) developed a process where abattoir wastewater containing 40-410 mg/L fat was hydrolysed in a CSTR at 30°C for 2 days, then the solids settled out and the liquid fraction fed to an upflow anaerobic filter (UAF). The stirred aerobic conditions stimulated the growth of Bacillus spp. to become the dominant microflora. They hypothesised that stirring improved degradation efficiency by hydrolytic enzymes such as proteases and lipases produced by the Bacillus strains. The UAF achieved COD removal efficiencies of 90 to 92%. 18

A treatment system handling a high-fat (>2,000 mg/L) wastewater from a rendering plant in Queensland employs an initial vessel with a retention time of approximately 2 days prior to a covered lagoon with a retention time of 50 days (S. Boyers pers comm. February 2014). During the initial acidogenesis stage, pH reduces from 7.0 to about 6.4. No problems of fat accumulation under the cover have been reported but the fat is finely emulsified and the high wastewater temperature of 38-40°C may assist in maintaining it in that state where it can be more easily digested.

5.2 Temperature-phased digestion Temperature-phased anaerobic digestion is a relatively new technology first reported in the 1990s and consists of an initial thermophilic digester (operating at ~55°C) followed by a mesophilic digester operating at ~35°C (Lv et al. 2012). Although systems have not been widely applied industrially, researchers have demonstrated enhanced methane yield, process stability, shorter hydraulic retention time and decreased foaming. The thermophilic digester normally has a shorter retention time of 1 to 2 days, with the mesophilic phase being 4 to 5 time longer. The system has mainly been applied to sludge digestion with no applications to fatcontaining wastewaters.

5.3 Inverted anaerobic sludge blanket Almost all anaerobic digestion of meat processing plant wastewater in Australia takes place in open anaerobic ponds or CALs. These arrangements are not used in the colder climate of Europe and Asia where most of the research on anaerobic digestion has been undertaken. Hence most research is focused on high-rate anaerobic systems controlled at mesophilic (~35°C) or thermophilic (~55°C) temperatures. As existing reactor configurations are not robust when applied to industrial effluents with a high lipid content, Alves et al (2009) proposed a new concept. This reactor has the primary biomass retained by flotation and the secondary biomass retention through settling. This is now termed an inverted anaerobic sludge blanket (IASB) digester and the first full-scale reactor of 100 m3 capacity (Figure 2) has been installed in a fish factory in Portugal (Picavet & Alves 2013) but no performance data is yet available.

19

Figure 2: First full-scale inverted anaerobic sludge blanket reactor in Portugal (from Picavet & Alves, 2013)

6

Conclusions and recommendations

Fat in meat plant wastewater that is to be treated by anaerobic digestion is potentially a valuable resource provided it does not inhibit the process by forming a scum on the surface and can be completely digested. There is ample evidence to show that, although slower to digest, a higher fat level can improve the methane yield when digesting wastewater or sludge which could improve the economics of installing covered anaerobic lagoons. Finely emulsified fat particles appear to be more readily digested, especially if the temperature can be maintained in the upper mesophilic range (>35°C). However, abattoir CALs operate at closer to 25°C and the larger fat particles are more difficult to digest and more readily float to the surface. The level of FOG in normal abattoir wastewater could be up to 3,000 mg/L and should not be detrimental to the operation of a CAL provided it is available for digestion. 20

Practical evidence, however, suggests that scum formation will occur at these levels. Many CALs appear to be operating satisfactorily in Australia and New Zealand treating meat processing plant effluent but much is still to be learned regarding optimisation of their operation. Long-term evaluation of CALs treating fatty abattoir wastewater is required to determine whether the development of a fatty layer on the surface under the cover increases over time or reaches a state of equilibrium and whether a surface fat layer inhibits biogas generation. At this stage, pre-treatments appear to offer limited benefits when evaluated against an increase in operating costs, however, ongoing efforts to minimise scum formation and maximise biogas production will lead to advances in digester operation and reduce the frequency of maintenance issues. Two-stage treatment, where the acidogenesis step takes place in a separate vessel prior to the main digestion tank or CAL, appears to offer potential benefits for fatty waste waters and warrants further investigation under Australian conditions. There is also potential to optimise the design and configuration of typical Australian CALs by evaluating parameters such as hydraulic retention time, length to breadth ratio, method of introducing effluent, sludge removal and sludge recirculation rate. There is a paucity of fundamental information on the anaerobic digestion of real Australian abattoir wastewater. It is recommended that data be gathered from laboratory trials with effluent of different fat levels to determine the limiting level of FOG in the wastewater. The results of these trials along with results from existing full-scale CALs may indicate that: 1. fat must be removed from the effluent to a specific level; or 2. pre-treatment of fatty effluent is required; or 3. a different configuration of the process is needed to reap the full benefit of the digestion of fat.

21

Abbreviations BOD CALs COD CSTR DAF FFA FOG IASB MINTRAC

Carbonaceous biochemical oxygen demand Covered anaerobic lagoons Chemical Oxygen Demand Continuous stirred tank reactor Dissolved air floatation Free fatty acid Fat, oil and grease Inverted anaerobic sludge blanket National Meat Industry Training Advisory Council Limited Meat & Livestock Australia Oil and grease tonne hot standard carcass weight Long chain fatty acids Upflow anaerobic filter Up flow anaerobic sludge blanket University of New South Wales Volatile fatty acid

MLA O&G tHSCW LCFAs UAF UASB UNSW VFA

22

References Affes, R., Palatsi, J., Flotats, X., Carrere, H., Steyer, J.P. & Battimelli, A. (2013). Saponification pretreatment and solids recirculation as a new anaerobic process for the treatment of slaughterhouse waste. Bioresource Technology 131, 460-467. Alves, M.M., Pereira, M.A., Sousa, D.Z., Cavaleiro, A.J., Picavet, M. Smidt, H & Stams, A.J.M. (2009). Waste lipids to energy: how to optimize methane production from long-chain fatty acids (LCFA). Microbial Biotechnology 2 (5), 538-550. Bouchy, L., Perez, A., Camacho, P., Rubio, P., Silvestre, G., Fernandez, B., Cano, R., Polanco, M. & Diaz, N. (2012). Optimization of municipal sludge and grease co-digestion using disintegration technologies. Water Science & Technology 65, 214-220. Boyers, S. (A. J Bush & Sons). Telephone conversation 12 February 2014. Battimelli, A., Carrere, H. & Delgenes, J-P. (2009). Saponification of fatty slaughterhouse wastes for enhancing biodegradability. Bioresource Technology 100, 3695-3700. Carrere, H., Dumas, C., Battimelli, A., Batstone, D.J., Delgenes, J.P., Steyer, J.P. & Ferrer, I. (2010). Pretreatment methods to improve sludge anaerobic degradability: A review. Journal of Hazardous Materials 183, 1-15. Cirne, D.G., Paloumet, X., Bjornsson, L., Alves, M.M. & Mattiasson, B. (2007). Anaerobic digestion of lipid-rich waste – effects of lipid concentration. Renewable Energy 32, 965-975. Damesceno, F.R.C., Cammarota, M.C. & Freire, D.M.G. (2012). The combined use of a biosurfactant and an enzyme preparation to treat effluent with a high fat content. Colloids and Surfaces B: Biointerfaces 95, 241-246. Donoso-Bravo, A. & Fdz-Polanco, M. (2013). Anaerobic co-digestion of sewage sludge and grease trap: Assessment of enzyme addition. Process Biochemistry 48, 936-940. GHD (2003). Assessment of hydrocyclones for fat removal from meat processing wastewater streams. Report to Meat & Livestock Australia on Project PRENV.022. Gannoun, H., Bouallagui, H., Okbi,A., Sayadi, S. & Hamdi, M. (2009). Mesophilic and thermophilic digestion of biologically pretreated abattoir wastewaters in an upflow anaerobic filter. Journal of Hazardous Materials 170, 263-271. 23

Green, J. (1992). Effluent treatment ponds. In Abattoir Waste Water & Odour Management. (ed) B McDonald, CSIRO Meat Research Laboratory, Cannon Hill, Qld, Australia pp 55-68. Greenfield, P.F. & Johns, M.R. (1992). High-rate anaerobic treatment of abattoir effluents. In Abattoir Waste Water & Odour Management. (ed) B McDonald, CSIRO Meat Research Laboratory, Cannon Hill, Qld, Australia pp 69-93. Husband, P. (1992). Primary effluent treatment. In Abattoir Waste Water & Odour Management. (ed) B McDonald, CSIRO Meat Research Laboratory, Cannon Hill, Qld, Australia pp 27-38. Hwu, C-S, Tseng, S-K, Yuan, C-Y, Kulik, Z & Lettinga, G. (1998). Biosorption of long-chain fatty acids in UASB treatment process. Water Research 32, 1571-1579. Jensen, P. & Batstone, D. (2012). Energy and nutrient analysis on individual waste streams. Report to Meat & Livestock Australia on Project A.ENV.0131. Johns, M.R. (1995). Developments in wastewater treatment in the meat processing industry: a review. Bioresource Technology 54, 203-216. Kim, S-H., Han, S-K. & Shin, H-S. (2004). Kinetics of LCFA inhibition on acetoclastic methanogenesis, propionate degradation and beta-oxidation. Journal of Environmental Science and Health A: Toxic/Hazardous Substances and Environmental Engineering 39, 1025-1037. Kim, S-H. & Shin, H-S. (1010). Enhanced lipid degradation in an upflow anaerobic sludge blanket reactor by integration with an acidogenic reactor. Water Environment Research 82, 267-272. Li, C., Champagne, P. & Anderson,B.C. (2013). Biogas production performance of mesophilic and thermophilic anaerobic co-digestion with fat, oil and grease in semi-continuous flow digesters: effects of temperature, hydraulic retention time and organic loading rate. Environmental Technology 34, 2125-2133. Leal, M.C.M.R., Freire, D.M.G., Magali, C.C. & Sant’Anna, G.L. (2006) Effect of enzymatic hydrolysis on anaerobic treatment of dairy wastewater. Process Biochemistry 41, 1173-1178. Long, J.H., Aziz, T.N., de los Reyes, F.L. & Ducoste, J.J. (2012). Anaerobic co-digestion of fat, oil, and grease (FOG): a review of gas production and process limitations. Process Safety and Environmental Protection 90, 231245.

24

Lv, W., Schanbacher, F.L, & Yu, Z. (2010). Putting microbes to work in sequence: Recent advances in temperature-phased anaerobic digestion processes. Bioresource Technology 101, 9409-9414. Masse, L., Masse, D.I., Kennedy, K.J. & Chou, S.P. (2002) Neutral fat hydrolysis and long-chain fatty acid oxidation during anaerobic digestion of slaughterhouse wastewater. Biotechnology and Bioengineering 79, 43-52. Masse, L., Masse, D.I. & Kennedy, K.J. (2003). Effect of hydrolysis pretreatment on fat degradation during anaerobic digestion of slaughterhouse wastewater. Process Biochemistry 38, 1365-1372. MINTRAC (2012). Wastewater management in the Australian red meat processing industry. Prepared for Australian Meat Processor Corporation, Sydney, Australia. MLA (2007). Wastewater. In Environmental Best Practice Guidelines for the Red Meat Processing Industry. (eds) M. Johns, S. McGlashan & A. Rowlands for Meat and Livestock Australia Ltd, Sydney, Australia. MLA (2014). January 2014 co-products monitor – summary. http://www.mla.com.au/Prices-and-markets/Trends-and-analysis/Beef/Coproducts. Accessed 16 Feb 2014. Peng, L, Bao, M., Wang, Q., Wang, F. & Su, H. (2014). The anaerobic digestion of biologically and physicochemically pretreated oily wastewater. Bioresource Technology 151, 236-243. Pereira, E.B., Castro, H.F., Spiller, V.R.& Furigo, A. (2006). Degradation of fat and grease in slaughterhouse wastewater by a commercial microbial lipase. Brazilian Archives of Biology and Technology 49, 21-28. Picavet, M.A. & Alves, M.M. (2013). IASB-Inverted anaerobic sludge blanket reactor: background, history and development. Proc. of 13th World Congress on Anaerobic Digestion, Santaigo de Compostela, Spain 25-28 June 2013 p.1-4. http://repositorium.sdum.uminho.pt/handle/1822/26329, accessed 12 Feb 2014. Pittaway, P. (2011). Using covered anaerobic ponds to treat abattoir wastewater, reduce greenhouse gases and generate bio energy. Report to Meat & Livestock Australia for Project A.ENV.0107. Sayed, S., van der Zanden, J., Wijffels, R. & Lettinga, G. (1988). Anaerobic degradation of various fractions of slaughterhouse wastewater. Biological Wastes 23, 117-142. Spooncer, W.F. (Kurrajong Meat Technology). Telephone conversation 16 February 2014. 25

Tetreault, A. (2003). Electrocoagulation process for wastewater treatment. Report to Meat & Livestock Australia for project PRENV.011. UNSW – CRC for Waste Management & Pollution Control (1998). Treatment of abattoir wastewater using a covered anaerobic lagoon. Report to Meat & Livestock Australia for Project M.665. Li, C., Champagne, P. & Anderson,B.C. (2013). Biogas production performance of mesophilic and thermophilic anaerobic co-digestion with fat, oil and grease in semi-continuous flow digesters: effects of temperature, hydraulic retention time and organic loading rate. Environmental Technology 34, 2125-2133. Valladao, A.B.G., Cammarota, M.C. & Freire, D.M.G. (2010). Performance of an anaerobic reactor treating poultry abattoir wastewater with high fat content after enzymatic hydrolysis. Environmental Engineering Science 28, 299-307. Van Oostrom, A. (Albert van Oostrom & Associates), email 12 February 2014. Vidal, G., Carvalho, A., Mendez. R. & Lema, J.M. (2000). Influence of the content in fats and proteins on the anaerobic biodegradability of dairy wastewaters. Bioresource Technology 74, 231-239. White, T., Johns, M. & Butler, B. (2013). Methane recovery and use at a meat processing facility. RIRDC Publication 13/027. RIRDC Project No. PRJ005673. Yuan, C-Y. (1995). Treatment of lipid ingredients in slaughterhouse wastewater by upflow anaerobic sludge bed reactor. M.Sc. Thesis, Graduate Institute of Environmental Engineering, national Taiwan University, Taipei, Taiwan, R.O.C. (in Chinese). Zawadzki, R.A.F., Prado, M.R., Mitchell, D.A. & Krieger, N. (2013). Continuous enzymatic prehydrolysis treatment of high-fat wastewater. Food, Technology and Biotechnology 51, 293-300.

26