Waste Management xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Anaerobic digestion of food waste using yeast Jutarat Suwannarat, Raymond J. Ritchie ⇑ Tropical Plant Biology, Faculty of Technology and Environment, Prince of Songkla University, Phuket Campus, Kathu, Phuket 83120, Thailand
a r t i c l e
i n f o
Article history: Received 17 December 2014 Accepted 22 April 2015 Available online xxxx Keywords: Food waste Anaerobic digestion Yeast (S. cerevisiae)
a b s t r a c t Fermentative breakdown of food waste seems a plausible alternative to feeding food waste to pigs, incineration or garbage disposal in tourist areas. We determined the optimal conditions for the fermentative breakdown of food waste using yeast (Saccharomyces cerevisiae) in incubations up to 30 days. Yeast efficiently broke down food waste with food waste loadings as high as 700 g FW/l. The optimum inoculation was 46 106 cells/l of culture with a 40 °C optimum (25–40 °C). COD and BOD were reduced by 30–50%. Yeast used practically all the available sugars and reduced proteins and lipids by 50%. Yeast was able to metabolize lipids much better than expected. Starch was mobilized after very long term incubations (>20 days). Yeast was effective in breaking down the organic components of food waste but CO2 gas and ethanol production (1.5%) were only significant during the first 7 days of incubations. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Food waste from restaurants and hotels is an increasing environmental problem, particularly in tourist areas. Restaurant waste consists of restaurant discards, waste from food preparation, large amounts of oils and fats with some paper (serviettes) and wood (chopsticks and toothpicks) but would be largely free of heavy metal contaminants (lead, cadmium, mercury) (Han and Shin, 2002; Cirne et al., 2006; Forster-Carneiro et al., 2007). Food waste can be defined as any edible waste from food production, transportation, distribution and consumption. It is also referred to as garbage, swill and/or kitchen refuse, solid and liquid by-product wastes. They are generated throughout food production and processing sectors. In total, food waste generated from food preparation may constitute as much as 20% of the total human food supply from the stage of processing to the point of consumption (Westendorf et al., 1998). The low heavy metal content of restaurant waste distinguishes it from household garbage which often has high heavy metal levels rendering the breakdown residues of biological digestion processes too contaminated with heavy metals
Abbreviations: FW/l, fresh weight per litre; BOD, Biochemical Oxygen Demand; COD, Chemical Oxygen Demand; TS, total solid; °C, degree celsius; BSA, bovine serum albumin; GC, gas chromatography; CO2, carbon dioxide; mmol, millimoles; ml, millilitre; EtOH, ethanol. ⇑ Corresponding author. E-mail address: [email protected] (R.J. Ritchie).
to be readily useable. Zhang et al. (2007) identified food waste as an excellent feedstock for fermentation digestion processes. In Thailand a large component of the waste is cooked rice (as found for Korean restaurant waste, Han and Shin, 2002) which would be partially hydrolysed from cooking but would be expected to require further treatment to mobilize the starches. Davis (2008) used amylase treatment to mobilize the starch in food waste. Similar methods have been used for preparing corn paste wastes (Akpan et al., 2008) and sweet potato processing waste (Qian et al., 2008) but acid hydrolysis has also been used to mobilize both starches and cellulosic material such as corn cobs, peanut shells and newspaper (Akpan et al., 2005, 2008). Sawayama et al. (1997) used high temperature and pressure to thermally hydrolyse food waste prior to fermentation treatment. The advantage of acid hydrolysis is that it mobilises both starches and some of the cellulose and many other organic compounds. The obvious drawbacks to acid hydrolysis are that it is expensive, involves bulk handling of acids which are highly corrosive and might mobilise undesirable toxic metals and finally the acid hydrolysate has to be neutralized before being inoculated with microorganisms. Han and Shin (2002) used microbial populations from ruminants as a source of acidophile microbes to break down cellulose and starches from Korean restaurant waste, however the main products from such fermentation were acetic and butyric acids. These are not high value products. Lipids of various kinds (fats and oils) are a major component of restaurant wastes (Cirne et al., 2006) and present particular problems for both physical reasons and in their metabolism. Yeasts and methanogen bacteria can
http://dx.doi.org/10.1016/j.wasman.2015.04.028 0956-053X/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Suwannarat, J., Ritchie, R.J. Anaerobic digestion of food waste using yeast. Waste Management (2015), http://dx.doi.org/ 10.1016/j.wasman.2015.04.028
2
J. Suwannarat, R.J. Ritchie / Waste Management xxx (2015) xxx–xxx
use hydrolysis products of fats and lipids but they are generally thought not to be able to rapidly ferment unprocessed lipids. There are also physical problems arising from fats and greases in food waste: fat saturated materials aggregate into grease balls and fats and greases deposit themselves on the walls of plumbing and storage vats. Physical problems arising from fats and grease balls may prove to be a major limitation to using microbial methods to process restaurant waste. Food waste is a complex biomass containing various components such as starchy, fatty, and cellulosic materials. Without some sort of processing, these organic polymer materials may be difficult for ethanol producing microorganisms such as Saccharomyces cerevisiae (common yeast) to utilize. Food waste generated in Korea is rich in carbohydrate, as high as 65% of total solids due to its high proportion of cooked rice (Kim et al., 2008). Food waste is difficult to dispose of by incineration. Most food waste has been placed in landfills together with other wastes (Han and Shin, 2002). Food waste is the major component of organic matter in garbage and so when it is disposed of in landfills it is the major source of methane gas produced in the landfills and a major contributor to the organic matter in leachates. Adding restaurant food waste to the waste stream exacerbates such problems. Forster-Carneiro et al. (2007) studied improvements in the efficiency of semidry anaerobic digestion and dry fermentation (20– 35% TS), where little or no water, or sludge is added to the organic fraction of the municipal solid waste to produce an inert biosolid final product with methane production from methanogenic bacteria. The anaerobic digestibility and biogas and methane yields of the food waste were evaluated using batch anaerobic digestion tests performed at 50 °C. The results of the study indicated that the food waste is a highly desirable substrate for anaerobic digesters with regards to its high biodegradability and methane yield. Anaerobic digestion is the method of choice for the treatment of organic waste. This method has advantages of low-level sludge production, low-level energy consumption and potentially useful amounts of methane or ethanol production. However, solid organic materials such as kitchen garbage and sewage sludge are not digested quickly anaerobically. This is mainly a result of the unfavourable surface area/volume ratio limiting the surface area accessible to the microbes. Formation of grease balls worsens this surface area/volume problem and accessibility of the substrates to the bacteria. A liquidisation step is needed to speed up the anaerobic digestion. Analysis of the energy balance of thermal/pressure liquidisation followed by anaerobic digestion treatment was better than direct incineration (Sawayama, 1997). The physical and chemical characteristics of the organic waste are important information for designing and operating anaerobic digesters. They affect biogas/ethanol production and process stability during anaerobic digestion and the costs of handling the feedstock material (Zhang et al., 2007). Yeasts on the other hand are extremely reliable organisms to grow in culture and backup cultures are easily maintained. This is a crucial advantage over methanogen digestion. Yeasts are the mainstay of the fermentation industry (Madigan et al., 1997). Food waste contains high enough levels of proteins and amino acids and ammonia from the hydrolysis and breakdown of foodstuffs to adequately supply fixed nitrogen for the growth of yeasts. Yeasts also have a requirement for phosphorus, which is assimilated as a dihydrogen phosphate ion, and sulphur, which can be assimilated as a sulphate ion or as organic sulphur compounds such as the amino acids methionine and cysteine. Some metals, like magnesium, iron, calcium, and zinc and potassium are also required for good growth of the yeast (Shimoda, 2004). Food waste is unlikely to lack the essential minerals required by yeasts or their vitamin requirements (Ritchie and Raghupathi, 2008). In previous
studies we have shown that yeast very efficiently removed ammonia from food waste but was not efficient at removal of phosphate (Suwannarat and Ritchie, 2013a,b). S. cerevisiae is used for stable ethanol fermentation around the world, but it is important that yeasts lack the full range of amylolytic enzymes (a-amylase, b-amylase and glucoamylase) required to fully break down starches to glucose. Yeast has only two genes for amylases, YIL099W (SGA1) and YIR019C (FLO11, MUC1 and STA4) (KEGG, 2013). These enzymes are both a-glucoamylases (EC:3.2.1.3) (KEGG, 2013). Watanabe et al. (2009) evaluated the culture conditions and material compositions for efficient ethanol production from rice washing drainage: they used rice bran as a cheap source of amylolytic enzymes (especially a-, b-amylase) because rice washing drainage from rice polishing contained only a-glucosidase. The alternative of digesting starches by acid hydrolysis has practical limitations already pointed out above. The efficiency of yeast in digesting bulk lipids and fats is not well documented but yeast is known to produce emulsifying agents which would help in breaking up aggregates of oils, greases and sludges (Barriga et al., 1999). Watanabe et al. (2009) studied yeast-based anaerobic batch fermentation of rice waste (main culture: net volume 30–36 ml), using rice washing drainage (30 ml) as the substrate with lactic acid (final concentration: 100 mM) as the bactericidal agent. Different weights of rice bran were mixed in a 50 ml centrifuge tube, and then 1.0 ml of pre-culture yeast broth was inoculated. Fermentation processes were terminated after 14 days. The concentration of ethanol and sugars was analysed using an HPLC. The maximum ethanol concentration attained was 6.2% (V/V) (Watanabe et al., 2009). Currently, ‘‘spent yeast’’ has a low value and is used as a protein supplement in animal feed (Barriga et al., 1999). The brewing industry is a ready bulk source of ‘‘spent yeast’’ which could be used to digest food waste at minimal cost. In using yeast to breakdown food wastes the incubations can be run at higher temperatures than used in brewing to maximize the digestion rate because the ‘‘flavor’’ of the product is not relevant in the case of using yeast to break down food waste (Fleet et al., 2009). There are several additional reasons why yeast was selected for the study. Yeast is completely sequenced (KEGG, 2013). It can be genetically transformed allowing for either the addition of new genes or deletion through homologous recombination. A complete set of yeasts with single knockout mutations is available. The major advantage of yeast remains, however, its ease of cultivation and ready availability. Many different strains are available and standard procedures could be used to select a strain most suitable for anaerobic breakdown of food waste. 2. Materials and methods 2.1. Microorganism and culture conditions S. cerevisiae, in this study was obtained from the Biology Laboratory of Prince of Songkla University, Phuket campus. It is a baker’s yeast strain. 2.2. Inoculum yeast culture Stocks of S. cerevisiae were maintained on agar plates in medium containing 6 g of peptone, 3 g of yeast extract, 6 g of dextrose, 15 g of agar added to 300 ml of distilled water. Liquid cultures were grown in the same medium with the agar omitted. The standard autoclaving routine was 121 °C for 15 min. Liquid cultures were grown out without aeration. The backup yeast cultures were maintained in the refrigerator at 4 °C until required (Akpan et al., 2008).
Please cite this article in press as: Suwannarat, J., Ritchie, R.J. Anaerobic digestion of food waste using yeast. Waste Management (2015), http://dx.doi.org/ 10.1016/j.wasman.2015.04.028
J. Suwannarat, R.J. Ritchie / Waste Management xxx (2015) xxx–xxx
2.3. Food waste Food waste used in this study was collected from stalls, trolleys, restaurants, and canteens in Phuket province. Batches of about 1 kg were collected and then mixed together. First plastic, chopsticks and toothpicks, paper and bones were removed and then the samples were homogenized in a blender and then stored frozen for experiments. Samples were thawed and diluted with distilled water and then autoclaved before inoculating with yeast (S. cerevisiae). Cultures were grown anaerobically in culture bottles with magnetic spin bars on magnetic stirrers. The gas exit pipe of the culture bottles was fitted with a simple pneumatic trough to ensure anaerobic conditions (Fig. 1) (Suwannarat and Ritchie, 2013b). 2.4. Analysis of food waste The major components of the restaurant food waste were identified and classified (Suwannarat, 2014). Homogenized food waste was analysed for characteristics of food waste such as pH, total solid, moisture, phosphate, ammonia, total sugar, reducing sugar, starch, wet-dry ratio, lipid, protein, carbohydrate, ammonia, COD and BOD were measured (Suwannarat, 2014). In the present study, the analyses of composition of autoclaved food waste before and after fermentation included: 2.4.1. Reducing sugar and total sugar were assayed using the Benedict assay (APHA 1998) 2.4.2. Starch was measured using the Iodine assay (APHA, 1998). 2.4.3. Lipid and oil were measured using a UV Spectrophotometer assay (APHA 1998).
An anaerobic digestion using yeast (S. cerevisiae) fermentation to produce CO2.
Yeast Culture
3
2.4.4. COD (Chemical Oxygen Demand), and BOD (Biochemical Oxygen Demand) were analysed according to Standard Methods (APHA, 1998). 2.4.5. Protein was measured using the standard Lowry method (Lowry et al., 1951) using BSA (bovine serum albumin) as standard. 2.4.6. Ammonia was assayed using the Solorzano assay (Solorzano, 1969). 2.4.7. CO2 was measured volumetrically using a simple pneumatic trough (Fig. 1). Volumes of gas were converted into moles of CO2 using the universal gas law.
PV ¼ nRT PV )n¼ RT
ð1Þ
where P is the standard atmospheric pressure (101.3 kPa), V is the volume of gas measured, R is the universal gas constant (8.3143 J K1 mol1), T is the temperature (25 °C or 298.15 K), n is the number of moles of CO2. 2.4.8. Ethanol assay by gas chromatography Alcohol (Ethanol) was measured using a gas chromatograph (Agilent 7890A GC) at Prince of Songkla University-Suratthanee, Suratthanee, Thailand using standard methods of operation as described by Amonsit and Petsom (1995). Ethanol samples in water with known concentrations of ethanol were placed in crimp sealed vials and equilibrated to 60 °C. A sample needle was then inserted into the gas phase and injected into the GC and a standard curve prepared which was used to estimate ethanol content of samples from the yeast fermentations. The Agilent software correctly identified the ethanol standards as ethanol. 2.5. Fermentation protocol Anaerobic digestions were carried out using a baker’s yeast strain of yeast (S. cerevisiae) at 25, 30, 35, 40 °C. The standard experimental culture volumes were 175 ml inoculated with 10, 15, 20, 25, 30, 35 ml of S. cerevisiae culture in 250 ml culture bottles. Food waste was sterilized at 121 °C for 30 min and experiments set up using of food waste 700 g FW (fresh weight)/l and using 10, 15, 20, 25, 30, 35 ml of S. cerevisiae culture. Each culture was fitted with a simple pneumatic trough to collect fermentation gases and to ensure anaerobic conditions (Fig. 1) (Suwannarat and Ritchie, 2013b). The starter yeast cultures used for inoculating the food waste contained about 5 106 cells/ml. Incubations were carried out for 7, 20, 25 or 30 days and the standard feedstock was 700 g FW/l because preliminary experiments had shown that yeast could easily break down a food waste feedstock loading of 700 g FW/l. 3. Results and discussion
Pneumatic trough to collect gas
Fig. 1. Experimental set-up for anaerobic fermentation of food waste and measurement efficiency yeast digests food waste. The culture was grown in a culture bottle with a magnetic stirrer. The gas exhaust pipe ran into a simple pneumatic trough to collect the gas and prevent entry of oxygen.
Food waste composition (protein, reducing sugar, total sugar, lipid and starch) were measured at the start of the incubation and at intervals during the course of the fermentation over a period of up to 30 days. Based on the standard APHA assay methods the major components of the autoclaved food waste used in the present study on a dry weight basis were: total non-reducing sugar 51%, protein 20%, lipid 8%, reducing sugar 15% and starch 6% (based on, Suwannarat and Ritchie 2013a and Suwannarat 2014). A range of inoculation volumes from 10 to 35 ml were used. Fig. 2 shows digestion values found after 30 day incubations. Yeast removed about 90–97% of total sugars (reducing and non-reducing sugar). A consistent removal rate of about 50–60% was found for reducing sugars. Most of the sugar present in the food waste was not present
Please cite this article in press as: Suwannarat, J., Ritchie, R.J. Anaerobic digestion of food waste using yeast. Waste Management (2015), http://dx.doi.org/ 10.1016/j.wasman.2015.04.028
4
J. Suwannarat, R.J. Ritchie / Waste Management xxx (2015) xxx–xxx
Optimising temperature for efficient digestion of food waste
Optimising volume of food waste for efficient digestion
100
lipid
120 100
80
protein
60
reducing sugar
40 20
total sugar
0 10
15
20
25
30
%Removed
% Removed
120
starch
80
25 C 60
35 C
20
40 C
35
Volume of yeast Fig. 2. Percent removal of measured parameters of food waste using yeast (S. cerevisiae) digestion. Error bars are ±standard errors using n = 6 replicates.
Lipid
Starch
Fig. 4. Percent removal of lipid, protein, total sugar, reducing sugar and starch at different fermentation temperatures (25, 30, 35, 40 °C). Error bars are ±standard errors using n = 6 replicates.
Volume of CO 2 (mmol/Day)
Gas production (CO 2 ) vs. Inoculation Volume 8
30
7
25
6 20
5 4
15
3
10
2 5
1
0
0 0
1
2
3
4
5
6
7
10 ml 15 ml 20 ml 25 ml 30 ml 35 ml
Day Fig. 5. Volume of CO2 produced from food waste using different of inoculums of yeast. The inoculum contained 5 106 cells/ml. The daily production of CO2 (solid lines) is on the left (Y) scale and the total cumulative gas production scale is on the right YY scale (dotted lines).
Volume of CO2 (mmol/Day)
Gas production (CO 2 ) vs. Incubation temperature 25
7 6
20
5 4
15
3
10
2 5
1 0
0 0
100 90 80 70 60 50 40 30 20 10 0
Protein Total sugar Reducing sugar
2
4
6
CO2 (mmol cumulative total)
Optimal of yeast inoculum
0
CO2 (mmol cumulative total)
as reducing sugar (Suwannarat, 2014). Yeast was much more efficient at removing non-reducing sugars than reducing sugars and there may have been some accumulation of reducing sugars such as glucose from the breakdown of starch. Lipid removal was also a consistent 30–40%. Protein removal was more variable, ranging from 30% to 80% but this variability was not a function of inoculation volume. In the case of protein, lipid, starch, and reducing sugar the best of inoculation volume of yeast was 10 ml of yeast/175 ml of culture. In terms of BOD and COD, the inoculation volume made little difference to the overall BOD for 30 day incubations, averaging about a 40% reduction in BOD (Fig. 3). On the other hand COD reduction was lower for lower inoculation volumes (
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Anaerobic digestion of food waste using yeast Jutarat Suwannarat, Raymond J. Ritchie ⇑ Tropical Plant Biology, Faculty of Technology and Environment, Prince of Songkla University, Phuket Campus, Kathu, Phuket 83120, Thailand
a r t i c l e
i n f o
Article history: Received 17 December 2014 Accepted 22 April 2015 Available online xxxx Keywords: Food waste Anaerobic digestion Yeast (S. cerevisiae)
a b s t r a c t Fermentative breakdown of food waste seems a plausible alternative to feeding food waste to pigs, incineration or garbage disposal in tourist areas. We determined the optimal conditions for the fermentative breakdown of food waste using yeast (Saccharomyces cerevisiae) in incubations up to 30 days. Yeast efficiently broke down food waste with food waste loadings as high as 700 g FW/l. The optimum inoculation was 46 106 cells/l of culture with a 40 °C optimum (25–40 °C). COD and BOD were reduced by 30–50%. Yeast used practically all the available sugars and reduced proteins and lipids by 50%. Yeast was able to metabolize lipids much better than expected. Starch was mobilized after very long term incubations (>20 days). Yeast was effective in breaking down the organic components of food waste but CO2 gas and ethanol production (1.5%) were only significant during the first 7 days of incubations. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Food waste from restaurants and hotels is an increasing environmental problem, particularly in tourist areas. Restaurant waste consists of restaurant discards, waste from food preparation, large amounts of oils and fats with some paper (serviettes) and wood (chopsticks and toothpicks) but would be largely free of heavy metal contaminants (lead, cadmium, mercury) (Han and Shin, 2002; Cirne et al., 2006; Forster-Carneiro et al., 2007). Food waste can be defined as any edible waste from food production, transportation, distribution and consumption. It is also referred to as garbage, swill and/or kitchen refuse, solid and liquid by-product wastes. They are generated throughout food production and processing sectors. In total, food waste generated from food preparation may constitute as much as 20% of the total human food supply from the stage of processing to the point of consumption (Westendorf et al., 1998). The low heavy metal content of restaurant waste distinguishes it from household garbage which often has high heavy metal levels rendering the breakdown residues of biological digestion processes too contaminated with heavy metals
Abbreviations: FW/l, fresh weight per litre; BOD, Biochemical Oxygen Demand; COD, Chemical Oxygen Demand; TS, total solid; °C, degree celsius; BSA, bovine serum albumin; GC, gas chromatography; CO2, carbon dioxide; mmol, millimoles; ml, millilitre; EtOH, ethanol. ⇑ Corresponding author. E-mail address: [email protected] (R.J. Ritchie).
to be readily useable. Zhang et al. (2007) identified food waste as an excellent feedstock for fermentation digestion processes. In Thailand a large component of the waste is cooked rice (as found for Korean restaurant waste, Han and Shin, 2002) which would be partially hydrolysed from cooking but would be expected to require further treatment to mobilize the starches. Davis (2008) used amylase treatment to mobilize the starch in food waste. Similar methods have been used for preparing corn paste wastes (Akpan et al., 2008) and sweet potato processing waste (Qian et al., 2008) but acid hydrolysis has also been used to mobilize both starches and cellulosic material such as corn cobs, peanut shells and newspaper (Akpan et al., 2005, 2008). Sawayama et al. (1997) used high temperature and pressure to thermally hydrolyse food waste prior to fermentation treatment. The advantage of acid hydrolysis is that it mobilises both starches and some of the cellulose and many other organic compounds. The obvious drawbacks to acid hydrolysis are that it is expensive, involves bulk handling of acids which are highly corrosive and might mobilise undesirable toxic metals and finally the acid hydrolysate has to be neutralized before being inoculated with microorganisms. Han and Shin (2002) used microbial populations from ruminants as a source of acidophile microbes to break down cellulose and starches from Korean restaurant waste, however the main products from such fermentation were acetic and butyric acids. These are not high value products. Lipids of various kinds (fats and oils) are a major component of restaurant wastes (Cirne et al., 2006) and present particular problems for both physical reasons and in their metabolism. Yeasts and methanogen bacteria can
http://dx.doi.org/10.1016/j.wasman.2015.04.028 0956-053X/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Suwannarat, J., Ritchie, R.J. Anaerobic digestion of food waste using yeast. Waste Management (2015), http://dx.doi.org/ 10.1016/j.wasman.2015.04.028
2
J. Suwannarat, R.J. Ritchie / Waste Management xxx (2015) xxx–xxx
use hydrolysis products of fats and lipids but they are generally thought not to be able to rapidly ferment unprocessed lipids. There are also physical problems arising from fats and greases in food waste: fat saturated materials aggregate into grease balls and fats and greases deposit themselves on the walls of plumbing and storage vats. Physical problems arising from fats and grease balls may prove to be a major limitation to using microbial methods to process restaurant waste. Food waste is a complex biomass containing various components such as starchy, fatty, and cellulosic materials. Without some sort of processing, these organic polymer materials may be difficult for ethanol producing microorganisms such as Saccharomyces cerevisiae (common yeast) to utilize. Food waste generated in Korea is rich in carbohydrate, as high as 65% of total solids due to its high proportion of cooked rice (Kim et al., 2008). Food waste is difficult to dispose of by incineration. Most food waste has been placed in landfills together with other wastes (Han and Shin, 2002). Food waste is the major component of organic matter in garbage and so when it is disposed of in landfills it is the major source of methane gas produced in the landfills and a major contributor to the organic matter in leachates. Adding restaurant food waste to the waste stream exacerbates such problems. Forster-Carneiro et al. (2007) studied improvements in the efficiency of semidry anaerobic digestion and dry fermentation (20– 35% TS), where little or no water, or sludge is added to the organic fraction of the municipal solid waste to produce an inert biosolid final product with methane production from methanogenic bacteria. The anaerobic digestibility and biogas and methane yields of the food waste were evaluated using batch anaerobic digestion tests performed at 50 °C. The results of the study indicated that the food waste is a highly desirable substrate for anaerobic digesters with regards to its high biodegradability and methane yield. Anaerobic digestion is the method of choice for the treatment of organic waste. This method has advantages of low-level sludge production, low-level energy consumption and potentially useful amounts of methane or ethanol production. However, solid organic materials such as kitchen garbage and sewage sludge are not digested quickly anaerobically. This is mainly a result of the unfavourable surface area/volume ratio limiting the surface area accessible to the microbes. Formation of grease balls worsens this surface area/volume problem and accessibility of the substrates to the bacteria. A liquidisation step is needed to speed up the anaerobic digestion. Analysis of the energy balance of thermal/pressure liquidisation followed by anaerobic digestion treatment was better than direct incineration (Sawayama, 1997). The physical and chemical characteristics of the organic waste are important information for designing and operating anaerobic digesters. They affect biogas/ethanol production and process stability during anaerobic digestion and the costs of handling the feedstock material (Zhang et al., 2007). Yeasts on the other hand are extremely reliable organisms to grow in culture and backup cultures are easily maintained. This is a crucial advantage over methanogen digestion. Yeasts are the mainstay of the fermentation industry (Madigan et al., 1997). Food waste contains high enough levels of proteins and amino acids and ammonia from the hydrolysis and breakdown of foodstuffs to adequately supply fixed nitrogen for the growth of yeasts. Yeasts also have a requirement for phosphorus, which is assimilated as a dihydrogen phosphate ion, and sulphur, which can be assimilated as a sulphate ion or as organic sulphur compounds such as the amino acids methionine and cysteine. Some metals, like magnesium, iron, calcium, and zinc and potassium are also required for good growth of the yeast (Shimoda, 2004). Food waste is unlikely to lack the essential minerals required by yeasts or their vitamin requirements (Ritchie and Raghupathi, 2008). In previous
studies we have shown that yeast very efficiently removed ammonia from food waste but was not efficient at removal of phosphate (Suwannarat and Ritchie, 2013a,b). S. cerevisiae is used for stable ethanol fermentation around the world, but it is important that yeasts lack the full range of amylolytic enzymes (a-amylase, b-amylase and glucoamylase) required to fully break down starches to glucose. Yeast has only two genes for amylases, YIL099W (SGA1) and YIR019C (FLO11, MUC1 and STA4) (KEGG, 2013). These enzymes are both a-glucoamylases (EC:3.2.1.3) (KEGG, 2013). Watanabe et al. (2009) evaluated the culture conditions and material compositions for efficient ethanol production from rice washing drainage: they used rice bran as a cheap source of amylolytic enzymes (especially a-, b-amylase) because rice washing drainage from rice polishing contained only a-glucosidase. The alternative of digesting starches by acid hydrolysis has practical limitations already pointed out above. The efficiency of yeast in digesting bulk lipids and fats is not well documented but yeast is known to produce emulsifying agents which would help in breaking up aggregates of oils, greases and sludges (Barriga et al., 1999). Watanabe et al. (2009) studied yeast-based anaerobic batch fermentation of rice waste (main culture: net volume 30–36 ml), using rice washing drainage (30 ml) as the substrate with lactic acid (final concentration: 100 mM) as the bactericidal agent. Different weights of rice bran were mixed in a 50 ml centrifuge tube, and then 1.0 ml of pre-culture yeast broth was inoculated. Fermentation processes were terminated after 14 days. The concentration of ethanol and sugars was analysed using an HPLC. The maximum ethanol concentration attained was 6.2% (V/V) (Watanabe et al., 2009). Currently, ‘‘spent yeast’’ has a low value and is used as a protein supplement in animal feed (Barriga et al., 1999). The brewing industry is a ready bulk source of ‘‘spent yeast’’ which could be used to digest food waste at minimal cost. In using yeast to breakdown food wastes the incubations can be run at higher temperatures than used in brewing to maximize the digestion rate because the ‘‘flavor’’ of the product is not relevant in the case of using yeast to break down food waste (Fleet et al., 2009). There are several additional reasons why yeast was selected for the study. Yeast is completely sequenced (KEGG, 2013). It can be genetically transformed allowing for either the addition of new genes or deletion through homologous recombination. A complete set of yeasts with single knockout mutations is available. The major advantage of yeast remains, however, its ease of cultivation and ready availability. Many different strains are available and standard procedures could be used to select a strain most suitable for anaerobic breakdown of food waste. 2. Materials and methods 2.1. Microorganism and culture conditions S. cerevisiae, in this study was obtained from the Biology Laboratory of Prince of Songkla University, Phuket campus. It is a baker’s yeast strain. 2.2. Inoculum yeast culture Stocks of S. cerevisiae were maintained on agar plates in medium containing 6 g of peptone, 3 g of yeast extract, 6 g of dextrose, 15 g of agar added to 300 ml of distilled water. Liquid cultures were grown in the same medium with the agar omitted. The standard autoclaving routine was 121 °C for 15 min. Liquid cultures were grown out without aeration. The backup yeast cultures were maintained in the refrigerator at 4 °C until required (Akpan et al., 2008).
Please cite this article in press as: Suwannarat, J., Ritchie, R.J. Anaerobic digestion of food waste using yeast. Waste Management (2015), http://dx.doi.org/ 10.1016/j.wasman.2015.04.028
J. Suwannarat, R.J. Ritchie / Waste Management xxx (2015) xxx–xxx
2.3. Food waste Food waste used in this study was collected from stalls, trolleys, restaurants, and canteens in Phuket province. Batches of about 1 kg were collected and then mixed together. First plastic, chopsticks and toothpicks, paper and bones were removed and then the samples were homogenized in a blender and then stored frozen for experiments. Samples were thawed and diluted with distilled water and then autoclaved before inoculating with yeast (S. cerevisiae). Cultures were grown anaerobically in culture bottles with magnetic spin bars on magnetic stirrers. The gas exit pipe of the culture bottles was fitted with a simple pneumatic trough to ensure anaerobic conditions (Fig. 1) (Suwannarat and Ritchie, 2013b). 2.4. Analysis of food waste The major components of the restaurant food waste were identified and classified (Suwannarat, 2014). Homogenized food waste was analysed for characteristics of food waste such as pH, total solid, moisture, phosphate, ammonia, total sugar, reducing sugar, starch, wet-dry ratio, lipid, protein, carbohydrate, ammonia, COD and BOD were measured (Suwannarat, 2014). In the present study, the analyses of composition of autoclaved food waste before and after fermentation included: 2.4.1. Reducing sugar and total sugar were assayed using the Benedict assay (APHA 1998) 2.4.2. Starch was measured using the Iodine assay (APHA, 1998). 2.4.3. Lipid and oil were measured using a UV Spectrophotometer assay (APHA 1998).
An anaerobic digestion using yeast (S. cerevisiae) fermentation to produce CO2.
Yeast Culture
3
2.4.4. COD (Chemical Oxygen Demand), and BOD (Biochemical Oxygen Demand) were analysed according to Standard Methods (APHA, 1998). 2.4.5. Protein was measured using the standard Lowry method (Lowry et al., 1951) using BSA (bovine serum albumin) as standard. 2.4.6. Ammonia was assayed using the Solorzano assay (Solorzano, 1969). 2.4.7. CO2 was measured volumetrically using a simple pneumatic trough (Fig. 1). Volumes of gas were converted into moles of CO2 using the universal gas law.
PV ¼ nRT PV )n¼ RT
ð1Þ
where P is the standard atmospheric pressure (101.3 kPa), V is the volume of gas measured, R is the universal gas constant (8.3143 J K1 mol1), T is the temperature (25 °C or 298.15 K), n is the number of moles of CO2. 2.4.8. Ethanol assay by gas chromatography Alcohol (Ethanol) was measured using a gas chromatograph (Agilent 7890A GC) at Prince of Songkla University-Suratthanee, Suratthanee, Thailand using standard methods of operation as described by Amonsit and Petsom (1995). Ethanol samples in water with known concentrations of ethanol were placed in crimp sealed vials and equilibrated to 60 °C. A sample needle was then inserted into the gas phase and injected into the GC and a standard curve prepared which was used to estimate ethanol content of samples from the yeast fermentations. The Agilent software correctly identified the ethanol standards as ethanol. 2.5. Fermentation protocol Anaerobic digestions were carried out using a baker’s yeast strain of yeast (S. cerevisiae) at 25, 30, 35, 40 °C. The standard experimental culture volumes were 175 ml inoculated with 10, 15, 20, 25, 30, 35 ml of S. cerevisiae culture in 250 ml culture bottles. Food waste was sterilized at 121 °C for 30 min and experiments set up using of food waste 700 g FW (fresh weight)/l and using 10, 15, 20, 25, 30, 35 ml of S. cerevisiae culture. Each culture was fitted with a simple pneumatic trough to collect fermentation gases and to ensure anaerobic conditions (Fig. 1) (Suwannarat and Ritchie, 2013b). The starter yeast cultures used for inoculating the food waste contained about 5 106 cells/ml. Incubations were carried out for 7, 20, 25 or 30 days and the standard feedstock was 700 g FW/l because preliminary experiments had shown that yeast could easily break down a food waste feedstock loading of 700 g FW/l. 3. Results and discussion
Pneumatic trough to collect gas
Fig. 1. Experimental set-up for anaerobic fermentation of food waste and measurement efficiency yeast digests food waste. The culture was grown in a culture bottle with a magnetic stirrer. The gas exhaust pipe ran into a simple pneumatic trough to collect the gas and prevent entry of oxygen.
Food waste composition (protein, reducing sugar, total sugar, lipid and starch) were measured at the start of the incubation and at intervals during the course of the fermentation over a period of up to 30 days. Based on the standard APHA assay methods the major components of the autoclaved food waste used in the present study on a dry weight basis were: total non-reducing sugar 51%, protein 20%, lipid 8%, reducing sugar 15% and starch 6% (based on, Suwannarat and Ritchie 2013a and Suwannarat 2014). A range of inoculation volumes from 10 to 35 ml were used. Fig. 2 shows digestion values found after 30 day incubations. Yeast removed about 90–97% of total sugars (reducing and non-reducing sugar). A consistent removal rate of about 50–60% was found for reducing sugars. Most of the sugar present in the food waste was not present
Please cite this article in press as: Suwannarat, J., Ritchie, R.J. Anaerobic digestion of food waste using yeast. Waste Management (2015), http://dx.doi.org/ 10.1016/j.wasman.2015.04.028
4
J. Suwannarat, R.J. Ritchie / Waste Management xxx (2015) xxx–xxx
Optimising temperature for efficient digestion of food waste
Optimising volume of food waste for efficient digestion
100
lipid
120 100
80
protein
60
reducing sugar
40 20
total sugar
0 10
15
20
25
30
%Removed
% Removed
120
starch
80
25 C 60
35 C
20
40 C
35
Volume of yeast Fig. 2. Percent removal of measured parameters of food waste using yeast (S. cerevisiae) digestion. Error bars are ±standard errors using n = 6 replicates.
Lipid
Starch
Fig. 4. Percent removal of lipid, protein, total sugar, reducing sugar and starch at different fermentation temperatures (25, 30, 35, 40 °C). Error bars are ±standard errors using n = 6 replicates.
Volume of CO 2 (mmol/Day)
Gas production (CO 2 ) vs. Inoculation Volume 8
30
7
25
6 20
5 4
15
3
10
2 5
1
0
0 0
1
2
3
4
5
6
7
10 ml 15 ml 20 ml 25 ml 30 ml 35 ml
Day Fig. 5. Volume of CO2 produced from food waste using different of inoculums of yeast. The inoculum contained 5 106 cells/ml. The daily production of CO2 (solid lines) is on the left (Y) scale and the total cumulative gas production scale is on the right YY scale (dotted lines).
Volume of CO2 (mmol/Day)
Gas production (CO 2 ) vs. Incubation temperature 25
7 6
20
5 4
15
3
10
2 5
1 0
0 0
100 90 80 70 60 50 40 30 20 10 0
Protein Total sugar Reducing sugar
2
4
6
CO2 (mmol cumulative total)
Optimal of yeast inoculum
0
CO2 (mmol cumulative total)
as reducing sugar (Suwannarat, 2014). Yeast was much more efficient at removing non-reducing sugars than reducing sugars and there may have been some accumulation of reducing sugars such as glucose from the breakdown of starch. Lipid removal was also a consistent 30–40%. Protein removal was more variable, ranging from 30% to 80% but this variability was not a function of inoculation volume. In the case of protein, lipid, starch, and reducing sugar the best of inoculation volume of yeast was 10 ml of yeast/175 ml of culture. In terms of BOD and COD, the inoculation volume made little difference to the overall BOD for 30 day incubations, averaging about a 40% reduction in BOD (Fig. 3). On the other hand COD reduction was lower for lower inoculation volumes (
