- Email: [email protected]
Ethanol production from kitchen waste using the flocculating yeast Saccharomyces cerevisiae strain KF-7
ARTICLE IN PRESS BIOMASS AND BIOENERGY
32 (2008) 1037 – 1045
Available at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
Ethanol production from kitchen waste using the flocculating yeast Saccharomyces cerevisiae strain KF-7 Yue-Qin Tanga, Yoji Koikeb, Kai Liua, Ming-Zhe Ana, Shigeru Morimuraa,, Xiao-Lei Wuc, Kenji Kidaa a
Graduate School of Science and Technology, Kumamoto University, 2-39-1 Kurokami, Kumamoto 860-8555, Japan Tokyo Gas Co., Ltd., 1-7-7 Suehiro-cho, Tsurumi-ku, Yokohama City, Kanagawa 230-0045, Japan c Department of Energy and Resources Engineering, College of Engineering, Peking University, Beijing 100871, China b
art i cle info
ab st rac t
Article history:
A process for producing ethanol from kitchen waste was developed in this study. The
Received 28 July 2007
process consists of freshness preservation of the waste, saccharification of the sugars in
Accepted 21 January 2008
the waste, continuous ethanol fermentation of the saccharified liquid, and anaerobic
Available online 2 April 2008
treatment of the saccharification residue and the stillage. Spraying lactic acid bacteria (LCB)
Keywords: Kitchen waste Ethanol fermentation Methane fermentation Saccharomyces cerevisiae Flocculating yeast
on the kitchen waste kept the waste fresh for over 1 week. High glucose recovery (85.5%) from LCB-sprayed waste was achieved after saccharification using Nagase N-40 glucoamylase. The resulting saccharified liquid was used directly for ethanol fermentation, without the addition of any nutrients. High ethanol productivity (24.0 g l
1
h 1) was
obtained when the flocculating yeast strain KF-7 was used in a continuous ethanol fermentation process at a dilution rate of 0.8 h
1
. The saccharification residue was mixed
with stillage and treated in a thermophilic anaerobic continuous stirred tank reactor (CSTR); a VTS loading rate of 6 g l
1
d
1
with 72% VTS digestion efficiency was achieved.
Using this process, 30.9 g ethanol, and 65.2 l biogas with 50% methane, was produced from 1 kg of kitchen waste containing 118.0 g total sugar. Thus, energy in kitchen waste can be converted to ethanol and methane, which can then be used as fuels, while simultaneously treating kitchen waste. & 2008 Elsevier Ltd. All rights reserved.
1.
Introduction
In Japan, the annual generation of organic waste from the food industry and kitchen garbage is about 20 million tons per year. Most of this waste is directly incinerated with other combustible waste, and the residual ash is disposed of in landfills. However, incineration is costly and energy consuming, and since these wastes have a high moisture content, they must be burnt with other drier wastes, resulting in the production of dioxins. Recently, more economically attractive treatment methods such as composting and anaerobic digestion have been used to treat some of this waste and Corresponding author. Tel./fax: +81 96 342 3669.
E-mail address: [email protected] (S. Morimura). 0961-9534/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2008.01.027
produce compost or biogas, but imbalances in supply and demand for recycled products hinder the wider application of these alternative approaches. Since these wastes are rich in sugars that could be converted to more valuable products, attention is being directed to biorefinery processing of these wastes to produce biomass materials such as lactic acid [1–3]. In addition, ethanol can also be produced from these sugars. Ethanol, a major fuel additive and a promising future energy alternative, is now produced mainly from corn in America and China and from sugarcane in Brazil. However, since corn is a major food source, its use as a fuel source has been criticized. Moreover, using corn as a feedstock is a major
ARTICLE IN PRESS 1038
BIOMASS AND BIOENERGY
contributor to the significant cost of ethanol production, so biomass wastes such as corn fiber, waste wood and food wastes are far more attractive as cheap feedstocks for ethanol production. In this study, we focus on ethanol production from kitchen waste collected from dining halls. As kitchen waste is highly
Table 1 – Representative composition of kitchen waste used for ethanol fermentation Results (ww 1%)
Test parameters Total solid (TS) Volatile total solid (VTS) Ash Moisture Total sugarsa (based on wet weight) Total sugarsa (based on dry weight) Holocellulose (based on dry weight) Lignin (based on dry weight) Protein (based on dry weight) Lipid (based on dry weight)
19.7 18.8 1.9 80.3 11.8 59.8 1.6 0.8 21.8 15.7
a
Total sugars referred to starch sugars. Protein contents were calculated with total nitrogen contents by multiplication as coefficient 6.25. Results were mean values of 3 analyses.
Kitchen waste Spraying of lactic acid bacteria
perishable, and most of the sugars in the waste are starchbased sugars (Table 1), a process for producing ethanol from kitchen waste was developed, as shown in Fig. 1. The process consists of freshness preservation of the waste with the aid of lactic acid bacteria (LAB), saccharification of sugars in the waste using glucoamylase, continuous ethanol fermentation of the saccharified liquid using flocculating yeast, and anaerobic treatment of the saccharification residue and stillage. This process converts energy in the waste first to ethanol, and then to methane, which can then be used as a fuel; in addition, the waste is simultaneously treated. Although there have been several reports regarding the production of lactic acid from food waste [1–3], there have been only very limited studies regarding the production of ethanol from food waste [4,5]. The present paper is the first report regarding ethanol fermentation from kitchen waste.
2.
Methods
2.1.
Microorganism strains
The flocculating yeast Saccharomyces cerevisiae KF-7 was used for batch and continuous ethanol fermentation tests. This yeast strain was constructed by protoplast fusion of the flocculating yeast strain IR-2 and the thermotolerant yeast strain EP-1 [6]. LAB strain 4, closely related to Lactobacillus paracasei (98% rRNA gene sequence similarity), was isolated from a vegetable beverage and used for the preservation of kitchen waste.
2.2.
Storing
Saccharification
Solid-liquid separation Liquid Ethanol fermentation
2.3. Distilling
Ethanol
Waste water Methane fermentation
Sewage plant
Cultivation of LCB
LCB cells grown on a 1% YPD (1% glucose, 1% yeast extract, 2% peptone, 2% agar) slant were transferred to 1% YPD liquid medium and cultured at 37 1C for 24 h with mixing at 100 rpm using a rotary shaker. The resultant pre-cultivation broth was used for jar fermentor cultivation using tofu compression liquid (pH 5.85; glucose 190 mg l 1; suspended solid (SS) 3.38 g l 1; volatile suspended solid (VSS) 2.79 g l 1) as the medium. Cultivation was carried out at 37 1C and 100 rpm for 24 h. The culture broth, with approximately 1010 cell ml 1, was stored in a spray bottle and used in kitchen waste storage experiments. The LCB broth could be stored for 1 month without obvious change in the number of viable cells.
Water
Solid
32 (2008) 1037 – 1045
Biogas
Treated water
Fig. 1 – Flow chart of the ethanol production process from kitchen waste.
Storage of kitchen waste
Kitchen waste used in this study was collected from the students’ dining hall of Kumamoto University, Japan. Large pericarps were removed, the remaining waste was stored in a plastic basket at room temperature, and LCB broth was evenly sprayed on the surface of the waste once a day when necessary. A water weight was placed over the waste in order to produce anaerobic conditions. Representative characteristics of the collected kitchen waste are shown in Table 1. Kitchen waste stored at room temperature for 3 days with and without LCB treatment was used for lab-scale experiments, and LCB-sprayed waste stored for 1 week was used for bench-scale experiments.
ARTICLE IN PRESS BIOMASS AND BIOENERGY
2.4.
Saccharification of the kitchen waste
2.4.1.
Lab-scale experiment
pH meter
Wet gas meter
One kilogram of waste was chopped into small pieces using a fruit mixer, then transferred into a 3 l jar fermentor containing 0.5 kg tap water. Two kinds of glucoamylase, Glucochimu #20000 (protein content 28.3 w w 1%) and Nagase N-40 (protein content 80.1 ww 1%) (Nagase ChemteX Corporation, Osaka, Japan) were used for saccharification, and the results were compared. Enzyme was added based on its protein content to a final concentration of 170 mg protein kg 1 wet waste. Saccharification was performed at 50 1C or 60 1C, 150 rpm for 6 h. Saccharified liquid was separated by centrifugation at 10,000 rpm for 10 min, then used for batch and continuous ethanol fermentation.
2.4.2.
2.5. Batch ethanol fermentation of saccharified liquid prepared from kitchen waste Preparation of inoculum
Yeast cells grown on 2% YPD (2% glucose, 1% yeast extract, 2% peptone, 2% agar) were inoculated into 100 ml of 5% YPD (5% glucose, 1% yeast extract, 2% peptone) medium in a 500ml flask. Pre-cultivation was performed aerobically at 30 1C for 16 h with mixing at 160 rpm using a rotary shaker, then the resulting pre-cultivation broth was used as inoculum for ethanol fermentation experiments.
2.5.2.
Fermented mash WV, 0.45L
Thermostated water
Bench-scale experiment
Kitchen waste (130.0 kg) was crushed using a compact chopper (MKBC-32; Masuko Sangyo Co. Ltd., Saitama, Japan), transferred to a 300 l reactor (CNVM1-E085-6, Sumitomo Heavy Industries, Ltd., Tokyo, Japan), and 65.0 kg tap water was added. Nagase N-40 was added to a final concentration of 160 mg protein kg 1 wet waste. Saccharification was performed at 60 1C, 150 rpm for 2 h. Saccharified liquid (134.0 l, 64.0 g glucose l 1) was separated from the solid residue (52.0 kg) using 6 filter press units (RF Fully Automatic Filter Press; Kurita Machinery Mfg. Co., Ltd., Osaka, Japan). The recovered saccharified liquid was used directly for continuous ethanol fermentation.
2.5.1.
1039
32 (2008) 1037 – 1045
Air P-2 P-1 Saccharified liquid Fig. 2 – Schematic diagram of the continuous ethanol fermentation process. Working volume of reactor is 0.45 l.
sterilized using a NaClO solution, as described previously [7]. The reactor was inoculated with about 0.45 l of precultivation broth. The unsterilized saccharified liquid medium was fed directly into the bottom of the reactor by peristaltic pump P-1 (Fig. 2), and the fermented mash overflowed from the top of the reactor via a solid–liquid–gas separation unit. The liquid in the reactor was circulated using roller pump P-2. The continuous fermentation process used a chemostat method to supply the feedstock, i.e., the dilution rate was increased in a stepwise fashion to increase the feed rate.
2.7. Thermophilic anaerobic treatment of the saccharification residue and stillage mixture
Batch ethanol fermentation
For batch fermentation tests, a 10-ml inoculum of precultivation broth was added to 90 ml of sterilized fermentation medium in a 300-ml flask. The flask was immersed in a thermostated water bath and the medium was stirred using a magnetic stirrer. Batch fermentation was carried out at 30 1C for 24 h.
2.6. Continuous ethanol fermentation of saccharified liquid
2.7.1.
2.7.2. Fig. 2 shows a schematic diagram of our continuous ethanol fermentation system, described previously [7]. The towershaped reactor, with a working volume of 0.45 l, had a natural solid–liquid–gas separation unit in the upper portion of the reactor. Air was supplied continuously to the bottom of the reactor through a ball filter at an aeration rate of 0.025 vvm. Prior to the fermentation experiment, the reactor was
Preparation of the mixture prior to anaerobic digestion
All saccharification residues and stillage from the bench saccharification and ethanol fermentation experiment were mixed together and used as feed for the 10-l anaerobic fermentation reactor. To accelerate the methane fermentation rate, Ni2+, Co2+, and Fe2+ were added at concentrations of 4.57, 4.57 and 61.11 mg l 1, respectively, to the mixture [8]. Table 2 shows the characteristics of the waste mixture.
Anaerobic treatment of the mixture
Fig. 3 shows a schematic diagram of the thermophilic methane fermentation apparatus. A CSTR-type (completely stirred tank reactor-type) reactor (Jar Fermentor MBF; Eyela Tokyo Rikakikai Co. Ltd., Tokyo, Japan) with a working volume of 8 l was used as the methane fermentation reactor. Eight liters of anaerobic digested sludge, provided by Kumamoto Hokubu sewage works (Kumamoto, Japan), was added to the
ARTICLE IN PRESS 1040
BIOMASS AND BIOENERGY
Table 2 – Characteristics of the mixture of saccharification residue and stillage Results (mg l 1)
Test parameters Total solid (TS) Volatile total solid (VTS) Suspended solid (SS) Volatile suspended solid (VSS) CODCr BOD5 Lactic acid Acetic acid NH+4 -N NO3 -N
65,000.0 57,000.0 40,000.0 38,000.0 105,150.0 30,571.0 8703.8 2106.0 109.5 27.8
1 Pump P-1
Wet gas meter
3
Aeration
4
Feeding 2
Drawing
32 (2008) 1037 – 1045
was analyzed by UV using an F-Kit-Glucose (Roche Diagnostics, Germany). The protein content of the kitchen waste was calculated as 6.25 nitrogen content, and the nitrogen was determined using a carbon, hydrogen, nitrogen (CHN) coder MT-3 (Yanako, Tokyo, Japan) [10]. Lignin, lipid, and holocellulose were analyzed by methods used for testing pulpwood [11]. Total solid (TS), total volatile solid (TVS), SS, VSS and BOD5 were analyzed in accordance to Standard Methods [10]. CODCr and NH+4 -N and NO3 -N were measured using the HACH method (Hach Co., Loveland, CO, USA). The moisture content was determined by the decrease in weight following drying overnight at 105 1C, and the ash was determined at 600 1C by combustion for 2 h. Ethanol was assayed by gas chromatography using an internal standard (isopropanol) as described previously [7]. Volatile fatty acids (VFAs) including lactic acid, acetic acid, propionic acid, butyric acid, isobutyric acid, valeric acid, and isovaleric acid were analyzed as described previously [12]. Soluble total organic carbon (S-TOC) was analyzed using a TOC auto analyzer (TOC-500; Shimadzu, Kyoto, Japan). The methane content of the biogas was measured by gas chromatography using a thermal conductively detector (TCD) (KOR-2G; GL Science Co., Tokyo, Japan) equipped with a packed column (Porapack Q, GL Science). Hydrogen sulfide was measured using Kitagawa precision gas detector tubes (Komyo Kitagawa, Tokyo, Japan). The number of total yeast cells and viable cells were calculated by the methylene blue method using a hematitometer.
Ball filter
3. Thermophilic methane fermentor (53°C;WV 8 L)
5-L Erlenmeyer flask with 3 L tap water
Fig. 3 – Schematic diagram of the methane fermentation process used for the treatment of the saccharification residue and stillage mixture.
reactor, and the saccharification residue and distilling wastewater mixture was treated by the draw-and-fill method. The pH, temperature and agitation speed were maintained at 7.0, 53 1C, and 200 rpm, respectively. Under non-aeration conditions, the generated biogas was channeled directly through line 1 into a gas meter (Fig. 3). In order to reduce the concentration of hydrogen sulfide in the biogas and in the reactor, at a TS volumetric loading rate of 4 g l 1 d 1 the reactor was micro-aerated by continuously supplying air at 7.5% of the amount of biogas being produced in the methane reactor. Under these conditions, the generated biogas was channeled first into a 5-l flask with tap water through line 2, and then into a gas meter through line 4. A portion of the biogas was recirculated into the methane reactor though line 3 using pump P-1.
2.8.
Analytical methods
After hydrolyzing a sample of kitchen waste with 25% HCl, the total sugar concentration in the waste was assayed by the Somogyi–Nelson method [9]. The concentration of glucose
Results and discussion
3.1. Effect of spraying LAB on glucose recovery from kitchen waste Kitchen waste stored without being sprayed with LAB produced an unpleasant odor within 24 h, and the entire surface of the waste was covered with putrefying fungi after 3 days. In contrast, waste sprayed with LAB and then stored produced no offensive odor and remained fresh for more than 1 week. These results indicated that lactic acid fermentation could inhibit the growth of putrefactive bacteria and fungi, making the preservation and deodorization of kitchen waste possible. Kitchen waste that was either sprayed or not sprayed with LAB, then stored for 3 days, was subjected to saccharification at 50 1C in order to determine the effect of LAB treatment on the recovery of glucose (Table 3). Fresh kitchen waste without storing (hereafter called fresh waste) was used as a control. In the case of fresh waste treated with Glucochimu #20000, high glucose recovery (97.5%) was achieved and the glucose concentration in the saccharified liquid was approximately 74.0 g l 1. Using the same glucoamylase, glucose recovery efficiencies were 72.0% and 60.1% for waste stored for 3 days with and without LAB spray treatment, respectively. Although lower than the recovery efficiency from fresh waste, waste sprayed with LAB showed much higher glucose recovery than unsprayed waste, suggesting that spraying with LAB could conserve more sugar that could then be used for ethanol fermentation. When Nagase N-40 was used as the glucoamylase, higher glucose
ARTICLE IN PRESS BIOMASS AND BIOENERGY
1041
32 (2008) 1037 – 1045
Table 3 – Comparison of sugar recoveries from stored kitchen wastea in the saccharification step with or without waste treatment with lactic acid bacteria (LCB) Sample
Fresh waste Waste stored for three days
Enzyme
Glucochimu #20000 Glucochimu #20000 Nagase N-40
Spraying of LCB
No spraying of LCB
Glucose in saccharified liquid (g l 1)
Glucose recovery (%)
Glucose in saccharified liquid (g l 1)
Glucose recovery (%)
–
–
74.1
97.5
56.6
72.0
47.3
60.1
60.5
77.0
51.2
65.1
a Kitchen waste was 1.5-times diluted before saccharification and the total sugars of the mixture was 78.0 g l 1. Saccharification enzyme was added based on its protein content to a concentration of 170 mg protein kg 1 wet waste. Saccharification was carried out in a 50 1C incubator shaker at 100 rpm for 6 h.
recoveries (77.0% and 65.0%) were achieved for waste sprayed or not sprayed with LAB, respectively, indicating that Nagase N-40 was better at saccharification than Glucochimu #20000 at 50 1C. As the pH of the waste after storage was below 4.0, these higher glucose recoveries might be due to the higher acid tolerance of Nagase N-40.
Table 4 – Effect of saccharification temperature on glucose recovery from kitchen wastea Saccharification temperature (1C)
50
3.2. Effect of saccharification temperature on glucose recovery from kitchen waste 60
To reduce the population of active bacteria present in the nonsterilized saccharified liquid (used directly for ethanol fermentation, the next step of the process), saccharification at 60 1C was studied using Glucochimu #20000 and Nagase N-40 (Table 4). In the case of Glucochimu #20000, glucose recovery efficiency decreased from 72.0% to 25.4% when the saccharification temperature increased from 50 to 60 1C. However, in the case of Nagase N-40, glucose recovery increased from 75.8% to 85.5% when the saccharification temperature was increased to 60 1C. Nagase N-40 was therefore shown to be more thermo-resistant than Glucochimu #20000 and was used at 60 1C in all subsequent saccharification experiments.
3.3. Effect of nitrogenous nutrients and pH on ethanol fermentation of saccharified liquid from kitchen waste Nitrogenous nutrients of yeast extract and (NH4)2SO4 were added to the saccharified liquid to investigate their effect on ethanol fermentation at pH 4.5 (Table 5). It was found that ethanol concentrations in fermented media containing extra nitrogen-based nutrients were approximately 30.0 g l 1, the same as in the absence of these additives. The ethanol yield was 0.47 (92% of the theoretical yield of 0.51), suggesting that the nutrients already present in the saccharified liquid were sufficient for the growth and maintenance of the yeast. The pH of the saccharified liquid was affected by the storage time of the waste, but was generally in the range 3.6–3.9. In order to enhance ethanol production, the pH of the medium was usually adjusted to an optimum pH prior to fermentation. Although the yeast strain KF-7 is somewhat acid tolerant, the effect of pH on ethanol fermentation was studied using saccharified liquids with different pHs (Table 5).
Enzyme
Glucose in saccharified liquid (g l 1)
Glucose recovery (%)
Glucochimu #20000 Nagase N-40
56.6
72.0
59.6
75.8
Glucochimu #20000 Nagase N-40
20.0
25.4
67.2
85.5
a
Kitchen waste was 1.5-times diluted before saccharification and the total sugars of the mixture was 78.6 g l 1. Saccharification enzyme was added based on its protein content to a concentration of 170 mg protein kg 1 wet waste. Saccharification was carried out in a 50 or 60 1C incubator shaker at 100 rpm for 6 h.
Table 5 – Effect of nitrogenous nutrients and pH on ethanol fermentation of saccharified liquid prepared from kitchen wastea Medium
Saccharified liquid+1% yeast extract Saccharified liquid+1% (NH4)2SO4 Saccharified liquid Saccharified liquid Saccharified liquid
Initial pH of the medium
Ethanol concentration (g l 1)
4.5
30.1
4.5
29.5
4.5 4.0 3.5
29.9 29.7 25.0
a Glucose concentration of saccharified liquid was 64 g l 1. pH of mediums were adjusted using 1 N NaOH solution. Ethanol fermentation was carried out at 30 1C for 24 h using 300-ml Erlenmeyer flask.
ARTICLE IN PRESS 1042
BIOMASS AND BIOENERGY
Ethanol concentration decreased significantly to 25.0 g l 1 at pH 3.5. However, the concentration of ethanol obtained at pH 4.0 was the same as that at pH 4.5, indicating that efficient fermentation could be achieved at pH of 4.0 using saccharified liquid.
3.4. Continuous ethanol fermentation of lab-scalegenerated saccharified liquid
32 (2008) 1037 – 1045
inside the reactor. The ethanol concentration in the reactor was approximately 30.0 g l 1, and the ethanol yield was about 0.47 (92% of the theoretical value). Microscopic observation of yeast cells sampled from the reactor confirmed that the flocculating character of the cells did not change throughout the entire fermentation period.
3.5. Study on ethanol productivity of continuous fermentation of bench-scale-generated saccharified liquid
Saccharified liquid (pH 3.9; glucose 64.0 g l 1; lactic acid 6200 mg l 1; acetic acid 1300 mg l 1) was prepared on a lab scale and, without adjusting the pH, was fed directly into the reactor and fermented continuously for 2 weeks at a dilution rate of 0.2 h 1 (Fig. 4). The concentrations of lactic acid and acetic acid in the reactor were almost the same as in the feed medium, suggesting no growth of acid-producing bacteria
To study the ethanol productivity, saccharified liquid (pH 3.85; glucose 64.0 g l 1; lactic acid 8000 mg l 1; acetic acid 1200 mg l 1) was prepared using bench-scale equipments consisting of a compact chopper, a saccharification reactor and a filter press. Prepared, unadjusted saccharified liquid was used as the feed for continuous ethanol fermentation for 1 month.
7000
30
6000
25
5000
20
4000
15
3000
10
2000
5
1000
0
, Acetic acid concentration (mg l-1)
35
, Lactic acid concentration (mg l-1)
Dilution rate (h-1)
Δ , Ethanol concentration (g l-1) , pH (-)
0.2
0 10
5
0
15
Time (d)
50
100
40
80
30
60
20
40
10
20
0
, Survival ratio of yeast cells (%)
, Ethanol concentration (g
×, Number of viable cell (108 ml-1)
l-1)
, Ethanol productivity (g l-1 h-1)
Fig. 4 – Continuous ethanol fermentation at 30 1C and an aeration rate of 0.025 vvm, using saccharified liquid medium prepared on a lab scale.
0 0.0
0.2
0.4
0.6
0.8
1.0
Dilution rate (h-1) Fig. 5 – Effect of dilution rate on ethanol productivity and viable cell number during continuous fermentation at 30 1C and an aeration rate of 0.025 vvm, using saccharified liquid medium prepared on bench scale equipments.
ARTICLE IN PRESS BIOMASS AND BIOENERGY
The dilution rate was initially 0.2 h 1, and was then increased to 0.4, 0.6, and finally to 0.8 h 1. Fig. 5 shows the ethanol concentration, ethanol productivity, survival ratio of yeast cells, and viable yeast cell number during continuous fermentation at different dilution rates. Ethanol fermentation was successfully conducted at all dilution rates studied. The ethanol concentration in the reactor averaged 30 g l 1 and the ethanol yield was approximately 0.47 throughout the fermentation period, almost the same as when saccharified liquid prepared on a lab scale was used (Fig. 4). High ethanol productivity of 24.0 g l 1 h 1 was achieved at a dilution rate of 0.8 h 1 though the glucose concentration in the feed was only 64.0 g l 1. We achieved similar ethanol productivity when using a molasses medium, or an acid hydrolysate from wood biomass with 150 g l 1 total sugar at a dilution rate of 0.4 h 1 [7,13]. In these two previous studies, the reactors could not be successfully operated at dilution rates of 0.5 or 0.6 h 1. However, in the present study, we successfully operated the system at a dilution rate of 0.8 h 1. This contrast with earlier studies could be due to the high survival rate of yeast cells and the high number of viable cells in the reactor (greater than 10 108 cell/ml) resulting from the excellent flocculating characteristics of strain KF-7
Kitchen waste
1043
32 (2008) 1037 – 1045
(Fig. 5). The present study is the first to achieve ethanol fermentation even at this high dilution rate. The acetic acid concentration in the reactor was the same as that in the feed medium. However, the lactic acid concentration in the reactor was approximately 9000 mg l 1, slightly higher than that in the feed medium, indicating some growth of LAB in the reactor regardless of the dilution rate used. As approximately 8000 mg l 1 of lactic acid was contained in prepared saccharified liquid, a small amount of sugar in the kitchen waste was converted to lactic acid and not the target product, ethanol. Since it is desirable from an economic standpoint that all sugars in the wastes be fermented to ethanol, a yeast strain capable of utilizing lactic acid to produce ethanol will be required in future studies. As the ethanol yield was approximately 0.47, 4.02 kg of ethanol could be produced from 134.0 kg of bench-scaleprepared saccharified liquid; since a total of 130.0 kg of kitchen waste was used, 30.9 g of ethanol could be obtained from 1 kg of kitchen waste by the ethanol fermentation process described above. However, if the calculation is based on the amount of glucose in saccharified broth before filter press separation, 45.0 g of ethanol should be produced from
130.0 kg (total sugar content: 118.0 g kg-1 kitchen waste)
65.0 L water
Saccharification efficiency: Saccharification
195.0 L (glucose content: 64.0 g l-1)
64.0 × 195.0 118.0 ×130.0
Glucose recovery efficiency in solid-liquid separation step:
Solid-liquid separation using filter press unit
64.0 × 134.0 64.0 × 195.0 52.0 kg
134.0 L
Solid residue
× 100% = 81.3 %
Saccharifiedliquid (glucose content: 64.0 g l -1 )
× 100% = 68.7%
Total glucose recovery efficiency: 64.0 × 134.0
× 100% =55.9%
118.0 ×130.0 65.0 L water By adding rinse operation Solid residue
65.0 L rinsed water
Glucose recovery efficiency in solid-liquid separation step: (glucose content: 52.0 g l -1)
81.3. ×134.0 64.0 × 195.0
× 100% = 87.2 %
Total glucose recovery efficiency:
Used as dilution water in saccharification step
81.3×134.0 118.0 ×130.0
× 100% =71.0%
134.0 L Saccharifiedliquid
(glucose content: 81.3 g l-1)
Fig. 6 – Mass balance in the saccharification and saccharified liquid separation processes.
ARTICLE IN PRESS 1044
BIOMASS AND BIOENERGY
1 kg of kitchen waste under conditions providing an ethanol yield of 0.47 (Fig. 6). The lower ethanol quantity was actually obtained because approximately 31% of the glucose in the saccharified broth was lost after the filter press separation (Fig. 6). As shown in Fig. 6, to recover the glucose remaining in the solid residue, a single rinse operation was carried out using 65.0 l of water, and the glucose recovery efficiency was determined. The glucose content in the rinse water was 52.0 g l 1. If this rinse water was used as dilution water in the saccharification step, the glucose content in the recovered saccharified liquid would increase from 64.0 to 81.3 g l 1, and the glucose recovery efficiency in the filter press separation step would increase from 68.7% to 87.2%. Therefore, the total glucose recovery efficiency in the saccharification and separation processes would increase from 55.9% to 71.0% (Fig. 6). Although we did not investigate these conditions in this study, we predict that if saccharified liquid with a glucose content of 81.3 g l 1 were used as the feedstock for ethanol fermentation, about 38.1 g of ethanol would be produced from 1 kg of kitchen waste.
3.6. Thermophilic methane fermentation of the mixture comprised of saccharification residue and stillage from the bench-scale experiment The total quantity of the mixture of saccharification residue and stillage from the bench-scale experiment was approxi-
32 (2008) 1037 – 1045
mately 186.0 l. As shown in Table 2, the concentration of VTS and VSS in the mixture was 57 and 38 g l 1, respectively. An anaerobic CSTR was required to treat this potent, high-solidcontent waste. Fig. 7 shows the time course of effluent quality, digestion efficiency, gas evolution rate, and the methane and H2S concentrations in the biogas under VTS volumetric loading rates from 1.0 to 8.0 g l 1 d 1. VTS digestion efficiencies were approximately 75% when the VTS loading rate was between 1.0 and 3.0 g l 1 d 1. The gas evolution rate was 2500 and 3400 ml l 1 d 1, which corresponds to 1000 ml (g-digested VTS) 1, at VTS loading rates of 2.0 and 3.0 g l 1 d 1, respectively. Propionic acid accumulated temporarily, but stopped when the system was operated longer at a VTS loading rate of 3.0 g l 1 d 1. However, when the VTS loading rate was increased to 4.0 g l 1 d 1, the accumulation of propionic acid reoccurred, accompanied by an increase in TOC concentration. Simultaneously, VTS digestion efficiency decreased to 73% and H2S concentration in the biogas increased from 800 to 1200 ppm. To reduce the repressive effect of H2S on methanogenesis micro-organisms [8], air at 7.5% of the amount of biogas evolved was supplied continuously into the upper region of the reactor from the 65th day of operation, resulting in a decrease in H2S content in the biogas to below 50 ppm and a decrease in TOC concentration from 2900 to 1800 mg l 1 following 20 days of micro-aeration (Fig. 7A, B). VTS digestion efficiency remained at approximately 75% under these micro-aeration conditions when the
VTS loading rate (g l-1 d-1) 4
6
8
Microaeration 7.5% of biogas produced
5000
100
4000
80
3000
60
2000
40
1000
20
0 0
20
40
60 80 Time (d)
100
120
0 140
5000
100
4000
80
3000
60
2000
40
1000
20
0
0 0
20
40
60 80 Time (d)
100
120
−, VTS digestion efficiency (%) ×, VSS digestion efficiency (%)
3
×, pH in reactor (-)
2
Δ, CH4 concentration in biogas (%)
, Biogas evolution rate (ml l-1 d-1) , H2S concentration in biogas (ppm)
, Propionate concentration (mg l-1) Δ, Acetate concentration (mg l-1) , Lactate concentration (mg l-1) , TOC concentration (mg l-1)
1
140
Fig. 7 – Thermophilic methane fermentation of the saccharification residue and stillage mixture.
ARTICLE IN PRESS BIOMASS AND BIOENERGY
VTS loading rate was 4 g l 1 d 1. When the VTS loading rate was increased to 6 g l 1 d 1, the VTS digestion efficiency decreased to 72% but the concentration of VFA was the same as that at a VTS loading rate of 4 g l 1 d 1. The gas evolution rate was approximately 4800 ml l 1 d 1 (1100 ml (g-digested VTS) 1) and the treatment was performed successfully. The methane concentration in the biogas under this condition was approximately 50%. However, the VTS digestion efficiency and the gas evolution rate decreased significantly, and the concentration of VFA accumulated to over 5000 mg l 1 after the VTS loading rate was increased to 8 g l 1 d 1. Therefore, the maximum VTS loading rate under microaeration conditions was considered to be 6 g l 1 d 1. The hydraulic retention time (HRT) at this VTS loading rate was about 9.5 d. In our previous study, when artificial kitchen waste was used as the feed, a higher VTS loading rate of 8 g l 1 d 1 was achieved using a similar treatment system [8], perhaps due to the balance of carbohydrate, protein, and lipid content in the feedstock. At a VTS loading rate of 6 g l 1 d 1, we calculate that a total of 8.48 m3 of biogas with 50% methane could be produced from the 186.0 kg of waste mixture. Therefore, 65.2 l of biogas could be obtained from 1 kg of kitchen waste by the anaerobic treatment process described above.
4.
Conclusions
A process for producing ethanol from kitchen waste was studied. Spraying LCB on kitchen waste kept the waste fresh and therefore increased the amount of sugar recovered and available for ethanol fermentation. Higher glucose recovery (85.5%) in the saccharification step was achieved at 60 1C using Nagase N-40 glucoamylase compared to other conditions tested. The resulting saccharified liquid was shown to be rich in nutrients supporting the growth of yeast. High ethanol productivity (24.0 g l 1 h 1) was obtained when the flocculating yeast strain KF-7 was used in continuous ethanol fermentation at a dilution rate of 0.8 h 1. The mixture of saccharification residue and distillation wastewater was subjected to anaerobic treatment; the treatment was performed stably at a VTS loading rate of 6 g l 1 d 1 with 72% VTS digestion efficiency under micro-aeration conditions. From 1 kg of kitchen waste containing 118.0 g total sugar, 30.9 g ethanol and 65.2 l biogas with 50% methane was produced. Therefore, kitchen waste is a promising biomass resource with realistic, practical applications. However, several problems must be solved to achieve more effective conversion of kitchen waste to ethanol. Using a saccharification process followed by solid–liquid separation, the loss of some sugar present in kitchen waste cannot be avoided. To make more sugar in kitchen waste available for
32 (2008) 1037 – 1045
1045
conversion to ethanol, a simultaneous saccharification and fermentation (SSF) process is being studied in our lab. Another problem is the accumulation of lactic acid produced during waste storage. The conversion of this lactic acid into ethanol must be addressed in future. R E F E R E N C E S
[1] Ohkouchi Y, Inoue Y. Impact of chemical components of organic wastes on L(+)-lactic acid production. Bioresource Technology 2007;98:546–53. [2] Sakai K, Ezaki Y. Open L-lactic acid fermentation of food refuse using thermophilic Bacillus coagulans and fluorescence in situ hybridization analysis of microflora. Journal of Bioscience and Bioengineering 2006;101:457–63. [3] Sakai K, Yamanami T. Thermotolerant Bacillus licheniformis TY7 produces optically active L-lactic acid from kitchen refuse under open condition. Journal of Bioscience and Bioengineering 2006;102:132–4. [4] Ozmihci S, Kargi F. Kinetics of batch ethanol fermentation of cheese-whey powder (CWP) solution as function of substrate and yeast concentrations. Bioresource Technology 2007;98:2978–84. [5] Rao K, Chaudhari V, Varanasi S, Kim DS. Enhanced ethanol fermentation of brewery wastewater using the genetically modified strain E. coli K011. Applied Microbiology and Biotechnology 2007;74:50–60. [6] Kida K, Morimura S, Sonoda Y. Repeated-batch fermentation process using a thermotolerant flocculating yeast constructed by protoplast fusion. Journal of Fermentation and Bioengineering 1992;74:169–73. [7] Tang YQ, An MZ, Liu K, Nagai S, Shigematsu T, Morimura S, et al. Ethanol production from acid hydrolysate of wood biomass using the flocculating yeast Saccharomyces cerevisiae strain KF-7. Process Biochemistry 2006;41:909–14. [8] Ikbal, Tang YQ, Shigematsu T, Morimura S, Kida K. Methanogenic activity and repression of hydrogen sulfide evolved during high rate thermophilic methane fermentation of municipal solid waste. Japanese Journal of Water Treatment Biology 2003;39:17–24. [9] Somogyi M. Notes on sugar determination. Journal of Biological Chemistry 1952;195:19–23. [10] Japanese Industrial Standards Committee. Testing methods for industrial wastewater. JIS K 0102-1986, 1986. [11] Japanese Industrial Standards Committee. Testing methods for lignin and wood of pulp, JIS P 8008-1976; Testing methods for alcohol-benzene solubility of pulpwood, JIS P 8010-1976; Testing methods for holocellulose in pulpwood, JIS P 8012-1976, 1976. [12] Kida K, Morimura S, Sonoda Y. Accumulation of propionic acid during anaerobic treatment of distillery wastewater from barely-shochu making. Journal of Fermentation and Bioengineering 1993;75:213–6. [13] Kida K, Yamadaki M, Asano SI, Nakata T, Sonada Y. The effect of aeration on stability of continuous ethanol fermentation by a flocculating yeast. Journal of Fermentation and Bioengineering 1989;68:109–11.
32 (2008) 1037 – 1045
Available at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
Ethanol production from kitchen waste using the flocculating yeast Saccharomyces cerevisiae strain KF-7 Yue-Qin Tanga, Yoji Koikeb, Kai Liua, Ming-Zhe Ana, Shigeru Morimuraa,, Xiao-Lei Wuc, Kenji Kidaa a
Graduate School of Science and Technology, Kumamoto University, 2-39-1 Kurokami, Kumamoto 860-8555, Japan Tokyo Gas Co., Ltd., 1-7-7 Suehiro-cho, Tsurumi-ku, Yokohama City, Kanagawa 230-0045, Japan c Department of Energy and Resources Engineering, College of Engineering, Peking University, Beijing 100871, China b
art i cle info
ab st rac t
Article history:
A process for producing ethanol from kitchen waste was developed in this study. The
Received 28 July 2007
process consists of freshness preservation of the waste, saccharification of the sugars in
Accepted 21 January 2008
the waste, continuous ethanol fermentation of the saccharified liquid, and anaerobic
Available online 2 April 2008
treatment of the saccharification residue and the stillage. Spraying lactic acid bacteria (LCB)
Keywords: Kitchen waste Ethanol fermentation Methane fermentation Saccharomyces cerevisiae Flocculating yeast
on the kitchen waste kept the waste fresh for over 1 week. High glucose recovery (85.5%) from LCB-sprayed waste was achieved after saccharification using Nagase N-40 glucoamylase. The resulting saccharified liquid was used directly for ethanol fermentation, without the addition of any nutrients. High ethanol productivity (24.0 g l
1
h 1) was
obtained when the flocculating yeast strain KF-7 was used in a continuous ethanol fermentation process at a dilution rate of 0.8 h
1
. The saccharification residue was mixed
with stillage and treated in a thermophilic anaerobic continuous stirred tank reactor (CSTR); a VTS loading rate of 6 g l
1
d
1
with 72% VTS digestion efficiency was achieved.
Using this process, 30.9 g ethanol, and 65.2 l biogas with 50% methane, was produced from 1 kg of kitchen waste containing 118.0 g total sugar. Thus, energy in kitchen waste can be converted to ethanol and methane, which can then be used as fuels, while simultaneously treating kitchen waste. & 2008 Elsevier Ltd. All rights reserved.
1.
Introduction
In Japan, the annual generation of organic waste from the food industry and kitchen garbage is about 20 million tons per year. Most of this waste is directly incinerated with other combustible waste, and the residual ash is disposed of in landfills. However, incineration is costly and energy consuming, and since these wastes have a high moisture content, they must be burnt with other drier wastes, resulting in the production of dioxins. Recently, more economically attractive treatment methods such as composting and anaerobic digestion have been used to treat some of this waste and Corresponding author. Tel./fax: +81 96 342 3669.
E-mail address: [email protected] (S. Morimura). 0961-9534/$ - see front matter & 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biombioe.2008.01.027
produce compost or biogas, but imbalances in supply and demand for recycled products hinder the wider application of these alternative approaches. Since these wastes are rich in sugars that could be converted to more valuable products, attention is being directed to biorefinery processing of these wastes to produce biomass materials such as lactic acid [1–3]. In addition, ethanol can also be produced from these sugars. Ethanol, a major fuel additive and a promising future energy alternative, is now produced mainly from corn in America and China and from sugarcane in Brazil. However, since corn is a major food source, its use as a fuel source has been criticized. Moreover, using corn as a feedstock is a major
ARTICLE IN PRESS 1038
BIOMASS AND BIOENERGY
contributor to the significant cost of ethanol production, so biomass wastes such as corn fiber, waste wood and food wastes are far more attractive as cheap feedstocks for ethanol production. In this study, we focus on ethanol production from kitchen waste collected from dining halls. As kitchen waste is highly
Table 1 – Representative composition of kitchen waste used for ethanol fermentation Results (ww 1%)
Test parameters Total solid (TS) Volatile total solid (VTS) Ash Moisture Total sugarsa (based on wet weight) Total sugarsa (based on dry weight) Holocellulose (based on dry weight) Lignin (based on dry weight) Protein (based on dry weight) Lipid (based on dry weight)
19.7 18.8 1.9 80.3 11.8 59.8 1.6 0.8 21.8 15.7
a
Total sugars referred to starch sugars. Protein contents were calculated with total nitrogen contents by multiplication as coefficient 6.25. Results were mean values of 3 analyses.
Kitchen waste Spraying of lactic acid bacteria
perishable, and most of the sugars in the waste are starchbased sugars (Table 1), a process for producing ethanol from kitchen waste was developed, as shown in Fig. 1. The process consists of freshness preservation of the waste with the aid of lactic acid bacteria (LAB), saccharification of sugars in the waste using glucoamylase, continuous ethanol fermentation of the saccharified liquid using flocculating yeast, and anaerobic treatment of the saccharification residue and stillage. This process converts energy in the waste first to ethanol, and then to methane, which can then be used as a fuel; in addition, the waste is simultaneously treated. Although there have been several reports regarding the production of lactic acid from food waste [1–3], there have been only very limited studies regarding the production of ethanol from food waste [4,5]. The present paper is the first report regarding ethanol fermentation from kitchen waste.
2.
Methods
2.1.
Microorganism strains
The flocculating yeast Saccharomyces cerevisiae KF-7 was used for batch and continuous ethanol fermentation tests. This yeast strain was constructed by protoplast fusion of the flocculating yeast strain IR-2 and the thermotolerant yeast strain EP-1 [6]. LAB strain 4, closely related to Lactobacillus paracasei (98% rRNA gene sequence similarity), was isolated from a vegetable beverage and used for the preservation of kitchen waste.
2.2.
Storing
Saccharification
Solid-liquid separation Liquid Ethanol fermentation
2.3. Distilling
Ethanol
Waste water Methane fermentation
Sewage plant
Cultivation of LCB
LCB cells grown on a 1% YPD (1% glucose, 1% yeast extract, 2% peptone, 2% agar) slant were transferred to 1% YPD liquid medium and cultured at 37 1C for 24 h with mixing at 100 rpm using a rotary shaker. The resultant pre-cultivation broth was used for jar fermentor cultivation using tofu compression liquid (pH 5.85; glucose 190 mg l 1; suspended solid (SS) 3.38 g l 1; volatile suspended solid (VSS) 2.79 g l 1) as the medium. Cultivation was carried out at 37 1C and 100 rpm for 24 h. The culture broth, with approximately 1010 cell ml 1, was stored in a spray bottle and used in kitchen waste storage experiments. The LCB broth could be stored for 1 month without obvious change in the number of viable cells.
Water
Solid
32 (2008) 1037 – 1045
Biogas
Treated water
Fig. 1 – Flow chart of the ethanol production process from kitchen waste.
Storage of kitchen waste
Kitchen waste used in this study was collected from the students’ dining hall of Kumamoto University, Japan. Large pericarps were removed, the remaining waste was stored in a plastic basket at room temperature, and LCB broth was evenly sprayed on the surface of the waste once a day when necessary. A water weight was placed over the waste in order to produce anaerobic conditions. Representative characteristics of the collected kitchen waste are shown in Table 1. Kitchen waste stored at room temperature for 3 days with and without LCB treatment was used for lab-scale experiments, and LCB-sprayed waste stored for 1 week was used for bench-scale experiments.
ARTICLE IN PRESS BIOMASS AND BIOENERGY
2.4.
Saccharification of the kitchen waste
2.4.1.
Lab-scale experiment
pH meter
Wet gas meter
One kilogram of waste was chopped into small pieces using a fruit mixer, then transferred into a 3 l jar fermentor containing 0.5 kg tap water. Two kinds of glucoamylase, Glucochimu #20000 (protein content 28.3 w w 1%) and Nagase N-40 (protein content 80.1 ww 1%) (Nagase ChemteX Corporation, Osaka, Japan) were used for saccharification, and the results were compared. Enzyme was added based on its protein content to a final concentration of 170 mg protein kg 1 wet waste. Saccharification was performed at 50 1C or 60 1C, 150 rpm for 6 h. Saccharified liquid was separated by centrifugation at 10,000 rpm for 10 min, then used for batch and continuous ethanol fermentation.
2.4.2.
2.5. Batch ethanol fermentation of saccharified liquid prepared from kitchen waste Preparation of inoculum
Yeast cells grown on 2% YPD (2% glucose, 1% yeast extract, 2% peptone, 2% agar) were inoculated into 100 ml of 5% YPD (5% glucose, 1% yeast extract, 2% peptone) medium in a 500ml flask. Pre-cultivation was performed aerobically at 30 1C for 16 h with mixing at 160 rpm using a rotary shaker, then the resulting pre-cultivation broth was used as inoculum for ethanol fermentation experiments.
2.5.2.
Fermented mash WV, 0.45L
Thermostated water
Bench-scale experiment
Kitchen waste (130.0 kg) was crushed using a compact chopper (MKBC-32; Masuko Sangyo Co. Ltd., Saitama, Japan), transferred to a 300 l reactor (CNVM1-E085-6, Sumitomo Heavy Industries, Ltd., Tokyo, Japan), and 65.0 kg tap water was added. Nagase N-40 was added to a final concentration of 160 mg protein kg 1 wet waste. Saccharification was performed at 60 1C, 150 rpm for 2 h. Saccharified liquid (134.0 l, 64.0 g glucose l 1) was separated from the solid residue (52.0 kg) using 6 filter press units (RF Fully Automatic Filter Press; Kurita Machinery Mfg. Co., Ltd., Osaka, Japan). The recovered saccharified liquid was used directly for continuous ethanol fermentation.
2.5.1.
1039
32 (2008) 1037 – 1045
Air P-2 P-1 Saccharified liquid Fig. 2 – Schematic diagram of the continuous ethanol fermentation process. Working volume of reactor is 0.45 l.
sterilized using a NaClO solution, as described previously [7]. The reactor was inoculated with about 0.45 l of precultivation broth. The unsterilized saccharified liquid medium was fed directly into the bottom of the reactor by peristaltic pump P-1 (Fig. 2), and the fermented mash overflowed from the top of the reactor via a solid–liquid–gas separation unit. The liquid in the reactor was circulated using roller pump P-2. The continuous fermentation process used a chemostat method to supply the feedstock, i.e., the dilution rate was increased in a stepwise fashion to increase the feed rate.
2.7. Thermophilic anaerobic treatment of the saccharification residue and stillage mixture
Batch ethanol fermentation
For batch fermentation tests, a 10-ml inoculum of precultivation broth was added to 90 ml of sterilized fermentation medium in a 300-ml flask. The flask was immersed in a thermostated water bath and the medium was stirred using a magnetic stirrer. Batch fermentation was carried out at 30 1C for 24 h.
2.6. Continuous ethanol fermentation of saccharified liquid
2.7.1.
2.7.2. Fig. 2 shows a schematic diagram of our continuous ethanol fermentation system, described previously [7]. The towershaped reactor, with a working volume of 0.45 l, had a natural solid–liquid–gas separation unit in the upper portion of the reactor. Air was supplied continuously to the bottom of the reactor through a ball filter at an aeration rate of 0.025 vvm. Prior to the fermentation experiment, the reactor was
Preparation of the mixture prior to anaerobic digestion
All saccharification residues and stillage from the bench saccharification and ethanol fermentation experiment were mixed together and used as feed for the 10-l anaerobic fermentation reactor. To accelerate the methane fermentation rate, Ni2+, Co2+, and Fe2+ were added at concentrations of 4.57, 4.57 and 61.11 mg l 1, respectively, to the mixture [8]. Table 2 shows the characteristics of the waste mixture.
Anaerobic treatment of the mixture
Fig. 3 shows a schematic diagram of the thermophilic methane fermentation apparatus. A CSTR-type (completely stirred tank reactor-type) reactor (Jar Fermentor MBF; Eyela Tokyo Rikakikai Co. Ltd., Tokyo, Japan) with a working volume of 8 l was used as the methane fermentation reactor. Eight liters of anaerobic digested sludge, provided by Kumamoto Hokubu sewage works (Kumamoto, Japan), was added to the
ARTICLE IN PRESS 1040
BIOMASS AND BIOENERGY
Table 2 – Characteristics of the mixture of saccharification residue and stillage Results (mg l 1)
Test parameters Total solid (TS) Volatile total solid (VTS) Suspended solid (SS) Volatile suspended solid (VSS) CODCr BOD5 Lactic acid Acetic acid NH+4 -N NO3 -N
65,000.0 57,000.0 40,000.0 38,000.0 105,150.0 30,571.0 8703.8 2106.0 109.5 27.8
1 Pump P-1
Wet gas meter
3
Aeration
4
Feeding 2
Drawing
32 (2008) 1037 – 1045
was analyzed by UV using an F-Kit-Glucose (Roche Diagnostics, Germany). The protein content of the kitchen waste was calculated as 6.25 nitrogen content, and the nitrogen was determined using a carbon, hydrogen, nitrogen (CHN) coder MT-3 (Yanako, Tokyo, Japan) [10]. Lignin, lipid, and holocellulose were analyzed by methods used for testing pulpwood [11]. Total solid (TS), total volatile solid (TVS), SS, VSS and BOD5 were analyzed in accordance to Standard Methods [10]. CODCr and NH+4 -N and NO3 -N were measured using the HACH method (Hach Co., Loveland, CO, USA). The moisture content was determined by the decrease in weight following drying overnight at 105 1C, and the ash was determined at 600 1C by combustion for 2 h. Ethanol was assayed by gas chromatography using an internal standard (isopropanol) as described previously [7]. Volatile fatty acids (VFAs) including lactic acid, acetic acid, propionic acid, butyric acid, isobutyric acid, valeric acid, and isovaleric acid were analyzed as described previously [12]. Soluble total organic carbon (S-TOC) was analyzed using a TOC auto analyzer (TOC-500; Shimadzu, Kyoto, Japan). The methane content of the biogas was measured by gas chromatography using a thermal conductively detector (TCD) (KOR-2G; GL Science Co., Tokyo, Japan) equipped with a packed column (Porapack Q, GL Science). Hydrogen sulfide was measured using Kitagawa precision gas detector tubes (Komyo Kitagawa, Tokyo, Japan). The number of total yeast cells and viable cells were calculated by the methylene blue method using a hematitometer.
Ball filter
3. Thermophilic methane fermentor (53°C;WV 8 L)
5-L Erlenmeyer flask with 3 L tap water
Fig. 3 – Schematic diagram of the methane fermentation process used for the treatment of the saccharification residue and stillage mixture.
reactor, and the saccharification residue and distilling wastewater mixture was treated by the draw-and-fill method. The pH, temperature and agitation speed were maintained at 7.0, 53 1C, and 200 rpm, respectively. Under non-aeration conditions, the generated biogas was channeled directly through line 1 into a gas meter (Fig. 3). In order to reduce the concentration of hydrogen sulfide in the biogas and in the reactor, at a TS volumetric loading rate of 4 g l 1 d 1 the reactor was micro-aerated by continuously supplying air at 7.5% of the amount of biogas being produced in the methane reactor. Under these conditions, the generated biogas was channeled first into a 5-l flask with tap water through line 2, and then into a gas meter through line 4. A portion of the biogas was recirculated into the methane reactor though line 3 using pump P-1.
2.8.
Analytical methods
After hydrolyzing a sample of kitchen waste with 25% HCl, the total sugar concentration in the waste was assayed by the Somogyi–Nelson method [9]. The concentration of glucose
Results and discussion
3.1. Effect of spraying LAB on glucose recovery from kitchen waste Kitchen waste stored without being sprayed with LAB produced an unpleasant odor within 24 h, and the entire surface of the waste was covered with putrefying fungi after 3 days. In contrast, waste sprayed with LAB and then stored produced no offensive odor and remained fresh for more than 1 week. These results indicated that lactic acid fermentation could inhibit the growth of putrefactive bacteria and fungi, making the preservation and deodorization of kitchen waste possible. Kitchen waste that was either sprayed or not sprayed with LAB, then stored for 3 days, was subjected to saccharification at 50 1C in order to determine the effect of LAB treatment on the recovery of glucose (Table 3). Fresh kitchen waste without storing (hereafter called fresh waste) was used as a control. In the case of fresh waste treated with Glucochimu #20000, high glucose recovery (97.5%) was achieved and the glucose concentration in the saccharified liquid was approximately 74.0 g l 1. Using the same glucoamylase, glucose recovery efficiencies were 72.0% and 60.1% for waste stored for 3 days with and without LAB spray treatment, respectively. Although lower than the recovery efficiency from fresh waste, waste sprayed with LAB showed much higher glucose recovery than unsprayed waste, suggesting that spraying with LAB could conserve more sugar that could then be used for ethanol fermentation. When Nagase N-40 was used as the glucoamylase, higher glucose
ARTICLE IN PRESS BIOMASS AND BIOENERGY
1041
32 (2008) 1037 – 1045
Table 3 – Comparison of sugar recoveries from stored kitchen wastea in the saccharification step with or without waste treatment with lactic acid bacteria (LCB) Sample
Fresh waste Waste stored for three days
Enzyme
Glucochimu #20000 Glucochimu #20000 Nagase N-40
Spraying of LCB
No spraying of LCB
Glucose in saccharified liquid (g l 1)
Glucose recovery (%)
Glucose in saccharified liquid (g l 1)
Glucose recovery (%)
–
–
74.1
97.5
56.6
72.0
47.3
60.1
60.5
77.0
51.2
65.1
a Kitchen waste was 1.5-times diluted before saccharification and the total sugars of the mixture was 78.0 g l 1. Saccharification enzyme was added based on its protein content to a concentration of 170 mg protein kg 1 wet waste. Saccharification was carried out in a 50 1C incubator shaker at 100 rpm for 6 h.
recoveries (77.0% and 65.0%) were achieved for waste sprayed or not sprayed with LAB, respectively, indicating that Nagase N-40 was better at saccharification than Glucochimu #20000 at 50 1C. As the pH of the waste after storage was below 4.0, these higher glucose recoveries might be due to the higher acid tolerance of Nagase N-40.
Table 4 – Effect of saccharification temperature on glucose recovery from kitchen wastea Saccharification temperature (1C)
50
3.2. Effect of saccharification temperature on glucose recovery from kitchen waste 60
To reduce the population of active bacteria present in the nonsterilized saccharified liquid (used directly for ethanol fermentation, the next step of the process), saccharification at 60 1C was studied using Glucochimu #20000 and Nagase N-40 (Table 4). In the case of Glucochimu #20000, glucose recovery efficiency decreased from 72.0% to 25.4% when the saccharification temperature increased from 50 to 60 1C. However, in the case of Nagase N-40, glucose recovery increased from 75.8% to 85.5% when the saccharification temperature was increased to 60 1C. Nagase N-40 was therefore shown to be more thermo-resistant than Glucochimu #20000 and was used at 60 1C in all subsequent saccharification experiments.
3.3. Effect of nitrogenous nutrients and pH on ethanol fermentation of saccharified liquid from kitchen waste Nitrogenous nutrients of yeast extract and (NH4)2SO4 were added to the saccharified liquid to investigate their effect on ethanol fermentation at pH 4.5 (Table 5). It was found that ethanol concentrations in fermented media containing extra nitrogen-based nutrients were approximately 30.0 g l 1, the same as in the absence of these additives. The ethanol yield was 0.47 (92% of the theoretical yield of 0.51), suggesting that the nutrients already present in the saccharified liquid were sufficient for the growth and maintenance of the yeast. The pH of the saccharified liquid was affected by the storage time of the waste, but was generally in the range 3.6–3.9. In order to enhance ethanol production, the pH of the medium was usually adjusted to an optimum pH prior to fermentation. Although the yeast strain KF-7 is somewhat acid tolerant, the effect of pH on ethanol fermentation was studied using saccharified liquids with different pHs (Table 5).
Enzyme
Glucose in saccharified liquid (g l 1)
Glucose recovery (%)
Glucochimu #20000 Nagase N-40
56.6
72.0
59.6
75.8
Glucochimu #20000 Nagase N-40
20.0
25.4
67.2
85.5
a
Kitchen waste was 1.5-times diluted before saccharification and the total sugars of the mixture was 78.6 g l 1. Saccharification enzyme was added based on its protein content to a concentration of 170 mg protein kg 1 wet waste. Saccharification was carried out in a 50 or 60 1C incubator shaker at 100 rpm for 6 h.
Table 5 – Effect of nitrogenous nutrients and pH on ethanol fermentation of saccharified liquid prepared from kitchen wastea Medium
Saccharified liquid+1% yeast extract Saccharified liquid+1% (NH4)2SO4 Saccharified liquid Saccharified liquid Saccharified liquid
Initial pH of the medium
Ethanol concentration (g l 1)
4.5
30.1
4.5
29.5
4.5 4.0 3.5
29.9 29.7 25.0
a Glucose concentration of saccharified liquid was 64 g l 1. pH of mediums were adjusted using 1 N NaOH solution. Ethanol fermentation was carried out at 30 1C for 24 h using 300-ml Erlenmeyer flask.
ARTICLE IN PRESS 1042
BIOMASS AND BIOENERGY
Ethanol concentration decreased significantly to 25.0 g l 1 at pH 3.5. However, the concentration of ethanol obtained at pH 4.0 was the same as that at pH 4.5, indicating that efficient fermentation could be achieved at pH of 4.0 using saccharified liquid.
3.4. Continuous ethanol fermentation of lab-scalegenerated saccharified liquid
32 (2008) 1037 – 1045
inside the reactor. The ethanol concentration in the reactor was approximately 30.0 g l 1, and the ethanol yield was about 0.47 (92% of the theoretical value). Microscopic observation of yeast cells sampled from the reactor confirmed that the flocculating character of the cells did not change throughout the entire fermentation period.
3.5. Study on ethanol productivity of continuous fermentation of bench-scale-generated saccharified liquid
Saccharified liquid (pH 3.9; glucose 64.0 g l 1; lactic acid 6200 mg l 1; acetic acid 1300 mg l 1) was prepared on a lab scale and, without adjusting the pH, was fed directly into the reactor and fermented continuously for 2 weeks at a dilution rate of 0.2 h 1 (Fig. 4). The concentrations of lactic acid and acetic acid in the reactor were almost the same as in the feed medium, suggesting no growth of acid-producing bacteria
To study the ethanol productivity, saccharified liquid (pH 3.85; glucose 64.0 g l 1; lactic acid 8000 mg l 1; acetic acid 1200 mg l 1) was prepared using bench-scale equipments consisting of a compact chopper, a saccharification reactor and a filter press. Prepared, unadjusted saccharified liquid was used as the feed for continuous ethanol fermentation for 1 month.
7000
30
6000
25
5000
20
4000
15
3000
10
2000
5
1000
0
, Acetic acid concentration (mg l-1)
35
, Lactic acid concentration (mg l-1)
Dilution rate (h-1)
Δ , Ethanol concentration (g l-1) , pH (-)
0.2
0 10
5
0
15
Time (d)
50
100
40
80
30
60
20
40
10
20
0
, Survival ratio of yeast cells (%)
, Ethanol concentration (g
×, Number of viable cell (108 ml-1)
l-1)
, Ethanol productivity (g l-1 h-1)
Fig. 4 – Continuous ethanol fermentation at 30 1C and an aeration rate of 0.025 vvm, using saccharified liquid medium prepared on a lab scale.
0 0.0
0.2
0.4
0.6
0.8
1.0
Dilution rate (h-1) Fig. 5 – Effect of dilution rate on ethanol productivity and viable cell number during continuous fermentation at 30 1C and an aeration rate of 0.025 vvm, using saccharified liquid medium prepared on bench scale equipments.
ARTICLE IN PRESS BIOMASS AND BIOENERGY
The dilution rate was initially 0.2 h 1, and was then increased to 0.4, 0.6, and finally to 0.8 h 1. Fig. 5 shows the ethanol concentration, ethanol productivity, survival ratio of yeast cells, and viable yeast cell number during continuous fermentation at different dilution rates. Ethanol fermentation was successfully conducted at all dilution rates studied. The ethanol concentration in the reactor averaged 30 g l 1 and the ethanol yield was approximately 0.47 throughout the fermentation period, almost the same as when saccharified liquid prepared on a lab scale was used (Fig. 4). High ethanol productivity of 24.0 g l 1 h 1 was achieved at a dilution rate of 0.8 h 1 though the glucose concentration in the feed was only 64.0 g l 1. We achieved similar ethanol productivity when using a molasses medium, or an acid hydrolysate from wood biomass with 150 g l 1 total sugar at a dilution rate of 0.4 h 1 [7,13]. In these two previous studies, the reactors could not be successfully operated at dilution rates of 0.5 or 0.6 h 1. However, in the present study, we successfully operated the system at a dilution rate of 0.8 h 1. This contrast with earlier studies could be due to the high survival rate of yeast cells and the high number of viable cells in the reactor (greater than 10 108 cell/ml) resulting from the excellent flocculating characteristics of strain KF-7
Kitchen waste
1043
32 (2008) 1037 – 1045
(Fig. 5). The present study is the first to achieve ethanol fermentation even at this high dilution rate. The acetic acid concentration in the reactor was the same as that in the feed medium. However, the lactic acid concentration in the reactor was approximately 9000 mg l 1, slightly higher than that in the feed medium, indicating some growth of LAB in the reactor regardless of the dilution rate used. As approximately 8000 mg l 1 of lactic acid was contained in prepared saccharified liquid, a small amount of sugar in the kitchen waste was converted to lactic acid and not the target product, ethanol. Since it is desirable from an economic standpoint that all sugars in the wastes be fermented to ethanol, a yeast strain capable of utilizing lactic acid to produce ethanol will be required in future studies. As the ethanol yield was approximately 0.47, 4.02 kg of ethanol could be produced from 134.0 kg of bench-scaleprepared saccharified liquid; since a total of 130.0 kg of kitchen waste was used, 30.9 g of ethanol could be obtained from 1 kg of kitchen waste by the ethanol fermentation process described above. However, if the calculation is based on the amount of glucose in saccharified broth before filter press separation, 45.0 g of ethanol should be produced from
130.0 kg (total sugar content: 118.0 g kg-1 kitchen waste)
65.0 L water
Saccharification efficiency: Saccharification
195.0 L (glucose content: 64.0 g l-1)
64.0 × 195.0 118.0 ×130.0
Glucose recovery efficiency in solid-liquid separation step:
Solid-liquid separation using filter press unit
64.0 × 134.0 64.0 × 195.0 52.0 kg
134.0 L
Solid residue
× 100% = 81.3 %
Saccharifiedliquid (glucose content: 64.0 g l -1 )
× 100% = 68.7%
Total glucose recovery efficiency: 64.0 × 134.0
× 100% =55.9%
118.0 ×130.0 65.0 L water By adding rinse operation Solid residue
65.0 L rinsed water
Glucose recovery efficiency in solid-liquid separation step: (glucose content: 52.0 g l -1)
81.3. ×134.0 64.0 × 195.0
× 100% = 87.2 %
Total glucose recovery efficiency:
Used as dilution water in saccharification step
81.3×134.0 118.0 ×130.0
× 100% =71.0%
134.0 L Saccharifiedliquid
(glucose content: 81.3 g l-1)
Fig. 6 – Mass balance in the saccharification and saccharified liquid separation processes.
ARTICLE IN PRESS 1044
BIOMASS AND BIOENERGY
1 kg of kitchen waste under conditions providing an ethanol yield of 0.47 (Fig. 6). The lower ethanol quantity was actually obtained because approximately 31% of the glucose in the saccharified broth was lost after the filter press separation (Fig. 6). As shown in Fig. 6, to recover the glucose remaining in the solid residue, a single rinse operation was carried out using 65.0 l of water, and the glucose recovery efficiency was determined. The glucose content in the rinse water was 52.0 g l 1. If this rinse water was used as dilution water in the saccharification step, the glucose content in the recovered saccharified liquid would increase from 64.0 to 81.3 g l 1, and the glucose recovery efficiency in the filter press separation step would increase from 68.7% to 87.2%. Therefore, the total glucose recovery efficiency in the saccharification and separation processes would increase from 55.9% to 71.0% (Fig. 6). Although we did not investigate these conditions in this study, we predict that if saccharified liquid with a glucose content of 81.3 g l 1 were used as the feedstock for ethanol fermentation, about 38.1 g of ethanol would be produced from 1 kg of kitchen waste.
3.6. Thermophilic methane fermentation of the mixture comprised of saccharification residue and stillage from the bench-scale experiment The total quantity of the mixture of saccharification residue and stillage from the bench-scale experiment was approxi-
32 (2008) 1037 – 1045
mately 186.0 l. As shown in Table 2, the concentration of VTS and VSS in the mixture was 57 and 38 g l 1, respectively. An anaerobic CSTR was required to treat this potent, high-solidcontent waste. Fig. 7 shows the time course of effluent quality, digestion efficiency, gas evolution rate, and the methane and H2S concentrations in the biogas under VTS volumetric loading rates from 1.0 to 8.0 g l 1 d 1. VTS digestion efficiencies were approximately 75% when the VTS loading rate was between 1.0 and 3.0 g l 1 d 1. The gas evolution rate was 2500 and 3400 ml l 1 d 1, which corresponds to 1000 ml (g-digested VTS) 1, at VTS loading rates of 2.0 and 3.0 g l 1 d 1, respectively. Propionic acid accumulated temporarily, but stopped when the system was operated longer at a VTS loading rate of 3.0 g l 1 d 1. However, when the VTS loading rate was increased to 4.0 g l 1 d 1, the accumulation of propionic acid reoccurred, accompanied by an increase in TOC concentration. Simultaneously, VTS digestion efficiency decreased to 73% and H2S concentration in the biogas increased from 800 to 1200 ppm. To reduce the repressive effect of H2S on methanogenesis micro-organisms [8], air at 7.5% of the amount of biogas evolved was supplied continuously into the upper region of the reactor from the 65th day of operation, resulting in a decrease in H2S content in the biogas to below 50 ppm and a decrease in TOC concentration from 2900 to 1800 mg l 1 following 20 days of micro-aeration (Fig. 7A, B). VTS digestion efficiency remained at approximately 75% under these micro-aeration conditions when the
VTS loading rate (g l-1 d-1) 4
6
8
Microaeration 7.5% of biogas produced
5000
100
4000
80
3000
60
2000
40
1000
20
0 0
20
40
60 80 Time (d)
100
120
0 140
5000
100
4000
80
3000
60
2000
40
1000
20
0
0 0
20
40
60 80 Time (d)
100
120
−, VTS digestion efficiency (%) ×, VSS digestion efficiency (%)
3
×, pH in reactor (-)
2
Δ, CH4 concentration in biogas (%)
, Biogas evolution rate (ml l-1 d-1) , H2S concentration in biogas (ppm)
, Propionate concentration (mg l-1) Δ, Acetate concentration (mg l-1) , Lactate concentration (mg l-1) , TOC concentration (mg l-1)
1
140
Fig. 7 – Thermophilic methane fermentation of the saccharification residue and stillage mixture.
ARTICLE IN PRESS BIOMASS AND BIOENERGY
VTS loading rate was 4 g l 1 d 1. When the VTS loading rate was increased to 6 g l 1 d 1, the VTS digestion efficiency decreased to 72% but the concentration of VFA was the same as that at a VTS loading rate of 4 g l 1 d 1. The gas evolution rate was approximately 4800 ml l 1 d 1 (1100 ml (g-digested VTS) 1) and the treatment was performed successfully. The methane concentration in the biogas under this condition was approximately 50%. However, the VTS digestion efficiency and the gas evolution rate decreased significantly, and the concentration of VFA accumulated to over 5000 mg l 1 after the VTS loading rate was increased to 8 g l 1 d 1. Therefore, the maximum VTS loading rate under microaeration conditions was considered to be 6 g l 1 d 1. The hydraulic retention time (HRT) at this VTS loading rate was about 9.5 d. In our previous study, when artificial kitchen waste was used as the feed, a higher VTS loading rate of 8 g l 1 d 1 was achieved using a similar treatment system [8], perhaps due to the balance of carbohydrate, protein, and lipid content in the feedstock. At a VTS loading rate of 6 g l 1 d 1, we calculate that a total of 8.48 m3 of biogas with 50% methane could be produced from the 186.0 kg of waste mixture. Therefore, 65.2 l of biogas could be obtained from 1 kg of kitchen waste by the anaerobic treatment process described above.
4.
Conclusions
A process for producing ethanol from kitchen waste was studied. Spraying LCB on kitchen waste kept the waste fresh and therefore increased the amount of sugar recovered and available for ethanol fermentation. Higher glucose recovery (85.5%) in the saccharification step was achieved at 60 1C using Nagase N-40 glucoamylase compared to other conditions tested. The resulting saccharified liquid was shown to be rich in nutrients supporting the growth of yeast. High ethanol productivity (24.0 g l 1 h 1) was obtained when the flocculating yeast strain KF-7 was used in continuous ethanol fermentation at a dilution rate of 0.8 h 1. The mixture of saccharification residue and distillation wastewater was subjected to anaerobic treatment; the treatment was performed stably at a VTS loading rate of 6 g l 1 d 1 with 72% VTS digestion efficiency under micro-aeration conditions. From 1 kg of kitchen waste containing 118.0 g total sugar, 30.9 g ethanol and 65.2 l biogas with 50% methane was produced. Therefore, kitchen waste is a promising biomass resource with realistic, practical applications. However, several problems must be solved to achieve more effective conversion of kitchen waste to ethanol. Using a saccharification process followed by solid–liquid separation, the loss of some sugar present in kitchen waste cannot be avoided. To make more sugar in kitchen waste available for
32 (2008) 1037 – 1045
1045
conversion to ethanol, a simultaneous saccharification and fermentation (SSF) process is being studied in our lab. Another problem is the accumulation of lactic acid produced during waste storage. The conversion of this lactic acid into ethanol must be addressed in future. R E F E R E N C E S
[1] Ohkouchi Y, Inoue Y. Impact of chemical components of organic wastes on L(+)-lactic acid production. Bioresource Technology 2007;98:546–53. [2] Sakai K, Ezaki Y. Open L-lactic acid fermentation of food refuse using thermophilic Bacillus coagulans and fluorescence in situ hybridization analysis of microflora. Journal of Bioscience and Bioengineering 2006;101:457–63. [3] Sakai K, Yamanami T. Thermotolerant Bacillus licheniformis TY7 produces optically active L-lactic acid from kitchen refuse under open condition. Journal of Bioscience and Bioengineering 2006;102:132–4. [4] Ozmihci S, Kargi F. Kinetics of batch ethanol fermentation of cheese-whey powder (CWP) solution as function of substrate and yeast concentrations. Bioresource Technology 2007;98:2978–84. [5] Rao K, Chaudhari V, Varanasi S, Kim DS. Enhanced ethanol fermentation of brewery wastewater using the genetically modified strain E. coli K011. Applied Microbiology and Biotechnology 2007;74:50–60. [6] Kida K, Morimura S, Sonoda Y. Repeated-batch fermentation process using a thermotolerant flocculating yeast constructed by protoplast fusion. Journal of Fermentation and Bioengineering 1992;74:169–73. [7] Tang YQ, An MZ, Liu K, Nagai S, Shigematsu T, Morimura S, et al. Ethanol production from acid hydrolysate of wood biomass using the flocculating yeast Saccharomyces cerevisiae strain KF-7. Process Biochemistry 2006;41:909–14. [8] Ikbal, Tang YQ, Shigematsu T, Morimura S, Kida K. Methanogenic activity and repression of hydrogen sulfide evolved during high rate thermophilic methane fermentation of municipal solid waste. Japanese Journal of Water Treatment Biology 2003;39:17–24. [9] Somogyi M. Notes on sugar determination. Journal of Biological Chemistry 1952;195:19–23. [10] Japanese Industrial Standards Committee. Testing methods for industrial wastewater. JIS K 0102-1986, 1986. [11] Japanese Industrial Standards Committee. Testing methods for lignin and wood of pulp, JIS P 8008-1976; Testing methods for alcohol-benzene solubility of pulpwood, JIS P 8010-1976; Testing methods for holocellulose in pulpwood, JIS P 8012-1976, 1976. [12] Kida K, Morimura S, Sonoda Y. Accumulation of propionic acid during anaerobic treatment of distillery wastewater from barely-shochu making. Journal of Fermentation and Bioengineering 1993;75:213–6. [13] Kida K, Yamadaki M, Asano SI, Nakata T, Sonada Y. The effect of aeration on stability of continuous ethanol fermentation by a flocculating yeast. Journal of Fermentation and Bioengineering 1989;68:109–11.