- Email: [email protected]
Food and processing residues in California: Resource assessment and potential for power generation
Bioresource Technology 98 (2007) 3098–3105
Food and processing residues in California: Resource assessment and potential for power generation Gary C. Matteson *, B.M. Jenkins Department of Biological and Agricultural Engineering, One Shields Avenue, University of California, Davis, CA 95616, United States Received 20 August 2005; received in revised form 26 October 2006; accepted 26 October 2006 Available online 31 January 2007
Abstract The California agricultural industry produces more than 350 commodities with a combined yearly value in excess of $28 billion. The processing of many of these crops results in the production of residue streams, and the food processing industry faces increasing regulatory pressure to reduce environmental impacts and provide for sustainable management and use. Surveys of food and other processing and waste management sectors combined with published state data yield a total resource in excess of 4 million metric tons of dry matter, with nearly half of this likely to be available for utilization. About two-thirds of the available resource is produced as high-moisture residues that could support 134 MWe of power generation by anaerobic digestion and other conversion techniques. The other third is generated as low-moisture materials, many of which are already employed as fuel in direct combustion biomass power plants. The cost of energy conversion remains high for biochemical systems, with tipping or disposal fees of the order of $30–50 Mg1 required to align power costs with current market prices. Identifying ways to reduce capital and operating costs of energy conversion, extending operating seasons to increase capacity factors through centralizing facilities, combining resource streams, and monetizing environmental benefits remain important goals for restructuring food and processing waste management in the state. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Biomass; Bioenergy; Anaerobic; Thermochemical; Food
1. Introduction Food and processing residuals in California presently are estimated to amount to more than 4 million dry metric tons (or Mg) per year. The fraction of this stream thought to be technically feasible for use in electric power generation could potentially support a total generating capacity of 248 MWe across multiple units and sites. Conversion, whether to energy or biobased products, can add value to otherwise waste materials and reduce environmental impacts of disposal. This paper examines vegetable crops, food and fiber processing facilities, and municipal solid waste facilities. For this study, food processing facilities are confined to
*
Corresponding author. Tel.: +1 530 219 6777. E-mail address: [email protected] (G.C. Matteson).
0960-8524/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2006.10.031
centralized food processing, packaging, and storage facilities. About half of the food and processing residuals in California flow from vegetable crop production activities, and food processing and handling facilities. The remaining half is found in municipal solid waste streams going to landfills or diverted to composting facilities. California, like several other states across the US, has recently enacted a renewable portfolio standard (RPS) calling for increased electricity from renewable resources, including biomass. A survey of major sources was undertaken to identify quantities and properties in order to characterize the potential contributions that food and food processing biomass might make toward the RPS. For the purposes of making estimates of the electricity generating potential, the residues were segregated by moisture class. Materials with moistures at or above 50% wet basis in their typical production or handling state were classified as high-moisture materials, and they were assumed to
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
be converted using biochemical approaches (e.g., anaerobic digestion). Materials below 50% moisture were assumed to be converted using thermochemical methods (e.g., combustion). The classification is to some degree arbitrary, but reflects the practical limits of combustion above 60% moisture (Jenkins and Ebeling, 1985) and the potential integration or expansion of existing waste treatment facilities for handling some materials. The major sources for high-moisture food residues include vegetable crop residues, food processing wastes, and municipal food wastes. The focus on materials with high-moisture content is due to present under-utilization of this biomass in conversion applications, and to the availability of existing and new biochemical and thermochemical conversion technologies. Present conversion technologies that are being applied to high-moisture biomass mainly include anaerobic digestion, ethanol fermentation, and thermal pyrolysis technologies. New technologies are being developed, and they include high-rate anaerobic digesters for solid organic materials (Zhang and Zhang, 2002). The low-moisture biomass, such as rice hulls, nut shells, and fruit pits, are generally better suited for thermal conversion technologies, such as combustion and gasification. Many of the low-moisture food residues already find their way to thermal biomass-to-energy conversion units. In the United States, the present disposal rate of municipal solid waste (MSW) in landfills, excluding construction and demolition wastes, is estimated at approximately 240 million tons per year, i.e., about 0.8 tons per capita (Kaufman et al., 2004). By comparison, California generates about 0.9 tons and New York generates about 0.7 tons. Contained in this waste stream is food waste at 12.7%, but for California, food waste makes up 15.7% of the landfill stream (SCS Engineers, 1992). Properties of food waste in MSW are in aggregate about 30% total solids as received, 21% volatile solids, 4% fixed carbon, and 5% ash (Tchobanoglous et al., 1993). Material in landfills decomposes under mostly anaerobic conditions, resulting in a gas that can be approximated as a mixture of 50% CH4 and 50% CO2 (Franklin Associates, 1998) plus small amounts of H2S, non-methane reactive organic gases (ROG) and siloxanes. Out of the entire mixture of wastes in the landfill, approximately 35% of the carbon is released into the atmosphere as gases (if not captured and recycled into chemicals and materials) while the remainder is assumed to be indefinitely stored (Mann and Spath, 2001). Most of the high-moisture vegetable crop residues are left in the fields at the time of vegetable harvesting and tilled into the soil, with a small fraction being collected and fed to animals. A literature search confirmed that no high-moisture vegetable crop residue to energy conversion systems are presently operating in the United States. An estimated 1.6 million tons of sugar beets are grown in California (California Department of Food and Agriculture, 2002), but 64% of this crop is grown specifically for sugar production, and 34% is grown for cattle feed (Mar-
3099
tin, 2005). Residual beet pulp is also used as animal feed, thus residues from this crop are not counted as potential energy feedstock. 2. Food and processing residue resource estimates An assessment of food and food processing residues in California was conducted by review of state data and by direct industry survey. Estimates were made of the quantities of materials produced and the potential for generating electricity from these resources. The categories considered include (1) food processing solid waste, (2) meat processing solid waste, (3) vegetable crop residues, and (4) food waste in municipal solid waste (MSW). The latter category was divided into the fraction currently disposed in landfills and the fraction diverted to other uses, such as composting. Vegetable crop residues were included, due to the potential to integrate collection into vegetable packing and other management operations. A full assessment of food processing wastewater resource potential has not yet been completed and is not included here. A recent survey conducted by the author in the Sacramento area of California revealed food processing wastewater to be very important. First, potential energy recovery from the nutrients in wastewater is significant. Second, food processors are frequently discharging these waste streams to wastewater treatment plants at a significant cost (Matteson et al., 2005). Power generation potentials from high-moisture materials were estimated assuming conversion by anaerobic digestion to produce biogas with subsequent use of the biogas as engine fuel. The annual electrical energy potential associated with each resource type, EAD;i (GW h y1) was computed as: EAD;i ¼ where qi fvs bi g cCH4 QCH4 ge
1 q fvs bi gcCH4 QCH4 ge 3600 i
ð1Þ
annual available resource of biomass type i, (Mg y1 dry matter) ratio of volatile solids to total solids (–) volatile solids biodegradability for biomass type i (–) biogas yield (m3 kg1 VS destroyed) volume concentration of methane in biogas (m3 m3) volumetric heating value of methane (MJ m3) engine-generator efficiency on biogas (–)
For thermochemical power generation, the annual electrical energy, ETC;i (GW h y1) was estimated as: ETC;i ¼ where Qi
1 qQg 3600 i i TC
ð2Þ
heating value of biomass type i (MJ kg1 dry matter)
3100
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
gTC overall thermochemical conversion efficiency (–) An overall conversion efficiency was also computed for the anaerobic digestion assumptions as: gAD ¼
3600EAD;i qi Q i
ð3Þ
Power generating potential, P i (MWe) was estimated from EAD;i or ETC;i as: Pi ¼
1000 Ej;i 8760 h
ð4Þ
where h = annual capacity factor (–) and the subscript j denotes either the anaerobic digestion (AD) or thermochemical conversion (TC) class. For all the materials assumed converted by anaerobic digestion, the fraction of volatile solids to total solids used was assumed to be 0.8, biodegradability was 0.67, intrinsic biogas yield was 0.75 m3 kg1 VS destroyed, methane concentration was 65%, and the engine-generator efficiency was 30%. The heating value of methane is 36.3 MJ m3 at standard conditions. Heating values of the biomass materials were estimated from data of Jenkins (1993) and von Bernath et al. (2004) and are listed in Table 1. The efficiency of thermochemical conversion was assumed to be 25%, based on new biomass direct combustion units operating in California. The annual capacity factor was in all cases assumed to be 85%. The potential also exists to utilize heat from power generation in combined heat and power (CHP) applications. The annual heat energy potential, H i (GW h y1), was computed as: Hi ¼
Ej;i ð1 gj;i Þfh gj;i
where gj;i fh
ð5Þ
conversion efficiency for either anaerobic digestion or thermochemical conversion (–) fraction of recoverable reject heat
For anaerobic digestion, the engine efficiency was used in Eq. (5) and not the overall efficiency of the biochemical biomass-to-energy conversion. Thermal power potentials were estimated in a manner similar to Eq. (4). Conversion assumptions and methodology are included in the full California biomass database (California Biomass Collaborative, 2005). 3. Resource potentials Results of the industry surveys and other data sources are listed in Table 1. The high-moisture categories include animal meat processing waste, food processing waste, prepared food waste, and vegetable crop residues. Biomass in these residue streams amount to 3.3 million Mg dry matter per year, with approximately 1.3 million dry Mg y1 available for biomass conversion. The electrical capacity that might be achieved using anaerobic technologies is esti-
mated at 134 MWe. The data in Table 1 are drawn from a recent study compiled by the California Biomass Collaborative (von Bernath et al., 2004), information provided by California Integrated Waste Management Board (Carr, 2004a,b), and direct surveys. Data on the available fraction was drawn from von Bernath et al. (2004). The availability estimates were developed from considerations of systems that will be used to harvest and handle biomass, and social and political constraints that may affect access and use. Details about each category are described below. 3.1. Vegetable crop residues Most of the high-moisture vegetable crop residues are tilled back into the soil or fed to animals. According to von Bernath et al. (2004), about 1 million dry tons per year of residues are estimated to be produced, but less than 10% is likely to be used for energy conversion. The latter yields an electric capacity of 10 MWe via anaerobic digestion (Table 1). 3.2. Processing residues The estimate of non-meat food processing residuals is based on the data collected by interviews of processing plant managers. Two industry groups provided data— non-grape food processors and grape processors. Data for the non-grape portion of the food processors came from the League of California Food Processors (Yates, 2003), and the contact information for the grape industry came from the Wine Institute (Lee, 2004). Responses were received from approximately 68% of non-grape food-processing operations in California, representing 58,166 Mg y1 dry weight of residuals. From this sample, a total of approximately 85,000 Mg y1 dry weight has been extrapolated. The data collected from the grape industry was from thirty of the wineries that run high-volume crushing operations. Survey data totaled 73,727 Mg y1 dry weight. This is about 61% of the total residue available from the crushing process. The California Department of Food and Agriculture reports annually on the state’s grape crush (California Agricultural Statistics Service, 2004). The survey revealed that approximately 4.6% on a dry-weight basis of this crush ends up as pomace. Thus, the 2003 crush report of 2.94 million tons (2.66 million Mg) yields an estimated total dry weight of 122,000 Mg y1, including grape seeds. Combining the non-grape and grape residuals gives a total in excess of 207,000 Mg y1 from the non-meat food processing residuals category (Table 1). With an availability of 65%, this would be sufficient to generate about 14 MWe under the assumptions used for anaerobic digestion. The primary contributors to this segment of the waste stream are the highly seasonal tomato pomace (60,000 dry Mg y1) and grape pomace (122,000 dry Mg y1) which
Table 1 Food and processing residue resources and energy potentials Gross resource (dry Mg/y)
Vegetable crop Food processing High-moisture Low-moisture Meat processing Grain and fiber processing Food waste in MSW (landfilled) Food waste in MSW (diverted)
1,098,477
Total
Total *
Available resource (dry Mg/y)
Assumed conversion technique*
Approximate Heating value (MJ/kg)
Conversion efficiency (%)
Potential electrical energy (GW h y1)
8
91,187
AD
18
16
72
207,703 433,377 65,304 454,170
65 75 70 80
135,007 326,577 45,713 363,336
AD TC AD TC
20 20 20 16
14 25 14 25
106 454 36 404
1,676,650
50
838,325
AD
14
20
660
295,879
50
147,940
AD
17
17
116
4,231,561
46
1,948,085
–
MW h/Mg
kWe/Mg/y
Assumed conversion technique* Vegetable crop Food processing High-moisture Low-moisture Meat processing Grain and fiber processing Food waste in MSW (landfilled) Food waste in MSW (diverted)
Available fraction (%)
Potential generating capacity (MWe)
1,847 CHP potential (GW h y1)
CHP potential (TJ/y)
CHP potential generation (MWt)
AD
10
0.8
0.11
100
362
13
AD TC AD TC
14 61 5 54
0.8 1.4 0.8 1.1
0.11 0.19 0.11 0.15
149 635 50 565
535 2,286 181 2,035
20 85 7 76
AD
89
0.8
0.11
923
3,325
124
AD
16
0.8
0.11
163
587
22
248
0.9
0.13
2,586
9,310
347
–
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
Category
AD = anaerobic digestion, TC = thermochemical conversion (e.g. combustion). 1 Mg = 1.1 short tons = 2,205 lbs.
3101
3102
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
account for about 88% for the stream flowing from this category. A major issue when dealing with processing residuals is the seasonality of production. Most of the grape crush occurs in a short period between August 15 and September 15, and most of the tomato processing occurs between August 1 and September 30 (Fig. 1). These processing residues are presently fed to livestock or composted then incorporated into the soil. The short operating seasons constitute a disincentive to financial investment and result in high costs of energy due to low-capacity factors when conversion facilities are constrained to operate only for the same short periods of time. The economic impacts of seasonal operations are discussed later. Identifying partner companies or other sources of feedstock that can extend the operating season is important for increasing development of these utilization systems. Suitable partners would include operations that are physically close to processing facilities, those that require significant amounts of power, and operate year round. An underlying rationale for this study was to geographically match the processing streams to increase facility capacity factor. This effort is still underway. Other processing residues include nut shells and hulls, fruit and olive pits, rice hulls, and cotton gin trash. Shells and pits total more than 400,000 Mg y1 (low-moisture food processing residues in Table 1), while rice hulls and cotton gin trash add about 450,000 Mg y1 (grain and fiber processing). Many of these are already used as fuel in biomass direct combustion facilities in the state. Total potential generating capacity from these lower-moisture sources is estimated at 115 MWe by thermochemical conversion. 3.3. Meat processing waste California annually slaughters approximately 1.1 million head of cattle, 2.4 million hogs and pigs, 600,000 sheep and lambs, 240 million broiler chickens, and 18.7 million
70.000
turkeys (CDFA, 2002). The estimated total annual animal live weight processed in the state is about 1.6 million Mg. Interviews with the managers of the meat processing facilities in California result in an estimated meat processing waste of at least 65,000 Mg dry weight per year. This could provide an electrical capacity of 4 MW for rumen contents and 1 MW for rendering (totaling 5 MW). Horse hair, wool, cow skin were all considered as saleable animal products, not residues, and so are not included in the inventory. 3.4. Food waste in municipal solid waste The amount of food scraps and uneaten food disposed as municipal solid wastes in landfill is estimated at between 1.6 and 2 million Mg dry matter per year (von Bernath et al., 2004; Carr, 2003, 2004a,b). About half of this food waste is considered recoverable for energy conversion and constitutes a potential by anaerobic digestion of close to 90 MWe. Food waste observed in municipal solid waste is from uneaten food and food preparation from residences, commercial food preparation establishments (such as restaurants), grocery stores, institutional sources (such as school cafeterias), and industry (factory lunchrooms). According to an analysis performed by Britton (2003), the meat scraps and bone present in municipal food waste were estimated to be 25% and 4%, respectively. Most of the food waste is disposed of in landfills. According to a report on the assessment of California’s compost- and mulch-producing infrastructure, published by the California Integrated Waste Management Board (Cotton, 2004), ‘‘Only a handful of facilities surveyed reported processing food waste or liquid wastes’’. Interviews with operators of California’s waste recycling centers conclude that only about 15% of food waste is diverted from the landfills to conversion of bioproducts. Therefore, the potential is great for diverting increasing amounts of food waste in the municipal solid waste stream into energy and other valuable products. The 15% diverted waste represents a potential of about 16 MWe additional power generation.
Residue Production (Mg/month)
Grapes
4. Cost of energy from residue conversion
60.000 Tomatoes
50.000 Stone Fruits
40.000 30.000
Garlic & Onion
20.000
Olives
10.000 0 January
Other
April
July Month
October
Fig. 1. Seasonal production levels (Mg month1 dry weight) for selected food processing residuals from surveys of operators in California.
The economics associated with energy production from processing residues are type and site specific. Some processing residues have long been used as fuel for direct combustion units, helping to defray the overall fuel cost for the facility from which the biomass can be obtained for essentially the cost of transportation, typically below $10– 15 Mg1. Each $10 Mg1 dry weight for transportation adds approximately $0.01 kW h1 to the cost of electricity generation at current conversion efficiencies. The overall cost of electricity from biomass combustion power plants without heat recovery is currently in the range of $0.05– 0.09 kW h1 (Jenkins, 2005). As regulatory and permitting requirements for agricultural operations are becoming more stringent, installations
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
0.80 0.70 COE ($/kWh, Constant)
of anaerobic digestion or other residue conversion systems are being considered primarily for environmental reasons, with energy providing a subsidiary benefit. The cost of energy depends strongly upon tipping fees that may be charged for waste handling and disposal. As an example, a 250-ton-per-day (227 Mg d1) digester plant proposed for California State University, Channel Islands, was to provide biogas to an on-site cogeneration plant. The project was determined to be marginally profitable (McOuat, 2004). Sensitivity analyses showed that tipping or disposal fees (including multiple transaction costs), certainty of feed stock supply, amount of methane production, and methane sales to be critical factors. An economic analysis for the city of Davis, California, 30-ton day1 (27 Mg d1) green waste digester-generator resulted in a cost of energy of $0.029 kW h1 (Matteson, 2003). The sensitivity analysis for this project showed tipping fees, and value of avoided utility purchases (on-site consumption of electricity and gas) to be critical factors. A 250-ton day1 (227 Mg d1) digester for the Sacramento Municipal Utility District in Sacramento, California, found that the system could achieve a cost of energy of $0.08 kW h1 under a base case which assumed tipping fees to be at $27 ton1 ($30 Mg1; Williams, 2005). Sensitivity analysis showed that tipping fees, capital costs, and operation and maintenance costs were critical factors. Without tipping fee, the cost of electricity (COE) from a 50,000 wet Mg y1 (160 Mg d1) food waste digestion system producing biogas for a 1.25 MWe engine-generator set
3103
0.60 0.50 0.40 0.30 0.20 0.10 0.00 -200
-100
0
100
200
300
Relative Change (%) Capital Cost
Debt Ratio
Debt Interest Rate
Costof Equity
Net Efficiency
Capacity Factor
Fig. 2. Sensitivity of cost of electricity (COE) to performance and financial factors for a stand-alone biogas power system. Base case assumptions are digester capital cost of $300 per input annual metric ton (wet) feed capacity, engine-generator set capital cost of $1,200 kW1 e installed, 85% capacity factor, 23.8% net generation efficiency from biogas, 1 no tipping fee, sludge sales worth $5 Mg , 85% debt ratio, 6% debt interest rate, 15% rate of return on equity, 20-year economic life, openloop biomass federal production tax credit for five years ($0.009 kW h1), five-year accelerated depreciation, and operating and maintenance costs escalated at 2.1% per year.
0.35 1. Stand alone, No CHP, Full Capital Cost ($13K/kWe) and O&M ($0.18/kWe)
0.30 2. With CHP ($6/MMBtu),Full Capital and O&M
0.25
COE ($/kWh)
3. With CHP, Capital costreduced by half
0.20 4. With CHP, Capital and O&M reduced by half
0.15
0.10
0.05
0.00 0
10
20
30
40
50
60
Disposal (Tipping) Fee ($/wet ton) Fig. 3. Sensitivity to disposal (tipping) fees for in-vessel anaerobic digestion of food and other municipal residues. Case 1 is based on a stand-alone system without heat recovery and with conservative estimates of capital costs and operating and maintenance (O&M) costs based on recent European experience and US projections (adapted from Williams, 2005). Other cases represent improvements in cost of electricity (COE) due to heat recovery (CHP) and indicated reductions in capital and O&M costs. For these assumptions, each additional $1/wet ton ($1.10/Mg) tip fee reduces COE by $0.005/kW h.
3104
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
and using similar base performance and financial factors (Williams, 2005) yields an estimated COE of $0.29 kW h1. Capital costs are based on an average $300/wet ton ($330/ wet Mg) of feedstock, with conservative operating and maintenance (O&M) costs amounting to nearly $0.18 kW h1 under these assumptions. The sensitivity of the COE to various financial and performance factors is illustrated in Fig. 2. To yield a COE equal to the current utility contract price of $0.053 kW h1 for biomass power in California would require a tipping fee of nearly $46 Mg1 (wet). A CHP operation without a tipping fee would yield a heat price equal to $0.02 kW h1 thermal ($6 MMBtu1) which is close to the current price of natural gas in California and decreases the COE by 10% to $0.26 kW h1. If the same system were to operate only seasonally with a capacity factor of 25% and without tipping fee, the COE would exceed $0.63 kW h1. Disposal or tipping fees charged on feedstock cause the cost of electricity from the facility to decline, as shown in Fig. 3. Each $1.10/wet Mg ($1.00/wet ton) increase in the tipping fee results in a reduction of $0.005/kW h in the COE. With greater industry experience, O&M costs can likely be substantially reduced, compared with the estimates above. A combination of heat utilization and increased value for digester residues makes these systems competitive against retail electricity rates in California, at tipping fees between about $20–30 Mg1, and against wholesale rates with future reductions in capital and O&M costs, and increasing costs of natural gas. Storage of food and processing residues could resolve the seasonality problem, but at a cost. Seasonality is less problematic for dry materials in which conventional storage can be employed. To date, wet storage systems, other than lagoons for liquid wastes such as dairy manure, have not been deployed to any large extent in California. An opportunity for co-digestion of food wastes has been investigated (Matteson et al., 2005; El-Mashad and Zhang, 2006). While these two papers suggest the feasibility of co-digesting food wastes and dairy manure, more work needs to be done on the economic feasibility. Co-digestion in combination with a wet storage system could address the seasonal, food and processing residual streams resulting from grape pomace and tomato pomace, along with a number of other residues. 5. Conclusions Food processing residuals continue to be a significant environmental problem for many agricultural and processing operations. Present methods of food residual management, including landfilling and land application, can cause negative impacts on land use, and air and water quality. These residuals could provide more than 4 million metric tons of resource for biomass-to-energy conversion in California. Deployment of anaerobic digestion and other conversion technologies at the site of the residue generation will likely be implemented principally in association with envi-
ronmental impact mitigation. Opportunities to integrate food residual flows with other biomass materials, or consolidate the flows sufficiently, may justify the investment of capital in more centralized biomass-to-energy or biomass-to-product conversion systems. The cost of electricity from high-moisture materials in biochemical conversion systems is high when not supported by tipping fees or when seasonality of production results in low-capacity factors. The diversion of food and processing residuals away from low-value disposal and to higher-value products will benefit both the environment and the people of California. Acknowledgements This work was supported by the California Energy Commission under a Public Interest Energy Research Grant. Ed Yates, Senior Vice President of the California League of Food Processors, provided assistance on the Food Processor Survey. References Britton, W., 2003. Wood End Laboratory, Mt. Vernon, ME, Personal communication with G.C. Matteson. California Agricultural Statistics Service, 2004. Final grape crush report 2003, California Department of Food and Agriculture, Sacramento, California. California Biomass Collaborative, 2005.. California Department of Food and Agriculture, 2002. Resources Directory. Page 67. Office of Public Affairs, 1220N Street, 4th Floor, Sacramento, CA, 95814, (916) 654-0462, . Carr, N., 2003. Solid Waste Characterization, 1999. California Statewide Waste Disposal Characterization Study, California Waste Management Board.
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105 Kaufman, S.M., Goldstein, N., Millrath, K., Themelis, N.J., 2004. The state of garbage in America. BioCycle 45 (1), 31–41. Lee, W., 2004. Summary of 2002 and 2003 estimated shipments, selected California wineries, ranked by size. Wine Institute, 2125 Market Street, Suite 1000, San Francisco, California 94105. Mann, M.K., Spath, P.L., 2001. A life cycle assessment of biomass cofiring in a coal-fired power plant. Clean Prod. Processes (3), 83. Martin, M., 2005. Environmental Department, Imperial Sugar Company, Sugar Land, Texas, Personal communication with G.C. Matteson. Matteson, G.C., 2003. Final report: City of Davis citizens’ task force on energy issues, 32 tons/day anaerobic digester for City of Davis. Public Works Department, 1717 Fifth Street, Davis, CA, 95616. Matteson, G.C. et al., 2005. Food Residuals Survey for Sacramento Municipal Utility District. Biological and Agricultural Engineering Department, University of California, Davis, CA 95616, December. McOuat, W., 2004. California State University, Channel Island campus. Proposed digester project: preliminary economic analysis and risk assessment, HMH Resources, Inc., 100 Larkspur Landing Circle, Suite 213, Larkspur, CA, 94939.
3105
SCS Engineers, 1992. New York City waste composition study (1989– 1990), New York City Department of Sanitation, NY. Tchobanoglous, G., Theisen, H., Vigil, S., 1993. Integrated Solid Waste Management. McGraw-Hill, New York (Chapter 4). von Bernath, H.I., Matteson, G.C., Williams, R.B., Yan, L., Gildart, M.C., Jenkins, B.M., Zhang, Z., Tiangco, V., Sethi, P., Simons, G., Rosenberg, M., Spero, J., Shih, T.T., Duan, L., 2004. An assessment of biomass resources in California. Final report, CEC/UCD 500-01-016, California Biomass Collaborative, University of California, Davis, CA. Williams, R., 2005. Technology assessment for biomass power generation: summary report on economic analyses. Interim report to the Sacramento Municipal Utility District and California Energy Commission, Contract 500-00-034, Sacramento, CA. Yates, E., 2003. California League of Food Processors Report 2003, 980 9th Street, Suite 23, Sacramento, CA, 95814. Zhang, R.H., Zhang, Z., 2002. Biogasification of solid wastes by anaerobic phased solids digester system. US Patent 6,342,378.
Food and processing residues in California: Resource assessment and potential for power generation Gary C. Matteson *, B.M. Jenkins Department of Biological and Agricultural Engineering, One Shields Avenue, University of California, Davis, CA 95616, United States Received 20 August 2005; received in revised form 26 October 2006; accepted 26 October 2006 Available online 31 January 2007
Abstract The California agricultural industry produces more than 350 commodities with a combined yearly value in excess of $28 billion. The processing of many of these crops results in the production of residue streams, and the food processing industry faces increasing regulatory pressure to reduce environmental impacts and provide for sustainable management and use. Surveys of food and other processing and waste management sectors combined with published state data yield a total resource in excess of 4 million metric tons of dry matter, with nearly half of this likely to be available for utilization. About two-thirds of the available resource is produced as high-moisture residues that could support 134 MWe of power generation by anaerobic digestion and other conversion techniques. The other third is generated as low-moisture materials, many of which are already employed as fuel in direct combustion biomass power plants. The cost of energy conversion remains high for biochemical systems, with tipping or disposal fees of the order of $30–50 Mg1 required to align power costs with current market prices. Identifying ways to reduce capital and operating costs of energy conversion, extending operating seasons to increase capacity factors through centralizing facilities, combining resource streams, and monetizing environmental benefits remain important goals for restructuring food and processing waste management in the state. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Biomass; Bioenergy; Anaerobic; Thermochemical; Food
1. Introduction Food and processing residuals in California presently are estimated to amount to more than 4 million dry metric tons (or Mg) per year. The fraction of this stream thought to be technically feasible for use in electric power generation could potentially support a total generating capacity of 248 MWe across multiple units and sites. Conversion, whether to energy or biobased products, can add value to otherwise waste materials and reduce environmental impacts of disposal. This paper examines vegetable crops, food and fiber processing facilities, and municipal solid waste facilities. For this study, food processing facilities are confined to
*
Corresponding author. Tel.: +1 530 219 6777. E-mail address: [email protected] (G.C. Matteson).
0960-8524/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2006.10.031
centralized food processing, packaging, and storage facilities. About half of the food and processing residuals in California flow from vegetable crop production activities, and food processing and handling facilities. The remaining half is found in municipal solid waste streams going to landfills or diverted to composting facilities. California, like several other states across the US, has recently enacted a renewable portfolio standard (RPS) calling for increased electricity from renewable resources, including biomass. A survey of major sources was undertaken to identify quantities and properties in order to characterize the potential contributions that food and food processing biomass might make toward the RPS. For the purposes of making estimates of the electricity generating potential, the residues were segregated by moisture class. Materials with moistures at or above 50% wet basis in their typical production or handling state were classified as high-moisture materials, and they were assumed to
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
be converted using biochemical approaches (e.g., anaerobic digestion). Materials below 50% moisture were assumed to be converted using thermochemical methods (e.g., combustion). The classification is to some degree arbitrary, but reflects the practical limits of combustion above 60% moisture (Jenkins and Ebeling, 1985) and the potential integration or expansion of existing waste treatment facilities for handling some materials. The major sources for high-moisture food residues include vegetable crop residues, food processing wastes, and municipal food wastes. The focus on materials with high-moisture content is due to present under-utilization of this biomass in conversion applications, and to the availability of existing and new biochemical and thermochemical conversion technologies. Present conversion technologies that are being applied to high-moisture biomass mainly include anaerobic digestion, ethanol fermentation, and thermal pyrolysis technologies. New technologies are being developed, and they include high-rate anaerobic digesters for solid organic materials (Zhang and Zhang, 2002). The low-moisture biomass, such as rice hulls, nut shells, and fruit pits, are generally better suited for thermal conversion technologies, such as combustion and gasification. Many of the low-moisture food residues already find their way to thermal biomass-to-energy conversion units. In the United States, the present disposal rate of municipal solid waste (MSW) in landfills, excluding construction and demolition wastes, is estimated at approximately 240 million tons per year, i.e., about 0.8 tons per capita (Kaufman et al., 2004). By comparison, California generates about 0.9 tons and New York generates about 0.7 tons. Contained in this waste stream is food waste at 12.7%, but for California, food waste makes up 15.7% of the landfill stream (SCS Engineers, 1992). Properties of food waste in MSW are in aggregate about 30% total solids as received, 21% volatile solids, 4% fixed carbon, and 5% ash (Tchobanoglous et al., 1993). Material in landfills decomposes under mostly anaerobic conditions, resulting in a gas that can be approximated as a mixture of 50% CH4 and 50% CO2 (Franklin Associates, 1998) plus small amounts of H2S, non-methane reactive organic gases (ROG) and siloxanes. Out of the entire mixture of wastes in the landfill, approximately 35% of the carbon is released into the atmosphere as gases (if not captured and recycled into chemicals and materials) while the remainder is assumed to be indefinitely stored (Mann and Spath, 2001). Most of the high-moisture vegetable crop residues are left in the fields at the time of vegetable harvesting and tilled into the soil, with a small fraction being collected and fed to animals. A literature search confirmed that no high-moisture vegetable crop residue to energy conversion systems are presently operating in the United States. An estimated 1.6 million tons of sugar beets are grown in California (California Department of Food and Agriculture, 2002), but 64% of this crop is grown specifically for sugar production, and 34% is grown for cattle feed (Mar-
3099
tin, 2005). Residual beet pulp is also used as animal feed, thus residues from this crop are not counted as potential energy feedstock. 2. Food and processing residue resource estimates An assessment of food and food processing residues in California was conducted by review of state data and by direct industry survey. Estimates were made of the quantities of materials produced and the potential for generating electricity from these resources. The categories considered include (1) food processing solid waste, (2) meat processing solid waste, (3) vegetable crop residues, and (4) food waste in municipal solid waste (MSW). The latter category was divided into the fraction currently disposed in landfills and the fraction diverted to other uses, such as composting. Vegetable crop residues were included, due to the potential to integrate collection into vegetable packing and other management operations. A full assessment of food processing wastewater resource potential has not yet been completed and is not included here. A recent survey conducted by the author in the Sacramento area of California revealed food processing wastewater to be very important. First, potential energy recovery from the nutrients in wastewater is significant. Second, food processors are frequently discharging these waste streams to wastewater treatment plants at a significant cost (Matteson et al., 2005). Power generation potentials from high-moisture materials were estimated assuming conversion by anaerobic digestion to produce biogas with subsequent use of the biogas as engine fuel. The annual electrical energy potential associated with each resource type, EAD;i (GW h y1) was computed as: EAD;i ¼ where qi fvs bi g cCH4 QCH4 ge
1 q fvs bi gcCH4 QCH4 ge 3600 i
ð1Þ
annual available resource of biomass type i, (Mg y1 dry matter) ratio of volatile solids to total solids (–) volatile solids biodegradability for biomass type i (–) biogas yield (m3 kg1 VS destroyed) volume concentration of methane in biogas (m3 m3) volumetric heating value of methane (MJ m3) engine-generator efficiency on biogas (–)
For thermochemical power generation, the annual electrical energy, ETC;i (GW h y1) was estimated as: ETC;i ¼ where Qi
1 qQg 3600 i i TC
ð2Þ
heating value of biomass type i (MJ kg1 dry matter)
3100
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
gTC overall thermochemical conversion efficiency (–) An overall conversion efficiency was also computed for the anaerobic digestion assumptions as: gAD ¼
3600EAD;i qi Q i
ð3Þ
Power generating potential, P i (MWe) was estimated from EAD;i or ETC;i as: Pi ¼
1000 Ej;i 8760 h
ð4Þ
where h = annual capacity factor (–) and the subscript j denotes either the anaerobic digestion (AD) or thermochemical conversion (TC) class. For all the materials assumed converted by anaerobic digestion, the fraction of volatile solids to total solids used was assumed to be 0.8, biodegradability was 0.67, intrinsic biogas yield was 0.75 m3 kg1 VS destroyed, methane concentration was 65%, and the engine-generator efficiency was 30%. The heating value of methane is 36.3 MJ m3 at standard conditions. Heating values of the biomass materials were estimated from data of Jenkins (1993) and von Bernath et al. (2004) and are listed in Table 1. The efficiency of thermochemical conversion was assumed to be 25%, based on new biomass direct combustion units operating in California. The annual capacity factor was in all cases assumed to be 85%. The potential also exists to utilize heat from power generation in combined heat and power (CHP) applications. The annual heat energy potential, H i (GW h y1), was computed as: Hi ¼
Ej;i ð1 gj;i Þfh gj;i
where gj;i fh
ð5Þ
conversion efficiency for either anaerobic digestion or thermochemical conversion (–) fraction of recoverable reject heat
For anaerobic digestion, the engine efficiency was used in Eq. (5) and not the overall efficiency of the biochemical biomass-to-energy conversion. Thermal power potentials were estimated in a manner similar to Eq. (4). Conversion assumptions and methodology are included in the full California biomass database (California Biomass Collaborative, 2005). 3. Resource potentials Results of the industry surveys and other data sources are listed in Table 1. The high-moisture categories include animal meat processing waste, food processing waste, prepared food waste, and vegetable crop residues. Biomass in these residue streams amount to 3.3 million Mg dry matter per year, with approximately 1.3 million dry Mg y1 available for biomass conversion. The electrical capacity that might be achieved using anaerobic technologies is esti-
mated at 134 MWe. The data in Table 1 are drawn from a recent study compiled by the California Biomass Collaborative (von Bernath et al., 2004), information provided by California Integrated Waste Management Board (Carr, 2004a,b), and direct surveys. Data on the available fraction was drawn from von Bernath et al. (2004). The availability estimates were developed from considerations of systems that will be used to harvest and handle biomass, and social and political constraints that may affect access and use. Details about each category are described below. 3.1. Vegetable crop residues Most of the high-moisture vegetable crop residues are tilled back into the soil or fed to animals. According to von Bernath et al. (2004), about 1 million dry tons per year of residues are estimated to be produced, but less than 10% is likely to be used for energy conversion. The latter yields an electric capacity of 10 MWe via anaerobic digestion (Table 1). 3.2. Processing residues The estimate of non-meat food processing residuals is based on the data collected by interviews of processing plant managers. Two industry groups provided data— non-grape food processors and grape processors. Data for the non-grape portion of the food processors came from the League of California Food Processors (Yates, 2003), and the contact information for the grape industry came from the Wine Institute (Lee, 2004). Responses were received from approximately 68% of non-grape food-processing operations in California, representing 58,166 Mg y1 dry weight of residuals. From this sample, a total of approximately 85,000 Mg y1 dry weight has been extrapolated. The data collected from the grape industry was from thirty of the wineries that run high-volume crushing operations. Survey data totaled 73,727 Mg y1 dry weight. This is about 61% of the total residue available from the crushing process. The California Department of Food and Agriculture reports annually on the state’s grape crush (California Agricultural Statistics Service, 2004). The survey revealed that approximately 4.6% on a dry-weight basis of this crush ends up as pomace. Thus, the 2003 crush report of 2.94 million tons (2.66 million Mg) yields an estimated total dry weight of 122,000 Mg y1, including grape seeds. Combining the non-grape and grape residuals gives a total in excess of 207,000 Mg y1 from the non-meat food processing residuals category (Table 1). With an availability of 65%, this would be sufficient to generate about 14 MWe under the assumptions used for anaerobic digestion. The primary contributors to this segment of the waste stream are the highly seasonal tomato pomace (60,000 dry Mg y1) and grape pomace (122,000 dry Mg y1) which
Table 1 Food and processing residue resources and energy potentials Gross resource (dry Mg/y)
Vegetable crop Food processing High-moisture Low-moisture Meat processing Grain and fiber processing Food waste in MSW (landfilled) Food waste in MSW (diverted)
1,098,477
Total
Total *
Available resource (dry Mg/y)
Assumed conversion technique*
Approximate Heating value (MJ/kg)
Conversion efficiency (%)
Potential electrical energy (GW h y1)
8
91,187
AD
18
16
72
207,703 433,377 65,304 454,170
65 75 70 80
135,007 326,577 45,713 363,336
AD TC AD TC
20 20 20 16
14 25 14 25
106 454 36 404
1,676,650
50
838,325
AD
14
20
660
295,879
50
147,940
AD
17
17
116
4,231,561
46
1,948,085
–
MW h/Mg
kWe/Mg/y
Assumed conversion technique* Vegetable crop Food processing High-moisture Low-moisture Meat processing Grain and fiber processing Food waste in MSW (landfilled) Food waste in MSW (diverted)
Available fraction (%)
Potential generating capacity (MWe)
1,847 CHP potential (GW h y1)
CHP potential (TJ/y)
CHP potential generation (MWt)
AD
10
0.8
0.11
100
362
13
AD TC AD TC
14 61 5 54
0.8 1.4 0.8 1.1
0.11 0.19 0.11 0.15
149 635 50 565
535 2,286 181 2,035
20 85 7 76
AD
89
0.8
0.11
923
3,325
124
AD
16
0.8
0.11
163
587
22
248
0.9
0.13
2,586
9,310
347
–
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
Category
AD = anaerobic digestion, TC = thermochemical conversion (e.g. combustion). 1 Mg = 1.1 short tons = 2,205 lbs.
3101
3102
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
account for about 88% for the stream flowing from this category. A major issue when dealing with processing residuals is the seasonality of production. Most of the grape crush occurs in a short period between August 15 and September 15, and most of the tomato processing occurs between August 1 and September 30 (Fig. 1). These processing residues are presently fed to livestock or composted then incorporated into the soil. The short operating seasons constitute a disincentive to financial investment and result in high costs of energy due to low-capacity factors when conversion facilities are constrained to operate only for the same short periods of time. The economic impacts of seasonal operations are discussed later. Identifying partner companies or other sources of feedstock that can extend the operating season is important for increasing development of these utilization systems. Suitable partners would include operations that are physically close to processing facilities, those that require significant amounts of power, and operate year round. An underlying rationale for this study was to geographically match the processing streams to increase facility capacity factor. This effort is still underway. Other processing residues include nut shells and hulls, fruit and olive pits, rice hulls, and cotton gin trash. Shells and pits total more than 400,000 Mg y1 (low-moisture food processing residues in Table 1), while rice hulls and cotton gin trash add about 450,000 Mg y1 (grain and fiber processing). Many of these are already used as fuel in biomass direct combustion facilities in the state. Total potential generating capacity from these lower-moisture sources is estimated at 115 MWe by thermochemical conversion. 3.3. Meat processing waste California annually slaughters approximately 1.1 million head of cattle, 2.4 million hogs and pigs, 600,000 sheep and lambs, 240 million broiler chickens, and 18.7 million
70.000
turkeys (CDFA, 2002). The estimated total annual animal live weight processed in the state is about 1.6 million Mg. Interviews with the managers of the meat processing facilities in California result in an estimated meat processing waste of at least 65,000 Mg dry weight per year. This could provide an electrical capacity of 4 MW for rumen contents and 1 MW for rendering (totaling 5 MW). Horse hair, wool, cow skin were all considered as saleable animal products, not residues, and so are not included in the inventory. 3.4. Food waste in municipal solid waste The amount of food scraps and uneaten food disposed as municipal solid wastes in landfill is estimated at between 1.6 and 2 million Mg dry matter per year (von Bernath et al., 2004; Carr, 2003, 2004a,b). About half of this food waste is considered recoverable for energy conversion and constitutes a potential by anaerobic digestion of close to 90 MWe. Food waste observed in municipal solid waste is from uneaten food and food preparation from residences, commercial food preparation establishments (such as restaurants), grocery stores, institutional sources (such as school cafeterias), and industry (factory lunchrooms). According to an analysis performed by Britton (2003), the meat scraps and bone present in municipal food waste were estimated to be 25% and 4%, respectively. Most of the food waste is disposed of in landfills. According to a report on the assessment of California’s compost- and mulch-producing infrastructure, published by the California Integrated Waste Management Board (Cotton, 2004), ‘‘Only a handful of facilities surveyed reported processing food waste or liquid wastes’’. Interviews with operators of California’s waste recycling centers conclude that only about 15% of food waste is diverted from the landfills to conversion of bioproducts. Therefore, the potential is great for diverting increasing amounts of food waste in the municipal solid waste stream into energy and other valuable products. The 15% diverted waste represents a potential of about 16 MWe additional power generation.
Residue Production (Mg/month)
Grapes
4. Cost of energy from residue conversion
60.000 Tomatoes
50.000 Stone Fruits
40.000 30.000
Garlic & Onion
20.000
Olives
10.000 0 January
Other
April
July Month
October
Fig. 1. Seasonal production levels (Mg month1 dry weight) for selected food processing residuals from surveys of operators in California.
The economics associated with energy production from processing residues are type and site specific. Some processing residues have long been used as fuel for direct combustion units, helping to defray the overall fuel cost for the facility from which the biomass can be obtained for essentially the cost of transportation, typically below $10– 15 Mg1. Each $10 Mg1 dry weight for transportation adds approximately $0.01 kW h1 to the cost of electricity generation at current conversion efficiencies. The overall cost of electricity from biomass combustion power plants without heat recovery is currently in the range of $0.05– 0.09 kW h1 (Jenkins, 2005). As regulatory and permitting requirements for agricultural operations are becoming more stringent, installations
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
0.80 0.70 COE ($/kWh, Constant)
of anaerobic digestion or other residue conversion systems are being considered primarily for environmental reasons, with energy providing a subsidiary benefit. The cost of energy depends strongly upon tipping fees that may be charged for waste handling and disposal. As an example, a 250-ton-per-day (227 Mg d1) digester plant proposed for California State University, Channel Islands, was to provide biogas to an on-site cogeneration plant. The project was determined to be marginally profitable (McOuat, 2004). Sensitivity analyses showed that tipping or disposal fees (including multiple transaction costs), certainty of feed stock supply, amount of methane production, and methane sales to be critical factors. An economic analysis for the city of Davis, California, 30-ton day1 (27 Mg d1) green waste digester-generator resulted in a cost of energy of $0.029 kW h1 (Matteson, 2003). The sensitivity analysis for this project showed tipping fees, and value of avoided utility purchases (on-site consumption of electricity and gas) to be critical factors. A 250-ton day1 (227 Mg d1) digester for the Sacramento Municipal Utility District in Sacramento, California, found that the system could achieve a cost of energy of $0.08 kW h1 under a base case which assumed tipping fees to be at $27 ton1 ($30 Mg1; Williams, 2005). Sensitivity analysis showed that tipping fees, capital costs, and operation and maintenance costs were critical factors. Without tipping fee, the cost of electricity (COE) from a 50,000 wet Mg y1 (160 Mg d1) food waste digestion system producing biogas for a 1.25 MWe engine-generator set
3103
0.60 0.50 0.40 0.30 0.20 0.10 0.00 -200
-100
0
100
200
300
Relative Change (%) Capital Cost
Debt Ratio
Debt Interest Rate
Costof Equity
Net Efficiency
Capacity Factor
Fig. 2. Sensitivity of cost of electricity (COE) to performance and financial factors for a stand-alone biogas power system. Base case assumptions are digester capital cost of $300 per input annual metric ton (wet) feed capacity, engine-generator set capital cost of $1,200 kW1 e installed, 85% capacity factor, 23.8% net generation efficiency from biogas, 1 no tipping fee, sludge sales worth $5 Mg , 85% debt ratio, 6% debt interest rate, 15% rate of return on equity, 20-year economic life, openloop biomass federal production tax credit for five years ($0.009 kW h1), five-year accelerated depreciation, and operating and maintenance costs escalated at 2.1% per year.
0.35 1. Stand alone, No CHP, Full Capital Cost ($13K/kWe) and O&M ($0.18/kWe)
0.30 2. With CHP ($6/MMBtu),Full Capital and O&M
0.25
COE ($/kWh)
3. With CHP, Capital costreduced by half
0.20 4. With CHP, Capital and O&M reduced by half
0.15
0.10
0.05
0.00 0
10
20
30
40
50
60
Disposal (Tipping) Fee ($/wet ton) Fig. 3. Sensitivity to disposal (tipping) fees for in-vessel anaerobic digestion of food and other municipal residues. Case 1 is based on a stand-alone system without heat recovery and with conservative estimates of capital costs and operating and maintenance (O&M) costs based on recent European experience and US projections (adapted from Williams, 2005). Other cases represent improvements in cost of electricity (COE) due to heat recovery (CHP) and indicated reductions in capital and O&M costs. For these assumptions, each additional $1/wet ton ($1.10/Mg) tip fee reduces COE by $0.005/kW h.
3104
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105
and using similar base performance and financial factors (Williams, 2005) yields an estimated COE of $0.29 kW h1. Capital costs are based on an average $300/wet ton ($330/ wet Mg) of feedstock, with conservative operating and maintenance (O&M) costs amounting to nearly $0.18 kW h1 under these assumptions. The sensitivity of the COE to various financial and performance factors is illustrated in Fig. 2. To yield a COE equal to the current utility contract price of $0.053 kW h1 for biomass power in California would require a tipping fee of nearly $46 Mg1 (wet). A CHP operation without a tipping fee would yield a heat price equal to $0.02 kW h1 thermal ($6 MMBtu1) which is close to the current price of natural gas in California and decreases the COE by 10% to $0.26 kW h1. If the same system were to operate only seasonally with a capacity factor of 25% and without tipping fee, the COE would exceed $0.63 kW h1. Disposal or tipping fees charged on feedstock cause the cost of electricity from the facility to decline, as shown in Fig. 3. Each $1.10/wet Mg ($1.00/wet ton) increase in the tipping fee results in a reduction of $0.005/kW h in the COE. With greater industry experience, O&M costs can likely be substantially reduced, compared with the estimates above. A combination of heat utilization and increased value for digester residues makes these systems competitive against retail electricity rates in California, at tipping fees between about $20–30 Mg1, and against wholesale rates with future reductions in capital and O&M costs, and increasing costs of natural gas. Storage of food and processing residues could resolve the seasonality problem, but at a cost. Seasonality is less problematic for dry materials in which conventional storage can be employed. To date, wet storage systems, other than lagoons for liquid wastes such as dairy manure, have not been deployed to any large extent in California. An opportunity for co-digestion of food wastes has been investigated (Matteson et al., 2005; El-Mashad and Zhang, 2006). While these two papers suggest the feasibility of co-digesting food wastes and dairy manure, more work needs to be done on the economic feasibility. Co-digestion in combination with a wet storage system could address the seasonal, food and processing residual streams resulting from grape pomace and tomato pomace, along with a number of other residues. 5. Conclusions Food processing residuals continue to be a significant environmental problem for many agricultural and processing operations. Present methods of food residual management, including landfilling and land application, can cause negative impacts on land use, and air and water quality. These residuals could provide more than 4 million metric tons of resource for biomass-to-energy conversion in California. Deployment of anaerobic digestion and other conversion technologies at the site of the residue generation will likely be implemented principally in association with envi-
ronmental impact mitigation. Opportunities to integrate food residual flows with other biomass materials, or consolidate the flows sufficiently, may justify the investment of capital in more centralized biomass-to-energy or biomass-to-product conversion systems. The cost of electricity from high-moisture materials in biochemical conversion systems is high when not supported by tipping fees or when seasonality of production results in low-capacity factors. The diversion of food and processing residuals away from low-value disposal and to higher-value products will benefit both the environment and the people of California. Acknowledgements This work was supported by the California Energy Commission under a Public Interest Energy Research Grant. Ed Yates, Senior Vice President of the California League of Food Processors, provided assistance on the Food Processor Survey. References Britton, W., 2003. Wood End Laboratory, Mt. Vernon, ME, Personal communication with G.C. Matteson. California Agricultural Statistics Service, 2004. Final grape crush report 2003, California Department of Food and Agriculture, Sacramento, California. California Biomass Collaborative, 2005.
G.C. Matteson, B.M. Jenkins / Bioresource Technology 98 (2007) 3098–3105 Kaufman, S.M., Goldstein, N., Millrath, K., Themelis, N.J., 2004. The state of garbage in America. BioCycle 45 (1), 31–41. Lee, W., 2004. Summary of 2002 and 2003 estimated shipments, selected California wineries, ranked by size. Wine Institute, 2125 Market Street, Suite 1000, San Francisco, California 94105. Mann, M.K., Spath, P.L., 2001. A life cycle assessment of biomass cofiring in a coal-fired power plant. Clean Prod. Processes (3), 83. Martin, M., 2005. Environmental Department, Imperial Sugar Company, Sugar Land, Texas, Personal communication with G.C. Matteson. Matteson, G.C., 2003. Final report: City of Davis citizens’ task force on energy issues, 32 tons/day anaerobic digester for City of Davis. Public Works Department, 1717 Fifth Street, Davis, CA, 95616. Matteson, G.C. et al., 2005. Food Residuals Survey for Sacramento Municipal Utility District. Biological and Agricultural Engineering Department, University of California, Davis, CA 95616, December. McOuat, W., 2004. California State University, Channel Island campus. Proposed digester project: preliminary economic analysis and risk assessment, HMH Resources, Inc., 100 Larkspur Landing Circle, Suite 213, Larkspur, CA, 94939.
3105
SCS Engineers, 1992. New York City waste composition study (1989– 1990), New York City Department of Sanitation, NY. Tchobanoglous, G., Theisen, H., Vigil, S., 1993. Integrated Solid Waste Management. McGraw-Hill, New York (Chapter 4). von Bernath, H.I., Matteson, G.C., Williams, R.B., Yan, L., Gildart, M.C., Jenkins, B.M., Zhang, Z., Tiangco, V., Sethi, P., Simons, G., Rosenberg, M., Spero, J., Shih, T.T., Duan, L., 2004. An assessment of biomass resources in California. Final report, CEC/UCD 500-01-016, California Biomass Collaborative, University of California, Davis, CA. Williams, R., 2005. Technology assessment for biomass power generation: summary report on economic analyses. Interim report to the Sacramento Municipal Utility District and California Energy Commission, Contract 500-00-034, Sacramento, CA. Yates, E., 2003. California League of Food Processors Report 2003, 980 9th Street, Suite 23, Sacramento, CA, 95814. Zhang, R.H., Zhang, Z., 2002. Biogasification of solid wastes by anaerobic phased solids digester system. US Patent 6,342,378.