Evaluation of Laboratory Biochemical Methane Potentials as a Predictor of Anaerobic Dairy Manure Digester Biogas and Methane Production

Size: px

Start display at page:

Download "Evaluation of Laboratory Biochemical Methane Potentials as a Predictor of Anaerobic Dairy Manure Digester Biogas and Methane Production"

Bruno Sherman
5 years ago
Views:

1 An ASABE Meeting Presentation Paper Number: Evaluation of Laboratory Biochemical Methane Potentials as a Predictor of Anaerobic Dairy Manure Digester Biogas and Methane Production Gayle C. Bishop, Undergraduate Research Assistant Iowa State University, 3252 NSRIC, gbishop8@iastate.edu. Robert T. Burns, Ph. D., P.E., Professor Iowa State University, 3224 NSRIC, rburns@iastate.edu. Timothy A. Shepherd, Extension Associate Cornell University, 272 Morrison Hall, tas229@cornell.edu. Lara B. Moody, P.E., Extension Specialist Iowa State University, 3165 NSRIC, lmoody@iastate.edu. Curt A. Gooch, P.E., Senior Extension Associate Cornell University, 334 Riley-Robb Hall, cag26@cornell.edu. Robert Spajic, Fulbright Scholar Iowa State University, 3155 NSRIC, rspajic@iastate.edu. Jennifer L. Pronto, Research Support Specialist Cornell University, 325 Riley-Robb Hall, jlp67@cornell.edu. Written for presentation at the 2009 ASABE Annual International Meeting Sponsored by ASABE Grand Sierra Resort and Casino Reno, Nevada June 21 June 24, 2009 Abstract. Estimates of the quantity of biogas and methane produced by a dairy manure-based anaerobic digester are an important design parameter; they are used to size collection, transport, and biogas clean-up and utilization equipment prior to digester construction. They are also used to estimate potential return on the producer s investment. Current methane The authors are solely responsible for the content of this technical presentation. The technical presentation does not necessarily reflect the official position of the American Society of Agricultural and Biological Engineers (ASABE), and its printing and distribution does not constitute an endorsement of views which may be expressed. Technical presentations are not subject to the formal peer review process by ASABE editorial committees; therefore, they are not to be presented as refereed publications. Citation of this work should state that it is from an ASABE meeting paper. EXAMPLE: Author's Last Name, Initials Title of Presentation. ASABE Paper No St. Joseph, Mich.: ASABE. For information about securing permission to reprint or reproduce a technical presentation, please contact ASABE at rutter@asabe.org or (2950 Niles Road, St. Joseph, MI USA).

2 production estimation methods include stoichiometric methane production calculations based on manure Chemical Oxygen Demand (COD) content, an estimate of digester COD removal, and data from the past performance of other dairy manure digesters. However, these methods can overestimate the actual biogas and methane production. This paper compares measured anaerobic digester biogas and methane production to estimated production based on laboratory biochemical methane potential (BMP) data developed from manure samples collected at six New York State dairy farms operating anaerobic digesters. Laboratory BMP tests of each digester s influent (manure and food wastes) were compared to on-farm monitored biogas and methane quantities calculated from biogas methane content. These comparisons were used to determine the ability of laboratory BMPs to predict on-farm production from dairy manure digesters. The results suggest that BMP assays could provide useful information to estimate methane production for dairy manure anaerobic systems. The results showed that using BMPs to estimate biogas production may not be accurate, but that predicting methane production with BMPs may be feasible. The linear regression results did not show a relationship that could be used for predicting biogas production from BMPs. However, a relationship and statistical similarities were found for predicting methane production from BMPs. Keywords. anaerobic digestion, biochemical methane potential, dairy manure, biogas 3

3 Introduction With the advancement of knowledge concerning the effects of nonrenewable energy on the atmosphere, there is a push for renewable energy. Anaerobic digestion is the breakdown of organic wastes which produces a renewable energy source of biogas as well as reducing nuisance odors, controlling greenhouse gas emissions, and converting organic nitrogen into available nitrogen for plants (Cantrell et al., 2008). Manure digesters provide livestock operations with an alternative energy resource and can have a positive benefit for the environment. The biogas collection, transport, and utilization system components (blower, delivery system, boiler, etc.) associated with manure anaerobic digester systems, are sized based on estimates of biogas and methane production. More accurate biogas and methane estimates would be beneficial to dairy digester system designers. Two biogas and methane estimation methods were presented by Wright (2007); the first was to initially install only a boiler to determine actual biogas and methane production. The other option for estimating biogas and methane production was to determine the biological degradability of the substrate to be digested and the methane content of the biogas. This second approach is investigated in the study with the use of Biochemical methane potentials (BMPs) to predict anaerobic digester biogas and methane production. Biochemical methane potentials were designed to determine the anaerobic degradability of a given waste (Owen et al., 1979). Though using BMPs to estimate actual digester biogas and methane production is not what the method was intended for; it would provide a simple and inexpensive method of predicting biogas and methane production for designing anaerobic digester biogas handling and utilization systems, if it provided an accurate estimate of these parameters. Materials & Methods Samples were collected from six New York State dairy farms that currently operate on-farm manure digesters. These digesters are in the process of being monitored by Cornell University following the Association of State Energy Research and Technology Transfer Institutions (ASERTTI) protocol. Table 1 provides a list of the dairies where manure was collected for this study. The table lists the farm, digester, and operational information for each site as well as the sample name used. A BMP assay was performed in triplicate for each of the farms listed in the table. 4

4 Table 1. List of the New York dairies that have monitored anaerobic digesters and information about each operation. Dairy Operation Sample Name Digester Type Estimated Hydraulic Retention Time (days) Number of Cows Co- Digestion Noblehurst NH Plug-Flow No New Hope View NHV Plug-Flow No Patterson Farm PATT Mix Digester Food Waste Ridgeline RL Mix Digester Food Waste Sunny Knoll SK Plug-Flow No AA Dairy AA Plug-Flow Fryer Oil BMP Basics A BMP assay bottle contains manure to be analyzed, inoculum bacteria, and basil media. Inoculum is an anaerobic conditioned bacterium that is cultivated in an active anaerobic digester, which is explained below. The addition of anaerobic conditioned bacteria allows the BMP to be completed in a shorter amount of time than if no bacteria were added. The amount of manure (substrate) added to the BMP is based on lab analyses of Chemical Oxygen Demand (COD) and Volatile Solids (VS). The amount of substrate is added at a level that is calculated to provide sufficient biogas production. Care must be taken not to add to much manure to the 250 ml assay bottle or excessive biogas production will result. The quantity of dairy manure sample added to the BMP is calculated using the COD concentration of the manure. The conversion of 1 g COD destruction is equal to 395 ml CH 4 at 35 degrees Celsius (Speece, 1996) at 100% efficiency. Basil media is a supplement of inorganic nutrient media and alkalinity; it is used in the inoculum reactor as well as the BMP assays to insure optimal conditions. The ingredients used to prepare the basil media are suggested by Speece (1996) and shown in Table 2. When the BMP assay is analyzed we are only interested in the degradability and biogas production of the substrate, and not of the inoculum. Therefore, a blank is used to account for the biogas produced solely by the amount of COD added by the inoculum. The control contains only inoculum and basil media and is used to correct the biogas and methane production data when the assays are completed. Table 2. Contents of Basil Media and lab scale digester Substrate (Speece, 1996). Basil Media Constituent NH 4 Cl MgSO 4 7H 2 O KCl Na 2 S 9H 2 O CaCl 2 2H 2 O (NH 4 )2HPO 4 FeCl 2 4H 2 O CoCl 2 6H 2 O KI (NaPO 3 ) 6 MnCl 2 4H 2 O Concentration in Reactor (mg/l) Constituent NH 4 VO 3 CuCl 2 2H 2 O ZnCl 2 AlCl 3 6H 2 O NaMoO 4 2H 2 O NaMoO 4 2H 2 O H 3 BO 3 NiCl 2 6H 2 O NaWO 4 2H 2 O Na 2 SeO 3 NaHCO 3 Concentration in Reactor (mg/l)

BMP Assay Inoculum The inoculum used for the BMP assay is acquired from an active laboratory scale anaerobic digester, shown in Figure 1 a, in the Agricultural Waste Management Lab at Iowa State

The tipping bucket records biogas production and thus indicates if the digester is in good bacterial health.

5 BMP Assay Inoculum The inoculum used for the BMP assay is acquired from an active laboratory scale anaerobic digester, shown in Figure 1 a, in the Agricultural Waste Management Lab at Iowa State University. It is a 60-L, 35 C digester regularly fed substrate through automated controls. Digester biogas production is monitored with an inverted tipping bucket gas meter as shown in Figure 1 b. The tipping bucket records biogas production and thus indicates if the digester is in good bacterial health. The substrate fed to the inoculum digester is a combination of basil media (Table 2) and a high protein dog food. (a) Figure 1. Automated laboratory scale digester (a) and tipping bucket to monitor digester biogas gas production (b). (b) BMP Assay Loading Calculation The BMP procedure used for this study was first described by Owen et al. (1979) and has been adapted by the Agricultural Waste Management Lab (AWML) at Iowa State University for use with manures (Moody et al., 2009). Before performing the assay, the influent dairy manure was analyzed for TS and VS using Standard Method 2540 B and 2540 E (AWWA, 1998). Analysis for COD was conducted using method 8000 of the Hach DR/890 Colorimeter Procedures Manual. Analysis results from initial samples were used to determine the volume of inoculum and substrate to add for each BMP assay. The inoculum and substrate volumes used for this assay were determined based on 1 g inoculum Volatile Solids: 2 g substrate Volatile Solids. For dairy manure, VS analysis can be more accurately analyzed than COD; therefore, we use VS data in our calculations. The volume of substrate added to each bottle was calculated using Equation 1. The volume of substrate was calculated based on the desired methane production, the substrate COD concentration, and the assumed percent of COD removal efficiency. The calculated substrate volume was then used to calculate the inoculum volume with equation 2. 1mg COD 1000mL substrate ( ml) = mlch 4 % COD removal Eqn mLCH mg substratecod 4 6

VS substrate ratio inoculum( ml) = substrate( ml) Eqn.

volumes of inoculum and substrate. Basil media was added to the bottles to bring the total volume up to 200-mL.

Each bottle was then sealed with a rubber septum (Figure 2 b) and placed on an orbital shaker at 150 rpm in a 35 C

Biogas Production The biogas production was measured by inserting the needle of a gas syringe through the rubber septum and

The biogas volume displaced into the needle of the syringe was recorded daily.

6 VS substrate ratio inoculum( ml) = substrate( ml) Eqn. 2 VS inoculum Preparing the BMP Assay The BMP assays were conducted in 250-mL serum bottles filled with the calculated volumes of inoculum and substrate. Basil media was added to the bottles to bring the total volume up to 200-mL. The bottles were then purged with 70% nitrogen and 30% carbon dioxide gas to insure anaerobic conditions (Figure 2 a). Each bottle was then sealed with a rubber septum (Figure 2 b) and placed on an orbital shaker at 150 rpm in a 35 C incubator for 30 days (Figure 2 c). Biogas Production The biogas production was measured by inserting the needle of a gas syringe through the rubber septum and letting the biogas displace the wetted barrel of the syringe (Figure 2 d). The biogas volume displaced into the needle of the syringe was recorded daily. The biogas was analyzed for methane using an infrared gas analyzer (NDIR-CH 4 Gasanalyzer, University Kiel, Germany) and shown in Figure 2 e. The blank was also measured daily for biogas production and methane content. The biogas production from the blank was used to correct the biogas production data from the samples. (a) (b) (c) (d) (e) Figure 2. Procedure for preparing and monitoring a BMP assay. 7

7 Calculating the Results The biogas production was recorded daily, and the methane content was recorded daily as percent methane. The biogas volume production data was first corrected by subtracting the biogas production of the blank from the biogas production of the samples. The resulting BMP biogas volume and methane content was used to calculate methane production. Both biogas and methane yields were normalized on the basis of initial volatile solids content (ml biogas/g initial VS). The normalized BMP data was compared to biogas production data received from the monitored digesters which were also normalized on an initial VS basis. The theory that a bench scale model (BMP) could be used to estimate full scale production was tested by comparing the normalized BMP biogas and methane yields to the normalized anaerobic digester biogas and methane yields. Results After the results were normalized, the data was compared. Figure 3 shows the comparison between the normalized BMP and anaerobic digester biogas production for all the dairy sites. Figure 4 shows the same comparison between the BMP and anaerobic digester normalized methane production. It should be noted that the anaerobic digester data from Patterson dairy may not have been accurate. It was reported that the daily gas production was inaccurate due to equipment malfunction. For that reason the Patterson data was excluded from further result analysis. A single factor ANOVA was used to calculate the variance for the comparison of the BMP and anaerobic digester normalized data. These variances were used to calculate the degrees of freedom using the Welch-Satterthwaite equation. A t-test was then used to determine if t-values for the BMP and anaerobic digester were significantly different. The t-test was compared to a t- critical value found from the t-distribution table using a p-value of 0.05, and the degrees of freedom calculated from the Welch-Satterthwaite method. The results of this statistical analysis are in Table 3 and also shown in Figures 3 and 4. The normalized biogas t-test showed all the dairy sites were statistically different. However, the normalized methane t-test showed that New Hope View, Sunny Knoll, and AA dairies are statistically the same and the comparison between the BMP and anaerobic normalized data would be accurate. Table 3. This table shows the statistical analysis results of the comparison between the BMP and anaerobic digestion data. Dairy Normalized Biogas t-test Statistically Different Normalized Methane t-test NHV 10.8 Yes 0.17 No RL 25.9 Yes 3.11 Yes NH Yes 2.26 Yes SK 6.83 Yes 2.15 No AA 6.83 Yes 1.7 No t-critical 2.1 t-critical 2.16 Statistically Different 8

8 Figure 3. Comparing normalized biogas yields between BMPs and anaerobic digesters for each dairy site. Figure 4. Comparing normalized methane yields between BMPs and anaerobic digesters for each dairy site. 9

9 The data was also compared using regression analysis of the data. The averages of the BMP and anaerobic digestion biogas and methane production for each site were graphed against each other and the linear relationship was evaluated. Figures 5 and 6 show the linear relationship between the normalized biogas and methane data. The linear relationships were determined by applying a linear trendline to the data points. The comparison was considered to be accurate when the points are a 1:1 linear relationship and have a high R 2 value. For an ideal relationship, a slope of 1:1 would indicate that the BMP produces 1 ml biogas/g initial VS to every 1 ml biogas/ g initial VS for the anaerobic digester. The regression analysis shown in Figures 5 and 6 shows a linear regression trendline applied to the BMP and anaerobic digester normalized biogas and methane data points. The normalized biogas production regression line had an R 2 value of 3 and an equation of y=0.3x The conclusion from the biogas regression analysis is the BMPs did not provide a good relationship between the BMP and anaerobic digester biogas production. It is noted in Figure 3 that the BMP overpredicts the biogas production of the anaerobic digester for each of the sites. Therefore, based on these results, BMPs may not be the best method to predict biogas production volume from dairy manure anaerobic digesters. The results from the methane regression analysis did show a high R 2 value of 0.83 and a trendline equation of y=0.6x This is a much higher R 2 value that would indicate a better relationship between the BMP and anaerobic digester methane production. The BMP also overpredicts the methane production in Figure 5, with the exception of the Sunny Knoll dairy site. However, the methane production does not have as large of an over prediction as biogas production does. Figure 5. The linear relationship between the average normalized biogas yield of the BMP and anaerobic digester. 10

10 Figure 6. The linear relationship between the average normalized methane yield of the BMP and anaerobic digester. Discussion The results from comparing the BMP to the anaerobic digester normalized biogas and methane production both statistically and linearly show that while the biogas production may not be able to be predicted from BMPs, the use of BMPs for methane production may be useful. The results showed that statistically, three of the five dairies with representative anaerobic digester data had no significant difference when comparing methane production between BMPs and the anaerobic digesters. This is also seen in the linear relationship of the methane normalized data. The linear regression for the normalized methane production had a much higher R 2 value than the biogas yield regression. On average, the BMP over predicted the biogas production by 51.4%, while the BMP only over predicted the methane production by 1.2%. BMPs were designed to characterize the degradability of a given substrate. However, the inoculum bacteria used in the BMPs must adapt to each initially added substrate to the BMP. Initially, the methane content of the BMP assay biogas was low. This may have resulted from inoculum bacteria not yet optimized for the manure substrate. As the BMP assays progressed, methane content increased. Therefore the biogas contained greater concentrations of carbon dioxide and hydrogen at the beginning of the study than at the end. The increase in biogas methane content with time is shown in Figure 8. Because the bacteria in the full scale digesters were already optimized, the start-up effect seen in the BMPs was not evident. On a normalized basis, the BMPs produced a larger amount of biogas, but smaller methane content. This would explain the over prediction by the BMPs of the biogas production. 11

11 Figure 8. The methane content and other gases behavior of the BMP and anaerobic digester of New Hope View dairy. Another reason for differences in the BMP assay and full scale digester values may be related to scale. While all available precautions were taken, manure is very non-homogeneous and it can be difficult to obtain a representative sample and also measure small quantities. This can potentially lead to inaccuracies if more (or less) VS is added to an assay than is accounted for during the normalization of data. Inaccuracies can skew the data and may have contributed to the larger BMP standard deviations. Conclusion Anaerobic digestion is a process that is used to help control odor and recover energy from dairy manure. However, a better method is needed to size the biogas utilization system during the digester design process. This study evaluated the use of laboratory controlled biochemical methane potentials as an estimation tool for biogas and methane production. Results showed that it may be possible to use BMP assays as an estimation tool for sizing biogas utilization equipment for anaerobic digesters. The results showed that while predicting biogas production from BMPs is not accurate, but predicting methane production is. Predicting methane production from BMPs would be useful for designing biogas utilization equipment for dairy manure anaerobic digesters. While the biogas volume is needed to size some of equipment needed for anaerobic digestive systems, methane data can be used to estimate energy recovery values and size boilers on Btu capacity. Further comparisons are needed however with a larger data set. 12

12 References Cantrell, K., D. Thomas, S. Kyoung, P. Hunt Livestock waste to bioenergy generation opportunities. Bioresource Technology. 99: APHA AWWA - WPCF Standard methods for the examination of water and wastewater. Washington D.C.: American Public Health Association. Moody, L., R. Burns, W. Wu-Haan, R. Spajić Use of biochemical methane potential (BMP) assays for predicting and enhancing anaerobic digester performance. In proceedings of The 4 th International and 44 th Croatian Symposium of Agriculture. Optija, Croatia. Owen, W., D. Stuckey, J. Healy Jr., Young, P. McCarty Bioassay for monitoring biochemical methane potential and anaerobic toxicity. Water Research. 13: Speece, R Anaerobic biotechnology for industrial wastewaters. Archae Press. Nashville, Tennessee, U.S.A. Wright, P. (2007). Estimating biogas production from animal manures and other materials [PowerPoint slides]. Retrieved from Anaerobic Treatment of Agricultural Wastes conference notebook. 13

Differing Effects of Glycerin on Anaerobic Codigestion of Mixed Substrates in Bench-Scale Assays and Sub Pilot-Scale Reactors

Agricultural and Biosystems Engineering Conference Proceedings and Presentations Agricultural and Biosystems Engineering 8-2011 Differing Effects of Glycerin on Anaerobic Codigestion of Mixed Substrates