Cost-Effectiveness of Oxidation Catalyst Control of Hazardous Air Pollutant (HAP) Emissions From Stationary Combustion Turbines Prepared By the Combustion Turbine Work Group Of the Industrial Combustion Coordinated Rulemaking September 4, 1998 ------- MEMORANDUM: DATE: 4 September 1998 SUBJECT: Cost-Effectiveness of Oxidation Catalyst Control of Hazardous Air Pollutant (HAP) Emissions From Stationary Combustion Turbines FROM: Combustion Turbine Work Group TO: ICCR Coordinating Committee The Combustion Turbine Work Group (CTWG) formed a task group to develop a white paper on the cost-effectiveness of oxidation catalysts in controlling HAP emissions from combustion turbines. The attached document is the white paper developed by this task group. The CTWG concurs that this information may be valuable to EPA in developing regulations for combustion turbines and requests that the ICCR Coordinating Committee pass it to EPA as a Closure Item. Attachment: Cost-Effectiveness of Oxidation Catalyst Control of Hazardous Air Pollutant (HAP) Emissions From Stationary Combustion Turbines ------- Table of Contents I. Introduction 1 II. Summary of Base Case Assumptions 3 III. Baseline HAP Emissions from Combustion Turbines 4 A. Source of Baseline HAP Emissions Data 5 B. Criteria to Include Emission Test Data in Baseline Emissions .... 5 C. Emission Factors for Baseline HAP Emissions 6 D. Complicating Factors 9 IV. Oxidation Catalyst Costs 10 A. Cost Inputs 11 B. Costs Estimated by OAQPS Control Cost Manual 16 C. Summary of Base Case Cost Estimates 17 D. Complicating Factors 17 V. Performance of Oxidation Catalysts in Reducing HAP Emissions 23 A. HAP Emissions Test Data for Oxidation Catalysts 23 B. Engineering Estimates of Catalyst Performance on HAPs 26 C. Summary of Base Case Performance Estimates 27 D. Complicating Factors 31 VI. Cost-Effectiveness Results 32 VII. Conclusions and Recommendations 34 Appendix A List of Model Turbines Appendix B Description of ICCR Emissions Database for Turbines Appendix C QA\QC Review Criteria for Emission Tests Appendix D List of Emission Tests that do not meet Criteria Appendix E Cost Spreadsheets Appendix F Description of the SCONOx™ System Appendix G Cost-Effectiveness Estimates for Individual HAPs i ------- I. Introduction This paper presents the assessment of the Combustion Turbine Work Group (CTWG) with regard to the potential cost-effectiveness of oxidation catalysts used to control hazardous air pollutant (HAP) emissions from combustion turbines. This assessment is made in the context of the Coordinating Committee providing recommendations that contribute to EPA's evaluation of "above-the-floor" MACT options for existing combustion turbines. In accordance with Section 112(d) of the Clean Air Act, EPA must consider costs in evaluating above-the-floor options for MACT, along with any non-air quality health and environmental impacts and energy requirements. In previous materials, the Coordinating Committee recommended to EPA, based on available information, that it is not possible to identify a best performing subset of existing combustion turbines, and as a result, there is no MACT floor for the existing population of combustion turbines in the United States. Therefore, to determine MACT, EPA may evaluate emission reduction technologies above the floor for existing combustion turbines. The CTWG has reviewed emission reduction technologies for existing turbines to identify controls that may be considered in the above-the-floor MACT analysis. Based on the CTWG's review, oxidation catalysts for the reduction of carbon monoxide (CO) may reduce emissions of organic HAPs from combustion turbines. The CO oxidation catalyst is an add-on control device that is placed in the turbine exhaust duct and serves to oxidize CO and hydrocarbons to H20 and CO2. The catalyst material is usually a precious metal (platinum, palladium, or rhodium). The oxidation process takes place spontaneously, without the requirement for introducing reactants (such as ammonia) into the fuel gas stream (EPA, 1993a). Oxidation catalysts are used on turbines to achieve control of CO emissions, especially turbines that use steam injection, which can increase the concentrations of CO and unburned hydrocarbons in the exhaust (EPA 1993a, Chen et al., 1993). Therefore, EPA may evaluate oxidation catalysts as an "above-the floor" MACT option for existing combustion turbines. This 1 ------- paper addresses the costs and the HAP air emissions reductions that may be achieved with oxidation catalysts. The CTWG recognizes that EPA may consider other factors, such as non-air quality environmental impacts, energy requirements, and secondary pollutants, in assessing above-the-floor MACT The approach taken in this paper is to present a base case quantitative estimate of the cost-effectiveness of oxidation catalysts for model combustion turbine units, which range in size from 1.13 megawatts (MW) to 170 MW. To determine cost-effectiveness for the base case analysis, the CTWG developed quantitative estimates for the three inputs required to estimate cost-effectiveness: 1. the baseline HAP emissions of combustion turbines before emissions control, 2. the costs of acquiring and operating oxidation catalysts, and, 3. the performance of oxidation catalysts in reducing HAP emissions. For each of these inputs this paper presents the key factors that the CTWG considers important. In assessing these three areas the CTWG presents a base case quantitative estimate of the cost-effectiveness of oxidation catalysts for each model turbine. The quantitative cost-effectiveness for each model was calculated by dividing the total annual cost by the mass of annual HAP emission reductions. Cost-effectiveness is expressed as dollars per megagram of HAP emission reduction. A megagram (Mg) is one metric ton, or approximately 1.1 U.S. tons. The paper also presents a qualitative discussion of the CTWG's views on complicating factors that could cause the estimated cost-effectiveness base case to be different in real-world situations. Section II provides a summary of the base case assumptions. Sections III, IV, and V present the quantitative estimates and complicating factors for each of the three inputs for cost-effectiveness: baseline HAP emissions, control costs, and emission reduction. The range of cost-effectiveness values and the base case cost-effectiveness for each model turbine are presented in Section VI. The CTWG's conclusions and recommendations are presented in Section VII. 2 ------- II. Summary of Base Case Assumptions For the base case cost-effectiveness analysis, the CTWG selected seven model turbines that range in size from 1.13 megawatts (MW) to 170 MW: • Model 1 -GEPG7121EA, 85.4 MW • Model 2 -- GE PG 723 IF A, 170 MW • Model 7- GEPG6561B, 39.6 MW • Model 9 - GE LM2500, 27 MW • Model 13 — Solar Centaur 40, 3.5 MW • Model 15 - Solar Mars T12000, 9 MW • Model 17 - Solar Saturn T1500, 1.13 MW These seven model turbines were selected from the 32 model turbines developed by the CTWG to provide the basis to estimate the national impacts associated with any future combustion turbine MACT standard. A complete list of the 32 model turbines is provided as Appendix A. As originally developed, the list of model turbines incorporates the fuels used, the typical hours of operation for a unit, the industry sector that may use a turbine, the presence of a duct burner, and information about space limitations. For the base case analysis, the CTWG simplified the model turbines selected. The base case assumes that each turbine is operated for 8,000 hours annually and operates at 80% rated load or greater. The CTWG also limited the base case analysis to natural gas-fired model turbines. Natural gas is the predominant fuel used by combustion turbines in the ICCR database. 54.3% of the turbines in ICCR Inventory Database Version 3 were reported as firing natural gas exclusively. In addition, 14.5% were reported as being dual fuel units, and it is expected that these units primarily use natural gas. The CTWG has assembled quantitative information available on baseline emissions, catalyst costs and catalyst performance for natural gas-fired turbines. In addition, the CTWG decided to focus the quantitative analysis on natural gas-fired turbines because fuels other than natural gas introduce complicating factors. For example, a catalyst vendor indicated that for turbines 3 ------- that operate continuously on fuel oil, it is preferable to use a special catalyst formulation that is unaffected by sulfur exposure (Chen et al., 1993). The CTWG has no data on the specially formulated catalysts. In addition, the CTWG limited the base case quantitative analysis to uncomplicated retrofit installations. Although the CTWG identified a number of situations that would complicate a retrofit installation of an oxidation catalyst, especially complications due to space limitations, time did not permit the CTWG to develop quantitative estimates for these complications. Therefore, the base case includes only a qualitative description of retrofit complications, and no costs for retrofit complications are included in the cost- effectiveness values. Based on the experience of the CTWG members, most retrofit installations for existing turbines would involve some complicating factors and, therefore, the costs to retrofit the units with oxidation catalysts would be higher in general, and in some cases much higher, than the costs presented in this base case analysis. III. Baseline HAP Emissions from Combustion Turbines The CTWG used emissions data included in the ICCR Emissions Database to identify HAPs emitted by natural gas-fired combustion turbines and to estimate baseline emission rates. Only emissions tests that met the criteria established by the CTWG for this analysis were considered. Mass emissions for each HAP were calculated using emission factors (lb/MMBtu) from those emission tests that met the CTWG's criteria. Since the rate of emissions reported for natural gas-fired combustion turbines varies, the CTWG used two emission factors to estimate baseline emissions — the highest emission factor and the average emission factor. Further discussion of the baseline emissions data used in this analysis and complicating factors is provided below. 4 ------- A. Source of Baseline HAP Emissions Data The information available to the CTWG about the emissions of HAPs from combustion turbines is included in the ICCR Emissions Database. The CTWG believes that the emissions database adequately represents the turbine population, and that these source test data are a sufficient basis for emission factors for a cost-effectiveness analysis. The current version of the emissions database includes over 70 source tests collected by EPA, many of which involve replicate sampling and analysis runs. For each test report EPA has calculated consistent emission factors for measured HAPs based on the emissions concentration reported. A description of the development of the emissions database, including assumptions used in the calculations, is provided as Appendix B. Also, EPA and the CTWG have performed a quality assurance review of each test report and determined which reports should be considered adequate for general assessment of HAP emissions from combustion turbines. These review criteria are included in Appendix C. When possible, pertinent information identified as missing from test reports was obtained by contacting the tested facilities. Only those source test data considered appropriate for use in evaluating HAP emissions were used to calculate emission factors. B. Criteria to Include Emission Test Data in Baseline Emissions The CTWG identified a subset of combustion turbine emission tests from the ICCR Emissions Database to develop the baseline emission factors for this cost-effectiveness analysis, based on the following criteria: 1. Because the baseline emissions estimate is to be done only for natural gas, emission factors were included only from tests of combustion turbines firing natural gas. [42 of the 70 test reports in the database are for natural gas.] 2. Only test reports that were judged to be complete and to have met quality assurance criteria were included. [Of the 42 tests for natural gas, 8 reports were not complete or did not meet QA\QC criteria.] 3. Because combustion turbines typically operate near full load, emission factors were extracted only for combustion turbine tests that were 5 ------- conducted at above 80% of rated load. [Of the 42 tests for natural gas, 11 reports were conducted at less than 80% rated load.] A list of the tests excluded based on the above criteria is provided in Appendix D. C. Emission Factors for Baseline HAP Emissions For those test reports in the ICCR Emissions Database that met the criteria discussed above, emission factors were included in this cost-effectiveness analysis for those HAPs measured at concentrations above the test method's detection limit in at least one run. Therefore, none of the emission factors are based solely on non-detects. This criterion is consistent with the ICCR Testing and Monitoring Work Group's recommendations that regulatory decisions should not be based solely on non-detects (ICCR Testing and Monitoring Work Group, 1997). For natural gas-fired turbines, nine HAPs were measured above the detection limits in at least one run. Both the highest emission factor and the average emission factor were used for the base case analysis. The emission factors are presented in Table 1. Baseline annual emissions for each model turbine were calculated using these emission factors. The heat input was calculated by converting the model turbine rating (MW) to MMBtu/hr and dividing by the turbine efficiency, assumed to be 35%. The baseline annual emissions were then calculated using the heat input (MMBtu/hr), the emission factor (lb/MMBtu), and the annual operating hours (hr/yr). The baseline emissions (megagrams/year) for each model turbine are presented in Table 2. [Note: The emission estimates used in this analysis are presented as emissions at the stack outlet. The emissions estimates do not address ambient air dispersion of the pollutants, nor ground- level concentrations.] 6 ------- Table 1. HAPs Emission Factors for the Base Case Analysis Pollutant Highest Emission Factor Average Emission Factor Test (Ib/MMBtu) (Ib/MMBtu) No. of Tests Formaldehyde Test 316.1.1 5.61 E-03 7.13E-04 22 Tests Toluene Test 28 7.60E-04 1.42E-04 7 Tests Acetaldehyde Test 11 3.50E-04 9.12E-05 7 Tests Xylenes Test 18 1.20E-04 4.59E-05 5 Tests Ethylbenzene Test 18 4.10E-05 4.10E-05 1 Test Benzene Test 315.1 3.91 E-05 1.03E-05 11 Tests PAHs Test 7 7.32E-06 2.23E-06 4 Tests Acrolein Test 18 6.08E-06 5.49E-06 2 Tests Naphthalene Test 7 3.31 E-06 1.46E-06 3 Tests Source: ICCR Emissions Database for Combustion Turbines 7 ------- Table 2. Baseline Emissions (Mg/yr) for Each Model Turbine Baseline Emissions (Mg/yr)-- Highest Emission Faetor Model Turbine Formsildchvde Toluene Aeetsildehvde Xvlenes Fthvlbcn/cne Ben/ene PAHs Acrolein iNsiphthsilcnc Total IIAl's 2 170 MW 33.810 4.580 2.109 0.723 0.247 0.236 0.044 0.037 0.020 41.806 1 85.4 MW 16.984 2.301 1.060 0.363 0.124 0.118 0.022 0.018 0.010 21.001 7 39.6 MW 7.876 1.067 0.491 0.168 0.058 0.055 0.010 0.009 0.005 9.738 9 27 MW 5.370 0.727 0.335 0.115 0.039 0.037 0.007 0.006 0.003 6.640 15 9 MW 1.790 0.242 0.112 0.038 0.013 0.012 0.002 0.002 0.001 2.213 13 3.5 MW 0.696 0.094 0.043 0.015 0.005 0.005 0.001 0.001 < 0.001 0.861 17 1.13 MW 0.225 0.030 0.014 0.005 0.002 0.002 < 0.001 < 0.001 < 0.001 0.278 Baseline Emissions (Mg/yr) — Average Emission Factor Model Turbine Formsildchvde Toluene Aeetsildehvde Xvlenes Ftlivl benzene Benzene PAHs Acrolein Niiphthiilene Totiil IIAl's 2 170 MW 4.297 0.856 0.550 0.277 0.247 0.062 0.013 0.033 0.009 6.344 1 85.4 MW 2.159 0.430 0.276 0.139 0.124 0.031 0.007 0.017 0.004 3.187 7 39.6 MW 1.001 0.199 0.128 0.064 0.058 0.014 0.003 0.008 0.002 1.478 9 27 MW 0.682 0.136 0.087 0.044 0.039 0.010 0.002 0.005 0.001 1.008 15 9 MW 0.227 0.045 0.029 0.015 0.013 0.003 0.001 0.002 < 0.001 0.336 13 3.5 MW 0.088 0.018 0.011 0.006 0.005 0.001 < 0.001 0.001 < 0.001 0.131 17 1.13 MW 0.029 0.006 0.004 0.002 0.002 < 0.001 < 0.001 < 0.001 < 0.001 0.042 8 ------- D. Complicating Factors The emission factors used for the base case cost-effectiveness analysis, as presented in Table 1, represent a necessary simplification of actual HAP emissions which could be expected in the existing population of combustion turbines in the United States. The following complicating factors would change the baseline emissions of certain combustion turbines in some cases: 1. The use of the highest HAP emission factors reported tends to overestimate HAP baseline emissions. 2. For the "highest" case, the highest HAP emissions factors for each pollutant were used. It has not been shown that all these "highs" would occur simultaneously from a combustion turbine. In fact, it is not likely that all the "highs" for all pollutants would occur simultaneously. Therefore, total HAP emissions are overstated in the case where the highest emission factor from all the tests is used for each HAP. 3. HAP emissions may be different for combustion turbines using fuels other than natural gas. 4. HAP emission factors used in this base case analysis tend to overestimate HAP emissions for uncontrolled turbines, since a significant portion of the emissions tests in the ICCR Emissions Database for natural gas-fired turbines were conducted on units that use steam or water injection to reduce NOx emissions, and steam or water injection may result in increased HAP emissions due to the cooling of the combustion process. 5. For some pollutants there are very few emissions test reports available. In those cases where emission averages rely on very few tests, it is unclear whether the resulting emission factor is representative of the turbine population. 6. The baseline emissions included in this analysis may underestimate annual HAP emissions from turbines that operate at less than 80% load, since the emission factors included in this base case analysis do not include the higher emission rates that may occur when turbines are operated at low loads. 9 ------- IV. Oxidation Catalyst Costs The CTWG obtained information on the costs of acquiring, installing, and operating oxidation catalysts for HAPs reduction on combustion turbines from the following sources: • Quotes provided to EPA by catalyst vendors • Costs gathered by the Gas Research Institute (GRI) • Estimates provided by Work Group members The methodology to estimate the total annual costs for oxidation catalysts was obtained from the EPA "OAQPS Control Cost Manual" (EPA, 1990). The OAQPS methodology provides generic cost categories and default assumptions to estimate the installed costs of control devices. The CTWG relied on the OAQPS methodology to develop the cost- effectiveness analysis because the Work Group understands that this is the methodology that EPA has used in the past to assess cost-effectiveness. The GRI study (Ferry et al., 1998) also relied on the OAQPS methodology. The OAQPS cost manual requires direct cost inputs for certain key elements, such as control device capital costs, and then relies on default assumptions (percentages of the direct cost inputs) to estimate other costs, such as installation. The following sections describe the direct cost inputs into the OAQPS methodology and the costs estimated using the OAQPS default assumptions. A printout of the spreadsheet used to estimate costs is presented as Appendix E. The OAQPS manual uses five cost categories to describe the annual incremental cost incurred by installing a control device, such as an oxidation catalyst: • Purchased Equipment Costs (PEC) include the capital cost of the catalyst and auxiliary equipment, and the cost of instrumentation, sales tax, and freight. • Direct Costs for Installation (DCI) are the construction-related costs associated with installing the catalyst. 10 ------- • Indirect Costs for Installation (ICI) include expenses related to engineering and start up. • Direct Annual Costs (DAC) include catalyst replacement and disposal costs and the annual increases in utilities and operating and maintenance costs. • Indirect Annual Costs (IAC) are the annualized cost of the catalyst system and costs due to tax, overhead, insurance and administrative burdens. The cost used in the cost-effectiveness calculation is the total annual cost, which is the sum of the DAC and IAC. A. Cost Inputs The CTWG developed cost estimates for the following inputs: • Capital cost of the oxidation catalysts • Capital cost of the catalyst housing • Contingency for capital costs • Catalyst life and equipment life • Catalyst disposal costs • Interest rate for capital recovery • Direct annual operating & maintenance costs • Fuel penalty costs • Annual compliance test costs A description of the each cost input is provided below. Capital cost of the oxidation catalysts The CTWG used cost estimates from Engelhard, a catalyst vendor, for six turbine exhaust flows ranging from 28.4 lb/sec to 984.0 lb/sec to estimate the capital cost of the oxidation catalysts. The Engelhard costs were based on an oxidation catalyst that would achieve 90% CO conversion efficiency and 1" pressure drop across the catalyst panels (not total system pressure drop) and include the cost of 11 ------- an internal support frame and catalyst modules. Regression analysis on these cost data provided by the vendor suggested that there is a nearly linear relationship between catalyst cost and exhaust flow rate (r2 = 0.993, when Catalyst cost=1541.8*(lb/sec)+102370). In estimating catalyst costs for the seven model turbines, the CTWG relied on the equation based on the Engelhard cost quotes, where cost is a function of turbine exhaust flow. Additional cost information reviewed by the CTWG is discussed in complicating factors. Capital cost of the catalyst housing The capital cost of the catalyst housing was estimated as 30% of the total cost of the catalyst system (the catalyst plus housing). This estimate is based on estimates provided orally by catalyst vendors. The CTWG contacted catalyst installers to get additional information on the costs for catalyst housings, but the data was not made available in time to include it in the base case analysis. Contingency A contingency of 10% of the sum of the purchased equipment costs, direct costs of installation, and indirect costs of installation was incorporated in the base case analysis. The budgeted contingency would cover costs associated with equipment redesign and modifications, cost escalations, and delays in start-up. The OAQPS Control Cost Manual recommends a 3% contingency. However, the CTWG agreed that a contingency of at least 10 percent would be appropriate for the base case analysis since the analysis is based on a preliminary vendor quote, not a guaranteed quote. Based on CTWG experience, a contingency factor of 25 percent DCI and ICI (direct and indirect installation costs) is budgeted in the early planning stages of a project and a contingency factor of at least 10 percent is budgeted once the project is under contract. Catalyst life and equipment life For the base case, the lifetime of purchased equipment was assumed to be fifteen years, except for the catalyst. Two scenarios were used for the catalyst life: the 12 ------- vendor guaranteed life (three years) and the "typical" life (six years) reported by catalyst vendors and users. The guaranteed life of the catalyst was used by EPA in the cost-effectiveness analysis for a passive catalytic device (non-selective catalytic reduction, NSCR) in the Alternative Control Techniques (ACT) document for reciprocating internal combustion engines (EPA, 1993b). In the Turbine ACT document, EPA used 5 years as the catalyst life for Selective Catalytic Reduction (SCR) (EPA, 1993a). The Turbine ACT did not specify whether the catalyst life was guaranteed life or "typical" life for SCR. However, in general, EPA prefers to rely on the useful life of equipment for cost- effectiveness calculations. The CTWG determined that the base case should evaluate the costs using both the guaranteed life and the typical life to account for the uncertainty regarding the long-term performance of oxidation catalysts. Further discussion of the issues related to catalyst life are discussed as complicating factors. The cost of catalyst replacement is annualized by applying a capital recovery factor based on the catalyst lifetime and interest rate to the cost of the oxidation catalyst only (based on the Engelhard formula). Catalyst Disposal Costs For the base case analysis, costs for catalyst disposal were limited to the freight charge associated with shipping the spent modules back to the vendor. Based on the experience of CTWG members, catalyst vendors do not charge for catalyst disposal since the vendors can recover the noble metals from the spent catalysts. Interest Rate for Capital Recovery An interest rate of 7 percent was used in the base case to calculate capital recovery. The EPA Co-Chair of the ICCR Economics Work Group recommended this interest rate for the cost-effectiveness analysis. 13 ------- Direct annual operating and maintenance costs Operating labor costs were estimated using a factor of $25 per hour operating labor and an estimate of two hours per day incremental labor. The labor costs cover costs for operator duties likely to result from installing an oxidation catalyst and complying with MACT. Those duties include 1) inspection of the continuous parameter monitoring device, 2) collection and review of continuous parameter monitoring data, 3) inspection of the control device, and 4) recordkeeping and reporting assumed to be required by the MACT standard. In developing the labor estimates, the CTWG reviewed the EPA estimates for labor for NSCR for reciprocating internal combustion engines and for SCR for turbines included in the Alternative Control Techniques (ACT) documents (EPA, 1993a and 1993b). The CTWG agreed that the labor estimates for NSCR would more closely approximate the labor associated with an oxidation catalyst, since NSCR is essentially a passive catalytic device, like oxidation catalysts. The CTWG agreed that labor costs for SCR for turbines would be greater than the labor costs for oxidation catalysts, since SCR may require frequent inspection and adjustment of the ammonia feed system. Maintenance costs, including labor and materials, were estimated as 10% of the total purchased equipment cost, based on the ACT formula for NSCR. Maintenance costs cover catalyst washing (with water), maintenance of monitoring equipment, and labor for catalyst replacement (including removal and return of old catalyst and installation of replacement). Fuel penalty costs Increased pressure drop in the exhaust of a gas turbine will impact both heat rate and power output. For the base case analysis, fuel penalty costs are included to compensate for the increased heat rate as a result of the increased exhaust backpressure on the turbine that results from installing an oxidation catalyst. The fuel penalty is assessed as the cost of increased fuel, which is calculated by assuming a heat rate increase of 0.105% per inch of pressure drop (measured in inches of water column) and estimates of $2 per MMBtu and a 9,000 Btu/hp-hr baseline. The heat rate increase of 0.105% was drawn from the GRI study. The 14 ------- CTWG agreed that 0.105% is a very low estimate of the heat rate increase anticipated and most turbines would have higher increased heat rate due to backpressure from the catalyst. Other estimates of the heat rate increase are discussed in the complicating factors portion of this section. The estimate of $2 per MMBtu for natural gas was drawn from the GRI study. The CTWG agreed that this estimate is low compared to market value of natural gas at this time. The estimate of increased exhaust backpressure on the turbine from the catalyst was based on an assumption that the total pressure drop associated with the catalyst system is solely the pressure drop across the catalyst panels. The CTWG agreed that the total pressure drop would be higher than the pressure drop across the catalyst panels due to the pressure drop associated with the inlet and outlet ductwork for the catalyst system. Therefore, the increase in the exhaust backpressure and, therefore, the fuel penalty costs resulting from the increase in exhaust backpressure are understated in the base case analysis. The Turbine World Handbook indicates that exhaust backpressure may result in a loss of power. The costs for loss of power were not included in the base case quantitative analysis. These costs would increase the cost of control beyond the base case costs presented in this paper. The costs for loss of power are discussed in the complicating factors portion of this section. Annual Compliance Test Costs Costs to perform one annual emissions compliance test are included in the base case. The costs for this annual test are estimated at $5,000. The costs were estimated based on an assumption that no continuous emissions monitoring data would be required in a MACT standard for combustion turbines. Instead, it was assumed that the MACT would require continuous monitoring for an operating parameter, such as temperature at the catalyst, along with an annual emissions test. The costs also were based on an assumption that a surrogate criteria pollutant can be measured and that HAPs would not be speciated. 15 ------- B. Costs Estimated by OAQPS Control Cost Manual The methodology outlined in the OAQPS Control Cost manual was used by the CTWG to estimate costs for the following: • Capital cost for instrumentation (continuous parameter monitor) • Sales tax for equipment purchases • Freight for equipment purchases • Direct installation costs (DCI), including foundations & supports, handling & erection, electrical, piping, insulation for ductwork, and painting. • Indirect installation costs (ICI), including engineering, construction and field expenses, contractor fees, start-up, and performance tests. • Indirect annual costs (IAC), including annualized equipment costs, overhead, administrative costs, property taxes, and insurance. A description of the methodology to estimate these costs is provided below. Costs for instrumentation, taxes and freight are estimated by applying factors from the OAQPS cost manual to the capital cost of the catalyst and auxiliary equipment. These costs (catalyst capital cost, instrumentation, taxes, and freight) are then summed to estimate the total Purchased Equipment Costs (PEC). The components of the DCI (foundations and supports, erection and handling, electrical work, piping, painting and insulation) are then calculated by applying OAQPS cost manual factors to the PEC. Likewise, the components of the ICI (engineering, construction and field expenses, contractor fees, start-up, and initial performance test) are also calculated by applying factors to the PEC. Indirect Annual Costs (IAC) are the annualized cost of the catalyst housing and the costs for overhead, administrative tasks, property taxes, and insurance. The equipment costs are annualized by applying a capital recovery factor (based on the equipment life, 15 years, and interest rate) to the sum of the direct and the indirect equipment costs, excluding the cost of the catalyst modules. The cost of the catalyst modules is considered a direct annual cost (DAC), and is annualized separately. Factors applied to the sum of 16 ------- the direct and indirect equipment costs (including contingency) are used to estimate the overhead, administrative costs, property taxes, and insurance. C. Summary of Base Case Cost Estimates Table 3 presents the range of costs estimated for the seven model turbines included in the base case cost-effectiveness analysis. The costs for each model turbine are presented in Appendix E. The highest annual costs are for the largest model turbine and the lowest annual costs are for the smallest model. The $/MW are lower for the larger model turbines and higher for the smaller model turbines. Table 3. Range of Costs Estimated for Seven Model Turbines Cost Category Costs for 3-Year Catalyst Life* Costs for 6-Year Catalyst Life* Total Capital Cost $360,000 - $4,800,000 $360,000 - $4,800,000 Direct Annual Cost $96,000 - $980,000 $74,000 - $680,000 Indirect Annual Cost $65,000 - $700,000 $65,000 - $700,000 Total Annual Costs (DAC + IAC) $160,000 - $1,700,000 $140,000 - $1,400,000 *Costs are rounded. D. Complicating Factors This section presents the views of the CTWG with regard to factors that complicate the estimation of the costs of acquisition, installation, and operation of oxidation catalyst on combustion turbines. For discussion, these complicating factors are divided into five categories: • factors related to the cost of acquiring the oxidation catalyst, • costs associated with site installation complications, • costs associated with performance testing, • complicating factors associated with increased exhaust backpressure, and • costs associated with compliance monitoring. 17 ------- Factors Complicating the Estimation of Catalyst Acquisition Costs The catalyst costs used in this base case analysis are based on a formula that was derived from one vendor's cost quotes for six different sizes of combustion turbines. The vendor's cost quotes covered a range of turbine sizes that is similar to the turbine sizes represented in the seven model turbines used in this cost- effectiveness analysis. Exhaust flow rates for the vendor's cost quotes ranged from 28.4 lb/sec to 984 lb/sec, while exhaust flow rates for the seven model turbines ranged from 14.2 lb/sec to 986 lb/sec. The formula developed by the CTWG for this cost-effectiveness analysis represents a necessary simplification of the vendor's cost quotes to facilitate estimating costs for the seven model turbines used in this analysis. The CTWG had cost estimates for oxidation catalysts available from two other sources: 1) cost estimates provided by Mr. Marvin Schorr of General Electric (Schorr, 1998), and 2) cost estimates included in the GRI cost study (Ferry et al., 1998). Cost estimates were provided by General Electric for two large turbines (exhaust flow rates of 400 lb/sec and 1200 lb/sec). The formula calculated using the General Electric cost estimates is (0.85*(568.75*Exhaust Flow Rate (lb/hr) +172,500). For small turbines, the costs estimated using the General Electric formula are higher than the costs used in this base case analysis. For example, the General Electric formula estimates $153,490 for the catalyst for a 1.13 MW turbine, while the costs used in this base case analysis are $105,624. For a 3.5 MW turbine, the costs are similar, $166,446 estimated using the General Electric formula and $165,584 used in this analysis. For larger turbines, the costs estimated using the General Electric formula are lower than the costs used in this base case analysis. The differences in the costs estimated using the two different approaches increase with turbine size. For the 170 MW turbine, the General Electric formula estimates the cost of the catalyst as $623,294, while $1,622,585 was used in this cost-effectiveness analysis. [Note: the quote provided by Engelhard for a 170 MW turbine, exhaust flow 984.01b/sec was $1,550,000.] The CTWG agreed not to use the General Electric cost estimates for this base case 18 ------- analysis for the following reasons: 1) cost estimates were provided only for two large turbines, and 2) the costs seemed to underestimate the costs when compared with the quotes received directly from a catalyst vendor. The CTWG also reviewed the cost estimates included in the GRI study. In that case, GRI used cost quotes provided by two catalyst vendors for a 6,000 horsepower turbine. Vendors provided cost quotes for a range of VOC control estimates: 95 percent, 50 percent, 35 percent, and 22 percent. In comparing the cost quote in the GRI study for 95 percent VOC control and 98 percent CO control, the CTWG noted that the costs were similar to the costs for a 6,000 hp turbine estimated using the formula in this base case (assuming 90 percent CO control) — $204,500 in the GRI study, and $206,796 using the base case formula. The CTWG decided not to use the GRI costs for this analysis because there was insufficient information to develop a reliable cost formula that could be applied to a wide range of turbine models, ranging in size from 1.13 MW to 170 MW. The CTWG notes that vendor quotes that have been obtained are essentially for CO oxidation catalysts. As noted above, available emissions data indicates that CO/VOC oxidation catalysts should reduce organic HAP compounds. However, the CTWG is not aware of any actual industry experience in the acquisition of an oxidation catalyst specified to achieve a percentage reduction of formaldehyde, or the other HAPs. In the absence of such experience, the cost estimate for an oxidation catalyst designed to reduce organic HAPs from combustion turbines is uncertain. Uncertainty about the estimated cost for a HAP reduction catalyst is increased when considering that oxidation catalysts would be required for fuels other than natural gas. Oxidation catalysts for oil fired turbines may have to be formulated differently than for gas fired turbines, and may have different lifetime and degradation characteristics. Another key uncertainty in estimating oxidation catalysts costs is the assumption regarding catalyst life. Clearly, a catalyst that can be relied upon to function for 19 ------- many years will have lower annual costs than a catalyst that must be replaced more often. The issue of catalyst lifetime includes estimating the probability of complete failure of the catalyst, and also estimating the degradation of catalyst performance over time. The CTWG notes that there may be a difference between the expected useful life of an oxidation catalyst, and the period of the vendor's performance guarantee. This raises the question of which period should be used in calculating cost- effectiveness. As noted in another section, the CTWG has elected to present a number of cost-effectiveness estimates based on different assumptions about catalyst life and performance. Limited information was available to the CTWG on the life of the catalyst. Information from an emissions test conducted by GRI on a ten-year-old CO oxidation catalyst indicates that performance can degrade when the catalyst is used for an extended period of time (10 years in that case). The GRI test is described under Section V of this paper. Further information is not available that would allow the CTWG to estimate the expected rate of oxidation catalyst performance degradation, or the effect of maintenance (such as catalyst washing) on catalyst life. According to catalyst vendors, the degradation of catalyst performance over time is not linear. The CTWG has not obtained any information that would allow the Work Group to estimate the expected rate of performance degradation over the life of the catalyst. Costs associated with site installation complications Costs for retrofit complications were not available for the base case analysis. Site-specific factors can have a major impact on the cost of retrofitting a catalyst control system to an existing turbine installation. In general, the heat recovery unit (if one exists) must be altered, ductwork and piling supports must be added, and piping, electrical conduits and wiring must be lengthened. Some turbine installations have enough space between the turbine exhaust and the heat recovery 20 ------- unit to add the catalyst system. In cases where space is very limited, the heat recovery unit might have to be removed and replaced with a new vertical style unit. One of the work group members provided retrofit costs for adding a catalyst system to an ABB Type 11 gas turbine (gas flow = 580 lb/sec) (Allen, 1998a and 1998b). The retrofit costs totaled about $800,000, including $100,000 for ductwork. The cost of down time is also site specific. In the case described above, the cost cited by the work group member for down time was about $3.5 million based on a 35 day outage, a power sales price of $35/MWh, and a steam cost $4.5/thousand pounds of steam (Allen, 1998a). Costs Associated with Performance Testing Costs for performance testing were included in the base case quantitative analysis in accordance with the OAQPS Control Cost Manual. The costs for performance testing are estimated as 0.01% of the Purchased Equipment Costs (PEC). For the 170 MW turbine, $27,000 was calculated as the performance test costs using the OAQPS formula. For the 1.13 MW turbine, $2,095 was calculated as the performance test costs using the OAQPS formula. The CTWG agreed that the costs for stack emissions testing would be fixed, regardless of turbine size. The costs estimated for performance testing may have been underestimated for the base case analysis, especially for the small model turbines. Complicating Factors Associated with Increased Exhaust Backpressure For the base case quantitative analysis, fuel penalty costs were estimated assuming a 0.105% heat rate increase per inch of pressure resulting from installation of a catalyst system. The CTWG agreed that 0.105% is a very low estimate of the heat rate increase. The Gas Turbine World 1997 Handbook provides rough rule of thumb estimates of heat rate increase and power loss per inch pressure drop (Gas Turbine World 1997). For aeroderivative turbines, the Handbook indicates that every 4 inches outlet loss will increase heat rate 0.7% (0.175%) per inch) and reduce power output 0.7%. For heavy frame turbines, the Handbook indicates that every 4 inches outlet loss will increase heat rate 0.6% 21 ------- (0.15% per inch) and reduce power output 0.6%. Therefore, the heat rate increase due to increased pressure drop is understated in the base case analysis. To estimate pressure drop for the base case quantitative analysis, it was assumed that the total pressure drop associated with the catalyst system is solely the pressure drop across the panels. The CTWG agreed that the total pressure drop would be higher than the pressure drop across the catalyst panels alone due to the inlet and outlet ductwork. Therefore, the operating costs associated with the increase in exhaust backpressure are understated in the base case analysis. The fuel penalty costs associated with backpressure may be significantly higher when a more realistic estimate of the catalyst system pressure drop is used. In addition, implementing oxidation catalyst control may result in a reduction in turbine power output caused by increased exhaust backpressure on the engine. The costs associated with the power loss depend on site-specific factors (e.g., value of lost product or capital and annual costs for equipment required to make up for the power loss). The increase in exhaust backpressure results in a loss of power sales if the unit is operating at full load. One of the work group members provided information on the loss in annual sales at different selling prices for electrical power (Allen, 1998b). For a GE Frame 7 turbine, the annual cost (i.e., lost sales) per inch of water pressure drop may be estimated using the following equation: Annual Cost ($/inch) = 1,160 * Power Value ($/MWh) +100 For this example turbine unit, if electricity can be sold for $40 per MWh, the annual cost per each additional inch of water pressure drop caused by the catalyst would equal $46,500. These costs were not incorporated into the base case analysis. The cost associated with power loss would increase the costs for the control system. 22 ------- Costs Associated with Compliance Monitoring If the MACT would require speciated HAP emissions test data, the costs for the annual compliance test would increase significantly. Also, if compliance tests must be conducted more frequently than annually, the costs would increase. V. Performance of Oxidation Catalysts in Reducing HAP Emissions Oxidation catalysts have been installed on combustion turbines for the purposes of controlling emissions of carbon monoxide (CO) and some volatile organic compounds (VOC). The catalyst is designed to promote the oxidation of hydrocarbon compounds to carbon dioxide (CO2) and water (H20). It is expected that existing catalysts similar to those in use for CO and VOC control may oxidize organic HAPs. In order to estimate the quantitative performance of an oxidation catalyst the CTWG evaluated two emissions test reports and reviewed engineering estimates of potential oxidation catalyst performance. A. HAP Emissions Test Data for Oxidation Catalysts At present, no HAP emissions tests in the ICCR Emissions Database include before and after testing of a combustion turbine with an oxidation catalyst. Emissions test data on the performance of oxidation catalysts should be collected during the CTWG testing campaign. The CTWG identified two existing emission test reports that provide some information on the performance of oxidation catalysts in reducing HAP emissions. The two emission tests are still being evaluated and may be included in the database after review. One test was conducted by the Gas Research Institute(GRI), in cooperation with the American Petroleum Institute (API) and Southern California Gas (SoCal), in March 1998, on a combustion turbine using a passive oxidation catalyst system, similar to the catalyst used 23 ------- for this base case cost-effectiveness evaluation. A summary of this test has been provided to the CTWG and the complete test data will be provided to EPA when it is available (Gundappa, 1998). The complete test report will be required by EPA and the report will have to undergo review prior to being included in the ICCR Emissions Database. The oxidation catalyst installed on this turbine is a precious metal catalyst, similar to the catalyst technology used as the basis for this cost-effectiveness analysis. This type of oxidation catalyst may be used over a temperature range of 450°F to 1500°F (Chen et al., 1993). The second test was submitted to EPA for a new catalytic oxidation control system, called SCONOx™ (Bell and Finken, 1997). Although the SCONOx™ system relies on oxidation to reduce hydrocarbons, such as CO, or HAPs, such as formaldehyde, the SCONOx™ catalyst is a more complicated control system than the oxidation catalyst used for this base case cost-effectiveness evaluation. SCONOx™ may be operated over a temperature range of 300°F to 700°F (Goal Line Environmental Technologies, LLC). The cost and cost-effectiveness values presented in this paper were not based on costs for the SCONOx™ system. However, the CTWG included a discussion of the source test results as an indicator of the types of emission reductions that may be achievable for systems that rely on oxidation to reduce HAP emissions. A description of the SCONOx™ system is provided in Appendix F. The results from these two emissions tests are discussed below. GRI/API/SoCal Test The GRI/API/SoCal testing was conducted in March 1998. GRI, API, and SoCal added the emissions test to an existing emissions testing program in order to provide data to the CTWG on the performance of oxidation catalysts. Some members of the CTWG and EPA representatives witnessed the GRI/API/SoCal test. The test was performed on a 20 MW GE LM2500 turbine equipped with a Johnson Matthey CO oxidation catalyst. Three load conditions were tested, including full load (typical) and part loads (88% and 70% of rated load). Concentrations of HAPs, including formaldehyde, were measured before and after 24 ------- the oxidation catalyst. HAP and CO measurements were conducted with Fourier transform infrared (FTIR) sampling upstream and downstream of the oxidation catalyst. Aldehydes also were measured with the California Air Resources Board (CARB) Method 430, which relies on an aqueous 2,4-Dinitrophenylhydrazine solution. Complete results of the test were not available in time to incorporate them into the ICCR Emissions Database. However, the CTWG has been provided a summary of the results (Gundappa, 1998). Based on FTIR, formaldehyde emissions upstream of the catalyst were in the approximate range of 400 to 460 parts per billion by volume (ppbv) and CO emissions upstream of the catalyst were in the range of 10 to 17 parts per million by volume (ppmv). Both formaldehyde and CO emissions increased as the load decreased. With FTIR, the reduction in emissions across the oxidation catalyst was on the order of 10 to 30 percent for formaldehyde and 25 to 33 percent for CO, with the highest reduction at the lowest load condition. CARB 430 results did not agree with the FTIR data. In some cases, the CARB 430 results indicated that levels of aldehydes (formaldehyde and acetaldehyde) increased after the catalyst. SCONOx™ Test A unit equipped with a SCONOx™ catalyst system was tested on March 14, 1997, by Delta Air Quality Services (Bell and Finken, 1997). Samples were collected at the inlet to the catalyst and at the exhaust from the cogeneration unit (turbine exhaust stack) and analyzed for the following three HAPs: formaldehyde, acetaldehyde, and benzene. Formaldehyde and acetaldehyde reportedly were reduced by 97% and 94%, respectively, based on the catalyst inlet and turbine exhaust concentrations. No conclusion regarding the control efficiency for benzene could be drawn since the levels before and after the catalyst were both very low and within 0.05 parts per billion of each other. A subgroup of the CTWG reviewed the SCONOx™ report in greater detail to determine if the data from this test should be included in the emissions database. The subgroup was concerned with the accuracy of the catalyst inlet concentrations 25 ------- measured during the test since isokinetic sampling was not conducted nor was a multi-point probe used to collect the samples. However, the catalyst inlet concentrations were consistent with other source tests involving the same model turbine (GE LM 2500), using water injection. Also, even if the catalyst inlet concentrations were one-half to one-third of the average concentration measured during the source test, the efficiency of the SCONOx™ would still exceed 90% for formaldehyde. Therefore, the subgroup decided to support inclusion of the data from this test in the emissions database, with the caveat that EPA may want to retest this unit to address some of the specific concerns identified during the subgroup's review. Based on a review of the two emissions tests available, the CTWG concluded that organic HAPs, such as formaldehyde and acetaldehyde, may be reduced using after- treatment controls that rely on catalytic oxidation. The Work Group also concluded that, in some cases, a high percent reduction may be possible for certain pollutants. However, the CTWG noted that the limited data available is not sufficient to draw conclusions about the achievability of high emission reductions over the life of catalytic devices. In addition, the CTWG noted that although there is some data that suggests catalysts degrade over time, the rate and the extent of the degradation cannot be determined based on the limited data. B. Engineering Estimates of HAP Reduction Performance for Oxidation Catalysts The CTWG reviewed information available in the literature on the HAP reduction performance of oxidation catalysts on organic HAPs, such as formaldehyde. In particular, the Work Group reviewed an article prepared by Engelhard, the catalyst vendor that supplied the cost quotes for this base case cost-effectiveness analysis (Chen et al., 1993). In the article, Engelhard notes that oxidation catalysts for combustion turbines are typically designed to achieve between 80 and 95 percent CO removal. In addition, the article indicates the conversion level for each species of hydrocarbon will 26 ------- depend on its diffusion rate in the exhaust gas. In general, larger, heavier molecules will diffuse more slowly than smaller, lighter molecules. As the size of the hydrocarbon molecule increases, hydrocarbon conversion decreases due to decreased gas diffusivity. According to the article, an oxidation catalyst designed for 90 percent CO removal will achieve 77 percent reduction of formaldehyde, 72 percent reduction of benzene, and 71 percent reduction of toluene. The article notes that the relative conversion rates do not depend on geometry and that reduction for molecules larger than formaldehyde will be lower than rates achievable for formaldehyde. C. Summary of Base Case Performance Estimate The CTWG has agreed to use two performance values for the base case cost- effectiveness analysis — 80 percent emissions reduction and 50 percent emissions reduction. 80 percent emissions reduction is used for both the 3-year and 6-year catalyst life assumptions. 50 percent emissions reduction is evaluated for a 6-year catalyst life. The CTWG believes these levels of reduction represent appropriate levels of reduction for the base case cost-effectiveness analysis, covering both high and moderate levels of emission reduction. The Work Group relied on the Engelhard engineering estimates for formaldehyde to select 80% reduction as the catalyst performance in the base case analysis (77% rounded up to 80%). Although the Engelhard article indicates that emission reductions for larger molecules, such as PAHs, may be less than the reduction achieved for formaldehyde, the HAP reduction performance for the base case analysis was set to 80 percent for all pollutants. The Work Group selected 50% reduction as a moderate level of emission reduction to examine the sensitivity of the cost-effectiveness to any significant degradation of the catalyst performance that might occur over time. Additional emissions test data before and after oxidation catalysts would be necessary to determine whether the levels of reductions are achievable for combustion turbines, considering the full range of operating conditions and catalyst degradation. 27 ------- The emission reductions achieved for each model turbine assuming 80 percent reduction and 50 percent reduction are presented in Tables 4 and 5. 28 ------- Table 4. Emissions Reductions for Each Model Turbine Assuming 80% HAPs Reduction Performance Emissions Reductions (Mg/yr)— Highest Emission Factor — 80% HAPs Reduction Performance Model Turbine Forniiildchvde Toluene Aeetsildehvde Xvlenes Ftlivl ben/ene Ben/ene PAHs Acrolein Nil pli tliii lene Tohil IIAl's 2 170 MW 27.048 3.664 1.687 0.579 0.198 0.189 0.035 0.029 0.016 33.445 1 85.4 MW 13.587 1.841 0.848 0.291 0.099 0.095 0.018 0.015 0.008 16.801 7 39.6 MW 6.301 0.854 0.393 0.135 0.046 0.044 0.008 0.007 0.004 7.791 9 27 MW 4.296 0.582 0.268 0.092 0.031 0.030 0.006 0.005 0.003 5.312 15 9 MW 1.432 0.194 0.089 0.031 0.010 0.010 0.002 0.002 0.001 1.771 13 3.5 MW 0.557 0.075 0.035 0.012 0.004 0.004 0.001 0.001 < 0.001 0.689 17 1.13 MW 0.180 0.024 0.011 0.004 0.001 0.001 < 0.001 < 0.001 < 0.001 0.222 Emissions Reductions (Mg/yr)— Average Emission Factor — 80% HAPs Reduction Performance Model Turbine Formsildchvde Toluene Aeetsildehvde Xvlenes Ftlivl ben/ene Ben/ene PAHs Acrolein Nil plitliiilene Tohil IIAl's 2 170 MW 3.438 0.685 0.440 0.221 0.198 0.050 0.011 0.026 0.007 5.075 1 85.4 MW 1.727 0.344 0.221 0.111 0.099 0.025 0.005 0.013 0.004 2.549 7 39.6 MW 0.801 0.159 0.102 0.052 0.046 0.012 0.003 0.006 0.002 1.182 9 27 MW 0.546 0.109 0.070 0.035 0.031 0.008 0.002 0.004 0.001 0.806 15 9 MW 0.182 0.036 0.023 0.012 0.010 0.003 0.001 0.001 < 0.001 0.269 13 3.5 MW 0.071 0.014 0.009 0.005 0.004 0.001 < 0.001 0.001 < 0.001 0.104 17 1.13 MW 0.023 0.005 0.003 0.001 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.034 29 ------- Table 5. Emissions Reductions for Each Model Turbine Assuming 50% HAPs Reduction Performance Emissions Reductions (Mjj/yr)— Highest Emission Factor — 50% Reduction Performance Model Turbine Formsildchvde Toluene Aeetsildehvde Xvlenes Ftlivl benzene Benzene PAHs Acrolein Nil plitliiilene Tohil IIAPs 2 170 MW 16.905 2.290 1.055 0.362 0.124 0.118 0.022 0.018 0.010 20.903 1 85.4 MW 8.492 1.150 0.530 0.182 0.062 0.059 0.011 0.009 0.005 10.501 7 39.6 MW 3.938 0.533 0.246 0.084 0.029 0.027 0.005 0.004 0.002 4.869 9 27 MW 2.685 0.364 0.168 0.057 0.020 0.019 0.004 0.003 0.002 3.320 15 9 MW 0.895 0.121 0.056 0.019 0.007 0.006 0.001 0.001 0.001 1.107 13 3.5 MW 0.348 0.047 0.022 0.007 0.003 0.002 < 0.001 < 0.001 < 0.001 0.430 17 1.13 MW 0.112 0.015 0.007 0.002 0.001 0.001 < 0.001 < 0.001 < 0.001 0.139 Emissions Reductions (M^/vr)-- Average Emission Factor — 50% HAPs Reduction Performance Model Turbine Formsildchvde Toluene Aeetsildehvde Xvlenes Ftlivl benzene Benzene PAHs Acrolein N si plithii lene Totiil IIAPs 2 170 MW 2.149 0.428 0.275 0.138 0.124 0.031 0.007 0.017 0.004 3.172 1 85.4 MW 1.079 0.215 0.138 0.069 0.062 0.016 0.003 0.008 0.002 1.593 7 39.6 MW 0.500 0.100 0.064 0.032 0.029 0.007 0.002 0.004 0.001 0.739 9 27 MW 0.341 0.068 0.044 0.022 0.020 0.005 0.001 0.003 0.001 0.504 15 9 MW 0.114 0.023 0.015 0.007 0.007 0.002 < 0.001 0.001 < 0.001 0.168 13 3.5 MW 0.044 0.009 0.006 0.003 0.003 0.001 < 0.001 < 0.001 < 0.001 0.065 17 1.13 MW 0.014 0.003 0.002 0.001 0.001 < 0.001 < 0.001 < 0.001 < 0.001 0.021 30 ------- D. Complicating Factors This section presents the views of the CTWG with regard to factors that complicate the estimation of the performance of oxidation catalysts in the reduction of organic HAP in the exhaust of combustion turbines. Uncertainty About the Real World Performance of Oxidation Catalysts for HAPs As noted earlier in this paper, although there are oxidation catalysts installed on existing turbines for control of CO and some VOCs, there are not conclusive emissions data available regarding the HAP reduction performance of those oxidation catalysts over time. CO catalysts systems in use operate on far higher levels of CO than the expected concentration of HAPs. The cost-effectiveness estimates used for this base case analysis are derived from engineering judgement rather than actual data. It is possible that it may be more difficult than anticipated to achieve a consistent 80% reduction of HAPs across a real world population of combustion turbines running under various ambient conditions and operating points. Differential Performance for Various HAPs The assumption used in this base case analysis that oxidation catalysts will have the same HAP reduction performance for all organic HAPs was necessary because there was insufficient emissions data to estimate HAP reduction performance for specific species of HAPs. The CTWG is aware that this assumption is incorrect, based on engineering estimates performed by Engelhard, a catalyst vendor (Chen et al., 1993). Engelhard indicates that individual HAPs will be oxidized at different rates due to differences in the size of the hydrocarbons and that the HAP reduction performance for each HAP will depend on its diffusion rate. In general, larger, heavier molecules (like PAHs) will diffuse more slowly than smaller, lighter molecules (like CO). The CTWG notes that the assumptions used in this base case analysis tend to overestimate HAP reduction efficiencies for HAPs other than formaldehyde, especially HAPs like PAHs that are larger, heavier molecules. 31 ------- Decreased Catalyst Performance Over Time This effect was discussed as a part of the evaluation of catalyst life for costing purposes. A decline in catalytic activity also would impact the performance side of the equation in that fewer metric tons of HAPs would be removed from the turbine exhaust. Again, the CTWG does not have sufficient information to estimate the rate at which catalytic activity would decline in a real-world installation. VI. Cost-Effectiveness Results A breakdown of the total HAP reductions achieved for individual pollutants is provided in Tables 4 and 5. The cost-effectiveness values based on total HAP reductions are presented in Table 6 for each model turbine. The cost-effectiveness for total HAPs is provided to more fully demonstrate the benefit achieved in terms of total reduction of HAPs for the costs required to install oxidation catalysts. Cost-effectiveness for individual HAPs, calculated as the total annual costs by the mass emissions for each individual HAP, is presented in Appendix G. The cost-effectiveness for individual HAPs is presented to show the cost-effectiveness sensitivity for individual HAPs. In general, the cost per metric ton of reduced HAP emissions is higher for small turbines, because capital costs, on a per-megawatt basis, are highest for these units and the annual HAP emissions are low. The costs per metric ton also would increase for small and large turbines as operating hours decrease because capital costs remain unchanged while annual HAP emissions are lower. 32 ------- Table 6. Cost-Effectiveness Estimated for Each Model Turbine ~ Base Case Analysis Cost Effectiveness ($/Mg Total HAPs Reductions*) Model Plant Highest EF Average EF 3-Year Catalyst Life 80% Emissions Reduction 6-Year Catalyst Life 80% Emissions Reduction 6-Year Catalyst Life 50%) Emissions Reduction 3-Year Catalyst Life 80%o Emissions Reduction 6-Year Catalyst Life 80%o Emissions Reduction 6-Year Catalyst Life 50% Emissions Reduction Model 1 - 85.4 MW Turbine $69,000 $57,000 $91,000 $450,000 $380,000 $600,000 Model 2 — 170 MW Turbine $50,000 $41,000 $66,000 $330,000 $270,000 $440,000 Model 7 - 39.6 MW Turbine $81,000 $67,000 $110,000 $530,000 $440,000 $710,000 Model 9 — 27 MW Turbine $78,000 $66,000 $100,000 $520,000 $430,000 $690,000 Model 13 - 3.5 MW Turbine $290,000 $250,000 $400,000 $1,900,000 $1,700,000 $2,600,000 Model 15 — 9 MW Turbine $150,000 $130,000 $200,000 $1,000,000 $840,000 $1,400,000 Model 17 - 1.13 MW Turbine $730,000 $630,000 $1,000,000 $4,800,000 $4,100,000 $6,600,000 *Cost-effectiveness values were rounded. Annual costs estimated for each model turbine are presented in Appendix E. HAPs reductions estimated for each model turbine are presented in Tables 4 and 5. Cost-effectiveness values for individual HAPs are presented in Appendix G. 33 ------- VII. Conclusions and Recommendations The CTWG has assessed the various elements that are relevant to estimation of the cost- effectiveness of oxidation catalysts for control of organic HAPs emitted by combustion turbines. Base on this assessment the CTWG has reached the following conclusions. 1. Using a simplified base case, the annual costs associated with installation and operation of oxidation catalysts for the model turbines ranged from $160,000 for a 1.13 MW unit to $1,700,000 for a 170 MW unit, assuming a three-year catalyst life. Annual costs ranged from $140,000 for a 1.13 MW unit to $1,400,000 for a 170 MW unit, assuming a six-year catalyst life. 2. Based on quantified estimates of emissions, cost, and percent reduction for a simplified base case, the cost-effectiveness of oxidation catalysts for control of total HAPs from combustion turbines ranges from $41,000 per metric ton for a 170 MW unit to $1,000,000 per metric ton for a 1.13 MW unit, assuming emission rates based on the highest reported emission factors for all HAPs. The cost-effectiveness values range from $270,000 for a 170 MW unit to $6,600,000 for a 1.13 MW unit when the average emission factor is used. 3. Because of a variety of complicating factors, it is likely that the base case cost-effectiveness estimated range is lower than the actual cost- effectiveness which would be exhibited by actual application of oxidation catalysts to most combustion turbines in the United States. Key complicating factors include the catalysts life, problems with retrofitting ducts and the catalyst housing at existing facilities, differential effectiveness of the catalysts on various HAP compounds, and fuels that require pre-treatment to avoid fouling the catalyst. In addition, there is uncertainty regarding the HAPs reduction performance included in this base case analysis due to the limited emissions test data available to predict the performance of oxidation catalyst in reducing organic HAP emissions from combustion turbines. While experience with CO oxidation catalysts is useful for evaluating the potential HAP reduction performance, there may be important differences between the costs and performance of CO catalysts and the costs and performance of catalysts for reduction or organic HAPs. Most of the complicating factors that have not been quantified in the numerical estimates would tend to increase the catalyst costs, or decrease catalyst performance. Because of this, the CTWG views the base case quantitative estimate reported in this paper as a 34 ------- lower range estimate of the cost-effectiveness of oxidation catalysts for HAPs control on combustion turbines. The CTWG recommends that the Coordinating Committee forward this information to EPA and recommend that EPA consider the information presented in this paper in the Agency's assessment of above-the-floor MACT options for combustion turbines. This paper provides reasonable estimates, based on available information, of the costs and the HAP air emissions reductions that may be achieved with oxidation catalysts. The CTWG recognizes that EPA may consider other factors, such as non-air quality environmental impacts, energy requirements, and secondary pollutants (including possible CO/VOC control), in assessing above-the-floor MACT options. 35 ------- References Cited Allen, S. 1998a. E-mail to Sims Roy, U.S. EPA, to transmit retrofit cost information provided by Vogt, August 10, 1998. Allen, S. 1998b. Memorandum to transmit information regarding power loss/pressure drop relationship. Submitted to ICCR Combustion Turbine Work Group on July 30, 1998. Bell, A., and R. A. Finken. 1997. Formaldehyde and Acetaldehyde and Benzene Control Efficiency at Federal Cold Storage. Report prepared by Delta Air Quality Services, Orange, CA, April 2, 1997. Chen, J. M., B. K. Speronello, and R. M. Heck. 1993. "Catalytic Control of Unburned Hydrocarbon Emissions in Combustion Turbine Exhausts." Air & Waste Management Association. Paper prepared by Engelhard Corporation, Iselin, New Jersey, for presentation at the AWMA 86th Annual Meeting & Exhibition, Denver, Colorado, June 13-18, 1993. Engelhard. 1998. Facsimile to U.S. EPA, regarding CO catalysts/HAP control, April 27, 1998. EPA. 1990. OAQPS Control Cost Manual, Fourth Edition. EPA 450/3-90-006. Office of Air Quality Planning and Standards, Office of Air and Radiation, Research Triangle Park, NC, January 1990. EPA. 1993a. Alternative Control Techniques Document — NOx Emissions from Stationary Gas Turbines. Office of Air Quality Planning and Standards, Office of Air and Radiation, Research Triangle Park, NC, January 1993. EPA. 1993b. Alternative Control Techniques Document — NOx Emissions from Stationary Reciprocating Internal Combustion Engines. Office of Air Quality Planning and Standards, Office of Air and Radiation, Research Triangle Park, NC, July 1993. Ferry, K. R., W. C. Rutherford, and G. S. Shareef. 1998. Preliminary Study of Oxidation Catalyst Costs Applied to Gas Turbines for Control of Aldehydes. Gas Research Institute Topical Report GRI-98/0218. Report prepared by Radian International, LLC, Austin, TX, June 1998. Gas Turbine World. 1997. Gas Turbine World 1997 Handbook. Goal Line Environmental Technologies, LLC. Promotional materials regarding the SCONOx™ Catalytic Absorption System. ------- Gundappa. M. 1998. "Preliminary Results from Natural Gas-Fired Combustion Turbine Testing." American Petroleum Institute, Southern California Gas Company, and Gas Research Institute. Report prepared by Radian International, LLC, Austin, TX, August 1998. ICCR Testing & Monitoring Work Group. 1997. "September 1997 TMPWG Guidance on Interpreting and Using Emissions Database Containing Non-detection Values." MacDonald, R. J. and L. Debbage. 1997. "The SCONOx™ Catalytic Absorption System for Natural Gas Fired Power Plants: The Path to Ultra-Low Emissions." Power- Gen International '97, Dallas, TX, December 9-11, 1997. Schorr, M. E-mail to U.S. EPA, to provide catalyst costs, May 15, 1998 ------- Appendix A - List of Model Turbines Model Plant No. Unit Size Operating Hours Per Year Heat Recovery (YIN) Existing Application (YIN) Clean Fuel (YIN) Typical Applications Surrogate Turbine Output MW(ISO) Ex. Flow (lbs/sec) 1 Large 8000 Y Y Y existing utility/IPP generating station GE PG 7121EA 85.4 658 1A Large 8000 Y Y N existing unit with residual oil fuel GE PG 7121 EA 85.4 658 1B Large 8000 Y Y Y existing utility/IPP generating station (duct burner) GE PG 7121 EA 85.4 658 2 Large 8000 Y N Y new utility/IPP generating station GE PG 7231 FA 170 986 2A Large 8000 Y N N new unit with residual oil fuel GE PG 7231 FA 170 986 2B Large 8000 Y N Y new utility/IPP generating station (duct burner) GE PG 7231 FA 170 986 3 Large 2000 N Y Y existing utility/IPP generating station GE PG 7231 FA 170 986 3A Large 2000 N Y Y existing utility/IPP station (space constrained) GE PG 7231 FA 170 986 4 Large 2000 N N Y new utility/IPP generating station GE PG 7231 FA 170 986 5 Large 500 N Y Y existing utility/IPP peaking unit GE PG 7121 EA 85.4 658 6 Large 500 N N Y new utility/IPP peaking unit GE PG 7121 EA 85.4 658 7 Medium 8000 Y Y Y existing industrial power production GE PG 6561B 39.6 318 7A Medium 8000 Y Y N existing unit with residual oil fuel GE PG 6561B 39.6 318 7B Medium 8000 Y Y Y existing industrial power production (duct burner) GE PG 6561B 39.6 318 8 Medium 8000 Y N Y new industrial power production GE PG 6561B 39.6 318 8A Medium 8000 Y N N new unit with residual oil fuel GE PG 6561B 39.6 318 8B Medium 8000 Y N Y new industrial power production (duct burner) GE PG 6561B 39.6 318 9 Medium 8000 N Y Y existing pipeline compressor/ ind.- mech. drive GE LM2500 27 178 10 Medium 8000 N N Y new pipeline compressor/ ind. mech. drive GE LM2500 27 178 11 Medium 500 N Y Y existing utility/IPP peaking unit GE PG 6561B 39.6 318 12 Medium 500 N N Y new utility/IPP peaking unit GE PG 6561B 39.6 318 13 Small 8000 Y Y Y existing industrial process plant (food, nat'l gas) Solar Centaur 40 3.5 41 13A Small 8000 Y Y N existing landfill operation or residual oil fuel Solar Centaur 40 3.5 41 13B Small 8000 Y Y Y existing ind. process plant (duct burner) Solar Centaur 40 3.5 41 14 Small 8000 Y N Y new industrial process plant (food, nat'l gas) Solar Centaur 40 3.5 41 14A Small 8000 Y N N new landfill operation or residual oil fuel Solar Centaur 40 3.5 41 14B Small 8000 Y N Y new ind. process plant (duct burner) Solar Centaur 40 3.5 41 15 Small 8000 N Y Y existing pipeline compressor Solar Mars T12000 9 83.6 15A Small 8000 N Y Y existing offshore platform (space constrained) Solar Mars T12000 9 83.6 16 Small 8000 N N Y new pipeline compressor/offshore platform Solar Mars T12000 9 83.6 17 Small 200 N Y Y existing emergency power (hospital,university,etc) Solar Saturn T1500 1.13 14.2 18 Small 200 N N Y new emergency power (hospital, university, etc) Solar Saturn T1500 1.13 14.2 A-l ------- Appendix B - Description of ICCR Emissions Database MEMORANDUM DATE : SUBJECT: TO : FROM : March 6, 1998 Documentation on the Combustion Turbines Emissions Database Combustion Turbines Project File Ana Rosa Alvarez and Dan Herndon This memorandum provides a short description of the development of the emissions database for turbines, including assumptions used in the underlying calculations. Development of the Emissions Database The emission test reports were first carefully reviewed and summarized. Facility name, location, testing company, date of testing, make and model of turbine, manufacturer rating (and units), load, fuel type, application and control device (for emissions) were entered in a table named "Facilities." Pollutant name, sampling method, concentrations and units, detection limits and units, % oxygen, fuel factors, exhaust gas flow rates, stack temperature, fuel heating value and flow rate, % humidity, standard temperature, and pollutant molecular weight were entered in a table named "Test Data." Emission rates (lb/hr) and emission factors (lb/MMBtu) were also entered in that table for comparison with the emissions calculated in the database using the pollutant concentrations for each test run. Test reports included in the database were identified using the following scheme: numbers from 1 to 99 were assigned to tests containing only hazardous air pollutants (HAPs), and numbers greater than 100 were allocated for tests with only criteria pollutants or with both HAPs and criteria pollutants. Exceptions are the reports numbered 10 and 15. These test reports contain both HAPs and criteria pollutant test results. They are numbered as HAPs-only type reports because criteria pollutant data were identified in these reports after the first version of the database was posted on the TTN. Test reports containing more than one turbine, multiple load conditions, different fuels, control device inlet and outlet samples (criteria pollutant data only), or more than three sampling runs were assigned the same initial number followed by an extension (for example, 1.1 or 1.1.1). Some of the test reports in the database include an "x" symbol at the end of the test report number (e.g., test report 8x). The "x" symbol indicates that the test report does not meet the acceptance criteria developed by the CTWG. The data from these test reports are included in the database for informational purposes only. B-l ------- Construction of database reports (i.e., summaries of relevant data) required the complete separation of tests with HAPs-only data from tests with only criteria pollutant data and tests with both HAPs and criteria pollutant data. The "Test Data" table was consequently divided into three tables: "Test Data - HAPs," containing all HAP data in the Test Data table; "Test Data - Criteria Pollutants," containing all criteria pollutant data in the Test Data table, and "Test Data - HAPs + Criteria," containing the tests that include data for both HAPs and criteria pollutants. In the report section, a set of 6 different reports was built for each of the test data tables discussed above. These reports provide information about pollutant concentrations (corrected to 15% 02) and emissions in units of lb/hr, lb/MMBtu, and lb/MW-hr. Individual sets of reports were also developed for test summaries and pollutant summaries. Treatment of non-detected or non-reported concentrations Many pollutants, especially HAPs, were not detected in some or all of the sampling runs collected during a test. In these cases, concentrations were entered in the database as "ND." Although the test reports identified those pollutants not detected for a given testing run, the detection limit (DL) values were not always provided (i.e., ND was reported rather than a detection limit concentration). Often, review of the lab report and some additional calculations were necessary to determine the DL concentration. For example, in the case of formaldehyde, detection limits were usually given in micrograms or micrograms per milliliter in the lab report. Estimation of the DL in the same units as the test data (e.g., ppb) involved the use of the sample volume collected during the test and additional unit conversions (for example, micrograms/cubic meter to ppb). Unfortunately, the DL could not always be found or calculated based on the laboratory report. Whenever a pollutant was not detected in all three runs and the DL could not be determined, the pollutant was removed from the database. This procedure was used for report ID #1 for benzene and chromium (VI). Also, due to the calculations discussed above, two or three different DLs (one per testing run) were determined for the same pollutant in some tests. The protocol followed in these cases was to take the highest DL value. In some tests, only one or two runs were conducted, or runs were eliminated during test report preparation due to sampling problems encountered during the test. Missing runs were entered as NR (not reported) in the database. Other parameters missing from the test reports, such as exhaust gas flow rates, were also entered in the database as NR. The acronym NA sometimes appears in the DL field. This acronym is used in those cases when a pollutant was measured above the detection limit in all of the testing runs but a detection limit value was not reported in the test report. Equations Using raw test data (i.e., lab-reported pollutant concentrations and stack test parameters), calculations were performed to estimate emissions in lb/hr, lb/MW-hr and lb/MMBtu. Modules, small programs written in Visual Basic code, were built to perform the calculations. There are various modules in the emissions database that perform different tasks, but only the main modules are described in this memorandum. The equations used in the modules were taken from EPA sampling methods 19 and 20 in 40 CFR Part 60, „ 20.9-15 Cad] Cd 20.9-%O2 Appendix A. For example, for the correction of the dry pollutant concentration to 15% 02, Equation 20-4 from EPA method 20 is used: where %02 refers to the reported oxygen level during the testing and Cd to the pollutant dry concentration in ppb. B-2 ------- For the calculation of emission rates in lb/hr, lb/MW-hr, and lb/MMBtu, the following equations were used : 1. Pounds per hour: When the concentration of pollutant is given in ppb : MW M(lb/hr) = Cvvb*Q* 60* — * 1.369x io" T std + 460 where Cppb is the dry concentration of pollutant in ppb; Q is the exhaust gas flow rate in dry standard cubic feet per minute; 60 is the conversion factor from minutes to hours; MW is the pollutant molecular weight (in lb/lb-mol); Tstd is the standard temperature in degrees Fahrenheit used in the test report; 460 is the conversion factor from degrees Fahrenheit to degrees Rankine; and 1.369xl0"9 is the conversion factor from ppb to pounds per cubic feet. The conversion factor from ppb to pounds per cubic feet was derived from 40 CFR, App. A, Meth. 20, page 1026. When the concentration of a pollutant is given in units other than ppb or ppm, the equation is : M(lb/hr) = CP*Q*60* A where Cp is the concentration of pollutant in micrograms per dry cubic feet (ug/dscf), micrograms per dry cubic meter (ug/dscm), grams per dry cubic feet (g/dscf) or grams per dry cubic meter (g/dscm). For particulate matter, concentrations are in grains per dry cubic feet (gr/dscf), grains per dry cubic meter (gr/dscm), micrograms per dry cubic feet (ugr/dscf) and micrograms per dry cubic meter (ugr/dscm). Q is the exhaust gas flow rate in dry standard cubic feet per minute; 60 is the conversion factor from minutes to hours; and A is a conversion factor from the given units to lb/dscf. The values for A for the different units are: 1.1 For ug/dscf, A = 2.205xl0-8 1.2 For ug/dscm, A = 6.24xlO"10 1.3 For g/dscf and g/dscm, multiplying 1.1 and 1.2 by lxlO"6 1.4 For ugr/dscf, A= 1.43xlO"10. 1.5 For ugr/dscm, A = 4.043xl0"12. 1.6 For gr/dscf and gr/dscm, multiplying 1.4 and 1.5 by lxlO"6 2. Pounds per megawatt-hour: The emission factor is calculated by dividing the emissions rate in lb/hr by the turbine rating during the test. The manufacturer rating and the test load are necessary data for this calculation. When load was not available, it was assumed to be 100%. The equation is : B-3 ------- Mflb/MW-hr)-^^ R*L 100 where M(lb/hr) is the emission rate in lb/hr; R is the manufacturer rating for the turbine in MW; and L is the turbine testing load in %. The equation is : 20 9 MW M(lbfMMBtu) = CV* F* — —) 20.9-%O2 T std + 460 3. Pounds per million Btu: where Cp is the dry concentration of pollutant in any of the units already described for the calculation of emission factors (1.1 - 1.6); F is the fuel factor in dry standard cubic feet per minute per million Btu; the fraction 20.9/(20.9-%02) is an oxygen correction factor; and B is the conversion factor corresponding to the units in which the pollutant concentration is reported (see the units described in 1.1 - 1.6). The fraction MW/(Tstd+460) is a conversion factor used only when the pollutant concentration was provided in ppb. When the fuel factor or standard temperature was not available, defaults were used. These defaults are discussed in next section. A sample of the modules used for the calculations is provided in Appendix C-l. Defaults and Assumptions For the estimation of emission factors from the concentrations given in ppb, gaseous pollutants were assumed to have ideal gas behavior, so that the volume occupied by an ideal gas (22.4 liters/mol) could be used for calculation of a conversion factor. Not all of the reports contained the necessary information required for the calculation of emission factors. Important parameters are concentrations, units, detection limits, oxygen levels, exhaust gas flow rates, fuel factors, standard temperatures and molecular weights. In most cases, fuel factors and standard temperatures were missing. In some cases, exhaust gas flow rates were not provided in the report. Lack of gas flow rates still allows for the calculation of emission factors in pounds per million Btu. Consequently, tests lacking exhaust gas flow rates were kept in the database, but the emissions in pound per hour are shown as NR. For non-methane hydrocarbons (NMHC) and total hydrocarbons (THC), a molecular weight of 16 (as methane) was assumed. Test reports in the database indicated a molecular weight of 16 for THC and, in most cases, for NMHC. However, in some test reports, the molecular weight chosen to report emission factors for NMHC was the molecular weight of hexane. Fields with NR for fuel factors and standard temperatures were filled with default values based on Table 19-1 in 40 CFR Part 60, Appendix A. A default standard temperature of 68°F was used. This standard temperature was selected because EPA sampling methods rely on this value. As discussed earlier, some pollutants were not detected in one or more of the sampling runs conducted during a test. In these cases, the detection limit was used in the emission calculations. Reports generated in the emissions B-4 ------- database use a "<" sign in front of the sampling ran concentration, as well as the average concentration calculated for the three runs, to indicate when a pollutant was not detected in one or more of the runs. When a pollutant was not detected in all three runs, a"«" sign is shown in front of the average concentration presented in the database reports. The DL value was used in calculating the average concentration when a pollutant was not detected in one or more of the runs. B-5 ------- Appendix C-l Sample of modules used in the database The modules shown here are the modules for the calculation of emission factors in pounds per million Btu (Module Convert) and the module that handles the criteria for the use of detection limits (Module NonDetect). 1. Module for the calculation of emission factors in pounds per million Btu 1.1 Declaring the function that will perform the calculations and return the result to the query. The parameters r, s, t, u, v, w, z refer to concentration units (r), fuel factor (s), molecular weight (t), standard temperature (u), % oxygen (v), concentration (w), and a parameter (z, set to three in the database) used to limit the number of significant digits (utilizing another module) in the result. Function IbMMBtu (r, s, t, u, v, w, x, y, z) 1.2 Estimating the emission factor to return to the query that is calling this module. First the module identifies the units (r=ppb), then it makes sure that there are values in all necessary fields and finally performs the calculation. SigDig_ is calling another module that will perform the reduction of the result to a given number (z) of significant digits. Val calls for the numerical value of the field being processed. If(& = "ppb") And Not (s = "NR" Or t = "NR" Orv = "NR" Orw = "NR")) Then IbMMBtu = CStr(SigDig_((Val(s) * Val(t) * (00000000137/ (Val(u) + 460)) * (20.9/(20.9 - Val(v))) * Val(w)), z)) Elself ((r = "ug/dscm ") And Not (s = "NR" Orv= "NR" Orw = "NR")) Then IbMMBtu = CStr(SigDig_((Val(s) * Val(w) * .0283 * .000000002204 * (20.9/(20.9 - Val(v)))), z)) Elself ((r = "ug/dscf) And Not (s = "NR" Orv= "NR" Orw = "NR")) Then IbMMBtu = CStr(SigDig_((Val(s) * Val(w) * .000000002204 * (20.9/(20.9 - Val(v)))), z)) Elself ((r = "gr/dscf) And Not (s = "NR" Orv= "NR" Orw = "NR")) Then IbMMBtu = CStr(SigDig_((Val(s) * Val(w) * (20.9/(20.9 - Val(v¦))) / 7000), z)) Elself ((r = "ugr/dscm ") And Not (s = "NR" Orv= "NR" Orw = "NR")) Then IbMMBtu = CStr(SigDig_((Val(s) * Val(w) * .0283 * (20.9/(20.9 - Val(v))) * 0.000001/7000), z)) B-6 ------- Elself ((r = "gr/dscm") And Not (s = "NR" Orv= "NR" Orw = "NR")) Then IbMMBtu = CStr(SigDig_((Val(s) * Valfw) * .0283 * (20.9/(20.9 - Val(v))) / 7000), z)) 1.3 In any other case (units not recognized or necessary parameters were not reported) the function is returned with the value "NR" Else IbMMBtu = "NR" End If End Function 2. Module Handling the use of non-detected values 2.1 Declaring the function that will return the values to the query. The parameters x and y refer respectively to concentration and detection limit. Function Correction (x, y) 2.2 Identifying the concentration. If it is not reported, return the value "NR;" if it is not detected, take the value of the detection limit as the value for the concentration to be returned. Otherwise leave the value as it is. If(x = "NR") Then Correction = "NR" Elself If (x= "ND") Then Correction =y Else Correction = x End If End Function B-7 ------- Appendix C ~ QA\QC Review Criteria for Emissions Tests HAPS and Criteria Pollutant Source Test Checklist Source Test Source Test Report # Report # Date Date BASIC TURBINE INFORMATION Manufacturer Model # Rating (BHP or MW) Operating Cycle (Simple, Regenerative, etc.) FUEL DESCRIPTION Fuel Name(s) Fuel Analysis Summary Flowrate (or BTU/H, if available) OPERATING CONDITIONS Load (during test) Water or Steam Injection and/or Ammonia Mass Flowrate Firing Temperature or Turbine Inlet Temperature AMBIENT CONDITIONS Temperature Relative Humidity Barometric Pressure Altitude EXHAUST INFORMATION Temperature Flowrate (F-Factor or Measured) EMISSIONS TEST * Criteria Pollutants HAPS Oxygen or C02 Moisture Averaging Time METHODS USED CARB EPA Other QUALITY CONTROL DOCUMENTATION Calibration of Instruments Specialty Gases CEMs Dry Gas Meters MISCELLANEOUS Limits of Detection Reporting Supplemental Firing Details C-l ------- Appendix D Development of Emission Factors (lb/MMBtu) for Natural Gas Fired Turbines The emission factors (lb/MMBtu) presented in Table 1 were calculated for natural gas- fired turbines from 23 source test reports in the emissions database. Emission factors from test reports that did not meet acceptance criteria established by the CTWG were not used in the calculations (4.1.2x, 8x, lOx, 29.1, 29.2, and 29.3). In addition, only test reports where the testing was conducted at high loads (greater than 80%) were included in the analysis. Test reports in which the load was not specified in the test report or could not be estimated from fuel use data were excluded. The following test reports were used for the emission factor calculations: 2, 3.1, 4.2, 6.2, 7, 9, 11, 12.1, 13.1, 15.1, 17, 18, 22, 26, 27, 28, 313.1.1x, 313.2.1x, 314.1x, 315.lx, 316.1.1x, 316.2.1x, and 317. lx. Listed below are the source test reports that were excluded from the emission factor calculation with the reason for exclusion. Test Report ID# Reason for Exclusion 4.1.2x Formaldehyde data point appears to be an outlier. Retest of the same turbine generated formaldehyde data more consistent with other formaldehyde data in the database. 8x Report deemed inadequate by state and federal regulators according to telephone contact with the turbine operator. lOx Missing load and fuel usage data. 29.1, 29.2, 29.3 Only summary data provided; no raw data sheets, laboratory results, etc. 16, 21, 313.1.2x, 313.2.2x, 314.2x, 314.3x, 314.4x, 315.2x, 316.1.2x, 316.2.2x, 317.2x Testing occurred only at operating loads less than 80%. 23, 25 Load information not available. Test data for individual HAPs that were not detected in any of the sampling runs for a source test (i.e., where the concentration was ND in all three runs) were excluded from the emission factor calculation for that HAP. This exclusion was made on a pollutant basis such that data for a subset of the HAPs analyzed for in a particular source test may have been used. D-l ------- Appendix E — Cost Spreadsheets INPUTS AND CALCULATIONS Model Turbine Number Turbine Exhaust Flow (lb/sec) Turbine Rating (MW) Turbine Rating (hp) Heat Input, MMBtu/hr, including efficiency Hours of operation/yr Life of equipment Life of catalyst Interest rate (fraction) Capital Recovery Factor, Equipment, 15-yr Life Capital Recovery Factor, 3-yr Catalyst Life Capital Recovery Factor, 6-yr Catalyst Life Destruction Efficiency - 3-yr & 6- yr Catalyst Life Destruction Efficiency - 6-yr Catalyst Life w/Degradation VAPCCI Escalator Fuel Type (CLEAN OR DIRTY) Turbine Assumed Efficiency (fraction) Turbine Exhaust Temp (OF) 1 658 85.4 114523.1 832.5656 (Rating in MW/ .29307 MW/MMBTU/hr)/ Efficiency 8000 15 3 or 6 Years 0.07 0.109795 0.381052 0.209796 80 for emission reduction calculation 50 for emission reduction calculation CLEAN 0.35 for emission reduction calculation 1000 Catalyst Calculations: Vendor Estimate - Based on 80 Percent Reduction of Formaldehyde Catalyst, Frame & 1595574 Per Catalyst Vendors, assume housing is 30 percent of Total Catalyst Costs Housing Catalyst only 1116874 EPA formula based on Vendor Quotes Ductwork (No quantitative estimates available) E-l ------- COSTS (Patterned after the OAQPS Cost Manual (1S90) Thermal and Catalytic Incinerators Section) Direct Costs Purchased Equipment Costs (PEC) Catalyst + auxiliary equipment* (EC) Instrumentation** Sales Tax Freight 1 EC 0.1 EC 0.03 EC 0.05 EC 3-Year Costs 1595574 159557.4 47867.23 79778.72 6-Year Costs 1595574 159557.4 47867.23 79778.72 Total Purchased Equipment Cost, PEC 1.18 EC 1882778 1882778 Direct Installation Costs Foundations & supports 0.08 PEC 150622.2 150622.2 Handling & erection 0.14 PEC 263588.9 263588.9 Electrical 0.04 PEC 75311.11 75311.11 Piping 0.02 PEC 37655.56 37655.56 Insulation for ductwork 0.01 PEC 18827.78 18827.78 Painting 0.01 PEC 18827.78 18827.78 Direct Installation Cost 0.3 PEC 564833.3 564833.3 Site preparation Buildings Total Direct Cost, DC As required, SP As required, Bldg. 0 0 0 0 1.30 PEC + SP + Bldg. 2447611 2447611 Indirect Costs (installation) Engineering 0.1 PEC 188277.8 188277.8 Construction and Field Expenses 0.05 PEC 94138.89 94138.89 Contractor Fees 0.1 PEC 188277.8 188277.8 Start-up 0.02 PEC 37655.56 37655.56 Performance test 0.01 PEC 18827.78 18827.78 Total Indirect Cost, IC 0.28 PEC 527177.8 527177.8 Contingencies 0.1 DC+IC 297478.9 297478.9 Total Capital Cost (TCC) = DC + IC + ^ .61 PEC + SP + 3272268 3272268 Contingencies Bldg.+0.1 (DC+IC) E-2 ------- Direct Annual Cost (DAC) Fuel Fuel Penalty due to Penalty Pressure Drop Assume 1" backpressure Perf. Test Performance Test Not speciated HAPs Cat. Costs Freight to return catalyst for disposal Catalyst replacement Freight=.05*Catalyst only cost*[i/[(1+i)An i=interest rate, n=catalyst lifetime Catalyst only cost * CRFcat Operating Labor Operator 2 hours per day Supervisor ,15*OL Maintenance Labor & .10 PEC Materials Per Engine ACT-NSCR Per Engine ACT- NSCR 0.15 OL 0.1 PEC Total Direct Annual Cost (DAC) Indirect Annual Cost (IAC) Overhead 0.6 O&M costs Administrative 0.02 TCC Property Taxes 0.01 TCC Insurance 0.01 TCC Capital for catalyst: CRFequip(TCC -1,08(Cat only)) Recovery Total Indirect Annual Cost (IAC) Total Annual Cost (TAC) E-3 17320 17320 5000 5000 17370.28 7806.717 425586.9 234315.6 18250 18250 2737.5 2737.5 188277.8 188277.8 674542.4 473707.6 125559.2 125559.2 65445.36 65445.36 32722.68 32722.68 32722.68 32722.68 226840.5 226840.5 483290.3 483290.3 1157833 956997.9 ------- INPUTS AND CALCU .ATIONS Model Turbine Number Turbine Exhaust Flow (lb/sec) Turbine Rating (MW) Turbine Rating (hp) Heat Input, MMBtu/hr, including efficiency Hours of operation/yr Life of equipment Life of catalyst Interest rate (fraction) Capital Recovery Factor, Equipment, 15-yr Life Capital Recovery Factor, 3-yr Catalyst Life Capital Recovery Factor, 6-yr Catalyst Life Destruction Efficiency - 3-yr & 6- yr Catalyst Life Destruction Efficiency - 6-yr Catalyst Life w/Degradation VAPCCI Escalator Fuel Type (CLEAN OR DIRTY) Turbine Assumed Efficiency (fraction) Turbine Exhaust Temp (OF) 2 986 170 227973.4 1657.332 (Rating in MW/ .29307 MW/MMBTU/hr)/ Efficiency 8000 15 3 or 6 Years 0.07 0.109795 0.381052 0.209796 80 for emission reduction calculation 50 for emission reduction calculation CLEAN 0.35 for emission reduction calculation 1000 Catalyst Calculations: Vendor Estimate - Based on 80 Percent Reduction of Formaldehyde Catalyst, Frame & 2317985 Per Catalyst Vendors, assume housing is 30 percent of Total Catalyst Costs Housing Catalyst only 1622585 EPA formula based on Vendor Quotes Ductwork (No quantitative estimates available) E-4 ------- COSTS (Patterned after the OAQPS Cost Manual (1S90) Thermal and Catalytic Incinerators Section) Direct Costs Purchased Equipment Costs (PEC) Catalyst + auxiliary equipment* (EC) Instrumentation** Sales Tax Freight 1 EC 0.1 EC 0.03 EC 0.05 EC 3-Year Costs 2317985 231798.5 69539.54 115899.2 6-Year Costs 2317985 231798.5 69539.54 115899.2 Total Purchased Equipment Cost, PEC 1.18 EC 2735222 2735222 Direct Installation Costs Foundations & supports 0.08 PEC 218817.8 218817.8 Handling & erection 0.14 PEC 382931.1 382931.1 Electrical 0.04 PEC 109408.9 109408.9 Piping 0.02 PEC 54704.44 54704.44 Insulation for ductwork 0.01 PEC 27352.22 27352.22 Painting 0.01 PEC 27352.22 27352.22 Direct Installation Cost 0.3 PEC 820566.6 820566.6 Site preparation Buildings Total Direct Cost, DC As required, SP As required, Bldg. 0 0 0 0 1.30 PEC + SP + Bldg. 3555789 3555789 Indirect Costs (installation) Engineering 0.1 PEC 273522.2 273522.2 Construction and Field Expenses 0.05 PEC 136761.1 136761.1 Contractor Fees 0.1 PEC 273522.2 273522.2 Start-up 0.02 PEC 54704.44 54704.44 Performance test 0.01 PEC 27352.22 27352.22 Total Indirect Cost, IC 0.28 PEC 765862.2 765862.2 Contingencies 0.1 DC+IC 432165.1 432165.1 Total Capital Cost (TCC) = DC + IC + ^ .61 PEC + SP + 4753816 4753816 Contingencies Bldg.+0.1 (DC+IC) E-5 ------- Direct Annual Cost (DAC) Fuel Fuel Penalty due to Penalty Pressure Drop Assume 1" backpressure Perf. Test Performance Test Not speciated HAPs Cat. Costs Freight to return catalyst for disposal Catalyst replacement Freight=.05*Catalyst only cost*[i/[(1+i)An i=interest rate, n=catalyst lifetime Catalyst only cost * CRFcat Operating Labor Operator 2 hours per day .15 *OL Superviso r Maintenance Labor & .10 PEC Materials Per Engine ACT-NSCR Per Engine ACT- NSCR 0.15 OL 0.1 PEC Total Direct Annual Cost (DAC) Indirect Annual Cost (IAC) Overhead 0.6 O&M costs Administrative 0.02 TCC Property Taxes 0.01 TCC Insurance 0.01 TCC Capital for catalyst: CRFequip(TCC -1,08(Cat only)) Recovery Total Indirect Annual Cost (IAC) Total Annual Cost (TAC) E-6 34470 34470 5000 5000 25235.39 11341.53 618288.6 340411.5 18250 18250 2737.5 2737.5 273522.2 273522.2 977503.7 685732.7 176705.8 176705.8 95076.32 95076.32 47538.16 47538.16 47538.16 47538.16 329540.3 329540.3 696398.7 696398.7 1673902 1382131 ------- INPUTS AND CALCULATIONS Model Turbine Number Turbine Exhaust Flow (lb/sec) Turbine Rating (MW) Turbine Rating (hp) Heat Input, MMBtu/hr, including efficiency Hours of operation/yr Life of equipment Life of catalyst Interest rate (fraction) Capital Recovery Factor, Equipment, 15-yr Life Capital Recovery Factor, 3-yr Catalyst Life Capital Recovery Factor, 6-yr Catalyst Life Destruction Efficiency - 3-yr & 6- yr Catalyst Life Destruction Efficiency - 6-yr Catalyst Life w/Degradation VAPCCI Escalator Fuel Type (CLEAN OR DIRTY) Turbine Assumed Efficiency (fraction) Turbine Exhaust Temp (OF) 7 318 39.6 53104.39 386.0609 (Rating in MW/ .29307 MW/MMBTU/hr)/ Efficiency 8000 15 3 or 6 Years 0.07 0.109795 0.381052 0.209796 80 for emission reduction calculation 50 for emission reduction calculation CLEAN 0.35 for emission reduction calculation 1000 Catalyst Calculations: Vendor Estimate - Based on 80 Percent Reduction of Formaldehyde Catalyst, Frame & 846662.4 Per Catalyst Vendors, assume housing is 30 percent of Total Catalyst Costs Housing Catalyst only 592662.4 EPA formula based on Vendor Quotes Ductwork (No quantitative estimates available) E-7 ------- COSTS (Patterned after the OAQPS Cost Manual (1S90) Thermal and Catalytic Incinerators Section) Direct Costs Purchased Equipment Costs (PEC) Catalyst + auxiliary equipment* (EC) Instrumentation** Sales Tax Freight 1 EC 0.1 EC 0.03 EC 0.05 EC 3-Year Costs 846662.4 84666.24 25399.87 42333.12 6-Year Costs 846662.4 84666.24 25399.87 42333.12 Total Purchased Equipment Cost, PEC 1.18 EC 999061.6 999061.6 Direct Installation Costs Foundations & supports 0.08 PEC 79924.93 79924.93 Handling & erection 0.14 PEC 139868.6 139868.6 Electrical 0.04 PEC 39962.47 39962.47 Piping 0.02 PEC 19981.23 19981.23 Insulation for ductwork 0.01 PEC 9990.616 9990.616 Painting 0.01 PEC 9990.616 9990.616 Direct Installation Cost 0.3 PEC 299718.5 299718.5 Site preparation Buildings Total Direct Cost, DC As required, SP As required, Bldg. 0 0 0 0 1.30 PEC + SP + Bldg. 1298780 1298780 Indirect Costs (installation) Engineering Construction and Field Expenses Contractor Fees Start-up Performance test Total Indirect Cost, IC Contingencies Total Capital Cost (TCC) = DC + I Contingencies 0.1 PEC 99906.16 99906.16 0.05 PEC 49953.08 49953.08 0.1 PEC 99906.16 99906.16 0.02 PEC 19981.23 19981.23 0.01 PEC 9990.616 9990.616 0.28 PEC 279737.3 279737.3 0.1 DC+IC 157851.7 157851.7 + 1.61 PEC+ SP+ 1736369 1736369 Bldg.+0.1 (DC+IC) E-8 ------- Direct Annual Cost (DAC) Fuel Fuel Penalty due to Penalty Pressure Drop Assume 1" backpressure Perf. Test Performance Test Not speciated HAPs Cat. Costs Freight to return catalyst for disposal Catalyst replacement Freight=.05*Catalyst only cost*[i/[(1+i)An-1], i=interest rate, n=catalyst lifetime Catalyst only cost * CRFcat Operating Labor Operator 2 hours per day .15 *OL Superviso r Maintenance Labor & .10 PEC Materials Per Engine ACT-NSCR Per Engine ACT- NSCR 0.15 OL 0.1 PEC Total Direct Annual Cost (DAC) Indirect Annual Cost (IAC) Overhead 0.6 O&M costs Administrative 0.02 TCC Property Taxes 0.01 TCC Insurance 0.01 TCC Capital for catalyst: CRFequip(TCC -1,08(Cat only)) Recovery Total Indirect Annual Cost (IAC) Total Annual Cost (TAC) E-9 8030 8030 5000 5000 9217.431 4142.586 225835 124338.1 18250 18250 2737.5 2737.5 99906.16 99906.16 368976.1 262404.3 72536.2 72536.2 34727.38 34727.38 17363.69 17363.69 17363.69 17363.69 120367.2 120367.2 262358.1 262358.1 631334.2 524762.5 ------- INPUTS AND CALCU .ATIONS Model Turbine Number Turbine Exhaust Flow (lb/sec) Turbine Rating (MW) Turbine Rating (hp) Heat Input, MMBtu/hr, including efficiency Hours of operation/yr Life of equipment Life of catalyst Interest rate (fraction) Capital Recovery Factor, Equipment, 15-yr Life Capital Recovery Factor, 3-yr Catalyst Life Capital Recovery Factor, 6-yr Catalyst Life Destruction Efficiency - 3-yr & 6- yr Catalyst Life Destruction Efficiency - 6-yr Catalyst Life w/Degradation VAPCCI Escalator Fuel Type (CLEAN OR DIRTY) Turbine Assumed Efficiency (fraction) Turbine Exhaust Temp (OF) 9 178 27 36207.54 263.2233 (Rating in MW/ .29307 MW/MMBTU/hr)/ Efficiency 8000 15 3 or 6 Years 0.07 0.109795 0.381052 0.209796 80 for emission reduction calculation 50 for emission reduction calculation CLEAN 0.35 for emission reduction calculation 1000 Catalyst Calculations: Vendor Estimate - Based on 80 Percent Reduction of Formaldehyde Catalyst, Frame & 538310.4 Per Catalyst Vendors, assume housing is 30 percent of Total Catalyst Costs Housing Catalyst only 376810.4 EPA formula based on Vendor Quotes Ductwork (No quantitative estimates available) E-10 ------- COSTS (Patterned after the OAQPS Cost Manual (1S90) Thermal and Catalytic Incinerators Section) Cost Item Direct Costs Purchased Equipment Costs (PEC) Catalyst + auxiliary equipment* (EC) Instrumentation** Sales Tax Freight 1 EC 0.1 EC 0.03 EC 0.05 EC 3-Year Costs 538310.4 53831.04 16149.31 26915.52 6-Year Costs 538310.4 53831.04 16149.31 26915.52 Total Purchased Equipment Cost, PEC 1.18 EC 635206.3 635206.3 Direct Installation Costs Foundations & supports 0.08 PEC 50816.5 50816.5 Handling & erection 0.14 PEC 88928.88 88928.88 Electrical 0.04 PEC 25408.25 25408.25 Piping 0.02 PEC 12704.13 12704.13 Insulation for ductwork 0.01 PEC 6352.063 6352.063 Painting 0.01 PEC 6352.063 6352.063 Direct Installation Cost 0.3 PEC 190561.9 190561.9 Site preparation Buildings Total Direct Cost, DC As required, SP As required, Bldg. 0 0 0 0 1.30 PEC + SP + Bldg. 825768.2 825768.2 Indirect Costs (installation) Engineering Construction and Field Expenses Contractor Fees Start-up Performance test Total Indirect Cost, IC Contingencies Total Capital Cost (TCC) = DC + I Contingencies 0.1 PEC 63520.63 63520.63 0.05 PEC 31760.31 31760.31 0.1 PEC 63520.63 63520.63 0.02 PEC 12704.13 12704.13 0.01 PEC 6352.063 6352.063 0.28 PEC 177857.8 177857.8 0.1 DC+IC 100362.6 100362.6 + 1.61 PEC+ SP+ 1103989 1103989 Bldg.+0.1 (DC+IC) E-ll ------- Direct Annual Cost (DAC) Fuel Fuel Penalty due to Penalty Pressure Drop Assume 1" backpressure Perf. Test Performance Test Not speciated HAPs Cat. Costs Freight to return catalyst for disposal Catalyst replacement Freight=.05*Catalyst only cost*[i/[(1+i)An i=interest rate, n=catalyst lifetime Catalyst only cost * CRFcat Operating Labor Operator 2 hours per day .15 *OL Superviso r Maintenance Labor & .10 PEC Materials Per Engine ACT-NSCR Per Engine ACT- NSCR 0.15 OL 0.1 PEC Total Direct Annual Cost (DAC) Indirect Annual Cost (IAC) Overhead 0.6 O&M costs Administrative 0.02 TCC Property Taxes 0.01 TCC Insurance 0.01 TCC Capital for catalyst: CRFequip(TCC -1,08(Cat only)) Recovery Total Indirect Annual Cost (IAC) Total Annual Cost (TAC) E-12 5470 5470 5000 5000 5860.375 2633.826 143584.2 79053.24 18250 18250 2737.5 2737.5 63520.63 63520.63 244422.7 176665.2 50704.88 50704.88 22079.77 22079.77 11039.89 11039.89 11039.89 11039.89 76530.51 76530.51 171394.9 171394.9 415817.7 348060.1 ------- INPUTS AND CALCU .ATIONS Model Turbine Number Turbine Exhaust Flow (lb/sec) Turbine Rating (MW) Turbine Rating (hp) Heat Input, MMBtu/hr, including efficiency Hours of operation/yr Life of equipment Life of catalyst Interest rate (fraction) Capital Recovery Factor, Equipment, 15-yr Life Capital Recovery Factor, 3-yr Catalyst Life Capital Recovery Factor, 6-yr Catalyst Life Destruction Efficiency - 3-yr & 6- yr Catalyst Life Destruction Efficiency - 6-yr Catalyst Life w/Degradation VAPCCI Escalator Fuel Type (CLEAN OR DIRTY) Turbine Assumed Efficiency (fraction) Turbine Exhaust Temp (OF) 15 83.6 9 12069.18 87.74111 (Rating in MW/ .29307 MW/MMBTU/hr)/ Efficiency 8000 15 3 or 6 Years 0.07 0.109795 0.381052 0.209796 80 for emission reduction calculation 50 for emission reduction calculation CLEAN 0.35 for emission reduction calculation 1000 Catalyst Calculations: Vendor Estimate - Based on 80 Percent Reduction of Formaldehyde Catalyst, Frame & 330364.5 Per Catalyst Vendors, assume housing is 30 percent of Total Catalyst Costs Housing Catalyst only 231264.5 EPA formula based on Vendor Quotes Ductwork (No quantitative estimates available) E-13 ------- COSTS (Patterned after the OAQPS Cost Manual (1S90) Thermal and Catalytic Incinerators Section) Direct Costs Purchased Equipment Costs (PEC) Catalyst + auxiliary equipment* (EC) Instrumentation** Sales Tax Freight 1 EC 0.1 EC 0.03 EC 0.05 EC 3-Year Costs $236,584 $23,658 $7,098 $11,829 6-Year Costs $236,584 $23,658 $7,098 $11,829 Total Purchased Equipment 1.18 EC $279,169 $279,169 Cost, PEC Direct Installation Costs Foundations & supports 0.08 PEC $22,334 $22,334 Handling & erection 0.14 PEC $39,084 $39,084 Electrical 0.04 PEC $11,167 $11,167 Piping 0.02 PEC $5,583 $5,583 Insulation for ductwork 0.01 PEC $2,792 $2,792 Painting 0.01 PEC $2,792 $2,792 Direct Installation Cost 0.3 PEC $83,751 $83,751 Site preparation As required, SP $0 $0 Buildings As required, Bldg. $0 $0 Total Direct Cost, DC 1.30 PEC + SP + $362,920 $362,920 Bldg. Indirect Costs (installation) Engineering 0.1 PEC $27,917 $27,917 Construction and Field Expenses 0.05 PEC $13,958 $13,958 Contractor Fees 0.1 PEC $27,917 $27,917 Start-up 0.02 PEC $5,583 $5,583 Performance test 0.01 PEC $2,792 $2,792 Total Indirect Cost, IC 0.28 PEC $78,167 $78,167 Contingencies 0.1 DC+IC $44,109 $44,109 Total Capital Cost (TCC) = DC + IC + 1.61 PEC + SP + $485,196 $485,196 Contingencies Bldg.+0.1 (DC+IC) E-14 ------- Direct Annual Cost (DAC1 Direct Costs Purchased Equipment Costs (PEC) Catalyst + auxiliary equipment* (EC) 1 EC Instrumentation** 0.1 EC Sales Tax 0.03 EC Freight 0.05 EC Total Purchased Equipment 1.18 EC Cost, PEC Direct Installation Costs Foundations & supports 0.08 PEC Handling & erection 0.14 PEC Electrical 0.04 PEC Piping 0.02 PEC Insulation for ductwork 0.01 PEC Painting 0.01 PEC Direct Installation Cost 0.3 PEC Site preparation Buildings As required, SP As required, Bldg. Total Direct Cost, DC Indirect Costs (installation) Engineering Construction and Field Expenses Contractor Fees Start-up Performance test Total Indirect Cost, IC Contingencies 1.30 PEC + SP + Bldg. 0.1 PEC 0.05 PEC 0.1 PEC 0.02 PEC 0.01 PEC 0.28 PEC 0.1 DC+IC Total Capital Cost (TCC) = DC + IC + Contingencies 1.61 PEC + SP + Bldg.+0.1 (DC+IC) E-15 3-Year 6-Year Costs Costs 330364.5 330364.5 33036.45 33036.45 9910.934 9910.934 16518.22 16518.22 389830.1 389830.1 31186.41 31186.41 54576.21 54576.21 15593.2 15593.2 7796.602 7796.602 3898.301 3898.301 3898.301 3898.301 116949 116949 0 0 0 0 506779.1 506779.1 38983.01 38983.01 19491.5 19491.5 38983.01 38983.01 7796.602 7796.602 3898.301 3898.301 109152.4 109152.4 61593.15 61593.15 677524.7 677524.7 ------- Direct Annual Cost (DAC) Fuel Fuel Assume Penalty Penalty 1" due to backpress Pressure ure Drop Perf. Test Performan Not speciated HAPs ce Test 1 Cat. Costs Freight to return catalyst for disposal Catalyst replacement Freight=.05*Catalyst only cost*[i/[(1+i)An- i=interest rate, n=catalyst lifetime Catalyst only cost * CRFcat Operating Labor Operator 2 hours per day .15 *OL Superviso r Maintenance Labor & .10 PEC Materials Per Engine ACT-NSCR Per Engine ACT- NSCR 0.15 OL 0.1 PEC Total Direct Annual Cost (DAC) Indirect Annual Cost (IAC) Overhead 0.6 O&M costs Administrative 0.02 TCC Property Taxes 0.01 TCC Insurance 0.01 TCC Capital for catalyst: CRFequip(TCC -1,08(Cat only)) Recovery Total Indirect Annual Cost (IAC) Total Annual Cost (TAC) E-16 1820 1820 5000 5000 3596.76 1616.49 88123.72 48518.32 18250 18250 2737.5 2737.5 38983.01 38983.01 158511 116925.3 35982.31 35982.31 13550.49 13550.49 6775.247 6775.247 6775.247 6775.247 46965.64 46965.64 110048.9 110048.9 268559.9 226974.3 ------- INPUTS AND CALCULATIONS Model Turbine Number 13 Turbine Exhaust Flow (lb/sec) 41 Turbine Rating (MW) 3.5 Turbine Rating (hp) 4,694 Heat Input, MMBtu/hr, including 34 (Rating in MW/ .29307 MW/MMBTU/hr)/ efficiency Hours of operation/yr 8000 Life of equipment 15 Life of catalyst 3 or 6 Years Interest rate (fraction) 0.07 Capital Recovery Factor, 0.1098 Equipment, 15-yr Life Capital Recovery Factor, 3-yr 0.3811 Catalyst Life Capital Recovery Factor, 6-yr 0.2098 Catalyst Life Destruction Efficiency - 3-yr & 6- 80 for emission reduction calculation yr Catalyst Life Destruction Efficiency - 6-yr 50 for emission reduction calculation Catalyst Life w/Degradation VAPCCI Escalator Fuel Type (CLEAN OR DIRTY) CLEAN Turbine Assumed Efficiency 0.35 for emission reduction calculation (fraction) Turbine Exhaust Temp (OF) 1000 Catalyst Calculations: Vendor Estimate - Based on 80 Percent Reduction of Formaldehyde Catalyst, Frame & $236,584 Per Catalyst Vendors, assume housing is 30 percent of Total Catalyst Costs Housing Catalyst only $165,584 EPA formula based on Vendor Quotes Other catalyst - associated costs Ductwork (No quantitative estimates available) E-17 ------- COSTS (Patterned after the OAQPS Cost Manual (1S90) Thermal and Catalytic Incinerators Section) Direct Costs Purchased Equipment Costs (PEC) Catalyst + auxiliary equipment* (EC) Instrumentation** Sales Tax Freight 1 EC 0.1 EC 0.03 EC 0.05 EC 3-Year Costs $236,584 $23,658 $7,098 $11,829 6-Year Costs $236,584 $23,658 $7,098 $11,829 Total Purchased Equipment 1.18 EC $279,169 $279,169 Cost, PEC Direct Installation Costs Foundations & supports 0.08 PEC $22,334 $22,334 Handling & erection 0.14 PEC $39,084 $39,084 Electrical 0.04 PEC $11,167 $11,167 Piping 0.02 PEC $5,583 $5,583 Insulation for ductwork 0.01 PEC $2,792 $2,792 Painting 0.01 PEC $2,792 $2,792 Direct Installation Cost 0.3 PEC $83,751 $83,751 Site preparation As required, SP $0 $0 Buildings As required, Bldg. $0 $0 Total Direct Cost, DC 1.30 PEC + SP + $362,920 $362,920 Bldg. Indirect Costs (installation) Engineering 0.1 PEC $27,917 $27,917 Construction and Field Expenses 0.05 PEC $13,958 $13,958 Contractor Fees 0.1 PEC $27,917 $27,917 Start-up 0.02 PEC $5,583 $5,583 Performance test 0.01 PEC $2,792 $2,792 Total Indirect Cost, IC 0.28 PEC $78,167 $78,167 Contingencies 0.1 DC+IC $44,109 $44,109 Total Capital Cost (TCC) = DC + IC + 1.61 PEC + SP + $485,196 $485,196 Contingencies Bldg.+0.1 (DC+IC) E-18 ------- Direct Annual Cost (DAC1 Fuel Fuel Penalty due to 1.0 Penalty Pressure Drop Assume 1" backpressure Perf. Test Performance Test Not speciated HAPs Cat. Costs Freight to return Freight=.05*Catalyst only cost*[i/[(1+i)An- catalyst for disposal i=interest rate, n=catalyst lifetime Catalyst replacement Catalyst only cost * CRFcat Operating Labor Operator 2 hours Per Engine ACT-NSCR per day Supervisor ,15*OL 0.15 OL Maintenance Labor & .10 PEC Per Engine ACT- 0.1 PEC Materials NSCR Total Direct Annual Cost (DAC) Indirect Annual Cost (IAC1 Overhead 0.6 O&M costs Administrative 0.02 TCC Property Taxes 0.01 TCC Insurance 0.01 TCC Capital for catalyst: CRFequip(TCC -1,08(Cat only)) Recovery Total Indirect Annual Cost (IAC) Total Annual Cost (TAC) E-19 $710 $710 $5,000 $5,000 $2,575 $1,157 $63,096 $34,739 $18,250 $18,250 $2,738 $2,738 $27,917 $27,917 $120,286 $90,511 $29,343 $29,343 $9,704 $9,704 $4,852 $4,852 $4,852 $4,852 $33,637 $33,637 $82,388 $82,388 $202,673 $172,898 ------- INPUTS AND CALCULATIONS Model Turbine Number 17 Turbine Exhaust Flow (lb/sec) 14.2 Turbine Rating (MW) 1.13 Turbine Rating (hp) 1,515 Heat Input, MMBtu/hr, including 11 (Rating in MW/ .29307 MW/MMBTU/hr)/ efficiency Hours of operation/yr 8000 Life of equipment 15 Life of catalyst 3 or 6 Years Interest rate (fraction) 0.07 Capital Recovery Factor, 0.1098 Equipment, 15-yr Life Capital Recovery Factor, 3-yr 0.3811 Catalyst Life Capital Recovery Factor, 6-yr 0.2098 Catalyst Life Destruction Efficiency - 3-yr & 6- 80 for emission reduction calculation yr Catalyst Life Destruction Efficiency - 6-yr 50 for emission reduction calculation Catalyst Life w/Degradation VAPCCI Escalator Fuel Type (CLEAN OR DIRTY) CLEAN Turbine Assumed Efficiency 0.35 for emission reduction calculation (fraction) Turbine Exhaust Temp (OF) 1000 Catalyst Calculations: Vendor Estimate - Based on 80 Percent Reduction of Formaldehyde Catalyst, Frame & $177,564 Per Catalyst Vendors, assume housing is 30 percent of Total Catalyst Costs Housing Catalyst only $124,264 EPA formula based on Vendor Quotes Other catalyst - associated costs Ductwork (No quantitative estimates available) E-20 ------- COSTS (Patterned after the OAQPS Cost Manual (1S90) Thermal and Catalytic Incinerators Section) Direct Costs Purchased Equipment Costs (PEC) Catalyst + auxiliary equipment* (EC) Instrumentation** Sales Tax Freight 1 EC 0.1 EC 0.03 EC 0.05 EC 3-Year Costs $177,564 $17,756 $5,327 $8,878 6-Year Costs $177,564 $17,756 $5,327 $8,878 Total Purchased Equipment 1.18 EC $209,525 $209,525 Cost, PEC Direct Installation Costs Foundations & supports 0.08 PEC $16,762 $16,762 Handling & erection 0.14 PEC $29,334 $29,334 Electrical 0.04 PEC $8,381 $8,381 Piping 0.02 PEC $4,191 $4,191 Insulation for ductwork 0.01 PEC $2,095 $2,095 Painting 0.01 PEC $2,095 $2,095 Direct Installation Cost 0.3 PEC $62,858 $62,858 Site preparation As required, SP $0 $0 Buildings As required, Bldg. $0 $0 Total Direct Cost, DC 1.30 PEC + SP + $272,383 $272,383 Bldg. Indirect Costs (installation) Engineering 0.1 PEC $20,953 $20,953 Construction and Field Expenses 0.05 PEC $10,476 $10,476 Contractor Fees 0.1 PEC $20,953 $20,953 Start-up 0.02 PEC $4,191 $4,191 Performance test 0.01 PEC $2,095 $2,095 Total Indirect Cost, IC 0.28 PEC $58,667 $58,667 Contingencies 0.1 DC+IC $33,105 $33,105 Total Capital Cost (TCC) = DC + IC + 1.61 PEC + SP + $364,154 $364,154 Contingencies Bldg.+0.1 (DC+IC) E-21 ------- Direct Annual Cost (DAC1 Fuel Fuel Penalty due to Penalty Pressure Drop - Assume 1" backpressure Perf. Test Performance Test Not speciated HAPs Cat. Costs Freight to return Freight=.05*Catalyst only cost*[i/[(1+i)An- catalyst for disposal i=interest rate, n=catalyst lifetime Catalyst replacement Catalyst only cost * CRFcat Operating Labor Operator 2 hours Per Engine ACT-NSCR per day Supervisor ,15*OL 0.15 OL Maintenance Labor & .10 PEC Per Engine ACT- 0.1 PEC Materials NSCR Total Direct Annual Cost (DAC) Indirect Annual Cost (IAC1 Overhead 0.6 O&M costs Administrative 0.02 TCC Property Taxes 0.01 TCC Insurance 0.01 TCC Capital for catalyst: CRFequip(TCC -1,08(Cat only)) Recovery Total Indirect Annual Cost (IAC) Total Annual Cost (TAC) E-22 $230 $230 $5,000 $5,000 $1,933 $869 $47,351 $26,070 $18,250 $18,250 $2,738 $2,738 $20,953 $20,953 $96,453 $74,109 $25,164 $25,164 $7,283 $7,283 $3,642 $3,642 $3,642 $3,642 $25,247 $25,247 $64,977 $64,977 $161,431 $139,086 ------- Appendix F ~ Description of SCONOx™ System The SCONOx™ catalytic absorption system was described in a paper presented at the Power-Gen International '97 conference as follows: The SCONOx™ system uses a single catalyst for both CO & NOx control. It oxidizes CO to C02 and NO to N02, and the N02 is then absorbed onto the surface of the catalyst. Just as a sponge absorbs water and must be wrung out periodically, the SCONOx™ catalyst must be periodically regenerated. This is accomplished by passing a dilute hydrogen gas across the surface of the catalyst in the absence of oxygen. Nitrogen oxides are broken down into nitrogen and water, and this is exhausted up the stack instead of NOx. Source: "The SCONOx™ Catalytic Absorption system for Natural Gas Fired Power Plants: The Path to Ultra-Low Emissions," Robert J. MacDonald, P.E., and Lawrence Debbage, presented to Power-Gen International '97, December 9-11, 1997. F-l ------- Appendix G ~ Cost-Effectiveness for Individual HAPs Model 1 - 85.4 MW Turbine Pollutant 80% Reduction & 3-Yr Catalyst Life 80% Reduction & 6-Yr Catalyst Life 50% Reduction & 6-Yr Catalyst Life Highest EF Average EF Highest EF Average EF Highest EF Average EF Formaldehyde $85,213 $670,472 $70,432 $554,173 $112,692 $886,677 Toluene $629,008 $3,366,524 $519,902 $2,782,575 $831,843 $4,452,120 Acetaldehyde $1,365,847 $5,241,737 $1,128,930 $4,332,518 $1,806,289 $6,932,029 Xylenes $3,983,720 $10,414,955 $3,292,714 $8,608,402 $5,268,342 $13,773,443 Ethylbenzene $11,659,669 $11,659,669 $9,637,211 $9,637,211 $15,419,538 $15,419,538 Benzene $12,226,251 $46,412,275 $10,105,515 $38,361,714 $16,168,825 $61,378,743 PAHs $65,306,889 $214,370,595 $53,978,915 $177,186,393 $86,366,264 $283,498,228 Acrolein $78,626,057 $87,075,852 $64,987,772 $71,971,886 $103,980,436 $115,155,018 Naphthalene $144,424,903 $327,429,060 $119,373,310 $270,634,011 $190,997,296 $433,014,417 Total HAPs $68,914 $454,166 $56,961 $375,388 $91,137 $600,620 Model 2 — 170 MW Turbine Pollutant 80% Reduction & 3-Yr Catalyst Life Highest EF Average EF 80% Reduction & 6-Yr Catalyst Life Highest EF Average EF 50% Reduction & 6-Yr Catalyst Life Highest EF Average EF Formaldehyde $61,887 $486,938 $51,100 $402,062 $81,760 $643,299 Toluene $456,825 $2,444,978 $377,198 $2,018,804 $603,516 $3,230,087 Acetaldehyde $991,963 $3,806,874 $819,058 $3,143,314 $1,310,492 $5,029,302 Xylenes $2,893,224 $7,563,985 $2,388,918 $6,245,538 $3,822,269 $9,992,861 Ethylbenzene $8,467,973 $8,467,973 $6,991,956 $6,991,956 $11,187,129 $11,187,129 Benzene $8,879,461 $33,707,467 $7,331,718 $27,832,058 $11,730,750 $44,531,292 PAHs $47,429,906 $155,689,197 $39,162,595 $128,551,656 $62,660,151 $205,682,649 Acrolein $57,103,110 $63,239,874 $47,149,703 $52,216,793 $75,439,524 $83,546,868 Naphthalene $104,890,305 $237,799,252 $86,607,309 $196,349,447 $138,571,694 $314,159,115 Total HAPs $50,050 $329,844 $41,326 $272,350 $66,122 $435,760 G-l ------- Model 7 - 39.6 MW Turbine Pollutant 80% Reduction & 3-Yr Catalyst Life Highest EF Average EF 80% Reduction & 6-Yr Catalyst Life Highest EF Average EF 50% Reduction & 6-Yr Catalyst Life Highest EF Average EF Formaldehyde $100,204 $788,418 $83,289 $655,330 $133,262 $1,048,528 Toluene $739,661 $3,958,748 $614,803 $3,290,496 $983,685 $5,264,793 Acetaldehyde $1,606,121 $6,163,841 $1,335,001 $5,123,360 $2,136,002 $8,197,375 Xylenes $4,684,519 $12,247,109 $3,893,753 $10,179,747 $6,230,005 $16,287,596 Ethylbenzene $13,710,787 $13,710,787 $11,396,351 $11,396,351 $18,234,162 $18,234,162 Benzene $14,377,040 $54,576,921 $11,950,138 $45,364,117 $19,120,221 $72,582,586 PAHs $76,795,394 $252,081,741 $63,832,022 $209,529,327 $102,131,235 $335,246,924 Acrolein $92,457,612 $102,393,858 $76,850,395 $85,109,363 $122,960,632 $136,174,980 Naphthalene $169,831,505 $385,028,960 $141,163,263 $320,034,521 $225,861,221 $512,055,233 Total HAPs $81,038 $534,061 $67,358 $443,910 $107,773 $710,255 Model 9 — 27 MW Turbine Pollutant 80% Reduction & 3-Yr Catalyst Life Highest EF Average EF 80% Reduction & 6-Yr Catalyst Life Highest EF Average EF 50% Reduction & 6-Yr Catalyst Life Highest EF Average EF Formaldehyde $96,796 $761,608 $81,023 $637,504 $129,637 $1,020,007 Toluene $714,509 $3,824,133 $598,080 $3,200,990 $956,927 $5,121,583 Acetaldehyde $1,551,505 $5,954,241 $1,298,687 $4,983,997 $2,077,900 $7,974,395 Xylenes $4,525,223 $11,830,650 $3,787,838 $9,902,844 $6,060,540 $15,844,550 Ethylbenzene $13,244,556 $13,244,556 $11,086,354 $11,086,354 $17,738,167 $17,738,167 Benzene $13,888,154 $52,721,050 $11,625,077 $44,130,148 $18,600,124 $70,608,237 PAHs $74,183,991 $243,509,783 $62,095,700 $203,829,832 $99,353,120 $326,127,732 Acrolein $89,313,621 $98,911,988 $74,759,955 $82,794,267 $119,615,928 $132,470,827 Naphthalene $164,056,440 $371,936,175 $137,323,422 $311,329,127 $219,717,475 $498,126,604 Total HAPs $78,282 $515,901 $65,526 $431,835 $104,841 $690,935 G-2 ------- Model 13 - 3.5 MW Turbine Pollutant 80% Reduction & 3-Yr Catalyst Life Highest EF Average EF 80% Reduction & 6-Yr Catalyst Life Highest EF Average EF 50% Reduction & 6-Yr Catalyst Life Highest EF Average EF Formaldehyde $363,955 $2,863,658 $310,486 $2,442,953 $496,777 $3,908,725 Toluene $2,686,563 $14,378,788 $2,291,876 $12,266,378 $3,667,001 $19,626,205 Acetaldehyde $5,833,680 $22,388,026 $4,976,645 $19,098,966 $7,962,632 $30,558,345 Xylenes $17,014,900 $44,483,398 $14,515,214 $37,948,271 $23,224,342 $60,717,234 Ethylbenzene $49,799,706 $49,799,706 $42,483,553 $42,483,553 $67,973,684 $67,973,684 Benzene $52,219,641 $198,231,840 $44,547,971 $169,109,287 $71,276,753 $270,574,860 PAHs $278,932,781 $915,599,980 $237,954,325 $781,087,739 $380,726,920 $1,249,740,383 Acrolein $335,820,387 $371,910,374 $286,484,483 $317,272,433 $458,375,173 $507,635,893 Naphthalene $616,854,367 $1,398,484,901 $526,231,317 $1,193,031,273 $841,970,107 $1,908,850,037 Total HAPs $294,341 $1,939,794 $251,099 $1,654,815 $401,758 $2,647,705 Model 15 — 9 MW Turbine Pollutant 80% Reduction & 3-Yr Catalyst Life Highest EF Average EF 80% Reduction & 6-Yr Catalyst Life Highest EF Average EF 50% Reduction & 6-Yr Catalyst Life Highest EF Average EF Formaldehyde $187,550 $1,475,677 $158,509 $1,247,173 $253,614 $1,995,477 Toluene $1,384,418 $7,409,561 $1,170,045 $6,262,213 $1,872,072 $10,019,541 Acetaldehyde $3,006,165 $11,536,816 $2,540,669 $9,750,376 $4,065,071 $15,600,602 Xylenes $8,767,980 $22,922,824 $7,410,286 $19,373,296 $11,856,457 $30,997,274 Ethylbenzene $25,662,381 $25,662,381 $21,688,641 $21,688,641 $34,701,826 $34,701,826 Benzene $26,909,402 $102,151,226 $22,742,565 $86,333,426 $36,388,104 $138,133,482 PAHs $143,737,381 $471,819,563 $121,480,094 $398,759,771 $194,368,151 $638,015,633 Acrolein $173,052,241 $191,649,841 $146,255,640 $161,973,459 $234,009,023 $259,157,534 Naphthalene $317,872,394 $720,655,908 $268,650,843 $609,064,582 $429,841,348 $974,503,331 Total HAPs $151,677 $999,599 $128,191 $844,814 $205,105 $1,351,702 G-3 ------- Model 17 - 1.13 MW Turbine Pollutant 80% Reduction & 3-Yr Catalyst Life Highest EF Average EF 80% Reduction & 6-Yr Catalyst Life Highest EF Average EF 50% Reduction & 6-Yr Catalyst Life Highest EF Average EF Formaldehyde $897,899 $7,064,815 $773,614 $6,086,919 $1,237,782 $9,739,070 Toluene $6,627,912 $35,473,330 $5,710,491 $30,563,190 $9,136,785 $48,901,104 Acetaldehyde $14,392,037 $55,232,597 $12,399,923 $47,587,423 $19,839,877 $76,139,877 Xylenes $41,976,774 $109,743,200 $36,166,442 $94,552,789 $57,866,307 $151,284,462 Ethylbenzene $122,858,851 $122,858,851 $105,853,000 $105,853,000 $169,364,800 $169,364,800 Benzene $128,828,974 $489,049,793 $110,996,752 $421,356,601 $177,594,803 $674,170,562 PAHs $688,143,835 $2,258,839,853 $592,892,485 $1,946,176,230 $948,627,977 $3,113,881,969 Acrolein $828,488,959 $917,525,113 $713,811,348 $790,523,314 $1,142,098,156 $1,264,837,302 Naphthalene $1,521,816,578 $3,450,145,803 $1,311,170,089 $2,972,584,242 $2,097,872,142 $4,756,134,788 Total HAPs $726,157 $4,785,587 $625,644 $4,123,176 $1,001,030 $6,597,082 G-4 ------- |