United States Office of Air Quality EPA.450/3-81-001 Environmental Protection Planning and Standards January 1981 ey Research Triangle Park NC 2711 Air Urea Manufacturing Industry — Technical DocumentEPA-450/3-81-001 Urea Manufacturing Industry — Technical Document Emission Standards and Engineering Division Contract No. 68-02-3058 U.S. Environmental Protection A Fsaon oe (PL-12J) ae 7 West Jackson B Chicago, IL” GO6Oa-ggug* 12th Floor U.S. ENVIRONMENTAL PROTECTION AGENCY Office of Air, Noise, and Radiation Office of Air Quality Planning and Standards Research Triangle Park, North Carolina 27711 January 1981This report has been reviewed by the Emission Standards and Engineering Division, Office of Air Quality Planning and Standards, Office of Air, Noise, and Radiation, Environmental Protection Agency, and approved for puhlication. Mention of company or product names does not constitute endorsement by EPA. Copies may be obtained, for a fee, from the National Technical Information Service, 5285 Port Royal Poad, Springfield, VA. 22161.TABLE OF CONTENTS Chapter Page 1.0 Introduction and Summary... eee eee ee eee Ed 1.1 Purpose 1-1 1.2. Summary 1 2.0 The Urea Industry... ... ee cee cell 2.1 Industry Structure... .. 2.4 ee 21 2.2 Urea Products and End Uses... ee eee eee eee 2nd io (iciiss opaccodcuodo0douGgd46 207 3.0 Processes and Their Emissfons.........--...4..5 31 3.1 Introduction, ©. ee eee 31 3.2 Description of Processes and Emissions... . 2... BB 3.3 References. 2... 1. ee eee ee ee ee 337 4.0 Emission Control Techniques... 2... ee eee eee 4-1 4.1 Overview of Control Techniques. . . 4-1 4.2 Description of Control Techniques we 45 4.3. Emission Test Mata... ee ee ee ee ee poo Gai 4.4 Evaluation of Control hevice Performance. . 4-43 4.5 References... .. 50000000909 : 4-57 5.0 Model Plants and Control Alternatives. .......... 5el EB ie Ceaie Goo o gon ooucag00cd se Sel 5.2 Determination of Existing Control Levels 57 5.3 Control Options... 2... ee ee ee ee 518 Bo (iid) AbGotaites oo on 0nd 0000506 oo 5.30 5.5 References... . ee eee ee ee eee 5-32 CSET Sita bao oo oop ou od uGd on g6 0 6-1 6.1 Air Pollution Impact. 2... ee ee ee ee ee 6-1 (es2eeMa teriPo) lution inpac teers minecwds eat sar 6-4 6.3 Solid Waste Impact... 2... ee ee ee ee 6-6 (He Gaus ood goud du O So 6-6 6.5 Other Impacts... eee eee ee eee 6-8 6.6 References, ©... ee eee ee ee ee 6-9 7.0 Cost Analysis. ©. ee eee ee ee OG en 7-1 7.1 Cost Analysis of Control Alternatives... .... 7-1 7.2 Other Cost Considerations... eee... ce eee 7-20 7.3 References... eee eee eae bouoodouG 7-23 Appendix AL ee eee ee Peer) sae) ee oe Aol ppe nid) 9)8 sttcmeat ects oe tone ecteet Bed eee) oe et aces Bel 4aTable Ae 1-2 2-1 3-1 3-2 3-3 4e1 4-2 43 4-5 5-1 5-2 5-4 5-5 5-5 5-6 LIST OF TABLES Control Alternatives... 1. +s eee eee eee 1-3 Summary of Control Alternatives and their Effect on Product Price... ee ee eee ee ee THF Urea Producers - Plants, Locations, and Capacities . 2-2 Urea Production by Use... ....- eer de eo) Uncontrolled Emissions From Urea Facilities .... 3-5 Estimated Annual Uncontrolled Emissions From Typical Urea Plants ss es ee eee ee te eee 3-7 Uncontrolled Urea Particulate Emission Tests For Nonfluidized Bed Prill Towers... .....- 3-22 Summary of Use of Wet Scrubbers in The Urea findistryey ee ee eee 4-3 Summary of EPA Nass Emission Test Results... . . 4-28 Summary of EPA Visible Emission Test Results... . 4-29 Summary of Industry Mass Emission Test Results For Controlled Prill Towers... +--+ see 4-30 Summary of Cooler Controlled Emissions... . se 441 Model Urea Plants ...... ++ 2 Ee rie) esac Raw Material and Utility Requirements for [eT onoondogcogccGccc0d 0 5-8 Summary of Existing Emission Levels... .... ° 5-9 Emissions Standards Affecting Urea Plants... . . 5-12 Allowable Emissions by Plant Size (Metric Units) . 5-14 Allowable Emissions by Plant Size (English Units) 5-15 Control Equipment Performance Parameters... . 5-19Table 5-7 5-9 5-10 5-11 5-12 5-13 5-14 5-16 5-17 6-1 6-2 6-4 LIST OF TABLES (Continued) Emission Characteristics Control Options... .. Emission Characteristics Control Options... . + Emission Characteristics Control Options... .. Emission Characteristics Control Options... . . Emission Characteristics Control Options... . ee ee eee ee eee Emission Characteristics Control Options... .. ee ee eee eee Emission Characteristics Control Options... .. Emission Characteristics Control Options... .. Emission Characteristics Control Options... .. Emission Characteristics Control (Opt tons wetsuits eae w yee a Control Alternatives... 2... ee eee eae for Model for Model for Model for Model for Model for Model for Mode} Emission and Removal Factors for Control Alternatives... 6 4. Total Annual Reduction Over ELOC of Particulate Emissions for Control Alternatives Secondary Air Pollution Impacts Associated with the Application of Control Alternatives to Typical Urea Plants. . 2... 22. + eee Annual Energy Requirements for Urea Model Plants Control Alternatives. . . iv Page 5-20 5-21 5-22 5-23 5-29 5-30 5-31Table 7-1 7-3a 7-3 7-3¢ 7-4 7-5 7-7 7-8 7-10 Tell 7-1 7-12 LIST OF TABLES (Continued) Summary of Urea Model Plants and Control Alternatives Specifications for Particulate Control Systems. . Example of Major Equipment Requirements for Control of Prill Towers... 1 ee eee eee Example of Major Equipment Requirements for Control of Coolers... . Se Example of Major Equipment Requirements for Control of Granulators, 6. ee ee ee ee ee Purchased Equipment Costs Associated with Cie oo ooo GG0DoD0D000 Example of Purchased Equipment Cost Breakdown on Major Equipment... .-..--------6 Component Capital Cost Factors for a Wet Scrubber as a Function of Equipment Cost... .. Capital Costs of Control Alternatives for Model Plants 2... eee ee ee ee eee Bases for Scrubber Annualized Cost Estimates. . . Component Annualized Costs Net Annualized Costs for Control Options. .... Net Annualized Costs and Cost Effectiveness of Control Alternatives for Model Urea Facilities (Metric Units)... 1... ee ee . Net Annualized Costs and Cost Effectiveness of Control Alternatives for Model Urea Facilities (English Units) ..... ee eee Capital Costs of Uncontrolled Urea Plants... Page 7-2 7-5 7-6 1-7 7-8 7-10 Tel 7-12 7-13 7-14 7-16 7-17 7-18 7-19 7-21LIST OF FIGURES Figure Page 3-1 Urea Manufacturing... . . Se ee ee eee . 3-3 3-2 Total Recycle Urea Processes... .......04 oe 3-9 3-3 Air Swept Falling Film Evaporator... ... see 3-11 3-4 Two-Stage Vacuum Evaporator»... . ee. eee ee ee 3-12 3-5 Prill Tower - Nonfluidized Beds... ee... ee 3-13 3-6 Prill Tower - Fluidized Bed... .. 1. , See ee 14 3-7 SPUNNINGLBUCKet stecle eit eu Mt Mista esas tee ee eae 3-16 3-8 Multiple Spray Head Arrangement... .......004 3-17 3-9 Process Flow Diagram for Prill Tower... .. 2.000 3-19 3-10 Drum Granulator. ©... eee ee ee ee ee ee 3-23 3-11 Urea Drum Granulation Process... . eee eee ee ee 3-25 3-12 Pan Granulator. .. . ee eee ee eee ee ee 3-29 3-13 Process Flow Diagram for Pan Granulator..... see 3-30 3-14 Typical Countercurrent Direct Contact Air Chilled Rotary Cooler... . eee eee ee ee se 3-32 4-1 Typical Spray Tower Scrubber... ..... 2. ee . 4-8 4-2 Typical Packed Tower Scrubber... . 2... ..00. ae 4-10 4-3 Typical Mechanically Aided Scrubber... 2... 2.0.4 4-12 4-4 Uypicalliicayalypenscrubbere-leseiesus eaten ens nee . 4-13 4-5 Standard Fractional Efficiency for Tray Type Scrubber. . 4-15 4-6 Effect of Pressure Drop on Tray Type Scrubber ne eee eee Se ee eee 4-16LIST OF FIGURES (Cont inued) Figure Page 4-7 Typical Entrainment Scrubber... eee ee ee ees 4-17 4-8 Fractional Efficiency of Entrainment Scrubber Used in the Urea Industry as a Function of Particle Size and Pressure Drop. sss eevee eet tte eee 4-19 4-9 Typical Fibrous Filter Scrubber... . . Se a + 420 4-10 Fractional Efficiency of Wetted Fibrous Filter Scrubber. 4-22 4-11 Effect of Pressure Drop on Efficiency of Wetted Fibrous Filter Scrubber. . 6. see ee eee eee ee 423 4-12 Diagram of a Fabric Filter... ee eee eee eee 4-24 4-13 Particle Size Distribution of Uncontrolled NFB Prill Tower se Mnalis tae] antic) opepurtsemtenisrse sete nia sae teeters 4-32 4-14 Particle Size Distribution of Uncontrolled NFB Prill Tower Exhaust (Plant E). . 2... eee ee ee eee + 433 4-15 Particle Size Distribution of Uncontrolled Prill Tower Exhaust (Plant F). oe. eee eee eee te Ad 4-16 Histograms of Six Minute Opacity Averages for Controlled NFB Prill Tower Exhaust (Plant C)...... 4-35 4-17 Histograms of Six Minute Opacity Averages for Controlled NFB Prill Tower Exhaust (Plant E)...... 4-36 4-18 Histograms of Six Minute Opacity Averages for Controlled NFB Prill Tower Exhaust (Plant D)...... 4-48 4-19 Particle Size Distribution of Uncontrolled FB Prill Toworgeshiaust)(B]ant (0) sierente eye ruet tie emauenear 4-39 4-20 Particle Size Distribution of Uncontrolled Cooler Exhaust (Plant C)h. se eee eee eee ee eee se 682 4-21 Particle Size Distribution of Sub Micron Fraction Measured at Plant EB... ee eee ee ee see Add viiLIST OF FIGURES (Cont inued) Figure Page 4-22 Variation in Particle Size with Respect to Anbient Temperature see eee eee ee ee es re 4-49 4-23 Efficiency of Wetted Fibrous Filter as a Function of Ambient Temperature... te ee ee eee 4-50 4-24 Estimated Airflow Cutback as a Function of Ambient Temperature 2. ee te ee eee 4-52 4-25 Variation in Controlled Nonfluidized Prill Tower Emissions with Respect to Ambient Temperature... ... 4-53 4-26 Emission Levels from Uncontrolled Granulator Exhaust . . 4-55 Sel Process Diagrams for Model Plants 1-6.........+ 53 5-2 Process Diagrams for Model Plants 7-10... ..ee05 5-4 viii1.0 INTRODUCTION AND SUMMARY 1.1. PURPOSE The purpose of this document is to present and discuss technical information on the emissions, control techniques, and costs associated with control of emissions from processes in the domestic urea industry. Results of uncontrolled and controlled emissions testing are presented to quantify uncontrolled emissions and evaluate control device performance. 1.2 SUMMARY 1.2.1 Industry Structure The domestic urea industry produces urea in both solid and solution form, Solids are manufactured in two sizes. The smaller size is used for animal feed supplement. The larger sized solid is used for fertilizer applications and in the production of plastics and resins. Urea solutions are combined with other types of nitrogen solutions and used as fertilizers. There are 47 plants in the United States producing either urea solution alone or both solution and solids. In 1979 domestic urea production was 7.2 million Mg (9.9 million tons), a 19 percent increase over 1978. 1.2.2 Processes and Emissions Unit processes in the urea industry include urea solution synthesis, solution concentration, solids formation (prilling and granulation), solids cooling, solids screening, solids coating, and bagging and/or bulk shipping. Uncontrolled particulate emission rates range from 0.00241 kg/Mg of product (0.00482 1b/ton) for urea solution synthesis and concentration to 148.8 kg/Mg of product (297.6 1b/ton) for a solids producing process (granulation). The most effective control device used to control urea particulate emissions is a wet scrubber, 1.2.3 Model Plants and Control Alternatives Model plants were chosen to represent the existing domestic urea industry, These model plants have production capacities that range from 1-1182 Mg/day (200 tons/day) to 1090 Mg/day (1200 tons/day). Control devices that exhibit various levels of removal efficiency were identified for each source. Removal efficiencies for control devices applied to the model plants range from 57.9 percent for a spray tower to 99.9 percent for a wet entrainment scrubber. Several control alternatives were identified for each model plant. The control alternatives are based upon combinations of control devices applied to the sources within the plant. Three control alternatives were identified for prilling plants and one for granulation plants. Table 1-1 summarizes the control alternatives and corresponding emission factors for the model plants. 1.2.4 Economic and Environmental Impacts Table 1-2 presents a summary of impacts on urea product price due to the application of control alternatives. The increases in product price range from 2 to 8 percent based on a urea product price of $132/Mg ($120/ton). There are no water quality or solid waste impacts attributable to the use of wet scrubbers to control emissions. The primary air quality impact is the reduction in particulate emissions from sources in the urea industry. These reductions range from 58 to 98 percent for prill towers and 99.9 percent for granulators. Small secondary air impacts exist due to increased power plant particulate emissions resulting from the energy requirements of the control devices. The secondary ‘impact relative to plant-wide emission reductions range from 1 percent for a granulation plant to 3 percent for a prilling plant.eL TABLE 1-1. CONTROL ALTERNATIVES Control Alternatives = Control Entsston Factors, Alternatives g/g’ (1b/ton) Plant Extsston configuration Sources, 2 3 1 2 3 va Nonfluidized bed, Agrteul tural PrtT1 Tower + ” 0.900 0,385 0.138 grade pritl production Cooter ° © (1,800) (0.770) (0.276) at Fluidtzed bed, Agetcultural PrtTl Tower ‘ ” 0.600 0.470 0,062 grade pritT product ion (1.200) (0.930) (0.124) 7 Nonflutdized bed, Feed grade Pri Tower + ” 0.800 0,270 0.036 pritl product ion (1.600) (0.540) (0.072) 10 Granulator Granulator 0.118 = 5 (0.230) Legend: 0 - ELOC - defined tn Chapter 5 + Option I~ defined In Chapter 5 ++ option 2 - defined in chapter §el TABLE 1-2. SUMMARY OF CONTROL ALTERNATIVES AND THEIR EFFECT ON PRODUCT PRICE Wodel Plant Size Effect on Cost of Product $/Mg ($/ton) No. Ng/D Configuration Control Alternative (tons /0) al F A 1 181(200) Nonfluidized prit1 tower plant 6.35 7.20 10.31 producing agricultural grade (5.77) (6.53) (9.38) prills. 2 726( 800) 3.2 4,06 4.93 Ga) (3.68) (4.48) 3 1090(1200) 2193 3.68 4.39 (2.66) (3.34) (3:99) 4 181(200) —Flutdized bed pril] tower 6.62 6.64 9.92 plant. producing (6.02) (6.03) (9.02) 5 726(800) agricultural grade prits 3.73 4.79 5.77 (3:38) (4:35) (5124) 6 1090(1200) 3.64 4.81 5.45 (3.31) (4.36) (4.95) 3.86 3.37 6.43, 7 181(200) Pil Tower Plant (3:51) (3.06) (5.85) producing feed grade prills. 8 363( 400) fess ae 7 ({5.39]) A 9 726(800) Granulation Plant {6.03 a (fsa) 10 1090( 1200) {6.031 - “ ([5.68]) °control alternatives 2 and 3 are not presented for granulation plants. "values on brackets represent decreases in the product price.2.0 THE UREA INDUSTRY This chapter presents a description of the domestic urea industry. Section 2.1 will present information on the industry history, structure and growth. Section 2.2 will discuss urea products and end uses. 2.1 INDUSTRY STRUCTURE The domestic urea industry consists of 47 plants operated by 36 firms. Geographically, the industry production capacity distribution has shifted during recent years. Prior to 1966, capacity was fairly evenly distributed throughout the country. However, as of 1979 the primary concentrations of production capacity lay in the South-central states and Alaska, which together accounted for 41 percent of the total domestic capacity.! This shift is attributed to the availability of natural gas supplies (the basic feedstock for urea production) in these regions. Of the 36 urea producing firms, three firms account for over 39 percent of the total domestic urea production capacity. Table 2-1 presents a listing of all domestic producers, including their location, capacity, date of construction and product line. The majority of urea producers compete in the nitrogen fertilizer market with anhydrous ammonia, armonium nitrate, ammonia, nitrogen solutions, and nitric acid. Urea's share of the domestic nitrogen fertilizer market has been steadily increasing since 1970, In 1979, solid urea accounted for 12 percent of the nitrogen fertilizer applied in the United States. Historically, urea plants have operated at between 68 and 90 percent of their rated annual production capacity, depending on market conditions. Between 1966 and 1978 the average capacity utilization was 69.4 percent while in 1979 industry-wide capacity utilization increased to 90.2 percent.2*3 In 1979, 7.2 million Mg (7.9 million tons) of urea was produced, a 19 percent increase over the previous year. The projected 2-1TABLE 2-1, UREA PRODUCERS-~PLANTS, LOCATIONS, AND CAPACITIES —— Sapacity Date on company nae Plant location TMG) CO tons} Form of urea stream Air Proqucts and Chemicals Pensacola, FL a a Solutions 1965 Allied Chemical Corp. Helena, Ag a a Solutions 1967 Geismar, LA 285 me Solutions 1967 Omana, Ke ur 40 Solutions 1985 American Cyanami¢ Co. New Orleans, UA uo az Melamine 1966 ker Industry Corp.* Carlsbad, nM 169 6 Unspecified 1976 Bison Nitrogen Products Waodward, OK 104 ne Liquid feed 178 (corownea with Terra Chemical International) Borsen, Inc Geismar, Us 200 220 Pritts 1968 CF Industries, Ine. Donalasonvit) 738 367 Solutions, granular 1974 6 8 Solutions i365 a 5 Solutions, prills, 1967 Viguia fee Tunis, NC 180 165 Solutions 1968 Tyner, TH 3 $8 Solutions i363 ‘The Coastal Corp. wyeon chemical Co. cheyenne, Y 9 Solutions, pritis, 1966 Tiguia teed Coiumoia Nitrogen Corp. Augusta, GA 359 395 Solutions, pritis 1965 Comines American (rd. Camex, Ine. Borger, TX cy a Granular, pritls 1980 Wivak, ine. Kerens, TH 8 82 Unspecified Na Esgark, Ine. Estech General Chemicals Beaumont, TX 4s 50 Solutions 1967 Core. Farmland Industries, Ine. Dodge City. KS sa 64 Solutions 1975 Lawrence, KS wa 268 Solutions, granular, 1983, Viquia teed General American O11 Ca. of Texas Premier Petrocnenicals, subs. Pasadena, T% cy 7 Pritts Getty 011 Co. Clinten, 1A 5s 60 Solutions 1963 Hawkeye Chemica? Co. Gooapasture, Ine. dimes, 7 a a Solutions am WR, Grace and co. Memonis, 7 27 33 Pritts, crystal 1358 Hercules Ine. Louisiana, so 38 98 Solutions, ureaform 1958 tercilizer Kaiser Aluminum & Chemical Savannan, SA 320 ae Solutions 1986 te Scurrently smut gown. (continua)TABLE 2-1. (Continued) soacta date an Company sane Plant location CIOTRG) CIT TERS] form of urea stream “isstasipoi Chemical Co, Yazoo City, MS U7? CWO~«CSolutions, grils 1989 NeRen Corporation € Oubuaue, th 7 a Solutions, pritts 1a. Pryor, OK 18 13 Solutions 1370 alin Carp. lake Charles, A168 to, Pritts 1366 Pnitiips Pacttte chemical Finley, WA “ 37 Saluttons 1965 Pnitlips Petroleum Co. Beatrice, NB a 53 Solutions 1368 Tatemnold Chemicals St. Helens, oR 332 Mestly artis 1367 JR Staatot ce, Pocatalto, 10 “ 15 Solutions ws74 Stangara 941 of ca Chevron Chemical C3, Fort Madison, 1A 73 Solutions 1380 Stanaars 017 af onto ‘wsstren Corp. subs. Lima, OH 200° © «220 Solutions, prttis 1988 Tannessee Valley Authority Muscle Sheals, AL 8 so Solutions, granular 1972 Tara chenicai Port Neal, 1a 230253. Solutfons, granular, 1967 Tnearnat‘onal aris, aula fea Tevaa chemteais Jonatdsonvitie, (A 425 $83 PrilTs, melamine 1968 Tyler cars. Atlas Powder Corp., — aplin, HO a 74 Solutions, ari1ts 196 Union 211 of California rea, ca wo Solutions, prilts 1966 anal, aK 580748 Granular, arflls) 1969, US. Stee! Cors, Cherokee, AL 7 77 Solutions 3362 dallay Nitrogen Proqucars £1 Centro, cA HS MB Solutions, artis, 1968 Tiauia feed 410tans co, Agrica chemteat Biycnevitie, aR soo 330 Granular 1978 Qonalesonvitie, A 20S 228 Granular 1368 Veratgrisy aK 420 $28 Solutions 1375 243demand for urea in 1980 is 7.5 million Mg (8.2 million tons) which would represent a 4 percent increase in production over 1979. 2.2 UREA PRODUCTS AND END USES Urea has three basic uses: fertilizer, cattle feed, and as a component in the manufacture of plastics and resins. Table 2-2 presents the annual amount of urea used for fertilizer, feed, and plastics and other applications. Urea is marketed as a solution or in a variety of solid forms. Most urea solution is used in fertilizers, with a small amount going to animal feed supplements. Urea solution is never used alone as a fertilizer. It is always blended prior to application with another chemical such as ammonium nitrate or ammonium sulfate. Mixed urea-anmonium nitrate (UAN) solutions have a number of advantages over pure urea or ammonium nitrate solutions. UAN solutions are less corrosive than either of the individual components and do not decompose with time like pure urea solution. Most importantly, UAN solutions have a lower crystallization temperature than ammonium nitrate or urea solutions separately, which reduces the likelihood of a solution salting out during transfer and application. Solution fertilizers are currently becoming more popular than solid fertilizer because they are easier to transfer in tank cars and do not generate dust problems.°*© However, solid urea is still in demand for various appl ications. Urea solids are produced as prills, crystals, and granules. A prill is an air-cooled solid sphere that is produced in two sizes. The smaller of these, 0.35-1.7 mm (0.014 - 0.066 in) in diameter, is referred to as feed grade and the larger, 0.5 - 4.0 mm (0.020 - 0.157 in), as agricultural (or fertilizer) grade urea. Prills are used as a fertilizer, as a protein supplement in animal feeds, and in plastics manufacturing. Feed grade urea production has declined since 1960 when 12.9 percent of total urea production was feed grade urea. Sy 1978 feed grade urea represented only 6.7 percent of total urea production. The major urea based plastics are urea-formaldehyde resins and melamine. The domestic output of urea for use in the manufacture of plastics has grown steadily 2047 TABLE 2-2. UREA PRODUCTION BY USE 103 Mg Foret izer faim) Plastics career Teac other Total —E—_— Ee 138 ; Me wa 7 1987 ma ms mA 1988 ma me ry 1989 ue me ny 1360 223 6 3 586 (245) (95) (9h) (733) 1962 284 92 78. 836 (312) (ign) (38) (920) 1962 io oi 32 336 Ga ay 0) (2,008) 1963 prs us io $91 (375) (128) qn (2,090) 1964 334 308 Es 31986 (433) (ug) (103) (4,206) 1968 328 336 307 267 en ao) a) 3.284) 1966 08 75 is a7 (664) (as2) (64) (2,713) 1367 788 0 ies Tas? (823) (23) 160) (2,087) 1968 303 258 156 2208 (993) (282) (172) (2,423) wep 3,002 304 153 2698 (2,202) (334) a73) (2,964) asm 1.278 308 158 2923 1106) Gs) i) 3002) wn “Tis 28 $52 21235 (1,300) (273) (497) (3,8) we hus 308 rr 3s 28) GG) G80) ha 37 201 Baie (1,233) (40a) (41) (3,535) we Va 32 373 3a aap Ge) GT) 1975 1,309. 264 sue 3,445 (1,440) (230) (365) (3,780) 17s “Toes ee 56 3701 (2,232) (244) (612) (4,071) 7 Te fas aie 3600 ais70) «23 (675) (51060) 3378 Be, Na 438 Ger, 6183) 1973 $1236.00 34 “ e147 (5176000) _ G78) (61762) ———— — ANunners in parentin sare in 10? tons Pincluaes all preduets otner than ferti! Totals not exact due to rounding. 2-5since 1960 at an annual average growth rate of 11.8 percent.® Currently, 8 percent of the total urea produced is targeted for uses in plastics, resins and melamine.® Crystals are formed by the vacuum crystallization and drying of urea solution, These crystals may be used as is, or remelted for prilling. The major advantage of crystals is their Tower biuret content. Biuret is a urea decomposition product and a plant poison (see Chapter 3). Production of agricultural grade urea solids by granulation is on the increase compared to production by prilling. In granulation, seed particles are built up to granules by the addition of successive layers of molten material. Because of the nature of particle buildup, granulation can produce larger particles with greater abrasion resistance and particles with two or three times the crushing strength of standard prills. Another benefit of greater abrasion resistance is the reduction of solids dusting when the product is conveyed and bulk loaded, Granular product is not as spherical or as smooth as the prilled product, and smal] feed grade granules cannot be manufactured using present technology.? However, any of the larger desired product sizes, from fertilizer grade granules to even larger forestry grade granules can be manufactured. Large granules are preferred for forestry application because they are more massive and less likely to be caught in tree branches when being applied from the air.® 2-62.3 REFERENCES Stanford Research Institute. 1979 Directory of Chemical Producers. Menlo Park, California, SRI International, 1979. Search, W.J. and R.B8. Reznik. (Monsanto Research Corporation.) Source Assessment: Urea Manfacture. (Prepared for U.S. Environmental Protection Agency.) Washington, D. C. EPA Publication No. EPA- 600/2-77-107L. November 1977. 94 p. Bridges, J.D. Fertilizer Trends 1979. Muscle Shoals, Alabama, Tennessee Valley Authority. National Fertilizer Development Center. January 1980. p. 12. Chemical Profile:Urea. Chemical Marketing Reporter. 218(15):9,31. October 13, 1980. Trip report. Bornstein, M.I., GCA Corporation, to Noble, E.A., EPA:ISB. August 3, 1978. p. 9. Report of Visit to the Tennessee Valley Authority National Fertilizer Development Center in Muscle Shoals, Alabama. Harre, £.A. The Outlook for Nitrogen Fertilizers. Tennessee Valley Authority. Muscle Shoals, Alabama. (Presented at the Forest Fertilization Conference. Union, Washington. September 25- 27, 1979.) p. 10. Memo from Ramachandran, V., Research Triangle Institute, to Rader, R., Radian Corporation. January 6, 1981. 6p. Information about data stored in Triangle University Computing Center. Reference 1. Reference 2. 2-73.0 PROCESSES AND THEIR EMISSIONS This chapter presents a discussion of the processes and emissions found in the urea industry. Section 3.1 will present the urea process chemistry, a process overview, a description of the types of urea plants, and emissions overview. Section 3.2 will discuss in detail the individual urea production processes and their emissions. 3.1 INTRODUCTION 3.1.1 Process Chemistry Urea (also known as carbamide or carbonyl diamide) CO(NH,) is an organic, natural gas based chemical. The primary feedstocks of urea are ammonia and carbon dioxide. Urea is formed by reacting anmonia and carbon dioxide at 448-473 K (347-392°F) and 19.2-23.2 MPa (2,800-3,400 psi) to form ammonium carbamate. The carbamate is then dehydrated to form urea and water. These reactions are represented by the following equations. 2NH3 + CO;———>NHCOp HH (1) NH4CO,NHs————=NHCONH, + Hp0 (2) The carbamate formation step (1) is exothermic, releasing 150-160 kd (64500-68800 Btu) per mole of ammonium carbamate formed. This reaction is favored by high pressures. The dehydration step (2) is endothermic, consuming 32 kJ (13800 BTU) per mole of urea formed. This step is favored by high temperatures. Urea, as a solid, is a colorless crystal with a melting point of 406 K (271°F) and a specific gravity of 1.335 at 293 kK (68°F).293 Aqueous urea solutions begin to decompose at 333 K (140°F) to buiret and ammonia. Dry urea, however, is stable below 403K (26°F). Above this temperature dry urea decomposes to buiret and ammonia according to the following reaction. 3-12C0(NH>)) ————s— (HCO), NH + Ny Above 443K (338°F) the primary decomposition products of urea are cyanuric acid (HNCO), and ammonia. The biuret concentration in urea must be monitored, as it is a plant poison, and is also undesirable in industrial (plastics) applications. Biuret concentrations in urea solids are 0.1 percent or less in crystals, 0.3 percent in solids formed from crystal remelt, and 1.0 percent in solids formed from concentrated urea solution.* 3.1.2 Process Overview The process for manufacturing urea involves a combination of up to seven major unit operations. The basic arrangement of these operations is shown in the block diagram given in Figure 3-1, These major operations are: (1) solution synthesis (solution formation) (2) solution concentration (3) solids formation = prilling = granulation (4) solids cooling (5) solids screening (6) solids coating (7) bagging and/or bulk shipping The combinations of processing steps are determined by the desired end products. Plants producing urea solutions alone are comprised of only the first and seventh unit operations, solution formation and bulk shipping. Facilities producing solid urea employ these two operations and various combinations of the remaining five operations, depending upon the specific end product being produced. 3.1.3 Types of Urea Plants All urea plants produce an aqueous urea solution as depicted in the process diagram shown in Figure 3-1. In these plants, ammonia and carbon dioxide are reacted to form ammonium carbamate. The carbamate is then dehydrated to yield a 70 to 77 percent aqueous urea solution. The 3-2e-€ aporrive® conte? BAGGING RENIA ———"} sox urion sourioy |_| sous ‘soLtDs screenine carson ———+] FORMATION conc ENTRATION| FORMATION (COOLING DIOXIDE = SHIPPING rine OFFSIZE RECYCLE | : on SHIPPING These processes are opt ional depending on Individuat ma Figure 3-1. facturing practices. Urea manufacturing.solution can be sold as an ingredient in nitrogen solution fertilizers or can be further concentrated to produce solid urea. There are three methods of concentrating urea solution: vacuum evaporation, atmospheric evaporation and crystallization. Vacuum and atmospheric evaporation produce a urea melt containing from 99 to 99.9 percent urea at a nominal temperature of 413.7K (285°F). Crystallization is used primarily when product requirements dictate an extremely low biuret concentration in the final product. Urea solids are produced from the urea melt by two basic methods: prilling and granulation. In prilling there are two types of prill towers: fluidized bed and nonfluidized bed. Each of these is capable of producing both agricultural grade and feed grade urea prills. The major difference between these towers is that a separate solids cooling operation may be required when producing agricultural grade prills in a nonfluidized bed prill tower. The fluidized bed supplies the required cooling for agricultural prills in a fluidized bed prill tower. However, because the small feed grade prills exhibit better heat transfer properties, additional cooling external to the nonfluidized bed tower is not required. Prill towers are described in detail in Section 3.2.4. The other methods of solids formation used in the urea industry are drum and pan granulation. In drum granulation, solids are built up in layers on seed granules in a rotating drum granulator/cooler approx- imately 14 feet in diameter. Pan granulators also form the product in a Jayering process, but the equipment used is different from the drum granulator. There is only one pilot scale pan granulator operating in the domestic industry, providing 61,000 Ng/year (67,000 tons/year) of urea granules. Details of the granulation process are presented in Section 3.2.5. 3.1.4 Industry Emissions Overview Emissions from urea processes include particulate matter, ammonia and formaldehyde. Table 3-1 presents uncontrolled emission factors for each of the major processes in the urea industry. Table 3-2 provides an 3-4Se TABLE 3-1, UNCONTROLLED EMISSIONS FROM UREA FACILITIES?*® Particulate Anwonia Fornaldehyde Process Plant ka/tg (b/ton) ka/Mg (b/ton) g/g (b/ton) Solution Formation ‘and Concentration A e.00241 (0.00482) 1.89 (25.77) Solution Formation and Concentration 8 0.0150 (0.0317) 4.01 (8.02) Solution Formation and: Concentration D 0.0052 (0.0108) 14.40 (28.80) Drum Granulation a 148.8 (297.6) 1.08 (2.15) 0.00359 (0.0072) Drum Granulation 8 63.6 (127.2) 1.07 (2.13) 0.00555 (o.0111) Non-Flutdized Bed Pri! Tower E 1.90 (3,80) 0.433 (0.865) (Rgricaltural Grade) Fluidized Bed Prit Tower D 1.80 (3.60) 2.07 (a.ray 0.0020 (0.0040) (Feed Grade) Flutdized Bed Prit1 Tower D 3.2 (6.23) 1.42 (2.91) 0.0095, (0.0190) (Agricul tural Grade) Rotary Orun Cooler c 3.72 (7.45) 0.0255, (0.051) AIT data are from EPA test results (see Appendixestimate of the total annual emissions from sources in typical urea plants based on the emission factors in Table 3-1. Ammonia is emitted during urea synthesis (solution production) and solids production processes, Ammonia emissions range from 14.40 kg/Mg (28.80 1b/ton) for synthesis processes to 0.0255 kg/Mg (0.0051 1b/ton) for a rotary drum prill cooler, Amore detailed list of ammonia emission data are presented in Appendix A. Formaldehyde has been added to the urea melt in recent years for the purpose of reducing urea dust emissions and to prevent solid urea product from caking during storage. Formaldehyde is added to the urea melt in concentrations of 0.5 percent or less prior to solids formation. A further discussion on additives is contained in Section 3.2.8. The use of formaldehyde as an additive has resulted in formaldehyde emissions which range from 0.0095 kg/Mg (0.0190 1b/ton) of urea produced for a fluidized bed prill tower producing agricultural grade urea, to 0.0020 kg/Mg (0.0040 1b/ton) of urea produced for a fluidized bed prill tower producing feed grade urea solids. Amore detailed list of formaldehyde emissions is included in Appendix A. Particulate matter is the primary emission being addressed in this report. Table 3-1 includes a summary of uncontrolled particulate emissions form all urea processes. These particulate emissions range from 148.8 kg/Mg (297.6 1b/ton) of urea produced for a rotary drum granulator to 0.00241 kg/Mg (0.00482 1b/ton) of urea produced for a synthesis process. A more detailed list of particulate emissions is presented in Appendix A. In the following section each of the processing steps described earlier is reviewed. Several of these processes are comparatively small particulate emitters and/or are not expected to be built in the future because of changing production technology. For these processes, the sections will provide a brief description of the actual process operations. More detailed descriptions are provided for solids production and cooling processes which are large particulate emitters. 3-6TABLE 3-2. ESTIMATED ANNUAL UNCONTROLLED EMISSIONS FROM PROCESSES IN TYPICAL UREA PLANTS Mg (Tons) Plant Capactty recess nejaey, Ceons/aay) Partteulace enon Formaldenyae ion Formation > nd Concentration se (00) 0.32 (as ee (1288) a Solution Formation " {tnd Concentration v7 (00) 0.636 (0.755) 228e (2516) at Setution Fematton . {tnd Cancentraeion a0 (200) ozs (1132) 347 (37) ae rue Granulation 36 (400) nse (12889) ne, (128.8) 0.00831 (0.00935) Oru Granulation v7 (300) 22s (zeae) 232.9 (257.7) 0.0186 (0.0183) or Gramstacton oso (1200) aout (37348) 350.9 (386.5) e.czes (0.0275) Nonetiysdtzed Bey Pet « ‘ Tower (agrieatturat Graze) 182 (200) ton.8 (1168) ag.ge (26.04) Fusaized Sed Prt] € Toner (igrieuiaure! Grage) 383 (200), zor? (2zma) —aA.2_ (52.08) Son-Flutdtzed fed Pratt e 2 Toner {dgriea) tara’ Gre v7 (3800) mss (4575) a8. (06.18) NoneFlysdteed Bad feyt) Tower {igeteal ural Grae) 1090 (3200) s2a.2 (626.2) a (186.28) £ £ Flurdized bed rit] Toes 182 (200) 2.6 (178.3) 78.0 (8.7) oes (0.582) (igrscutturat Grave) Fate St zits ome 38 (409 RAP Ns) 182.9 87.8) (2.0826) geicattural Grace) Flutetaed fed 2rf11 Tor 727 (800) wart (5364) 308.0 334.77 (2.167) Cigricuttural Grace) Flusdised Sed 2rs11 Tower 1080 (1200) 730.5 (8085) 455.0 (502.2) 29s (3.281) (gescuteural Grace) FRegtged Se P01 Tor (2003 aL. (iol.t) 108s (9.8) owi9e (0.116) Flutéizes Sed Pert Tower 362 (400), wené (202.3) 217.0 (29.0) o.z08 (0.229) (Feed deaee) fBagtae tet Pitt Tome 787 (60) werd (4045) 408.0 (478.0) cate (0.688) (aed dense fBuigtaes ted iT} Tomer 1080 (100) $50.8 (606.8) 651.0 ousat (0.686) (Feed Grace fotary Oran Cooler we (200) nae (208) . cS Rotary drum Cooler 3s. (400) 254 (468.5) G 5 Rotary Oran Cooler re (800) 52.8 (938.1) 5 é « Aotary Oran Cooler oso (3200) va7e.2 (1408.7) aaa (9.21) ol cS conparison purposes only and do ucity of the teased process. Plant capacities are oresented f hot necessarily represent. the actual >. Faraldenyae ‘5 introduced to the aelt after solucton formation and concencration orocestes. ce Mot averTable, 3-73.2 DESCRIPTION OF PROCESSES AND EMISSIONS 3.2.1 Synthesis Processes There are numerous process designs for producing urea solution. These designs fal] into 3 categories, they are once-through processes, partial-recycle processes, and total recycle processes.° The older processes are the once-through and partial-recycle processes, dating back to the early 1950's. These processes represent Jess than 25 percent of current domestic urea production capacity.© The once-through process employs a reactor and a carbamate decomposer. The decomposer separates urea solution from a stream containing ammonia, carbon dioxide, and water. This ancillary stream is generally sent to another fertilizer-producing plant. The partial-recycle process provides a small refinement in that excess ammonia from the urea reactor is recovered and recycled to the reactor. Ammonia excesses as large as 200 percent are used to boost urea yields up to 80 percent.” The total recycle process is the most widely used of the basic processes since it provides the benefits of higher yields and Tower energy consumption. Major designers of urea synthesis plants have three types of total recycle process: (1) processes in which decomposed carbamate gases are separated and recycled to the reactor (Figure 3+-2a); (2) processes in which carbamate solution is recycled (Figure 3-2b); and (3) processes in which both gas and liquid recycling is used (Figure 3-2c). At least ten major companies provide designs within these total recycle process classifications.° Emission sources from synthesis processes are typically noncondensable vent streams from ammonium carbamate decomposers and separators. Emissions from synthesis processes are generally combined with emissions from the solution concentration process. Results of EPA testing on these emissions are presented in Table 3-1 and Appendix A. Based on EPA testing, combined particulate emissions from urea synthesis and concentration are smal] compared with particulate emissions froma typical solids producing urea plant. For this reason, emissions from synthesis processes will not be considered further in this report.Freep -L_J REACTOR UREA DECOMPOSER SOLUTION A, Basic gas recycle process CARBAMATE RECYCLE | ADDITIONAL i CONDENSER P-— ADDITION | TCASES REACTOR UREA I TOR pee UY SEPARATOR SOLUTION reep-— B. Basic liquid recycle process NH RECYCLE fT 1 i I | REACTOR UREA | SOLUTION rep oL CARBAMATE RECYCLE C. Basic gas / liquid recycle process — PRODUCT-CONTAINING STREAMS, ~~ RECYCLE, FEED, OR OTHER ANCILLARY STREAMS Figure 3-2, Total recycle urea processes.9 3-93.2.2 Solution Concentration The 70 to 77 percent urea solution resulting from the synthesis process mist be concentrated if a solid urea product is desired. The method of concentration depends upon the level of biuret impurity allowable in the end product. For low biuret urea, solution concen- tration is effected by means of continuous crystallization in an atmos- pheric or vacuum crystallizer, The solution is concentrated at moderate temperatures until urea is crystallized from solution. The crystals are separated from solution and are dried as a product or remelted for further processing. Vacuum is often developed by use of steam ejectors. At present, only five manufacturing plants employ crystallization, and at least one facility has eliminated the crystallization process. 10>11 Solution concentration to greater than 99 percent urea is more often accomplished by means of single or double stage evaporation. Evaporators operating at atmospheric pressure are commonly of the thin film or falling film variety as shown in Figure 3-3. Newer processes employ vacuum evaporators, typically of the thermosiphon and forced circulation design as illustrated in Figure 3-4, These evaporators operate at slightly higher temperatures than the crystallization process and provide a nearly pure urea melt to the solids formation process. Again, vacuum is provided by means of steam ejectors.!2+13+14 Noncondensable emissions from solution concentration processes are often combined with emissions from the synthesis process and vented in a common stack. Particulate emissions from concentration processes are smal1 compared to those from other plant processes. For this reason, emissions from concentration processes will not be discussed further in this report. 3.2.3. Prilling Prilling is a process by which solidnearly spherical particles are produced from molten urea. Molten urea is sprayed from a head tank into the top of a rectangular or circular tower (See Figures 3-5 and 3-6). As the droplets fall through a countercurrent air flow, they are cooled 3-10AIR AND WATER VAPOR OUT + SOLUTION IN AMBIENT AIR FALLING FILM EVAPORATOR aa STEAM AIR HEATER CONDENSATE | CONDENSATE PE BTED L_—.» concentrated Aik SOLUTION our Figure 3-3. Air swept falling film evaporator.wert venr CONDENSER sree stew eee conoensare eeron ReTuRNED To Process stem Cc Fuasw is srean area | conoensan CONCENTRATED UREA wactaee” “soutien ro Figure 3-4, Two-stage vacuun evaporator.AIR AIR OUTLET OUTLET INDUCED DRAFT FANS HEAD TANK SPRAY AREA (SPRAY HEADS OR SPINNING BUCKET) GRATED FLOOR PRILLS FORCED DRAFT FANS CONVEYOR 70.2056-1 Figure 3-5. Prill tower - nonfluidized bed.AIR, AIR ouTLeT ouTLeT INoUCED DRAFT FANS. HEAD TANK SPRAY AREA (SPRAY HEAOS OR SPINNING BUCKET) GRATED FLOOR AiR AiR INLET ries propuct | FLUIDIZED ma BED COOLER AiR nuer! AIR BLOWER eo Figure 3-6, Prill tower fluidized bed.and solidify into near-spherical particles. The solids are collected at the bottom of the tower for further processing. Prill towers typically range from 33,5 meters (110 feet) to 64 meters (210 feet) in height. Cross sectional areas range from 29.2 square meters (314 square feet) to 190 square meters (2040 square feet). The height and cross sectional area of a prill tower depend upon prilling rate, product grade produced and the amount of cooling required. Molten urea between 99.5 and 99.8 percent in concentration is sprayed into the prill tower at about 413K (284°F) by either a single spinning bucket (see Figure 3-7) or a spray head arrangement (see Figure 3-8). Natural, forced or induced draft may be used to provide air flow through the tower. The airflow rate for cooling of the prills depends upon ambient temperature and humidity, prilling rate and type of prills being produced.!° If the tower incorporates a fluidized bed cooler, the blower used to suspend the pril] bed supplies air to the main body of the tower as well. Uncontrolled emission rates from prill towers may be affected by the following factors: (1) product grade being produced (agricultural grade or feed grade) (2) air flow rate through the tower (3) type of tower bed (4) ambient air conditions and (5) melt spray conditions These factors are described in this section. Two different size prills are produced by industry: feed grade and agricultural grade. The hole diameter in the sprayhead or bucket determines the size of the pril1 being produced,!® which in turn deter mines the airflow rate required for cooling in the tower. Generally, 60 to 70 percent lower airflow rates are required when smaller sized feed grade urea is being produced. The smaller particle size results in better heat transfer because of the larger surface area per unit volume of urea, Although grain loadings are higher in the exhaust streamsMELT FROM HEAD TAMK PERFORATED BUCKET ROTATION + BUCKET MAY BE CONICAL SHAPE. HEAD TANK: SPINNING BUCKET PRILL TOWER 70.2000-1 Figure 3-7. Spinning bucket. 3-16HEAD TANK CONTAINING MELT 70.2045-1 MULTIPLE ‘SPRAY HEADS Figure 3-8. Multiple spray head arrangement. 3-17from feed grade prill towers, the total mass emissions per unit of feed grade prill production may be lower because of the lower airflows.27 Two different types of towers may be used to produce prills: fluidized and nonfluidized bed (see Figure 3-9). Fluidized bed prill towers incorporate a fluid bed cooler at the bottom of the prill tower, which provides additional cooling of agricultural grade prills without using an auxiliary rotary drum cooler. Higher airflow rates are used to suspend (fluidize) the prill bed and to provide supplementary cooling. The advantage of having a fluidized bed cooler at the bottom of tower is that the purchase of an additional large piece of equipment (a rotary drum cooler) is not necessary. The disadvantages of this type of tower are: (1) a large blower is required to suspend the prills at the bottom of the towers and (2) if the tower is also designed to produce feed grade prills, the additional cooling provided by the fluidized bed is more than required for proper solidification./® Nonfluidized bed towers may use an additional rotary drum cooler to provide the necessary cooling during production of agricultural grade prills. Alternately, prill tower height or prill tower air velocity could be increased. Prills collected at the bottom of nonfluidized bed towers are raked onto conveying belts for transport to the rotary drum cooler or storage. If a nonfluidized bed prill tower is also designed to produce feed grade prills, the rotary drum cooler is bypassed during the production of feed grade prills because the prill tower alone supplies sufficient cooling (see Figure 3-9). The advantages of a nonfluidized bed prill tower are: (1) airflow rates through the tower are generally lower than through a fluidized bed prill tower by as much as 20 percent (see Appendix A) and thus entrainment is reduced and (2) the cooler can be bypassed when making feed grade prills. The major disadvantage is that the production of agricultural grade prills usually requires the addition of a rotary drum cooler.6I-E Tame af a | NONFLUIDIZED BEO PrILL TOWER, 4 5-——goron cons me conctnrmaton FLUIDIZED BED PrILt TOWER Figure 3-9. Process flow diagram for prill tower. paAmbient air conditions can affect prill tower emissions. The ambient air temperature determines the required airflow rate through the tower, Theoreticatly, the required winter airflow rate is approximately 60 percent of that needed during summer operation. Anbient humidity can also affect prill tower emissions. If humidity is high, airflow rates must be adjusted. Higher airflow rates, in general, result in higher emissions, as noted for fluidized bed prill towers. Data supplied by industry indicates the particle size distribution of prill tower emissions is affected by ambient temperature.!9 1t appears that as the falling molten urea droplet is cooled by the toner airflow, urea is vaporized from the surface of the solidifying prill. This urea vapor then condenses in the cold tower airflow to form small diameter particles. Therefore, during colder weather the size distribution shifts toward smaller particles. Although it is reported! that uncontrolled emissions are reduced under these conditions, the grain loading remains constant due to the reduction in tower airflow. Additional data concerning this shift and its affect on control device performance is presented in Section 4 of Chapter 4. Melt spray conditions can also affect prill tower emissions. Theoretically, higher melt temperatures result in higher emission rates due to the increase in surface vapor pressure and associated increase in fuming.2° In addition, an increase in melt spray pressure could result in higher emissions due to increased atomization of the spray stream. Uncontrolled particulate emission rates from fluidized bed prill towers are higher than those from nonfluidized bed prill towers for agricultural grade prills, and approximately equal for feed grade prills. Emission testing was conducted by EPA on anew, large fluidized bed prill tower facility during the production of feed grade and agricultural grade prills. Airflow rates during testing were within normal operating limits and the melt temperature to the sprayheads did not vary more than 3k (56°F). EPA emission testing on a nonfluidized bed tower was conducted during agricultural grade production only. The nonfluidized bed pri1] tower tested was an older tower designed for lower production capacities. 3-20Particulate emissfons as measured during EPA testing from a fluidized bed tower producing feed grade prills were 1.80 kg/Mg (3.60 Ib/ton) of product. ?articutate emissions measured during EPA testing during the production of agricultural grade prills were 3.12 kg/Mg (6.23 Ib/ton) of product from a fluidized bed prill tower and 1.90 kg/Mg (3.80 1b/ton) of product from a nonfluidized bed prill tower, Table 3-3 presents data for tests of uncontrolled emissions from nonfluidized bed prill towers. Industry test data shows that uncontrolled emissions from a nonfluidized prill tower are slightly greater (13 percent) during feed grade production than uncontrolled emissions during agricultural grade production, However, due to differences in test methods and difficulties involved in measuring emissions from prill towers (see Chapter 5), this data may be misleading. Particulate emissions for a nonfluidized bed prill tower producing feed grade prills have not been tested by EPA. 3.2.4 Granulation Granulation has become the more popular means of producing solid urea for fertilizer uses. There are two methods currently being used to produce urea granules: drum granulation and pan granulation. Each of these processes are described in the following sections. 3.2.4.1 Drum Granulation, With one exception, all drum granulators operating in the United States are manufactured by one company and are essentially similar in design and operation, Presently, 18 of these granulators operate at five different urea plants in the United States. The production rate of each granulator is approximately 363 Mg/day (400 tons/day). The one exception is a larger drum granulator designed to produce 773 Mg/day (850 tons/day). The drum granulator (see Figure 3-10) consists of a rotating horizontal cylinder about 4.3m (14 ft) in diameter divided by a retaining dam into two sections, the granulating section and the cooling section. Both sections have lifting flights welded to the wall. A pipe running axially near the center of the granulating section emits a fine spray of liquid urea (including formaldehyde additive if used) in an upward direction. Seed urea particles are fed into the drum at the granulation end, As the drum rotates, the lifting flights pick up the urea seed 3-2122-8 TABLE 3-3, UNCONTROLLED UREA PARTICULATE EMISSION TESTS FOR NONFLUIDIZED BED PRILL TOWERS 22 Grain” EwissTon ties ion Capacity AlrgFlow Loadigg Rate Factor Tenperature Percent Data Product Ny/day iva 9/tw o/s koma Ko Moisture source p __"Twe (100) (1000 scr) __(qv/SCAM) _(1b/br) (ib/ton) Cy inate. _ F % 199.6 42.9 0.02 3,53 325, Industry (29:6) (99.9) (0.059) (28.0) 2) 6 198.2 ou 0.483 4.02 367 Industry ais} (19:2) (0.1935) (31:9) 209) 6 a 18.2 193.0 0.0295, 4.22 36 Industry (ou) (303) (ocoreay (33.5) as) 2.08 1 a 236.4 2.06 Industry, (260) (22:7) 1 a” 227.3 1.02 Industry (250) (e213) 3 6 500, 65.0 9.0156 3.87 a2 650) (180) (0.0199) (30-71) (02) 1.18 Industey K Fo 218,2 4.05 Industry (240) (32218) c 6 345 51 0.05 2.77 308, Industry (300) (108) (0.0230) (22:9) @5) Loa u AG 409 12,7 9.0699 9.44 Industey (450) (260) (0.0301) (67:1) t % 268.8 aa2 0.1522 5.01 19 a 1.07 tn (291.3) (30.9) (0.0665) (46.1) (3:8) (iat)‘SEED UREA PARTICLES ROTATING ORUM UREA / SPRAY RETAINING a \ AM exwausr ain e / To SCAUBBER / 8€0 OF UREA SY GRANULES CONCENTRATED _ | UREA SOLUTION —~ —— LIFTING FLIGHTS: COOLING AIR — Coo, | ING Seer; 'pRooUCT ow} 7o.2046.1 Figure 3-10. Drum granulator. 27particles and shower these particles down through the urea spray. AS the particles fall and roll, they become coated with molten urea. Particles gradually build up to product size by addition of successive layers of liquid, which solidify to give the granule an onion-skin-like (concentrically layered) structure. Particles in the granule bed will tend to segregate according to size, the smaller granules of urea settle to the bottom to be picked up by the lifting flights. The drum operates at a slight angle and materia migrates by gravity towards the cooling section. Larger particles at the top of the granule bed pass over the retaining dam into the cooling section. Throughout this operation granules are cooled with a countercurrent flow of air. An airflow velocity of 1.2 meter/sec (4.0 ft/sec) is used to minimize seed entrainment and disturbance of the fine melt spray. The cooling air, at this velocity, is chilled to an inlet temperature of about 283K (50°F) and has an exit temperature of about 358K (185°F).24%25 Urea spray in the granulating section is held at approximately 413K (285°F),2® but granules exiting the cooling section are approximately 310K (104°F).2” Cooled granules (Figure 3-11) are removed from the cooling section and undersized particles are separated and recycled as seed material. Oversize granules are either crushed and recycled as seed, dissolved and returned to the solution process, or both, The typical recycle to product ratio for a drum granulator is 2:1.28 Cooling air passing through the drum granulator entrains approximately 10-20 percent of the product.°? This air stream is controlled with a wet scrubber which is standard process equipment on all drum granulators. Emission rates from drum granulators may be affected by the following parameters :31 1, Number, Design and Location of lifting flights 2, Airflow rates through the drum 3. Melt spray pressure 4. Dam height 3-24Se-e FOB aout CONDENSATE FROM UREA SOLUTION PLANT wer SCRUBBER SCRUBBER LIQUOR TO UREA SOLUTION PLANT zs SCREEN EXHAUSTER STACK CRUSHER, UREA MELT Le exit ain Figure 3-11. Urea drum granulation process. CONVEYOR GRANULATOR TO STORAGE proouct ELEVATOR 305. Bed temperature 6. Recycle rate of seed material Te Product size. 8. Rotation rate of the drum The number, design and location of lifting flights directly affect the emission rate. Flights 1ift and drop granules through the moving air stream to cool the particles. When flights are located close to the air exit of the granulator, fine particles are entrained in the air stream leaving the granulator. Modifications have been made to many existing drum granulators to change the size and/or shape of the lifting flights, and to remove lifting flights at the air discharge end of the granulator because of excessive entrainment. This modification is also being made on new installations.2¢ Air velocities through the drum have been reported as high as 1.8 meters/sec (7 ft/sec).2? A greater air velocity will result in increased entrainment of small particles in the drum granulator and a subsequent increase in emissions. The pressure of the melt at the spray nozzles is maintained within a range of about 2.5 - 3.8 kPa (10-15 psig).°* Lower pressures cause the granules to take the shape of popcorn and higher pressures cause an increase in fine granules, which may increase emissions.?° The dam at the center of the granulator is used to separate the granulation zone and the cooling zone. Changing its height will result in changes in bed temperature, which could affect emission rates. The dam height is set by the manufacturer and is not normally changed. The bed temperature in the granulation zone is reported to be critical.%© only a relatively narrow temperature range can be tolerated. If the bed temperature drops too low, the granules will agglomerate. If the bed temperature is increased significantly and maintained for several hours, the bed will turn to dust and emissions will increase.?* 3-26The recycle rate of seed material affects the bed temperature and therefore the emission rate, An increase in seed material recycle rate will cool the bed, while a decrease will raise the bed temperature.?2 As mentioned previously, increased bed temperature results in increased particulate emissions. Drum granulators have an advantage over prill towers in that they are capable of producing very large particles without difficulty. Granulators also require less air for operation than do prill towers. A major disadvantage of granulators is their inability to produce the smaller feed grade granules economically. To produce smaller granules, the drum mst be operated at a higher seed particle recycle rate. It has been reported that although the increase in seed material results in a lower bed temperature, the corresponding increase in fines in the granulator causes a higher emission rate.2? Increasing the rotation rate of the drun may increase entrainment of urea in the airstream, with a corresponding increase in the loading of urea to the scrubber. However, once set by the manufacturer the rotation rate of the drum is not normally changed. The original granulator design of the granulators called for the drums to rotate at 9 rpm. But because of excessive wear, rotation rates have been decreased to 6 rpm, with no apparent effect on product quality.“ As previously stated, most granulators are a standard size and are operated in the same way with many of the parameters affecting emissions fixed by granulator design. Uncontrolled emissions from drum granul ators were determined at two different plants. The average particulate emission rate from each of these tested facilities were 63.6 and 148.8 kg/Mg (127.2 and 297.6 Ibs/ton) of product. The granulators tested Were of the same design. Airflow rates, melt temperatures, and melt pressures were within normal operating limits during EPA testing. 3.2.4.2 Pan Granulation. The pan granulation process operates on ‘the same principle as the drum granulator, forming granules by adding successive layers of molten urea to seed particles. The equipment, however, is quite different. It consists of a large, tilted rotating 3-27circular pan (see Figure 3-12), Seed material deposited near the top of the pan along with fine particles carried up by the rotating pan, fall through a fine spray of liquid urea, The newly sprayed partictes drop to the bottom of the pan. As in the case of the drum granulator, smaller particles tend to sift down toward the bottom of the granule bed on the lower part of the pan. The larger granules spill over the edge of the pan onto a conveyor belt. Pan granulation is a fairly recent development in urea processing and has yet to gain widespread use. It is still in the pilot plant stage with only one existing pan granulator in operation in the U.S. (see Figure 3-13). The pan granulator yields a product which is less spherical than either drum granules or prills and not quite as hard as a granule produced by a drum granulator.4! Pan granulation also tends to have a larger recycle to product ratio, as most of the required cooling in the pan is accomplished through heat absortion by the cooled recycle seed material. This mode of cooling is necessary since the air flowrate is only 20 percent of a drum granulator's air flowrate.4@+43 The pan granulator operates with an optimum bed temperature between 377K and 380K (220°F to 225°F).44 The temperature of recycled seed material is approximately 343k (158°F).4> The urea solution ( 99.0 percent urea) is Kept at approximately 413k (285°F) to assure even coating of particles before crystallization occurs.46 The recycle to product ratio will generally fall between 2:1 and 3:1.47 The recycled material serves to cool the granule bed and maintains the desired bed temperature. A decrease in recycle ratio will result in an increase in bed temperature. The pan granulator is followed by two rotary drum coolers. All of the material leaving the first cooler is screened. The oversized stream is either redissolved and returned to the solution concentration steps, or crushed and returned to the pan with the undersized material. The amount of crushed material used as seed is held to a minimum, as use of this material leads to the formation of agglomerates and weak granules. This recycle stream to the pan contributes to the cooling of the bed 3-28RECYCLE PAN ROTATION ENTERS HERE SPRAY a AREA COVERED BY SOLUTION SPRAYS LARGE GRANULES LEAVE PAN HERE DEEP. BED Figure 3-12. Pan granulator. 3-29oc-e SCREEN CRUSHER i SCREEN BAGGING COOLER tt SOLUTION FORMATION AND CONCENTRATION PAN GRANULATOR BULK LOADING COOLER | Figure 3-13. Process flow diagram for pan granulator.particles as noted previously. Product size urea is sent to the second cooler and then conveyed to the warehouse for shipment. The advantage of the pan granuiator over the drum granulator is that the airflow rate required for cooling is approximately one fifth of that required for the drum granulator. Although the existing plant (which is still experimental) uses two coolers, a new plant could be designed with only one cooler thus reducing the total system airflow needed for cooling. 48 Test data on the pan granulator cannot be directly correlated to EPA test data because of differences in test methods. However, uncontrolled particulate emissions were reported to be approximately 2.1 kg/Mg (4.2 1b/ton) of product.49 3.2.5. Solids Cool ing Supplementary cooling for the pan granulation process and for agricultural grade prills produced in nonfluidzed bed prill towers is provided by auxiliary coolers (see Figures 3-5 and 3-13.). All coolers currently in use in the urea industry are of the rotary drum type. The rotary drum cooler consists of a revolving cylindrical she11, horizontal or slightly inclined toward the outlet. Hot feed enters one end of the cylinders cooled material discharges from the other. As the shell rotates, internal flights 1ift the solids and shower them down through a countercurrent flow of air. A typical cooler is shown in Figure 3-14, A rotating shell made of sheet steel is supported on two sets of rollers and driven by gear and pinion, At the upper end is a hood which connects through a fan to a stack, Flights are welded inside the shell. At the lower end the cooled product discharges onto a conveyor. Just beyond the end of the rotary cooler is a set of chillers which cools the incoming air. The air is moved through the cooler by an induced draft fan which keeps the system under a slight vacuum. Emissions from coolers consist of urea particles that become entrained in the rotary cooler air stream. 3-31‘COOLER SHELL UFTING FLIGHTS SHELL. SUPPORTING ROLLERS END VIEW ) FEED coouer ‘ourter care ooLF CHILLER | P AIR i i incer TEES DISCHARGE FAN ain SHELLSUPPORTING COOLED SOLIDS DISCHARGE ROOLERS DISCHARGE HOOD ro-2087-+ SIDE View Figure 3-14. Typical countercurrent direct-contact air chilled rotary cooler. 3-32The following parameters affect emissions from rotary drum coolers: (1) Number, design and location of lifting flights (2) Air flowrate through the drum (3) Bed temperature (4) Speed of drum rotation Rotary drum coolers operate in mich the same manner as the cool ing section of drum granulators. Therefore, parameters will affect emissions in similar ways. The number, design, and location of lifting flights affect the amount of fine particles entrained in the cooling airstream. Likewise, the rotational speed of the drum may affect the entrainment of urea in the airstream, The airflow rate through the drum affects emissions from rotary drum coolers. Increased airflow rates increase the amount of fines entrained in the airstream, Also, with increased air flows, larger particles may be transported in the cooling air. The bed temperature in the rotary drum cooler can affect emissions indirectly. Higher bed temperatures require increased airflow rates in order to cool the prills. And, as discussed above, increased air flowrates cause higher emission rates. Testing of a rotary drum cooler cool ing agricultural grade prills was performed by EPA at one facility. Results of this test indicate an uncontrolled emission rate of 3.90 kg/Mg (7.80 1b/ton) of product. The cooler tested was of typical capacity for urea coolers. Airflow rates during testing varied within normal operating limits. 3.2.6 Solids Screening Solid urea is screened to remove offsize product. The of fsize material may be returned to the process in the solid phase, as is typically done in granulation plants, or it may be redissolved in water and returned to the solution formation end for reprocessing. This second option is usually performed at urea prilling facilities. Product specifications for the more typical grades and types of urea are presented below. 9025! +52 3-33Feed Grade 100 percent through a 10 mesh screen (U. S. Sieve) 90 percent caught on a 40 mesh screen (U. S. Sieve) Agricul tural Grade 98 percent through a § mesh screen (U. S. Sieve 98 percent caught on a 30 mesh screen (U. S. Sieve) Granular Grade 99 percent through a 6 mesh screen (U. S. Sieve) 99 percent caught on a 20 mesh screen (U. S. Sieve) Several types of screens are employed to separate product size from oversize and undersize material. Screening equipment commonly used in the urea manufacturing industry include shaking screens and vibrating screens. Dust is generated due to abrasion of urea particles and the vibration of the screening mechanisms. Therefore, almost all screening operations used in the urea manufacturing industry are enclosed or have a cover over the uppermost screen. Uncontrolled emissions from solids screening were not tested by EPA, Results of survey inspections conducted during this program indicated that this operation is a small emission source and in most cases no visible emissions were observed.°%*54+55 Therefore, particulate emissions from solids screening will not be considered further in this report. 3.2.7 Coating Operations /Additives Clay coatings are used in the urea industry to reduce product caking and urea dust formation. However, clay coatings also reduce the nitrogen content of the product and the coating operation itself creates of clay dust emissions. Presently, only three plants are still using coatings.° The popularity of coating has diminished considerably because of the increasingly common practice of injecting additives into the liquid or molten urea prior to solids formation.°7>98 additives reduce solids caking during storage and urea dust formation during transport and/or handling. Additives react with the urea to form 2 crystalline urea compound by a mechanism that is not clearly understood. °° 3434The resulting solid particle is harder than solids made without additives. Additives, therefore, have replaced coatings in a major portion of the urea industry, and this trend is expected to cont inue.©? The most common additive is formaldehyde which is incorporated into the liquid urea before solid formation.©!»®2 the formaldehyde content of the finished urea will generally fall between 0.3 and 1.0 percent.°9-64 Because addition of the additive involves a simple injection into the urea melt, no particulate emissions result from the process. Formal- dehyde emissions from EPA testing are reported in Table 3-1 and Appendix A. Emissions attributable to coating include entrained clay dust from loading, in-plant transfer, and from leaks around the seals of the coater. No emissions data are available to quantify this fugitive dust source. For this reason, coaters will not be considered further in this document. 3.2.8 Bagging and/or Bulk Shipping Solid urea product is either bagged or bulk shipped. The majority of product is bulk shipped; approximately 10 percent is bagged. Two types of bags are used: the open-top, sewn bag and the corner-fil1, valve-type bag. The open-top bag is held under the bagging machine which fills the bag to a predetermined weight. After filling, the top is pinched together and sewn. The corner-fill valve bag is "factory closed"; that is, the top and bottom are partially closed either by sewing or by pasting, and a small single opening or valve is left on one corner. Urea is discharged into the bag through the valve. The valve closes automatically due to the back pressure produced by the contents of the bag as soon as it is filled. Bagging operations are a source of particulate emissions. Dust is emitted from each bagging method during the final stages of filling when dust-laden air is displaced from the bag by urea. Bagging operations are conducted inside warehouses and are usually vented to keep dus out of the workroom area according to OSHA regulations. Mass emission tests were not conducted by EPA on an uncontrolled bagging operation. However, data provided by industry indicates that 3-35uncontrolled particulate emissions are approximately 0.095 kg/Mg (0.19 1b/ton) of product bagged. This emission rate was determined by weighing the amount of urea collected in a baghouse used to control the bagging operat ion. 6 On a national basis only a small fraction of urea produced is bagged (approximately 10 percent).° The major portion is bulk loaded in trucks or enclosed railroad cars. The actual method of product bulk loading varies from plant to plant. During bulk loading, long flexible chutes are used to convey the urea from the storage hopper to the tank truck or railroad car. Very few plants control their bulk loading operations. As discussed above, emissions vary with use of coatings. During this study, bulk loading of a coated urea product was not observed; however, the bulk loading of uncoated urea was observed. Generation of visible fugitive particulates was very slight.©7 3-363.3. REFERENCES Us 10. lL. 12, 13, 14, 15. 16. Search, W.J. and R.B. Reznik. (Monsanto Research Corporation. ) Source Assessment: Urea Manufacture. (Prepared for U.S. Environmental Protection Agency.) Washington, D.C. EPA Publication No. EPA-600/2- 77-107L. November 1977. 94 p. Reference 1, pp. 9-12. Considine, 0.M.(Ed.). Chemical and Process Technology Encyclopedia. New York, McGraw-Hi11 Book Company, 1974. p. 118. Reference 3, pp. 118 - 119. Memo from Stelling, J., Radian Corporation, to file. April 1, 1980 p. 5. Compilation of ammonia and formaldehyde emissions data. Reference 1, p. 13. Reference 1, p. 14. Reference 1, p. 15. Reference 1, p. 16. Memo from Bornstein, M., GCA Corporation, to file. December 6, 1979. p. 2. Solid urea manufacturing industry survey summary. Trip report. Jennings, M., Radian Corporation, to file. March 27, 1980. Report of visit to W.R, Grace and Company in Nemphis, Tennessee. Reference 10. Trip report. Bornstein, M.I. and $.V. Capone, GCA Corporation, to Noble, E.A., EPA:ISB. dune 23, 1978. p. 2. Report of visit to Agrico Chemical Company in Blytheville, Arkansas. Trip report. Capone, S.V. and M.I. Bornstein, GCA Corporation, to Noble, E.A., EPA:ISB. June 21, 1978. p. 2. Report of visit to Borden Chemical in Geismar, Louisiana. Letter and attachments from Killen, J.M., Vistron Corporation, to Goodwin, D.R., EPA:ESED. December 21, 1978. p. 9. Response to Section 114 letter on urea plants. Letter and attachments from Swanburg, J.D., Union Oi1 of California, to Goodwin, D.R., EPA:ESED. December 20, 1978. pp. 4-5. Response to Section 114 letter on urea plants. 3-3717, 18. 19. 20. 21. 22. 23. 24, 2s. 26. 27. 28. 29, 30. 31, 32, 33. Reference 16. Reference 15, p. 21. Cramer, J-H. Urea Pril] Tower Control Meeting 20% Opacity. Presented at the Fertilizer Institute Environmental Symposium. New Orleans. April 1980.) pp. 2-3. Reference 19, p. 4. Memo from Stelling, J., Radian Corporation, to file. dune 18, 1980. 43. Compilation of uncontrolled emissions from urea prill towers reported by industry. Reynolds, J.C. and R.M. Reed. Progress Report on Spherodizer Granulation 1975-1976. | C & I Girdier Incorporated. Louisville, Kentucky. (Presented at the Environmental Symposium of the Ferti- lizer Institute. New Orleans. January 15, 1976.) p. 11. Reference 1, p. 35. Ruskan, R.P. Prilling vs. Granulation for Nitrogen Fertilizer Production. Chemical Engineering. 83:114-118. June 7, 1976. p. 116. Letter and attachments from Alexander, J.P., Agrico Chemical Company, to Goodwin, D.R., EPA:ESED. December 21, 1978. p. 9. Response to Section 114 letter on urea plants. Trip report. Bornstein, M.I., GCA Corporation, to Noble, E.A., EPA:ISB. August 2, 1978. Report of visit to C & I Girdier Incor- porated in Louisville, Kentucky. (p. 24 of addendum.) Reference 25 p. 3. Reference 26, p. 3. Reference 28. Reference 26, addendum 1, p. 12. Reference 26, addendum 1, 25 p. Reference 26, p. 2. Reference 26, p. 2. 3-3834, 35. 36. 37. 38, 39. 40. an. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. Reference 26, addendum 1, p. 19. Reference 34. Reference 26, addendum 1, p. 6 - 7. Reference 36. Reference 26, addendum 1, p. 7. Reference 32. Reference 13. Reference 38. Reference 28. Reference 26, addendum 1, p. 2. McCamy, I.W. and M.M. Norton. Have You Considered Pan Granulation of Urea? Reprint from Farm Chemicals, January 1977 issue. p. 4. Hicks, G.C., I.W. McCamy and M.M. Norton. Studies of Fertilizer Granulation at TVA. In: Proceedings of the Second International Symposium on Agglomeration. New York, Anerican Institute of Mining, Metallurgical and Petroleum Engineers. March 6-10, 1977. p. 863. Tennessee Valley Authority. New Developments in Fertilizer Technology. (Presented at the 11th Denonstration. National Fertilizer Development Center. Muscle Shoals, Alabama. October 5-6, 1976.) pp. 80-81. Reference 45, p. 3. Reference 46, Harre, E.A. The Outlook for Nitrogen Fertilizers. Tennessee Valley Authority. Muscle Shoals, Alabama. (Presented at the Forest Fertilization Conference. Union, Washington. September 25-27, 1979.) p. 6. Reference 26. Reference 14. Reference 25. Reference 25, p. 1. 3-3954. 55. 56. 57, 58. 59, 60. 61. 62. 63. 64. 65, 66. 68. Reference 25, pp. 3-5. Letter and attachments from Picquet, N.E., W.R. Grace and Company, to Goodwin, D.R., EPA:ESED, December 14, 1978. p. 7. Response to Section 114 letters on urea plants. Reference 10. Reference 16. Trip report. Bornstein, M.I. and S.V. Capone, GCA Corporation, to Noble, E.A., EPA:ISB. June 22, 1978. p. 2. Report of visit to W.R. Grace and Company in Menphis, Tennessee. Barber, J.C. Energy Requirements for Pollution Abatement. Chemical Engineering Progress. 72:42-46. Decenber 1976. Letter and attachments from Homan, J.M., Terra Chemicals International, to Goodwin, D.R., EPA:ESED. December 14, 1978. pp. 2-22. Response to Section 114 letter on urea plants. Reference 50. p. 3. Reference 60, p. 11. Reference 60. Reference 55, p. Trip report. Curtin, T.L., GCA Corporation, to Noble, E.A., EPA:ISB. August 13-23, 1979. -p. 4. Report of visit to W.R. Grace and Company in Memphis, Tennessee. Reference 10. Reference 11. 3-404,0 EMISSION CONTROL TECHNIQUES This chapter discusses techniques used for controlling urea particulate emissions from prill towers, coolers, granulators, and bagging operations in the urea industry. ‘As mentioned in Chapter 3, ammonia and formaldehyde emissions are also generated in urea processes. However, the major objective of this study is to evaluate particulate emissions. Accordingly, this chapter concentrates on the effectiveness of various devices in controlling particulate matter. The majority of the data used in assessing control device effectiveness was generated by an EPA source testing program conducted in conjunction with this project. Testing involved particulate emission measurements at five urea plants utilizing test methods similar to the test method in Appendix B. Appendix A provides more detailed information on all test data used in this chapter. The chapter is organized in the following manner. Section 4.1 presents a general overview of control techniques used in the urea industry, Section 4.2 describes several of these control techniques in greater detail and outlines the factors that affect their performance. Section 4.3 reviews available industry and EPA emission test data. Finally, Section 4.4 evaluates this data and contro] device performance. 4.1 OVERVIEW OF CONTROL TECHNIQUES With the exception of bagging operations, urea emission sources are typically controlled with wet scrubbers. The preference toward scrubber systems as opposed to dry collection systems is primarily due to the ease of recycling dissolved urea collected in the device. Scrubber liquors are recycled back to the solution concentration process, eliminating potential waste disposal problems and recovering the urea collected. 4-1Concerning other potential control devices, fabric filters are not suitable for controlling emissions from many sources because the hygro- scopic nature of urea particulate combined with the moisture content of the gas streams could cause blinding of the bags. Dry cyclones offer lower collection efficiencies than scrubbers in urea particulate applications. Electrostatic precipitators are not currently in use in any urea industry applications. Fabric filters (baghouses) are used in the control of fugitive dust generated in bagging operations where humidities are lower and blinding js not a problem. Many bagging operations are uncontrolled. However, if a control device is used, baghouses are the typical method of control. Table 4-1 presents a summary of the present population of control devices being applied to prili towers, granulators, and coolers. As mentioned previously, these sources use wet scrubbers if a control device is used. The following subsections provide a brief description of how controls are applied to each urea emission source under consideration. 4.1.1 Nonfluidized Bed Pril] Tower Controls The majority of nonfluidized bed prill towers are uncontrolled. Of the seven prill towers which utilize control devices, one uses a spray tower scrubber, two use packed bed scrubbers, one uses a wetted fibrous filter, and three companies consider the type of scrubber used to be confidential information. Control devices vary considerably in the number used and in placement for various applications. The most common location for scrubber mounting on nonfluidized bed prill towers is on the top of the tower. Only two installations duct emissions to ground level.!>2 Tower mounting is usually more economical since long runs of corrosion resistant ducting are not required. However, scrubbers mounted on top of towers typically do not have the extended stacks necessary for suitable sampling locations. In addition, tower mounting may require a strengthened prill tower in order to withstand the additional weight and wind load of the scrubber. 4-2ep TACLE 4-1, SUMMARY OF USE OF HET SCRUBBERS IN THE UREA INDUSTRY46>47543 No. of Emission Spray Packed Mechanically — et vay Fibrous Type Emission Source Sources Tower Tower. Rided "Cyclone Type Entrainment Filter Unknow Uncontrolled PriN Towers wo 5 1 2 0 ° 0 0 , a a mw 3 1 0 ° 0 0 ’ 0 1 a Rotary w 0 1 0 0 0 18 0 0 0 Granulators, Pan Granulator 1 ° 0 ° ' 0 o 0 ° 0 Coolers 6 0 2 2 0 1 o ° ° 1The number of scrubbers used on a nonfluidized bed prill tower varies considerably. One system uses a single device, while other prill towers use up to four devices, The use of more than one scrubber allows for variability in airflow rates. A prill tower which produces both feed and agricultural grade product may need only 30 percent of the agricultural grade airflow during feed grade production. Seasonal changes in ambient temperatures may also dictate that flow rates be varied in order to maintain a reasonably constant prill temperature. Thus, a scrubber system needs the ability to be turned down to lower airflow rates while maintaining removal efficiencies. Multiple scrubbers allow units to be removed from service while maintaining normal airflow and pressure drops in the remaining operating scrubbers. The wetted Fibrous filter allows pressure drop to be adjusted readily while the unit is in operation, thus accomodating changes in airflow rates. 4.1.2 Fluidized Bed Prill Tower Controls Three fluidized bed prill towers are currently operating and all use some type of scrubber system. One manufacturer considers all information concerning their in-house designed scrubber system pro- prietary. Another manufacturer uses a spray tower scrubber with extensive internal baffles. The third fluidized bed prill tower uses multiple entrainment scrubbers. All fluidized bed prill towers use tower mounted control devices. As with nonfluidized bed prill towers, th’s mounting typically causes problems in prill tower emission testing. 4.1.3. Granulator Controls With one exception, all rotary drum granulators are controlled by nearly identical entrainment scrubbers. These are essentially the same scrubbers as used on the fluidized bed prill tower mentioned previouslys however, a higher pressure drop is used for granulator applications. The only exception to the use of entrainment scrubbers is the use of a packed tower at one drum granulator installation. This drum granulator is produced by a different company than the other 18 granulators.Since nearly all drum granulators are similar designs marketed by the same company, installations are fairly standard. One scrubber is used for each granulator and a testable outlet stack is typically provided. However, at least one installation uses a conmon outlet stack for two scrubbers.? The single pan granulator operating in the United States uses a wet cyclone scrubber. 4.1.4 Rotary Drum Cooler Controls Rotary drum coolers are used to cool agricultural grade prills when sufficient cooling is not provided in the prill tower. Coolers are not required in fluidized bed prill towers, during feed grade production, or when adequate cooling airflow is available in the prill tower. A wide variety of control devices are currently used to control cooler exhausts: packed towers (both moving bed and conventional bed), mechanically aided scrubbers and tray towers. One cooler is uncontrolled. 4.1.5 Bagging Operation Controls At some urea plants, a portion of the solid product is bagged, as discussed in Chapter 3. Bagging operations are hooded and vented to the atmosphere to reduce dust levels in the workroom air, in accordance with Occupational Safety and Health Administration (OSHA) standards. Emissions from the exhaust ventilation system for the bagging operation may be vented directly to the atmosphere or through an air pollution control device. The most conmonly used control device for bagging operations is a fabric filter (baghouse). Of the eleven urea plants conducting bagging operations, six use baghouses while one is reported to use a dry cyclone.* The remainder are uncontrolled. 4.2 DESCRIPTION OF CONTROL TECHNIQUES In this section, the various types of control devices used in the urea industry are reviewed. This review includes a description of the device, the collection mechanism, and the factors that affect performance. 4-54.2.1 Wet Scrubbers A wet scrubber is a device in which a gas stream is brought into contact with a liquid, usually water. Any device which introduces a liquid to clean an airstream may be termed a scrubber. Scrubbers are widely used to remove gaseous components as well as particulate matter. Scrubbers rely on a variety of collection mechanisms, however, the dominant mechanisms in al1 scrubbers used in the urea industry are impaction and interception. The scrubber provides water droplets and/or wetted surfaces which impact and intercept the particles. The particulate Jaden liquid is then separated from the gas stream and recycled or discharged as waste. Other collection mechanisms which may contribute to scrubber effectiveness include gravitational settling, diffusion (brownian motion), and condensation effects. However, the importance of these mechanisms are usually secondary in the types of scrubbers described in this chapter. Scrubber performance depends on the characteristics of the dust laden airstream being cleaned and on the design and operation of the scrubber. The most important airstream characteristics are particle size distribution and grain loading. Other factors being equal, larger particles are removed with greater efficiency than smaller ones. Likewise, higher grain loadings may lead to agglomeration, enhancing scrubber effectiveness. Two factors in the design and operation of a scrubber which may strongly influence performance are energy input and liquid flow rate. Increased energy input to the scrubber causes more turbulent gas-liquid contact and greater particulate removal. Similarily, increasing liquid flowrate in the scrubber usually enhances gas-liquid contact. In both factors, however, a level is reached where increases are no longer justified by the improvement in performance. The high velocity, turbulent flow of gas through the scrubber causes a decrease in the gas phase pressure head. This pressure drop 4-6across the scrubber is a convenient means of measuring the energy used in the scrubber. High pressure liquid sprays may also supply energy, however, the importance of this energy input is usually secondary for scrubbers used in the urea industry. 4.2.1.1. Spray Tower Scrubbers. Figure 4-1 depicts a spray tower scrubber system of the type presently being used to control a fluidized bed prill tower. The airflow from the prill tower travels upward ‘impinging on a baffle plate. As the gas flows around the baffle, several gas vortices are formed which increase the residence time in the scrubber. The gas stream passes through a jet of fine sprays which impacts the particles. Particle laden droplets fall to the bottom of the scrubber housing and are removed with the liquid stream. In general, the performance of spray towers is influenced by the surface area of the scrubbing droplets and the relative velocity between the droplets and the particles entrained in the gas stream. Small droplets can enhance performance since small droplets provide a large surface for particle impingement. On the other hand, large droplets can also enhance performance since large droplets fall at high terminal velocities, and thus provide a high relative velocity between particles and droplets. This high relative velocity usually increases the chances of a particle impacting a droplet. Depending on the particle size distribution of the incoming gas stream, an optimum droplet size which balances these two effects will provide best performance. This optimum size is reported to be in the range of 500 to 1000 microns over a wide range of particle sizes.® Droplet size is influenced by the nozzle configuration and the nozzle pressure, Nozzle pressures of 138 - 689 kPa (20 - 100 psig) are typical; however, high pressure sprays of 2760 kPa (400 psig) may also be used when a very Fine droplet size is desired.!° concerning liquid use in spray towers, a range of .0669 - 1.07 liters/m? of gas (.5 - 8 gat/1000 Ft)! has been reported. Pressure drops in spray towers are usually very low, typically less than .5 kPa (2 in, W.G.).22 4-7DEMISTER SPRAY HEADER Gas 4 y VORTEX \ \ eH | —— BAFFLE \ 4 > _——_ -_~> ~ oT PAC \Lg@ | wy — ‘ = rm re Bons INLET Figure 4-1. Typical spray tower scrubber. ? 4-8The spray tower currently used in urea prill tower applications is designed to operate at .25 - .5 kPa (1-2 in. W.G.) pressure drop with liquid to gas ratios of .134 - .268 liters/m? (1-2 ga1/1000 ft). Spray nozzle pressure is 689 - 1380 kPa (100 - 200 psig).'? Although efficiency curves for various particle sizes are not available, the manufacturer claims exit loadings of .0115 - .0344 g/m? (.005 - .015 gr/dscf) are achievable in urea prill tower applications.!* concerning visible emissions, the manufacturer has reported that opacities of 20 percent or less are achievable. 15 4.2.1.2 Packed Towers. In packed towers (Figure 4-2) the scrubber interior is packed with shaped elements or materials such as crushed rock. The packing is irrigated by water sprays to keep the packing wet and provide @ wet surface for particulate impingement. Particles impact the wetted packing and are subsequently flushed to the bottom of the scrubber. Gas flow may be concurrent, countercurrent, or crossflow to the liquid stream. The performance of a packed tower is directly influenced by the size, shape and type of packing material, Small packing material with high ratios of surface area to volume are usually desirable, al though clogging may be a problem with small, intricate packings. The depth of packing does not have a great effect on particulate removal once a minimum depth is provided. This minimum depth has been reported to be 10 - 12 times the major dimension of the packing pieces. 16 Velocity through the tower also affects performance. Higher velocities increase impingement of medium and large particles.!” for very small particles (less than .3 microns) a Tow velocity may be desirable to assist removal via diffusion. ‘8 Pressure drops depend on packing type and depth. Approximately 1123 kPa (.5 in. W.G.) per foot of bed is typical for most packings. Total pressure drop through the scrubber is typically between .5 kPa (2 in. W.G.) and 2.5 kPa (10 in. W.G.).19 Liquid use is normally .267 - 669 liters/m? of gas (2 - 5 gat/1000 ft*).*° 4-9Gas ourLer ‘CHEVRON DEMISTER t tft PACKED SECTIONS Liquor } yy ? “ cas INLET + SUMP roz04s.1 LauoR OUTLET Figure 4-2. Typical packed tower scrubber. 4-10The major disadvantage of packed beds is their susceptibility to clogging under high particulate loadings. A moving bed alleviates this problem to some extent by providing a semi-fluidized packing, usually of plastic spheres. The spheres rotate and jiggle slightly, constantly exposing clean areas which collect particles. Higher gas velocities, a result of fluidizing the packing material, increase turbulence and gas- liquid contact. Pressure drops for these scrubbers are about double a conventional packed bed.21 4.2.1.3 Mechanically Aided Scrubbers. Mechanically aided scrubbers rely on a motor driven device between the inlet and outlet of the scrubber body to effect particle removal. This device also serves as the fan which draws air through the scrubber. Particles are collected by impaction upon the fan blades as the gas flows through the scrubber. Liquid is typically introduced at the hub of the rotating fan blades. Some liquid atomizes upon fan impact, while some runs over the blades, washing them of collected particulate. This latter portion atomizes as it leaves the fan wheel. The liquid is recaptured by the fan housing, which drains into a sump. Figure 4-3 shows an example of a mechanical centrifugal scrubber. The performance of this scrubber is influenced by the total energy input to the fan and the liquid flow rate provided. Higher fan velocities generally cause greater impingement of particles on the fan blades. Likewise, increased liquid flow rates increase particle removal.”@ 4.2.1.4 Tray Type Scrubbers. A tray type scrubber is shown in Figure 4-4. It consists of a vertical tower containing one or more transversely mounted trays. Particulate laden gas enters the tower bottom and bubbles through valves, perforations, or other types of openings in each tray before exiting through the top of the tower. Scrubbing liquid is usually introduced at the top tray, and flows across each tray, over a restraining dam, and through a downcomer to reach the tray below. The particulate laden liquid exits the bottom of the tower. Gas passes through the openings in each tray and bubbles through the liquid flowing over the tray. Liquid-gas contacting causes the mass transfer and particle removal.WATER ‘SPRAYS 70.2036-1 aren 0.208. DRAINS Figure 4-3. Typical mechanically aided scrubber. 4-12CLEAN GAS OUTLET t LIQUOR INLET DIRTY GAS INLET Liquor OUTLET 7020641 Figure 4-4. Typical tray-type scrubber. 4-13As the diameter of the tray perforations decreases, the collection efficiency for smaller particles usually improves. A tray type scrubber does not have the same efficiency for all particle sizes, but instead exhibits a sharp efficiency drop at a specific particle size. This size is determined by the size of the tray perforations.2> The pressure drop through a plate type scrubber is determined by the size of the orifices, the number of trays, and the velocity of the gas stream through the scrubber. In general, higher pressure drops result in greater efficiencies.2¢ Although, some texts indicate that increasing the number of trays has little effect on particulate removal,“°manufacturer's performance curves show an increase in removal with more trays. Figure 4-5 illus- trates this effect for a tray type scrubber used in the urea industry for a variety of particle sizes. The efficiency on the vertical axis, termed standard efficiency, is for a standard .375 kPa (1.5 in. W.G.) pressure drop per tray. Figure 4-6 illustrates the effect of increasing the pressure drop across each tray. For any given standard efficiency at .375 kPa (1.5 in, W.G.) per tray obtained from Figure 4-5, the efficiency at higher pressure drops may be read from Figure 4-6. Liquid flow rate can have some effect on particle removal, however, an optimum flow rate is usually maintained which insures adequate liquid for particulate removal without blocking airflow through tray orifices. Typical liquid to gas ratios are .0669 - .401 liters/m? (.5 - 3 ga1/1000 #t3) at 172 kPa (25 psig) liquor pressure.7© 4.2.1.5 Entrainment Scrubbers. Entrainment scrubbers (also referred to as orifice type, self-induced spray, or impingement and entrainment scrubbers) utilize the velocity of the gas stream over the surface of a liquid, in combination with a sudden change in direction of the ges flow, to remove particulates. A common entrainment scrubber used in urea industry applications is shown in Figure 4-7. The gas stream enters the circular housing, where it is forced through a narrow gap formed between the surface of the sump liquor and an inner housing.Soe Standard Efficiency 100 99 98 97 9 8B 94 93 92 9 90 89 88 87 86 A ws 2 We 1s Particle Size, Microns Figure 4-5, Standard fractional efficiency for tray type scrubber. 10gI-p Standard Efficiency 100 98 94 92 90 88 86 84 82 80 90 91 92 B 94 95 96 97 Nigh Pressure Drop Efficiency Figure 4-6. Effect of pressure drop on tray type scrubber efficiency 98 99 100OUTLET SPINNER VANES INLET ——> Liquor INLET 70.2097-1 LIQUOR OUTLET Figure 4-7. Typical entrainment scrubber.°° 4-17A turbulent zone is established at this gap promoting spray droplet formation and dispersion as the gas abruptly changes directions. The moisture laden stream is then demisted by swirl vanes before exiting the scrubber. The primary factor influencing the performance of the entrainment scrubber is pressure drop across the device. The effect of pressure drop and particle size on scrubber efficiency is illustrated in the fractional efficiency curves presented in Figure 4-8. Entrainment scrubbers used in the urea industry operate at widely varying pressure drops depending on the application. 4.2.1.6 Fibrous Filter Scrubbers. A type of wetted fibrous filter scrubber has recently been installed and operated to control prill tower emissions. The device is depicted in Figure 4-9. The scrubber utilizes a filter installed over a perforated drum. The filter drum rotates slowly through a shallow liquor bath and is also irrigated by spray nozzles located throughout the drum chamber. Gases pass from the exterior of the drum, through the wetted filter, and into a mist eliminator section. The demister housing is a horizontal cylinder with an inclined demister elenent located near the scrubber exit. Flow through the prill tower and scrubber is induced by an axial fan mounted downstream from the demister section. The filter itself is a dense fibrous mat. For prill tower control, a Teflon® mat is used.27 As the particles travel at high velocity through the filter, they impact the wetted fibers and are held until they are washed either by the sprays or the bath at the bottom of the filter housing. The design of the wetted fibrous filter allows the pressure drop to be readily adjusted while the scrubber is in operation. This adjustment is possible through the use of a moving, semi-cylindrical baffle plate which may be used to cover a fraction of the filtration drum. By covering a portion of the drum face, the airflow is forced to travel through a smaller area on the drum which increases face velocities. These higher velocities result in greater impingement of particulate on the filter 4-18COLLECTION EFFICIENCY *. BY WEIGHT Figure 4-2. COLLECTION EFFICIENCY VS PARTICLE SIZE PARTICLE OUMETER IN MICRONS. Fractional efficiency of entrainment scrubber used in the urea industry as a function of particle size and pressure drop (courtesygf the Western Precipitation Division of Joy Nenufacturing Company) .° 4-19Oar Mist ELIMINATOR FILTER MEDIA women WNLET DEMISTER CHAMBER, — yo20001 ROTATING PERFORATED FILTRATION DRUM uauor SPRAYS Figure 4-9, Typical fibrous filter scrubber.°°mat and increase removal efficiency at the expense of higher pressure drop. The baffle may also be used to hold the pressure drop constant at various airflows through the scrubber. This feature allows collection efficiencies to be maintained while producing different grades of product which require different airflows. Figure 4-10 presents the efficiency of the wetted fibrous filter as a function of particle size. This curve was obtained during prill tower testing of a Teflon® Filter scrubber operating at 4.75 kPa (19 in. W.G.). According to the vendor, improvements have been made to the scrubber since this test, which allow this performance curve to be met with pressure drops in the range of 3.0 - 3.75 kPa (12 - 16 in. W.G.).28 The effect of pressure drop on particulate removal efficiency is illustrated in Figure 4-11. The plant where the wetted fibrous filter is used to control prill tower emissions uses a preconditioning system involving liquor injections in the ductwork prior to the scrubber. This preconditioning systen is reported to cause particle agglomeration prior to the scrubber and thus increase the scrubber's effectiveness. Details of the preconditioning system are considered proprietary by company personnel. The wetted fibrous Filter may be operated at from 2.5 - 7.5 kPa (10 - 30 in. W.G.) differential pressure drop across the device. Typical liquid recirculation requirenents (scrubber only) are .134 - .267 liters/m? of gas (1 - 2 ga/1000 #t3). Spray nozzle pressure is approximately 138 kPa (20 psig).22 4.2.2 Fabric Filters Fabric filters (baghouses) are high efficiency collection devices used quite extensively throughout the chemical processing industry. Design variables for baghouses include method of cleaning, choice of fabric, size of the unit, air-to-cloth ratio, and whether the baghouse is a pressure or suction unit. Figure 4-12 depicts a typical fabric filter system. In the type of design shown, the airstream enters the baghouse and is pulled up into 4-21ere 100 99 98 97 96 95 94 93 92 91 90 89 88 87 4.6.8 1.0 1.2 14 1.6 1.8 2.02.2 2.4 2.6 2.8 3.0 3.2 Particle Size (microns) Figure 4-10, Fractional Efficiency of Wetted Fibrous Filter Scrubber5® 3.43.6 3.8Overall Efficiency 100 98 7 6 95 94 93 92 a 90 89 88 87, 86 85 83 82 Figure 4-11. 10 15 20 25 30 Pressure Drop, in. W.G. Effect of pressure drop og efficiency of wetted fibrous filter scrubber. 97 4-23BRANCH HEADER CLEAN AIR, OUTLET BAFFLE PLATE PYRAMIDAL OR TROUGH HOPPERS ACCESS DOOR, 70-2025-1 Figure 4-12. Diagram of a Fabric Filter.°> 4-24fabric sleeves located throughout the baghouse. Air is pulled through these fabric sleeves and exhausted to the atmosphere while dust remains trapped in the weave of the fabric, forming a layer of dust on the bag. The pressure drop through the bag increases as this dust layer builds up. The dust is periodically removed from the bag by one of several bag cleaning methods. Two methods of cleaning are shaking (rapping) and reversing the airflow through the bag by air jets or pulses. Shaking consists of manually or automatically shaking the bag hangers or rapping the side of the baghouse to shake the dust free from the bags and into a receiving hopper below. In the jet pulse method, compressed air is released at regular intervals in to a group of bags, causing the bags to pulse and the dust to be released. Cleaning can be either continuous or intermittent. Intermittent cleaning consists of shutting down the baghouse or a section of the baghouse when it reaches its highest design pressure drop, For con- tinuous cleaning, individual bags are cleaned at regular, timed intervals. An important operating principle for fabric filters is that effective Filtering of the dusty airstream is accomplished not only by the fabric, but also by the dust layer which forms on the fabric. This dust layer bridges the gaps between adjacent fibers and increases the chances of impaction and interception of small particles. For this reason, too frequent cleaning can actually decrease efficiency by not allowing a dust layer to accumulate between cleaning cycles. The urea dust layer can cause problems in urea industry applications due to the hygroscopic nature of urea particulate. The dust layer can absorb moisture in the air and cause the formation of a sticky cake. This cake increases the pressure drop and can cause difficulties in cleaning. For this reason, use of baghouses in the urea industry is currently limited to process airstreams with low moisture contents, such as bagging operations. 4-25Materials available for bag construction are numerous. They include cotton, Teflon®, coated glass, orlon, nylon, dacron and wool. Temperatures, frequency of cleaning, ease of removing particles, resis- tance to chemical attack, and abrasion characteristics of the collected particles determines the type of bag fabric material. Factors affecting baghouse performance include air to cloth ratio, type of fabric used, method and interval of cleaning, pressure drop, and the properties of the dusty exhaust being cleaned. Air to cloth ratio is dimensionally equivalent to a velocity, and thus indicates the average face velocity of the gas stream through the effective area of the fabric. An excessive filter ratio results in excessive pressure loss, reduced collection efficiency, rapid bag blinding, and increased wear on the fabric. Too low an air to cloth ratio results in an over- size unit and can also reduce collection efficiency since an adequate filtering dust dayer may not be allowed to accumulate between cleaning cycles. Pressure drops in baghouses depend on a variety of factors including the air to cloth ratio, fabric type, and cleaning cycle. Pressure drops typically increase between cleaning cycles as the dust layer builds. Pressure drops of from .5 - 2 kPa (2-8 in. W.G.) are common for many applications.2° Air to cloth ratios range from 2 to 10 with 3 being the typical ratio reported in the urea industry. Methods used in the industry for cleaning baghouses include mechanical shaking, reverse pulse airflow, and vibration. Types of cloth material commonly used in the industry include cotton, dacron and polyester. #75s6 4.3 EMISSION TEST DATA Available data concerning control device performance is broken down into two basi types: data supplied by industry and state air pollution agencies (hereafter referred to as industry data), and data collected by EPA during source testing conducted for this study (hereafter referred to as EPA data). In general, the available industry data is very limited. Industry data presented in this section is confined to mass emission measurements of prill towers and coolers. It should be noted that 4-26the industry data vary widely in test procedures and sampling techniques. In particular, significant difficulties exist in sampling emissions from prill towers due to their design. Comparisons between the two types of data are not intended to imply that the sampling and test procedures are similar. Tables 4-2 and 4-3 present an overall summary of EPA mass emission test results and visible emission test results, respectively. Appendix A presents details of the test data and testing program. As can be noted in Table 4-2, control of ammonia emissions to a significant degree is not currently demonstrated in the urea industry. In fact, most test data indicates an increase in ammonia emissions across the control device. Control of formaldehyde emissions is quite variable, however, the level of formaldehyde in the control device inlet is usually quite low to begin with. Because ammonia control is not currently demonstrated in the industry and formaldehyde emissions are small, the following subsections will address control device performance in terms of particulate removal only. 4.3.1 Emission Data for Nonfluidized Bed Prill Towers Table 4-4 presents the available industry data for controlled nonfluidized bed prill towers consisting of four tests conducted at three plant sites. Two of these three plants were also tested by EPA. (Plant £ and Plant C.) Test results for Plant E are presented in Appendix A, Tables A-63 through A-65, Tests at Plant E represent measurements of a nonfluidized bed prill tower producing agricultural grade product. This plant uses a wetted fibrous filter. The tests (Appendix A, Tables A-63, and 64 versus Tables A-65 and 66) differ in the type of precondi- tioning sprays used in the ductwork prior to the scrubber. Tables A-65 and 66 represent full preconditioning as the plant normally operates and Tables A-63 and 64 represent testing with partial preconditioning. Preconditioning is used to encourage particle agglomeration prior to the scrubber. According to these tests, full preconditioning shows improvement in outlet mass emissions (0.22 g/kg and .044 1b/ton) compared to the partial preconditioning (.320 g/kg and .640 1b/ton). 4-278z-y TABLE 4-2, SUMMARY OF EPA MASS EMISSION TEST RESULTS Particulate Emissions __-famonia_ fmt sstons. Formaldehyde Emissions Outlet Collection ka/My Efficiency Control control Contre) Controt Controt Control Production Device’ —_pevice Device Device Device Device fate Inlet’ Outlet Collection —_—Inlet_ Outlet Collection —_Infet contro} Mo/day, Koy. kag Effevency —ka/Mg_ka/g.fFictency —_ka/Ma Process Device Plant (ton/day) __(Ih/ton)—_(Ib/ton) x (ibjten) —_(1b/ton) a (ib/ton) —_(1b/ton) NB Pri) Packed c 768 ne 188 Na ne 0.640 na ub NA Tower’ AG Tower. (25) (3975) Ginza) Product Serubbers Fa Pritt Wetted c 266 1.80 oe 98.3 0.326 2.18 <0 mw m Tower - AG Fibrous (293) (8.76) (C041) ([653) (4:36) Product "Filter fe rei entrain > 979 ae 392 876 1.39 3.25 <0 0510 000820 Tower = AG ment (1078) (6.24) (1785) (278) (6:50) (C0182) (000839) Product Scrubber Fo priv Entyaing «= 4020 1.80 240 6.7 1.98 1.04 7 9190 .900500, over = FS tent (23) 355) ("479) (3:98) (208) (00380) (000999) Product Serubber rum Entraine 356 149 15 99.9 1.00 3.07 <0 00359, 00136 Granulator_ went 92) (298) (:230) (26) (6.14) (C00717) (00871) Scrubber Oram Entrain- 8 sae 63.6 see 99.8 1.07 20.5 000380000125 Granutator went (272) (248) (213) (000560) (“o00260) Serutber inlet tests not concucted during outlet tests. Prommaldehyde tests rot conducted. Earlier inlet tests considered nonrepresentative (see Appendix A) production rate considered confidential by conpany. Data 1s averaged for tests A-1 and A-2 Legend: AG - Agricultural Grade Fe - Feed Crade HFG = Nonfluidized bed fh - Flutdized bed NA = Not available % WA NA 95.4 me 62.2 50.262-7 TABLE 4-3. SUMMARY OF EPA VISIBLE EMISSION TEST RESULTS Contro? Opacity Measurements (% Process Device Plant Observations@ Min. Max. Average NFB Prill Tower - Packed c 158 2 37.0 9.68 AG Product Tower NFB Prill Tower - Packed c 42 5.0 54.4 16.8 FG Product Tower NFB Prill Tower - Fibrous Je 179¢ 0 274 9.33 AG Product Filter FB Prill Tower - Entrainment D vb 10.0 41.2 25.3 AG Product Scrubber FB Prill Tower Entrainment D 106? 3.3 33.3 20.8 FG Product Scrubber Drum Entrainment A 79 0.0 5.0 2.92 Granulator Scrubber Drum Entrainment B 62 5.0 9.7 7.62 Granulator Scrubber Bagging Fabric D 35 0.0 1.0 +05 Operations Filter Rotary Drum Packed = E 10 1.0 44 3.0 Cooler Tower Rotary Drum Entrainment c so 15.0 27.0 22.0 Cooler Scrubber six minute average includes measurenents both on individual scrubbers and on pril] tower as a whole. “data is for both tests conducted at Plant E. Legend: NFB - Nonfluidized bed AG - Agricultural Grade FB - Fluidized bed FG - Feed Gradeoer TABLE 4-4, SUMMARY OF INDUSTRY MASS EMISSION TEST RESULTS FOR CONTROLLED PRILL Towers? *48 article Emission Emission Production Aigtiow —cComentrgtion “Rate” factor’ Device Tower ty/tay control ds/mtn fds gfmin, egg eFFictency Tyre Product (tons/day) Device ‘dserm) ——(ar/dser)—_—_(Mb/hr)__(ab/ton) % t mo AG zn Wetted 75.8.75 2,410 001s an 127 90.8 (toy Fis tori} (ass000) (0008) ($03) C051) Filter " woe 210 : 3,750 00506, 20 46 7 (vo towers) 240) (1327400) (00256) (2.5) C291) nro 909 - 10,200 00701 n.2 103 2 (1100) (959/000) (00206) (9.41) 205) wa AG 2 Packed 5 1,530 00701 1.0 0500 a6 G00) Tower @) (637500) Co0st0) (iis) 16) v 0 6 923 Entratrment 1.25 8,520 -oxei 23 426 6.3 (two products) (1017) Scrubber (01000) (0140) G6.) (852) 6 "6 946 Entratmment 1.25 12,470, 0198 27. a7 72.0 (1022) Serubber 5) (440200) (00867) G2 (753) we 8 - 629 Spray - 3,790 0190 185. 425 - (693) Tower (1337800) (C214) (20d) C889) resting method usedwas stamiard EPA Hettod resting done using wet inpinger train. Pressure drop daring test not avaiTable, Pressure dron given w: scared during EPA test. Device efficiency Iesed'an uncontrolled enlssions as given in Chapter 3. of control device, pressure drop, and uncontrolled enfsston rates are not available, Type of product, Cheoninetic sanpling used, however, analysts aethad 1s unkrom. Pressure drop not recorded during test. Pressure drop given measured during EPA test dieacer gas used to deternine flow rates and mintain tsokinetic sampling. Met Impinaer train used. tyne of product and pressure drop unavat lable NIB ~ Nontluidized bed TB = Muidized bedThe EPA test results for Plant C are presented in Appendix A, Tables A-26 and A-27, These results represent measurements of a nonfluidized bed prill tower producing agricultural grade product. The emission contro] system at this plant consists of four packed bed scrubbers operated in parallel. One scrubber was tested and the total emissions were determined by factoring the single emission measurement by four. This assumes that the tested scrubber is representative of the remaining three. Velocity traverses and visible emission observations of the untested scrubbers show this assumption to be reasonable (see Appendix A). Particle size tests were conducted at Plants C and £ on prill tower exhausts entering the scrubbers. At Plant C, tests were run during production of both agricultural and feed grade production. This data, presented in Figure 4-13, shows a shift toward larger particles during feed grade production as evidenced by a shift to the left of the cumulative distribution plot. At Plant £, the particle size distribution during agricultural grade production (Plant E does not produce feed grade urea) was measured. This data is presented in Figure 4-14. One industry particle size test for a nonfluidized bed prill tower is available and is presented in Figure 4-15. This data also shows a shift toward larger particles during feed grade production. Visible emission data were gathered by EPA during the tests at Plant C and E. Figure 4-16 presents histograms of the visible emission data collected at Plant C during both agricultural grade and feed grade tests. Opacity during feed grade production averaged somewhat higher. During feed grade production at Plant C, prill tower fans are shut off and the tower airflow is induced by natural draft only. This results in lower air velocities and pressure drops in the scrubbers, which may contribute to higher opacity readings during feed grade production, Figure 4-17 presents histograms of the visible emission data collected at Plant E, These tests both involved agricultural grade production and differ only in the preconditioning system used. Opacity readings were generally higher with full preconditioning. 4-31IAMETER, MICRONS PARTICLE 100.04 (© AGRICULTURE GRADE s004 4 FEED GRADE 4004 NOTE: THESE RESULTS REPRESENT AVERAGES OF cal THREE RUNS FOR EACH TEST 2004 1004 504 aod 204 204 v4 7 s+ 1 Ito —T +206 1 2 8 10 152 a 4 50 60 70 4085 (CUMLATIVE PERCENT LESS THAN SIZE Figure 4-13, Particle size distribution of uncontrolled NFB prill tower exhaust (Plant C). 4-32PARTICLE DIAMETER, MICRONS 10095 NOTE: THESE RESULTS REPRESENT AN soo AVERAGE OF THREE FUNS, 4004 00-4 wo 504 404 304 204 I 7 2 5 1 152 9 4 $0 Go 7 6085 9% 95 98 (CUMULATIVE PERCENT LESS THAN SIZE Figure 4-14. Particle size distribution of uncontrolled NFB prill tower exhaust (Plant E). 4-33PARTICLE DIAMETER, MICRONS. S004 4004 2004 2004 1004 50+ 40-4 204 AGRICULTURE GRADE —— Feep crave T T % 4 $0 6 7 wa 8 OB (CUMULATIVE PERCENT LESS THAN SIZE Figure 4-15. Particle size distribution of uncontrolled prill tower exhaust (Plant F - industry data)64 4-34No. of Observations No. of Observations 80 70 O- 5.1- 7 10 15 toh 12 Figure 4~16, Agricultural Grade Average Opacity = 9.68% VO.1- 15.1- 20.1- 25.1- 30.1- 35.1- 40.1- 45.1- 50.7 15 20 25 30 35 40 45 50 56 Intervals of Opacity (%) Feed Grade Average Opacity - 16.8% 9 6 7 LI 4 2 [] 9 C7 WO.t- 15.1- 20.1- 25.7 30.1- 35.1- 40.1- 45.1- 50.1- 15 20 25 30 36 40 45 50 55 Intervals of Opacity (3) Histograms of six minute opacity averages for controlled non-fluidized bed prill tower exhaust, Plant C.No. of Observations No. of Observations 45 40 0-5 45 40 35 0-5 Figure 4-17, 5.1- 10.1- 10 15 45 5.1- 10.1- 10 15 Partial Preconditioning Average Opacity - 7.16 15.1- 20.1- 25.1 30.1- 20 25 30 35 Intervals of Opacity (%) Full Preconditioning Average Opacity = 11. 4 15.1- 20.1- 25.1- 30.1- 20 2 30 35 Intervals of Opacity (%) Histograms of six minute opacity averages for controlled nonfluidized bed prill tower exhausts, Plant £. 4-364.3.2 Emission Data for Fluidized Bed Prill Towers Three fluidized bed prill towers are currently operating in the United States and all are controlled to some degree. Industry data on controlled emission levels at two of the three installations is sunmarized in Table 4-4, Industry data on uncontrolled emissions from Plant D's prill tower generally agrees with EPA data. EPA tests at Plant D are summarized in Appendix A, Tables A-40 through A-47. Both agricultural grade and feed grade production were tested. These tests involved measurements on two of eight scrubbers operated at this installation. The total emission rate from all operating scrubbers was calculated by factoring each scrubber emission valve by four. This assumes that the tested scrubbers are representative of the remaining untested scrubbers. Velocity and visible emission measurements show this assumption to be reasonable (see Appendix A). A histogram of the visible emission data collected during the testing at Plant D is shown in Figure 4-18, The distribution of readings changes slightly between agricultural and feed production, and the average opacity is 4.5 percent higher during agricultural grade production. Particle size distribution information was obtained at Plant D during testing. Two tests (each consisting of three runs) were made for both agricultural and feed grade production. This data is presented in Figure 4-19, In general, feed grade production shows increases in particle sizes over agricultural grade production similar to the trend noted on nonfluidized bed pril1 towers. 4.3.3. Emission Data for Rotary Drum Granulators Mass emission tests were conducted by EPA on three drum granulators. At Plant A, granulator "A" was tested twice and granulator "C" was tested once. During granulator "C" testing, a variety of factors which potentially could affect test method accuracy were investigated and no particle size or visible emissions were measured. At Plant B, one test was conducted which included uncontrolled, controlled, visible, and particle size emission measurements. 4-37No. of Observations No. of Observations . Agricultural Grace 38 Avg. Opacity = 2 0 9 8 2 23 2. 20 15 1 10 2 ° 3 oL a 7 a a a a Ws 1 ze 2919 alsa. Intervals of Opacity 3 Feed Grade x Avg. Opacity = 20.8 x0 se 25 20 13 3 z y 2 10] i 5 0 0 0 0 0 EO. 000 De 49 9.9 "9 18.828 299389389 a Intervals of Opacity Figure 4-18. Histograms of six-minute opacity averages for 4-38 controlled FB prill tower exhaust (Plant 0).PARTICLE DIAMETER, MICRONS. 000: 500: 100: m0: 200. 100. 50: 40. 30 20 © AGRICULTURE GRADE, SCRUBBER A 4 AGRICULTURE GRADE, SCRUBBER G FEED GRADE, SCRUBBER © FEED GRADE. scruBBERC NOTE, THESE RESULTS REPRESENT AVERAGES OF THREE RUNS FOR EACH TEST TI Stoo. 5 0 2 9 0 0 o 7 Oe (CUMULATIVE PERCENT LESS THAN SIZE Figure 4-19. Particle size distribution of uncontrolled FB prill tower exhausts (Plant D). 4-39In testing granulator "A" at Plant A, and Plant B very high removal efficiencies (above 99.8 percent) were demonstrated. One reason for the high efficiency of granulator scrubbers is the large particle sizes found in granulator exhausts where several particle size tests were conducted, showing that less than 1 percent of the total emissions in granulator exhausts were less than 5 microns in size Opacity measurements on all tests were low. Opacities during tests on granulator "A" at Plant A ranged from 0 to 5 percent. Opacities at Plant 8 were between 5 and 10 percent, 4.3.4 Emission Data for Rotary Coolers No EPA test data is available to determine controlled emission rates from any of the devices used to control cooler emissions. Industry has reported emission rates, however, and this data is summarized in Table 4-5. An average of this data results in an emission rate of .035 kg/Mg (.07 1b/ton) EPA tested the uncontrolled rotary cooler exhaust at Plant C and measured emissions of 3.73 kg/Mg (7.45 1b/ton). Plant C personnel have measured controlled cooler emissions of .01 kg/Mg (.02 1b/ton). According to this data, the mechanically aided scrubber used at Plant C is achieving an overall efficiency of 99.7 percent. A particle size test was also conducted on the uncontrolled cooler exhaust at Plant C. As can be seen in Figure 4-20, the particles are large, with less than 0.3 percent smaller than 10 microns Visible emissions measurements were conducted on the scrubber outlets of rotary drum coolers at plants C and £, These measurements are summarized in Table 4-3 and presented in Appendix A. A mechanically aided scrubber is used to contro] cooler emissions at Plant C with the average opacity reported to be 23 percent. Cooler emissions at Plant & are controlled to an average 3 percent opacity by a packed bed wet scrubber. 4.3.5 Emission Data for Bagging Operations Mass emission test data are not available for fabric filters controlling emissions from a urea bagging operation. However, regardless of the type, baghouses can attain collection efficiencies greater than 99 4-40lo-t TABLE 4-5, SUMMARY OF COOLER CONTROLLED EMISSIONS (INDUSTRY DATA) Reported Reported Plant Outlet Emission Pressure Type of Where Rate Drop Control Device Model Used “Ko7g STEER” = KPaS (in. WAG.) Packed Tower®? Buell Flyash H -01-.015 (.02-.03) 225 (1) Tray Type®! sly I a (.2) Not Available Packed Tower®” American Air E +02 (.04) 225 (19) Filter, Hydrofitter Mechanical ly? Anerican Air c +01 (02) Not Applicable® Nided Filter, Rotoclone Type W “Mechanically aided scrubber supplies power through intergral rotor.PARTICLE DIAMETER, MICRONS 100.04 s004 004 200-4 200. wo so4 404 20-4 204 NOTE: THESE RESULTS ARE AN AVERAGE OF THREE RUNS, Figure 4-20. a 1 mo > > he Oe CUMULATIVE PERCENT LESS THAN SIZE Particle size distribution of uncontrolled cooler exhaust (Plant C). 4-42percent even on submicron particle sizes.°* Testing conducted by EPA on baghouses used to control emissions in the non-metallic mineral industry demonstrated efficiencies of 99.8 percent or better with no visible emissions (zero percent opacity) .23934%35 Opacity measurements were made at Plant D to determine visible emission levels from fabric Filter controlled bagging operations. Visible emissions were usually nonexistant. The average opacity during this test was .05 percent. 4.4 EVALUATION OF CONTROL DEVICE PERFORMANCE This section presents an evaluation of the emission data presented in Section 4.3, This evaluation includes: 1) a general examination of the data to determine relative accuracy and representativeness, and 2) an assessment of the effects of changes in emission characteristics on control device performance. As in the previous section, the discussion is arranged by emission source. 4.4.1 Nonfluidized Bed Prill Towers Available EPA test data consist of tests of nonfluidized bed prill towers (producing agricultural grade urea) at Plants C and E as illustrated in Figure 4-21, Plant C was tested during both agricultural and feed grade production in April, 1979; however, the analysis was improperly conducted and the data unusable. During a subsequent retest it was possible to test emissions during agricultural grade production only. At Plant E, feed grade emissions could not be measured since this plant produces agricultural grade only. Thus, no EPA emission data for nonfluidized bed prill towers producing feed grade product is available. The two scrubber outlet emission tests (each consisting of three test runs) at Plant £ measured the lowest controlled emission level of any prill tower tested by EPA, The two tests measured emission rates of +0320 kg/Mg and .0220 kg/Mg (.0641 1b/ton and 0440 1b/ton). The wetted fibrous filter demonstrated an average removal efficiency of 98.7 Percent based on uncontrolled emissions measured simultaneously with the first outlet emission test. 4-43bey © Test Sampling Run 1 Test Average 5 1.0 4 +8 2 3 68 2 = = 42 428 5 é al a2 0 4 8 4 4 4 0 c El E-2 D-1 D-2 D-3 D4 Test ¢: Plant C, nonfluidtzed bed, agricultural grade packed Test O-1: Plant 0, fluidized bed, agricultural grade, bed’ scrubber entratntent scrubber "A" Test €-1: Plant E, nonfluidized bed, agricultural grade, Sane as 0-1 except scrubber *C* wetted Hbrous filter, partial preconditioning Same as D-1 except feed grade Test £2: Sane as E-1 except full preconditioning Same as 0-2 except feed grade Figure 4-21. mission levels from controlled prill tower, EPA testsThis removal efficiency is confirmed by the performance curve provided by the control device manufacturer. This fractional efficiency curve, based on pilot plant evaluations of a prototype wetted fibrous Filter unit and presented earlier as Figure 4-10, gives the particle renoval efficiencies for various size ranges of particles. Using these removal efficiencies and the uncontrolled particle size distribution measured at Plant E (Figure 4-14), a removal efficiency of 98.4 percent is predicted. This compares favorably with the 98.7 percent actually measured. An independent test conducted by company personnel at Plant E, under conditions similar to those during EPA testing, confirmed the results obtained during EPA testing. They measured controlled emissions of .0271 kg/Mg (.0541 1b/ton) (see Table 4-4), which is very close to the EPA measured emissions. A comparison of controlled emissions between Plant & and Plant ¢ reveals considerably higher emissions at Plant C. These higher emissions are believed to be the results of two factors. First, the wetted Fibrous filter used at Plant E operates at a much higher pressure drop and is specifically designed for control of particles less than 1 micron.36 In contrast, packed bed scrubbers are typically used for control of gaseous pollutants or in situations where the particles tend to be larger than 5 micron. Secondly, the prill tower exhaust at Plant C contains a much higher percentage of fine particles. A comparison of the internal air velocities between the two prill towers reveals considerable differences which might account for the change in particle size. The velocity in Plant C's tower is approximately .365 m/s (1.2 ft/sec),37 while the velocity in Plant E's tower is approximately 1.29 m/s (4.24 ft/sec).°8 Higher airflow increases the entrainment of particulates and may also increase the generation of small particles. A higher tower airflow would increase the turbulence near the molten prill resulting in the formation of small, solid urea particles. EPA tests are usually conducted on facilities which appear to be using the best available control technology. Plant C was selected 4-45for testing because at the time of the test, it was the only plant known to operate a nonfluidized bed prill tower which offered reasonable sampling locations. Mistakes in the sample analysis during the first test at Plant C resulted in erroneous data. By the time it was conclusively determined that this data was faulty, the test at Plant E was in the planning stage. Therefore, the retest at Plant C was limited to the minimum necessary to confirm the errors in the initial test. Therefore, only controlled emissions during agricultural grade production were retes ted. As mentioned earlier, no EPA mass emission data is available to quantify either controlled or uncontrolled nonfluidized bed prill tower emissions during feed grade production. However, a number of factors infer that feed grade emissions are easier to control compared to agricultural grade emissions. First, all available particle size data shows that larger particles are generated during feed grade production. Larger particles are more effectively removed by control systems than small particles. A comparison of the size distributions measured at Plant C during both grades of production (Figure 4-13) shows a clear shift toward larger particles during feed grade production. This shift is confirmed by a similar shift noted for fluidized bed prill towers (Figure 4-10) and industry particle size tests of a nonfluidized bed prill tower (Figure 4-15). Secondly, available data indicates feed grade emissions are probably lower than agricultural grade emissions. The only direct comparison between uncontrolled emission rates during both types of production using a nonfluidized bed prill tower is industry tests at Plant F (see Table 3-3). These tests measured 15 percent higher uncontrolled emissions during feed grade production. However, due to problems frequently encountered in testing prill towers and differences in test methods, industry data is difficult to quantify. The particle size data from this same plant indicates the control device efficiency would be improved because of the larger feed grade particulate. Increased efficiency would more than compensate for the slightly higher uncontrolled emissions during feed grade production. EPA tests of a 4-46fluidized bed pritl tower showed a 40 percent decrease in emissions in switching from agricultural to feed grade production. Since both types of prill towers are generally operated in a similar manner, with similar reductions in airflow during feed grade production, this decrease in emissions is believed to be typical for prill towers producing both product grades. Another variable believed to effect the uncontrolled emission characteristics of nonfluidized bed prill towers is anbient temperature. Available data indicates that colder temperatures promote the formation of smaller particles in the prill tower exhaust.29*49 since smaller particles are more difficult to remove, the efficiency of control devices used on prill towers tends to decrease with lower temperatures. This can lead to higher controlled emission levels while prill towers are operated during cold weather. The physical mechanism responsible for this shift toward smaller particles is not clearly understood. Available industry and EPA particle size data indicates the existence of two distinct populations of particles in prill tower exhausts. One population is greater than 5 micron in size and is composed of small micro prills and prill fragments, This population is believed to be formed as the molten urea separates into droplets at the melt distributor and as the semi-solid prills strike each other and the prill tower walls. The second population is smaller than 5 micron and is believed to result from the condensation of urea vapor into small crystals, Microscopic examination of the small particles reveals a crystalline structure, similar in appearance to a snowflake.*2 Urea vapor pressure data indicates that sufficient urea is present in the vapor state in a prill tower to account for the small particle emissions.4? The growth of urea crystals is directly affected by the rate at which the urea vapor is cooled, while this rate is directly affected by the temperature of the air drawn into the prill tower. During cold weather the vapor is quickly cooled from melt temperature to the exhaust duct temperature, while in warm weather the transition is more gradual. Rapid cooling does not allow time for the formation of 4-47larger crystals in the saturated urea vapor. Instead, many small crystals are formed as the saturated urea vapor cools. Personnel at Plant E conducted fifteen particle size tests during a control device evaluation program.*? This program involved the ducting of a slipstream from the prill tower exhaust to a ground mounted samp] ing location. Because the sampling technique did not account for segregation of particles within the slipstream due to flow direction changes caused by the ducting, only the fraction of particles 5 micron in size are in size considered in the following discussion. The effect of this exclusion on the final controlled emission levels presented later is negligible since prill tower control devices typically remove nearly ali particles of size larger than 5 micron. Ambient temperature during these fifteen tests ranged from 27 degrees F to 70 degrees F. Histograms of the two tests representing the highest and lowest temperatures are presented in Figure 4-21 and illustrate the increased fraction of particles in the smaller size ranges during cold ambient temperatures. This general trend is evident in all fifteen tests. Figure 4-22 presents a plot of mean particle size vs. ambient temperature for all fifteen tests. Although some scatter is evident, there is clear indication that temperature influences the size of particle in prill tower exhausts. The effect of this shift in particle size on the efficiency of the Wetted fibrous filter scrubber is illustrated in Figure 4-23. Two approaches were used in determining this trend. The first approach involves using each of the fifteen particle size tests in conjunction with the fractional efficiency data presented earlier in Table 4-13 to predict the efficiency for each specific particle size distribution. The individual data points in Figure 4-23 represent this approach. The second approach, represented by the solid line in Figure 4-23, involves assuming a log-normal distribution for the sub 5 micron particles in prill tower exhausts. Nearly all fifteen actual particle size distributions fit the log-normal distribution. Furthermore, several authors indicate that particle sizes are frequently log-normally distributed. 44945564 4-486o-b Mean Particle Size of Sub 5 ym Fraction (um) 20 30 40 50 60 70 80 90 Anbient Temperature (degrees F) Figure 4-22 Variation in particle size (sub 5 ym mean) with respect to ambient temperature (industry data)0s-p Efficiency (percent) 100 90 80 60 50 © Based on Actual Particle Size Tests — Based on Log-Normal Distribution 10 20 30 10 50 0 10 80 be Ambient Tenperature (degrees F) Figure 4-23, Efficiency of wetted fibrous filter as a function of ambient temperatureWith this assumption, it is possible to calculate the particle size distribution for any temperature by varying the mean in accordance with the trend line in Figure 4-22. The preceding discussion has centered on the effect of tenperature on particle sizes. Temperature may also affect the quantity of emissions. Particle concentration data obtained during the fifteen particle size tests at Plant E shows that concentration of particles in the prill tower exhaust is relatively constant with respect to temperature at approximately .039 o/m? (.017 gr/acfm). However, prill tower operators typically cut back airflows during cold weather conditions to save fan power since less of the colder air is required for adequate cooling of the prills. From heat transfer considerations, it is possible to predict this cutback. The results of these calculations are presented in Figure 4-24 and agree, in general, with conversations with industry.© Finally, Figure 4-25 presents estimated outlet emissions as a function of the ambient temperature. The prill tower configuration at Plant £ was used as a basis for this estimate. Once again, the data points represent actual particle size tests while the trend line was derived through the use of the log-normal distribution model. This plot indicates a considerable increase in controlled emissions as temperature drops. 4.4.2, Fluidized Bed Prill Towers Controlled and uncontrolled emissions during both agricultural and feed grade production were tested by EPA at the fluidized bed prill tower at Plant D and are illustrated in Figure 4-21. Emissions of .392 kg/Mg (.785 1b/ton) and .240 kg/Mg (.479 1b/ton) were measured during agricultural grade and feed production respectively. Control device efficiency in both tests was approximately 86 percent. The measured efficiencies confirm efficiencies predicted by particle size data and efficiency curves for the entrainment scrubbers. 5 A similar shift toward smaller particles during cold weather might be expected to occur in fluidized bed prill towers as discussed for nonfluidized bed prill towers. At present, it is impossible to quantify 4-51s-9 Relative Airflow* a Airflow at Given Temperature * Relative Airflow = —RrrFTow at 85 degrees F Canbrenty 10 20 30 40 50 60 70 80 90 Anbient Temperature (degrees F) Figure 4- 24. Estimated airflow cutback as a function of ambient temperatureeS-p rolled Emissions Con’ ® (1b/ton) -16 4 12 -10 08 -06 04 +02 04 03 -02 -O1 © Based on Actual Particle Size Tests 4 Measured Emissions at Plant E ~— Based on Log-Normal Distribution of Particle Sizes Reference Airflow = 100,000 dscfm @ 85 «agrees F ambient (cutbacks as per previous figure) Particle Conc. = .039 g/m? (.017 gr/dscf) (constant with temperature) 10 20 30 40 50 60 70 80 90 Aubient Temperature (degrees F) Figure 4-25, Variation in controlled nonfluidized prill tower to ambient temperature issions with respectto quantify a shift since particle size data for different temperatures is not available. 4.4.3 Rotary Drum Granulators Four EPA tests of three granulator/scrubber units at Plant A are illustrated in Figure 4-26. Two of these tests (granulator scrubber unit "A") measured controlled emission levels of .119 kg/Mg (.238 1b/ton), +111 kg/Mg (.221 1b/ton), and the third test (granulator/scrubber “B" had similar emissions of .114 kg/Mg (.228 1b/ton). The fourth test (granulator/scrubber unit "C") however, measured controlled emissions of -378 kg/Mg (.755 1b/ton). Unfortunately, no uncontrolled emission data are available for this test and thus it is impossible to determine whether the higher emissions are due to differences in uncontrolled emission characteristics or lower control device efficiency. According to Plant A personnel, the granulator/scrubber Unit A is virtually identical to the Unit C.°° However, differences in visible emissions between the units have been noticed since start-up with Unit C emitting a plume of higher opacity. The plant has not investigated the problem since both units are well below state emission standards and have a high rate of urea recovery (99.9 and 99.7 percent recovery). Concerning the reason(s) for the difference, plant personnel indicate lower scrubber efficiency is most likely. They speculated that the internal scrubber baffles may be misaligned, Based on these observations, it appears that units "A" and "B" at Plant A are representative of a granulator/scrubber unit operating at peak efficiency. 4.4.4 Rotary Drum Coolers Controlled mass emission data submitted by several plants operating coolers indicates controlled emissions that range from .01 to .1 kg/Mg (.02 to .2 1b/ton) with an average emission of .035 kg/Mg (.07 1b/ton). These emission levels are generally confirmed by predicted controlled emissions using uncontrolled EPA emission data and control device performance curves. A single tray type scrubber operating at a pressure drop of .375 kPa (1.5 in. W.G.) would remove approximately 98.9 percent of the particles in the cooler exhaust. An entrainment scrubber operating 4-54© Test Sample Run 60 1-4 PA Test Average 12 35 i .50 1.0 45 8 | 40 8 | ass 1 7 3 - 2.30 as & ust a 20 “4 . 3 a0 8 2 08 6 a ° 0 Bel Az Ae3 8 Test A-1: Plant A, Scrubber A, Test 1 Test A-2: Plant A, Scrubber A, Test 2 Test A-3: Plant A, Scrubber ¢ Test B: Plant B, Scrubber B Figure 4-26. Emission levels from controlled granulator exhausts. 4-55at an overall 1.25 to 1.50 kPa (5 to 6 in. W.G.) pressure drop would perform at approximately 99.6 percent efficiency. These high efficiencies are possible because of the large particles in cooler exhausts. Using these efficiencies and EPA uncontrolled mass emission measurements, controlled emissions of .043 and .016 kg/Mg (.086 1b/ton and .031 1b/ton) are estimated for a cooler controlled with a tray type and an entrainment scrubber respectively. 4-564.4 1, 5. ah 10, oa 12. 13, 14, 15. 16. wy. REFERENCES Telecon. Curtin, T., GCA Corporation, with J, Haggenmacher, Triad Chemical. February 22, 1979. Conversation about prill towers. Trip report. Jennings, M.S., Radian Corporation, to file. dune 11, 1980. 6 p. Report of January 31, 1980 visit to Reichhold Chemical in St. Helen, Oregon. p. 4. Trip report. Capone, S.V. and R.R. Hall, GCA Corporation, to Noble, E., EPA:ISB. May 25, 1978 5p. Report of visit to CF Industries in Donaldsonville, Louisiana. p. 4. Memo from Brown, P., GCA Corporation, to file. December 7, 1979. 6p. Summary of information about use of baghouses at urea plants. Reference 4, 6 p. Reference 4, 6 p. Telecon. Stelling, J., Radian Corporation, with Lerner, B.J., BECO Engineering. dune 30, 1980. Conversation about BECO scrubbers on prill tower emissions. Theodore, L. and A.J. Buonicore. Industrial Air Pollution Control Equipment for Particulates. Cleveland, CRC Press, 1976. p. 193, Bethea, RM. Air Pollution Control Technology. New York, Van Nostrand Reinhold Company, 1978. p. 275, Reference 8 p. 193. Reference 9, p. 275. Reference 8 p. 193. letter and attachments from Lerner, B.J., BECO Engineering Company, to Sherwood,oC., EPA:ISB. February 2, 1978 12 p. Information about the "V2" scrubber. pp. 3-4. Reference 13, p. 2. Reference 13, p. 2. Reference 9, pp. 294-295, Reference 9, p, 285, 4-5718. 1g. 20. ai. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31, 32. 33. Calvert, S. How to Choose a Particulate Scrubber. Chemical Engineering. 84:54-68. August 29, 1977. p. 58. Reference 9, pp. 289-290. Reference 9, p. 290. Reference 9, p. 290. The McIlvaine Scrubber Manual, Volume I. Northbrook, Illinois, The McIlvaine Company, 1974. Chapter III, Section 10.54, p. 70.0. Reference 18, p. 57. Reference 18, p. 57. Reference 18, p. 58. Impinjet Gas Scrubbers. Catalog No. 151. Cleveland, The W.W Sly Manufacturing Company. p. 2. Reference 2, p. 5. Telecon. Jennings, M., Radian Corporation, with Brady, J., Anderson 2000, Septenber 22, 1980. Conversation about fractional efficiency for CHEAF unit. The CHEAF System for Ultrafine Particle Emission Control. Bulletin No. 75-900048. Atlanta, Anderson 2000, Inc. October 12, 1978. p. 2 Kraus, M.N. Baghouses: Separating and Collecting Industrial Dusts. Chemical Engineering. 86(8):94-106. April 9, 1979. Letter and attachments from Cramer, J.H., Reichhold Chemicals, to Goodwin, D.R., EPA:ESED. December 1, 1978. 43 p. Response to section 114 letter. The National Air Pollution Control Administration. Control Techniques for Particulate Air Pollutants. (Prepared for U.S. Department of Health, Education and Welfare.) Washington, D.C. Publication No. AP-: January 1969. pp. 123-125. Clayton Environmental Consultants. Emission Study at a Feldspar Crushing and Grinding Facility. (Prepared for U.S. Environmental Protection Agency.) Research Triangle Park, N.C. EMB Report 76-NMM-1. September 27-29, 1976. 38 p. 4-5834. 35. 36. 37. 38. 39. 40. a. 42. 43. 44, 45. 46. 47. Engineering-Science, Inc. Air Pollution Emission Test at Kentucky Stone Company. (Prepared for U.S. Environmental Protection Agency.) Research Triangle Park, N.C. EMB Report 75-STN-3. 1975, 33 p. Jackson, B.L. and P.J. Marks (Roy F. Weston, Inc.) Source Emissions Test Report for Engelhard Minerals & Chemicals Corporation. (Prepared for U.S. Environmental Protection Agency.) Research Triangle Park, N.C. EMB Report 78-NMM-6. July 1978 34 p. Reference 29, p. 1. Letter and attachments from Swanburg, J.0., Union O11 Company, to Goodwin, D.R., EPA:ESED. Decenber 20, 1978. p. 9. Response to Section 114 letter. Reference 2, p. 3. Cramer, J.H. (Reichhold Chemicals, Inc.) Urea Prill Tower Control - Meeting 20% Opacity. (Presented at the Fertilizer Institute Environmental Symposium. New Orleans. Apri] 1980.) p. 2. Letter and attachments from Skinner, D., Radian Corporation, to Jennings, M.S., Radian Corporation. November 4, 1980. 114’. Information about particulate emissions from urea and ammonium nitrate prilling towers. Telecon. Jennings, M., Radian Corporation, with Cramer, J., Reichhold Chemical. August 11, 1980. Conversation about prill tower emissions. Reference 40, p. 5. Trip report. Jennings, M.S., Radian Corporation, to file. Novenber 7, 1980. 31p. Report of October 16, 1980 visit to Reichhold Chemical in St. Helens, Oregon. Kottler, F., The Distribution of Particle Sizes. Journal of the Franklin Institute. 250:339-356. October 1950. p. 350. Herdan, G. Small Particle Statistics, Second Edition. London, Butterworth and Company, 1960. Memo from Brown, P., GCA Corporation, to file. December 21, 1979. 4p. Summary of information about wet scrubbers at urea plants. Memo from Bornstein, M., GCA Corporation, to file. December 6, 1979. 3p. Solid urea manufacturing industry survey summary. 4-5948. 49, 50. 51. 52. 53, 54, 55, 56. 57. 58. 59. 60. 61. 62. Memo from Bornstein, M.I., GCA Corporation, to file. October 26, 1979. 6 p. Summary of urea manufacturing plant information. Letter and attachments from Thomson, J.W., Mississippi Chemical Corporation, to Bornstein, M.I., GCA Corporation. duly 17, 1978, 10 p. Information about MCC low-emission technology for ammonium nitrate neutralizers. Letter and attachments from Griec, J., American Air Filter, to Stelling, J., Radian Corporation. April 18, 1980. p. 3, Information about available equipment and efficiency. Reference 26, p. 2. Reference 26, p. 2. Western Precipitation Gas Scrubbers: Type "D" Turbulaire Scrubber. Publication No. S-100. Los Angeles, Joy Nanufacuting Company, 1978, p. 4 Reference 53, p. 2. Reference 29, p. 3. Brady, J. D., et al. A New Net Collector for Fine Particle Emission Control. (Presented at the 69th Annual Meeting of the Anerican Institute of Chemical Engineers. Chicago. dune 16, 1976. p. 57. Reference 56, p. 54, Roeck, D.R. and R. Dennis, (SCA Corporation.) Technology Assessment Report for Industrial Boiler Applications: Particulate Collection. (Prepared for U.S. Environmental Protection Agency.) Research Triangle Park, N.C. Publication No. EPA-600/7=79-178h. December 1979, p. 49. Letter and attachments from Bogatko, H.F., Atlas Powder Company, to Capone, S., GCA Corporation. July 31, 1978 16 p. Testing information. pp. 10-11, Trip report. Bornstein, M.I, and $.V. Capone, GCA Corporation, to Noble, E.A., EPA:ISB. dune 20, 1978 8 p. Report of visit to Mississippi Chemical Corporation in Yazoo City, Mississippi. Telecon. Bornstein, M., GCA Corporation, with Segar, T., N-Ren Corporation. dune 2, 1978. Conversation about two plants. Reference 31, p. 7. 4-6063. 64. 65. 66. Reference 37, p. 9. Stockham, J.D. and £.G. Fochtman (ed.s). Particle Size Analysis. Ann Arbor, Ann Arbor Science Publishers, 1977. Reference 2, pp. 3-5. Telecon. Stelling J., Radian Corporation, with Boggan, J., Agrico Company. July 30, 1980. Conversation about EPA tests result differences, as revised by J. Boggan's comments. 4-615.0 MODEL PLANTS AND CONTROL ALTERNATIVES Model urea plants and control alternatives are defined in this chapter. The model plants are chosen to be representative of solids producing plants in the urea industry. The model plants and control alternatives are used in subsequent chapters as the basis for analysis of the environmental and economic impacts associated with control of particulate emissions from sources in the urea industry. Section 5.1 describes the model plants in terms of process configuration, plant capacity, operating hours, raw material requirements, and utility requirements. Section 5.2 defines the existing level of control (ELOC) on each emission source. Section 5.3 describes the control options for each source, and Section 5.4 defines the control alternatives for each model plant. 5.1 MODEL PLANTS Process operations used in urea manufacturing include urea solution production, solution concentration, solids formation, solids finishing and solids handling. Urea plants differ in process configuration, plant capacity, and product type. In order to account for this variability, ten model plants were selected based on the present mix of process configurations and plant sizes in the industry, Table 5-1 identifies the solids formation process, plant capacity, and product type for each model plant. Process and control device flow diagrams for each model Plant are presented in Figures 5-1 and 5-2. further information con cerning the control options and emission characteristics of each model plant is contained in Section 5.2. 5.1.1 Process Configurations Four solids production techniques are currently in use in the urea industry: nonfluidized bed prilling, fluidized bed prilling, drum granulation, and pan granulation. However, only one pan granulator is 5-1ag TABLE 5-1, UREA MODEL PLANTS Plant Process TEmisston Model capacity Emission Dlauram Character Isties io. Myfday” Configuration Sources Figure Table (ton/day) 7 182 Won-futdized pritt Prat Tower a7 (200) tower plant producing agricultural” grade Cooter 5-H(a) 58 2 726 prills. Supplementary (800) Cooling required 59 3 1090 (1200) 4 ez, uldized bed pri 5-10 (200) tower plant. producing agricultural grade prits. — Pri11 Tower 51(b) sal 5 126 ta supplenentary cooling (200) required 6 090 S12 (1200) 7 182, Same as 4 except PrANT fower ——5-2(a) 513 (200) Feed grade 8 363 5-14 (400) 9 726, Granulation Plant Gramlator 5-2 515, (800) 10 090 5-16 (1200)LOC = 0.8 A5/H9 (1.6 To/ton) Cpeton I= 0:28 kg/ha {0.59 To/to Spelon 2 0168 SSng {0:08 IB/ten) Lac ~ 0,1 ayy (0.2 Th/ton) renous Fruren ear Te on erent sone scree fete te 1) nae i Eencentrat ior Froces Ucontrot ted neonate Enisetons essen 19 tare (3.8 tartan) SS fg (7.8 th7ton) nett ra rere, prurns .} ROTARY ORM nesnng 108 tag 155.55 bear rower coouer Storage ant Shiono 1% Blt store 9 Sod shies (9) Wonttutétzee bed pitt tower, agricultural grade Nose Plants 1-3 Froous FILTER oF ferubber Laer omar ieee, — es ts scmwsck Uncontratted Exisstons| Sima (6.2 torton) Pere mere ——| pestis (98.52 brea) ToweR 90% bate Storage a Shment (8) Flutdtzed bed prfT? tower, agricultural grade, Model Plants 66, Figure 5-1. Process diagrams for Model Plants 1-6. 5-3oc Sten #2 steno FILTER OR NTRATEAT crooner scrvtber Liqvor eeyee to Concantettton Process rcontrotied Enistion We aras (2.6 torton) tant ee reitis 10% map SRorage_ and C5e:58 ora) crs 30% Bute storage no ohipment (2) F411 comer, feed grade, Rodel Plame 7 fo - 0412 ta7my (0.23 1/ton) ‘eaten seq scnuesen feyele to Concentrate Process Eastons Ucontra eg {Eater (Gao orton) foTAay om a enaron Sauls 1088 ant 5s. wre) Shonen 90% mate storage na shipeent (0) Granuator, Moet Plants 8 = 10 Figure 5-2. Process diagrams for Model Plants 7-10.currently in operation in the United States and further expansion of granulator production is expected to be met through the use of drum granulators. Therefore, model plants were chosen to represent three types of solid production techniques: nonfluidized bed prilling, fluidized bed prilling, and drum granulation. Model Plants 1 through 7 represent prilling plants utilizing a single prill tower. Model Plants 1 through 3 and 7 utilize nonfluidized bed prill towers and Model Plants 4 through 6 use fluidized bed pril] towers. For the nonfluidized bed prilling plants, supplementary cool ing of the prills is assumed to be provided by a rotary drum cooler although some plants currently operating do not use a supplementary cool ing device. Except for Model Plant 7, al1 model prilling plants produce agricultural grade product. These plants may also produce feed grade product through a change in melt distributor and a reduction in air flow rates. However, feed grade towers are assumed to have lower emissions. For the purpose of presenting conservative economic and environmental impact analyses in subsequent chapters, it is assumed that Model Plants 1 through 6 produce agricultural products exclusively. Model Plant 7 represents a prilling facility dedicated to feed grade production only. This plant was selected to account for the possibility of granulators continuing to displace prill towers for agricultural grade production, and the subsequent need for prill towers to produce feed grade solids exclusively. This possibility could arise because, at the present time, feed grade solids are not produced by granulation. Node! Plants 8 through 10 use rotary drum granulators for solids formation. In contrast to prilling plants which use a single prill tower of varying capacity, granulator plants generally use multiple grenulator trains of a single uniform capacity to achieve total plant product ion, 5.1.2 Plant Capacities Currently, prilling plants producing agricultural grade urea range in capacity from 167 Ng/day (186 tons/day) to 1150 Mg/day (1270 tons/day). 5-5Using this range as a guide, a small model prilling plant of 182 Mg/day (200 tons/day) capacity and a large model prilling plant of 1090 Mg/day (1200 tons/day) capacity were selected. Because this size range is considerable, an intermediate size model prilling plant of 726 Ng/day (800 tons/day) was selected. The choice of an intermediate size nearer the larger plant size reflects the tendency for facilities to be large in order to take advantage of economies of scale. Only one size of model feed grade prilling plant was selected. The two plants currently producing feed grade exclusively are small, Both produce less than 217 Mg/day (240 tons/day). Therefore, a single small feed grade model plant of 182 Mg/day (200 tons/day) capacity was selected. Granulator plant sizes vary according to the number of granulators operated at any particular plant location. The number of granulators currently operated at one location varies between one and seven. However, many of these plants brought granulators on line in increments. Model Plants 8 through 10 utilize one, two and three 363 Mg/day (400 tons/day) capacity granulator trains respectively to represent a range of existing plant capacities. 5.1.3 Operating Hours Urea plants usually operate continuously except for scheduled maintenance shutdowns and unscheduled equipment failures. Total shut- down time is estimated at nine weeks per year. Thus, each model plant operates 43 weeks/year, 7 days/week, and 24 hours/day for a total production time of 7224 hours/year. 5.1.4 Raw Material and Utility Requirements In each of the model plants, ammonia and carbon dioxide are processed to form an aqueous urea solution which is then concentrated to 99+ percent urea, This molten urea is mixed with about 0.4 weight percent formaldehyde additive and then solidified either by prilling or granulation. The purpose of the formaldehyde additive is to prevent caking and breakage of the solid product. Because the model plants use the same types of solution production and concentration equipment and produce similar products (99+ percent solid urea with 0.4 percent formaldehyde additive), 5-6they have the same basic raw material requirements per unit of urea product. Thus, the annual raw material requirements for the four sizes of model plants given in Table 5-2 are used for all model plants. Also presented in the table are utility requirements for the various sizes of model plants. These requirements represent the total utility needs of the entire urea manufacturing plant, including solution synthesis and concentration processes. Electrical energy and steam requirements vary slightly between prilling plants and granulation plants. However, the difference is relatively small compared to the total plant energy usage. Therefore the utility requirements listed in Table 5-2 are used for all model plants. 5.2 DETERMINATION OF EXISTING CONTROL LEVELS The Existing Level of Control (ELOC) is that level of control which is currently applied to emissions from solid urea producing processes in the urea industry. Table 5-3 summarizes the ELOC chosen for the emission sources in the urea industry. Consideration is first given to current state regulations which apply to the emission sources. State regulations usually define the primary constraint on emissions. However, for many emission sources, a sizeable discrepency exists between actual emission levels and those levels allowed by state regulations. Several factors are responsible for the disparity between the allowable state emission levels and measured industry emissions. First, the test method used by state and industry personnel varies from state to state and may be considerably different from the method used in this study. The sampling procedures which have been endorsed and used by the States include EPA's Method 5, the American Society of Mechanical Engineers’ Performance Test Code (PTC) 27, and modifications of these procedures, The collection efficiencies of the various sampling procedures depend upon such factors as the type of filter used, the temperature of the filter, whether condens‘ble emissions are included, and the sample recovery and analytical procedures. Even when two state emission standards are identical, one standard can effectively be more stringent when the sampling procedure specified collects a higher percentage of emissions. 57TABLE 5-2. RAW MATERIAL AND UTILITY REQUIREMENTS FOR MODEL PLANTS! ~3 Carbon Plant Annual Aamonta Dioxide Fora denyle Steam Electricity 001 ing Cool ing size Production Consumption Consumption Consumption Use Use Vater ise® Natge Use™ Mg/day Ng/yr Malye Hg/yr Malye Nafyr Tye wh) 0 /yr wi/yp (tons/day) (tons/yr)—_(tons/yr) (tons/yr) (tons/yr) (tons/yr) ——(Ma-hr/yr) (gat xi0°7y (aa xi0°/yr) 182 54,700 31,700, 4,300 219 79,300 23.9 3.98 4090 (200) (60,200) (34,900) (45,500) (2a) (87,300) (6,620) (1,050) (1.08) 363 109, 000, 63,200 82,300 46 155,000, a7 79 ino a (400) (120,000) (69,600) (30,600) (4a0) (174,000) (13,200) (2,090) (2.16) © 726 219,000, 127,000, 165,000 075 317,000 95.8 15.9 1640 (800) (2av,000) (140,000) (182,000) (964) (349,000) (26,500) (4,190) (4,34) 1090 328,000, 190,000, 248,000 a 475,000 130.4 23.6 24600 (1200) (361,000) (209,000) (273,000) (yaaay (523,000) (39,700) (6.280) (6.50) ®pecirculated cooling water with 15°F rise across process. "consumptive use of cooling water6-5 TABLE 5-3. SUMMARY OF EXISTING EMISSION LEVELS Existing Emission Source Uncontrot led Controlled Required Emissions Emissions Control Equipment — Remaval kg/Mg_ kg/Mg Efficiency (1b/ton) (1b/ton) Pril] Towers Agricultural Grade, -80 Spray Tower 57.9% Nonfluidized Bed (1.6) Scrubber Agricultural Grade, -60 Spray Tower 80.6% Fluidized Bed (1.2) Scrubber Feed Grade -80 Spray Tower 55.5% (either bed type) (1.6) Scrubber Rotary Drum Granulators 103 +115 Entrainment 99.9% (207) (230) Scrubber Rotary Drum Coolers 3.9 -10 Tray-Type 97.4% (7.8) (.20) ScrubberFor example, a wet impinger collection device collects more particulate matter than a heated filter, which tends to vaporize some of the collected particles. Therefore, given equal emission standards, the standard requiring the wet impinger collector would be more stringent. Secondly, sampling problems may compromise the state agency's ability to determine compliance. As an example, prill towers commonly exhaust through horizontal vents near the top of the prill tower. Because of the tower height, an outlet stack is not needed to gain acceptable dispersion of pollutants. Therefore, sampling locations are either very poor or nonexistant. Faced with sampling sites which will not yield accurate data, state officials find it difficult to use source tests as a compliance tool. Instead, opacity readings are often used as a measure of compliance. Third, plants may find it to their advantage to control emissions to lower levels than are required by state regulations. Urea collected in scrubbers is recycled to the solution concentration process or to solution fertilizer make-up. This recovery of urea offsets much of the cost of control and may encourage a higher level of removal than is required by state regulations. Finally, in some cases it has been reported that opacity standards are more difficult to meet than mass emission standards.* In these cases, industry may use a control device to meet opacity standards which also reduces mass emissions well below the state regulation for mass emissions. A discussion of existing regulations is presented in Section 5.2.1. Existing levels of control are presented in the sections following for prill towers, rotary drum granulators, and rotary drum coolers. Where appropriate, the influence of the factors mentioned previously is addressed. 5.2.1 Existing Emissions Limitations Standards limiting particulate emissions are in effect in all 50 states. The regulations are of three types: opacity limits, exhaust gas particulate concentration limits, and particulate emission limits 5-10calculated from process weights. The process weight regulations can take the form of an allowable emission factor expressed as kg (1b) of particulate allowed per Mg (ton) of production. This section focuses on the regulations in effect in 23 states in which urea plants are presently located. In nearly all of the states, industrial source emissions are limited by both opacity standards and process weight standards. Table 5-4 presents a summary of opacity, concentration, and process weight standards in the 23 states.° by district. The regulation presented in Table 5-4 for California are those for the Los Angeles County Air Pollution Control District, which are the most stringent. Twenty of the 23 states in which urea plants are located have standards limiting the opacity of an exhaust stream to 20 percent. This is the most stringent opacity regulation affecting urea plants and is also the typical regulation, A variation of the 20 percent opacity standard is a standard which allows a source to exceed 20 percent for a certain time period (for example, 6 minutes in any one hour). Two of 23 states in which urea plants are located have standards limiting opacity to 40 percent and one state has a standard of 30 percent. Almost a11 states have particulate emission rate limits based on the amount of production. As shown in Table 5-4, particulate emissions from a 363 Mg/day (400 tons/day) process are limited to a range of 5.19 to 71.44 kg/hr (11.41 to 257.20 Ib/hr), or 0.34 to 4.72 kg/Mg (0.69 to 9.43 Ibs/ton). Illinois and the Los Angeles County Air Pollution Control District in California have the most stringent process weight regulation limiting a 363 Mg/day (400 tons/day) process to emissions of 5.19 and 5.97 kg/hr (11.41 and 13.14 1b/hr) respectively, or 0.34 to 0.40 kg/Mg (0.69 to 0.79 Ibs/ton). In addition to the process weight rate equation ranging from state to state, the method of enforcement also varies from state to state. Some states consider each process or stack as a source. Other states consider the allowable emission rate to apply to the combined emissions from all the processes or stacks at the entire plant or from one building In California, regulations differ 5-11TABLE 5-4. EMISSIONS STANDARDS AFFECTING UREA PLANTS Zig Mioable —— ANowble farcietate — taeteutate owactty eaten? et ceutation aR aogatgpet sate (reed) arb’ a tana recess, recess 0s on x» 3.9000 yy syns 20.54 or 20 0.0sep0- 603" 1.920423" 157.20 Tat Galltornta® 20 wate 5.97 Florida ~ 3,59(0)°82 17, a4qny 6 70.54 iad Georgta 2 esate) 7, anypyt6 20.56 9 Idaho 20 4.1000) ¢56,0(0)° 9) 40 27.00 war Tiiteots? 0 2.s4qe)-5 mM. 0-18 5.19 Tow 4.10(6)° 7 (55.0(9)°1)-40 27.00 wear Kansas 2 4.1016)" (55, 0(0)" 27.00 27 Lous 0 09 0,69 4.10(P)"87 (55, 0101" 27,00 12.27 Misstsstppt “o 4.20(ry-®7 (55, 0¢0)° 27.00 ae Missourt cy 0.3 0.69 4.10(7)°°7 (55, (9) 27.00 12,27 Nebraska 7 4.10(0)°°87 (55, 0(0)% My, 27.00 Mew York® 0 0.05, sar 3.90027 59, (990082 50 88 1.70 Worth Carol ina: 2 9.377(9)9- 9087 2222 ‘Ohio. 20 4.1000) (55, 0(7)").40 27.00 wa ‘Oklahoma 20 03 0.69 4, 204)°°57 (55.017) ")-40 27.00 W277 Oreyon 20 0.1 0.23 4,10 9)°97 655. (0)! )40 27.00 27 Tennessee 20 0.25, o.. 3.59¢0)°82 47,31 0-16 20.54 om Tens, 0 3.12 (9) 989 25,.4¢9) 287 49.05 22.65 Masninston » an 02 Hyomitng: 2 3,59(0)9°82 47.54 40)0'6 20.54 14 364 on 24 hour operat to Yorocess wetght regulations for Arkansas apply to production rates of 10010" tbyhr end 10% P-<10® Ibyhr, respectively. Sased on Lor Angeles county AFCO grocers selght rate table,at the plant. The most typical interpretation used is the first method which is also the less stringent of the two, Because the state process weight equations are nonlinear with respect to production, the allowable emission factor (kg/Mg or 1b/ton) changes with different plant sizes. Table 5-5 presents the allowable emissions for various sizes of plants by state, for the states with solid urea production capacity. Also included at the bottom of the table are the average emission limitations weighted by total solid production, prill production, and granule production respectively. If the state process weight regulations were the only factor affecting industry emissions, it would be expected that the existing level of contro} (ELOC) could be approximated by these weighted averages. 5.2.2 ELOC of Nonfluidized Bed Prill Towers Producing Agricultural Grade Prills Industry data presented in Chapter 3 indicates varying uncontrolled emission rates of .39 - 1.79 kg/Mg (.78 - 3.58 1b/ton) for nonfluidized bed prill towers. Although much of this variability can be attributed to actual differences in uncontrolled emissions, the difficulties involved in obtaining accurate emission measurements of prill towers also play a role. These difficulties include: Low particle concentration Poor or nonexistant stack sampling locations Hygroscopic nature of urea particles Dissociation of urea particles at high temperatures encountered in collection filters The variability in emissions, in conjunction with the variability in state regulations, results in differing levels of control in the industry. Seven of fifteen existing nonfluidized bed prill towers utilize control devices, while the rest are uncontrolled. A new pril] tower may or may not require a control device to meet applicable state regulations depending on the particular situation. For the purpose of this analysis, an emission level of 0.8 kg/Mg (1.6 1b/ton) was chosen to represent the ELOC for nonfluidized bed prill 5-13TABLE 5-5. ALLOWABLE EMISSIONS BY PLANT SIZE (Metric Units) 182 Mg/day 364 Ma/day 737 Mg/day 1091 Mg/day kg/hr kg Mg kg/hr “kg Mg kg/hr kg Mg kg/nr kg Mg abana 608.803 767 13.79 455 OATH Alaska Arkansas * 53.32 7,036 m5 4,717 95.8 2162.7 2.802 Caltfornta 495.654 6.01 397 7.06 233 779 Ae Georgia 6.08 202 9.06617 13.79 A550 OMT 328 Towa 771 1.019 12.27.10 18.59 614 20.3 446, fansas . . . . * . . . Louisiana, 5 - a : u . 2 oI Mississippi a . a . . . D Q Missourt 2 ? 2 . . . . * New York 7.36 170.772 18.61 64288 338 ‘ohio a 12.27 .810 18.59 614 20,3 446 Oregon . . . . . . . Tennessee 6.08 .a02 9.38 SIT 13.79 645500 14.71, 324 Texas 145 1,521 2.7 1.486 31.6 1.043 38.5 781 Wyoming 6.08 = .802 9.38617 13.79 45500 16.71 324 Straight Average @ 7.29.96 1.90.79 17.66 8 19.57.43 Weighted Average > 7.30 0.97 use 76 17.22 87 18.80 42 Weignted average 7.10.94 M16 74 16.56 55 18.10.40 Pritts only Jetanted Average d 7.65 3.01 12.15.80 18.28 81 20.03.44 Granules: only * This state defines an anfssion source as all process enissions for the entire plant while other states define an anission source as a single stack or process. Therefore, Arkansas was not included in the straight or weignted averages. = Straight aritmmetic average = Wetghted average based on percentage of solids production fn each state ¢ = Weighted average based on percentage of pritl production in each state 4 = Weighted average based on percentage of granule production in each state 5-14TABLE 5-5. ALLOWABLE EMISSIONS BY PLANT SIZE (English Units) 200 TPO. 300 TPO 800 TPO 1200 TPD Toye Tb/ton—ibymr_———Tby/ton—_b/hr_—_—_—iTb/ton_Ib/hr_—_‘Tb/ton abana 12.37 1,608 20,58 1233 90.34 90 32,37 647 Alaska Arkansas 7.3 Won 1872 9.433 210.8 6,323 250.2 5.004 california 1.291.307 13.22 .73 1553 0,468 12.13.39 Georgia 137 1.608 20.5¢ 1.233 30.4910 647 ova 16.97 2.037 27.00 1.620 122748, 382 Kansas . : . : : Louisiana : 0 ; 7 : : Wisstssipp . " " : " " " " Missourt : 5 d o : : New York 19 25,78 1544 40.958 122982, 73 1.075 ome 20m 27,00 620 40,09 1.227 44,58 292 regan C : : 5 : . Tennessee 14604 20,58 1.233034 91D 2.7 eras 3.022 49,85 2.991 69,49 2,085 78,06 yon ing 16604 20,56 1.233 0.389032. Straight werage® 16.08 1.92 38.85 3.05 86 Neighted Average” 16.05 1.93, 37.99 4.37 8 Meignted Average” 15.61 La7 38.43 33.83 80 Prttis Only. Meignted Average! 16.82 2002 28,73 1.60 OAL 06 8 Granules Only SThis state defines an antasion source as all process enisstons for the entire plant while other states define an emission source as 2 single stack or process. Therefore, Arkansas was not included in the straight or weighted averages. a= Straight arithmetic average b - Netgnted average based on percentage of solids production in each state c= Weighted average based on percentage of prilT production in aach state 4 = Weignted average based on percentage of granule production in each state 5-15towers producing agricultural grade prills. This is the average allowable mass emission rate, based upon state regulations for a 363 Mg/day (400 ton/day) plant. In addition, it is assumed that the uncontrolled emissions for a nonfluidized bed tower are 1.9 kg/Mg (3.8 1b/ton) based upon EPA testing. 5.2.3 ELOC for Fluidized Bed Prill Towers Producing Agricultural Grade Prilis Uncontrolled emissions for fluidized bed towers are higher than uncontrolled emissions from nonfluidized bed towers. EPA tests of a fluidized bed tower show that uncontrolled emissions are 3.1 kg/Mg (6.2 1b/ton) for a plant producing agricultural grade prills. Industry data for controlled emissions from fluidized bed towers vary from 0.38 - 0.43 kg/Mg (0.76 - 0.86 1b/ton). EPA test data show controlled emissions of 0.39 kg/Mg (0.79 1b/ton) for a fluidized bed tower producing agricul- tural grade urea prills. All three fluidized bed towers are currently controlled to levels required by state regulations. Based on this data, state regulations are being met by existing facilities and an average state regulation was used to establish the ELOC. For agricultural grade production, 0.6 kg/Mg (1.2 1b/ton) represents the allowable emission under an average state regulation for a typical size of fluidized bed prill tower (737 Mg/day or 800 ton/day). 5.2.4 §LOC for Feed Grade Pril1 Towers During the production of feed grade urea, nonfluidized and fluidized bed towers operate with approximately the same air flow and have com parable uncontrolled emissions. The limited industry data reports that uncontrolled emissions for nonfluidized bed prill towers producing feed grade product range from 1.61 - 1.76 kg/Mg (3.22 - 3.51 1b/ton). EPA data for a feed grade fluidized bed prill tower show uncontrolled emissions of 1.68 kg/Mg (3.36 1b/ton). Based on these emissions any new feed grade prill towers would have to control emissions to meet state regulations. A 0.8 kg/Mg (1.6 1b/ton) of product emission level was chosen as the ELOC for prill towers producing feed grade based upon the average state regulations for 363 Mg/day (400 ton/day) plants. This isthe same level selected for nonfluidized bed prill towers producing agricultural grade urea prills. The larger plant size used for estab- lishing the fluidized bed, agricultural grade ELOC is not used since plants producing feed grade tend to be smaller. 5.2.5 ELOC for Granulators Al] 19 existing granulators are controlled, 18 with entrainment scrubbers and one with a packed bed scrubber. A comparison of uncon- trolled emissions and applicable state regulations indicates that collection efficiencies of better than 99 percent are required. In addition, since uncontrolled emissions are high, process economics dictate control at the source. An emission level of 0.115 kg/Mg (0.230 1b/ton) of product was chosen as the ELOC for granulators. This level represents the emissions measured during EPA testing and is typical of existing industry practice. 5.2.6 ELOC for Rotary Drum Coolers Solids cooling is required, in some cases, during the production of agricultural grade prills in a nonfluidized bed prill tower. Rotary drum coolers are used when sufficient cooling is not available in the prill tower, Very little data is available to quantify typical uncontrolled emissions from coolers. One EPA test measured uncontrolled emissions of 3.9 kg/Mg (7.8 1b/ton). Industry data indicates four coolers have emission rates ranging from 0.01 to 0.1 kg/Mg (0.02 to 0.2 1b/ton). These levels are significantly below the allowable state regulations. No EPA data is available for controlled cooler emissions which would allow verification of the industry test data with an EPA approved method. However calculations based upon EPA particle size data in conjunction with manufacturers’ control device performance specifications confirm that the devices in use are capable of reducing emissions to the levels reported by industry. Thus, an ELOC emission level of 0.1 kg/Mg (0.2 1b/ton) was selected, This level represents the highest controlled emissions reported by industry.5.3. CONTROL OPTIONS This section presents control devices recommended for application to control particutate emissions from urea solids producing and finishing processes. The selection of control devices (hereafter referred to as control options) to achieve various control levels is based on per- formance data from EPA testing and vendor information. Table 5-6 presents a summary of control options for each source. Subsect fons 5.3.1, 5.3.2, and 5.3.3 presents control options for prilling, granu- Jation, and rotary drum cooling processes, respectively. Section 5.3.4 presents emission characteristics for each model plant. 5.3.1 Prill Towers To control particulate emissions to the ELOC for prill towers, a spray tower scrubber is recommended. This scrubber exhibits a removal efficiency of from 56 to 82 percent depending on the type of tower. To reduce emissions to a lower level, an entrainment scrubber is reconmended and designated as control option 1 for prill towers. The greatest degree of control is achieved by a wetted fibrous filter with a removal efficiency of 98 percent (control option 2 for prill towers). 5.3.2 Granulators Particulate emissions from granulators are currently well controlled to prevent excessive product loss. Since granulators are currently achieving the ELOC with an entrainment scrubber, no other control options are recommended, 5.3.3 Rotary Drum Coolers To meet the ELOC determined for rotary drum coolers, a plate impingement scrubber with a removal efficiency of 98 percent is recommended. Control options attaining greater levels of control are not defined. 5.3.4 Emissions Characteristics Tables 5-7 through 5-16 define emission characteristics for each model plant. This data is presented in terms of emission sources and control options, as discussed in the previous sections.61-s TABLE 5-6, CONTROL EQUIPMENT PERFORMANCE PARAMETERS Applicable Performance Parameters Emission Vodel Control ‘emoval Pressure Drop Liquid/Gas Ratio Plants Option Control Device Efficiency kPa (in.W.G.) 1/m3_ (gal/1000 Ft) 1-7 ELOC Spray Tower a a a 40 3.0 Tower 1-7 Option 1 Entrainment: 85% 1.3 5.0 87 6.5 Scrubber 17 Option 2 Wetted Fibrous 98% 3.1 12.0 227 2.0 Filter Granulator 8-10 ELOC Entrainment 99.9% 4.1 16.0 87 6.5 Scrubber Cooler 1-3 FLOc Plate Impingement 98% 1.3 5.0 +40 3.0 (Tray Type) Scrubber *Removal efficiency and pressure drop varies according to the specific Model Plant. Efficiencies range from 56 to 82 percent. ELOC- Existing level of control.02-3 TABLE 5-7. EMISSION CHARACTERISTICS FOR MODEL PLANT 1 CONTROL OPTIONS Total Stack Exhaust Particulate ‘Emtsston Figurate Temperature Concentyation Moisture Factor Emission Control dsm” /min K ‘g/m Content kg/Mg Source Level (dscfm) (F) (gr/dscf) % (1b/ton) Prill ELOC 2690 294 +0375 2.5 0.80 Tower (95000) (70) +0164 (1.60) Option 1 2690 294 +0133 2.5 +285 (95000) (70) (0058) (£570) Option 2 2690 294 0018 2.5 +038 (95000) (70) (.0078) (:076) Cooler ELOC 214 305 0586 45 +10 (7570) (90) (0256) (220)T2-s TABLE 5-8, EMISSION CHARACTERISTICS FOR MODEL PLANT 2 CONTROL OPTIONS Total Stack Exhaust Particulate Emission Figwrate Temperature Concentyation Moisture Factor Emission Control dsm” /min K g/m Content —_kg/Mg Source Level (dscfm) (F) (gr/dscf) % (1b/ton) Pri ELOC 6850 294 0.0587 2.5 0.80 Tower (242,000) (70) (0.0257) 1.60 Option 1 6850 294 0210 2.5 +285 (242,000) (70) (0092) (.570) Option 1 6850 294 +0028 2.5 +038 (242,000) (70) ( .0012) (.076) Cooler ELOC 858 305 +0586 4.5 +10 (30280) (90) ( .0256) (.20)22-s TABLE 5-9. EMISSION CHARACTERISTICS FOR MODEL PLANT 3 CONTROL OPTIONS Total Stack Exhaust Particulate Emission Figurate Temperature Concentyation Moisture Factor Emission Control dsm” /min k g/m Content —_kg/Mg Source Level (dscfin) (F) (gr/dscf) % (1b/ton) Prill ELOC 9360 294 +0629 2.5 0.80 Tower (340,000) (70) ( 0275) (1.60) Option 1 9360 294 +0224 2.5 +285 (340,000) (70) ( +0098) (2570) Option 2 9360 294 0030 2.5 +038 (340,000) (70) ( .0013) ( .076) Cooler ELOC 1290 305 +0586 4.5 +10 (45,400) (90) ( -0256) ( 220)2-5 TABLE 5-10. EMISSION CHARACTERISTICS FOR MODEL PLANT 4 CONTROL OPTIONS Total Stack Exhaust. Particulate Emi ssion Figwrate Temperature Concentration Moisture —-Factor. Emission Control dsm3/min K g/m Content, kg/Mg Source Level (dscfm) (°F) (gr/dscf) % (1b/ton) Pritt ELOC 2830 294 0268 2.5 0.60 Tower (100,000) (70) (.0117) (1.20) Option 1 2830 294 .0207 2.5 465 (100,000) (70) (00904 ) (930) Option 2 2830 294 00277 2.5 0620 (100,000) (70) (,00121) (124)2-5 TABLE 5-11. EMISSION CHARACTERISTICS FOR MODEL PLANT 5 CONTROL OPTIONS Total Stack Exhaust Particulate Emission Fipwrate Temperature Concentration Moisture — Factor Emission Control dsm3/min K g/m? Content, kg/Mg Source Level (dscfm) (°F) (gr/dscf) % (1b/ton) Prill ELOC 8890 294 0340 2.5 0.60 Tower (314,000) (70) (.0148) (1.20) Option 1 8890 294 0263 2.5 465 (314,000) (70) (.0115) (930) Option 2 8890 294 00353 2.5 0620 (314,000) (70) (00154) (.124)2-5 TABLE 5-12, EMISSION CHARACTERISTICS FOR MODEL PLANT 6 CONTROL OPTIONS Total Stack Exhaust Particulate Emi ssion | Figwrate Temperature Concentration Moisture ‘Factor Emission Control dsm°/min K g/m Content ko/Mg Source Level (dscfm) (°F) (gr/dscf) % (1b/ton) Pri] ELoc 12,900 294 0351 2.5 0.60 Tower (457,000) (70) (0153) (1.20) Option 1 12,900 294 0273 2.5 +465 (457,000) (70) (019) (2930) Option 2 12,900 294 00362 2.8 0620 (457,000) (70) (00158) (124)92-5 TABLE 5-13. EMISSION CHARACTERISTICS FOR MODEL PLANT 7 CONTROL OPTIONS Total Stack Exhaust Particulate Emission Figwrate Temperature Concentration Moisture Factor Emission Control dsm? /min K g/m Content, kg/Mg Source Level (dscfm) (°F) (gr/dscf) % (1b/ton) Pri] ELOC 1080 3 13 5.0 0.80 Tower (38,000) (100) (0491) (7.60) Option 1 1080 an 316 5.0 +270 (38,000) (100) (.0138) (540) Option 2 1080 301 00421 5.0 -0360 (38,000) (100) (00184) (<0720)Le-3 TABLE 5-14, EMISSION CHARACTERISTICS FOR MODEL PLANT 8 CONTROL OPTIONS Total Stack Exhaust Particulate. Emi ssion Flowrate Temperature Concentration Moisture Factor Emission Control dsn3/min K g/m Content. kg/Mg Source Level (dscfm) (°F) (gr/dscf) % (\b/ton) Granulator —ELOC 1360 311 20213 5.0 5 (48000) (100) (00932) (2230)82-5 EMISSION CHARACTERISTICS FOR MODEL PLANT 9 CONTROL OPTIONS TABLE 5-15, Total Stack Exhaust Particulate Emission Figwrate Temperature Concentration Moisture ‘Factor Emission Control dgn/min K g/m? Content kg/Mg Source Level (dscfm) (°F) (gr/dscf) % (1b/ton) Granulator —-ELoc 2720 301 20213 5.0 15 (96,000) (100) (00932) (:230)62-5 EMISSION CHARACTERISTICS FOR MODEL PLANT 10 CONTROL OPTIONS TABLE 5-16, Total Stack Exhaust Particulate Emission Fipwrate Temperature Concentration Moisture —‘Factor Emission Control dsm? /min K g/m Content. kg/Mg Source Level (dscfm) (°F) (gr/dscf) % (1b/ton) Granulator LOC 4080 EM 0213 5.0 5 (144,000) (100) (00932) (.230)5.4 CONTROL ALTERNATIVES 5.4.1 Approach Control alternatives for each model plant are summarized in Table 5-17, Each alternative is comprised of various control options (control devices) applied to each emission source in each model plant. In selecting ‘the control options, three basic levels of emission control are considered for each emission source. 1, ELOC - Controlling emissions to the ELOC as defined in Section 5.2. This level of control would typically be required under existing state regulations. 2. Controlling emissions to achieve the greatest degree of reduction. 3. Controlling emissions to an intermediate level. This is between the ELOC and the greatest degree of emission reduction. Selection of the intermediate and greatest levels of control is made on the basis of performance data in Chapter 4. For sources other than prill towers, the selection of control levels is limited to existing levels of control. Control alternatives will be referred to in subsequent chapters to facilitate economic and environmental impact comparisons. 5-30Tes TABLE 5-17. CONTROL ALTERNATIVES Nodel Plant Emission Control Plant Configuration Sources Alternatives No. 143 Nonfluidized bed, Agricultural Pri] Tower 0 + # grade production Cooler 0 0 0 4-6 Fluidized bed, Agricultural Prill Tower 0 + 4 grade production 7 Nonfluidized bed, Feed Prill Tower 0 + # grade production 8-10 Granulator Granulator 0 Legend: 0 - ELOC + - Option 1 ++ Option 25.3 REFERENCES i Shreve, R.N. and J.A. Brink. Chemical Process Industries, Fourth Edition. New York, McGraw-Hill Book Company, 1977. pp. 284-287. Kirk-Othmer (ed.). Encyclopedia of Chemical Technology, Volume 21. John Wiley & Sons, Inc., 1970. pp. 37-56. Chemical Engineering (ed.). Sources and Production Economics of Chemical Products, Second Edition. New York, McGraw-Hill Publishing Company, 1979. pp. 277-279. Trip report. Bornstein, M.I., GCA Corporation, to Noble, E.A., EPA:ISB. August 2, 1978. p. 2. Report of visit to C & I Gridler Incorporated in Louisville, Kentucky. Memo from Stelling, J., Radian Corporation, to file. July 6, 1980. Compilation of state regulations on particulate emissions from urea plants. 52326.0 ENVIRONMENTAL IMPACTS The purpose of this chapter is to present the environmental impacts of the control alternatives for particulate emissions from emission sources in the urea industry. The emission sources to be considered are prill towers, rotary drum coolers, and rotary drum granulators. The air quality, water pollution, solid waste, and energy impacts associated with the application of the control alternatives to the model plants are identified and discussed in Sections 6.1 to 6.4, respectively. Additional impacts are described in Section 6.5. 6.1 AIR POLLUTION IMPACT The impact of each control alternative on air quality is evaluated in this section. Two impacts are considered: primary impacts, or the reduction of particulates due to the control equipment used, and secondary impacts; the pollutants generated as a result of applying the control equipment. 6.1.1 Primary Air Quality Impacts The primary impact on air quality resulting from implementation of control alternatives is the reduction of particulate emissions into the atmosphere. Table 6-1 presents plant-wide (prill towers, granulators, coolers) emission and removal factors for the control alternatives and mode] plants presented in Chapter 5. Table 6-1 also presents the additional emissions reduction relative to the existing level of control (ELOC) for prill towers. The control alternatives for prill towers increase in their stringency from Alternative 1 (ELOC) to Alternative 3 (greatest degree of control). Using the emission reduction factors in Table 6-1, Table 6-2 presents the total annual emissions reduction for Control Alternatives 2 and 3 over the ELOC. 6-12-9 TABLE 6-1. EMISSION AND REMOVAL FACTORS FOR CONTROL ALTERNATIVES Emission Factors _kq/Mg_(1b/ton Absolute Reduction Over ELOC kg/Mg (1b/ton Model Plant ‘ontroT ATternat ive ‘Control ATternat ive Number Plant Configuration 1 2 a 1 2 3 13 Nonfluidized bed 0.900 0.385, ° 0.515 0.762, Prill tower, cooler, (1.800) (0.770) (0) (1.030) (1.524) Agricultural grade or Sax atx 46 Fluidized bed 0.600 0.470 0,062 0 0.135 0.538 Pril] toner» (1.200) (0.930) (0.174) (0) (0,270) (1.076) Agricultural grade O25) 90% 7 Pri] Towers 0.800, 0.270 0.036, o 0.530 0.764, Feed grade (12600) (0-540) (0.072) (0) (1,060) (1:528) Ox 6x 96% 8-10 Granulator, 0.115 (0.230)e-9 TABLE 6-2. TOTAL ANNUAL REDUCTION OVER THE ELOC OF PARTICULATE EMISSIONS FOR CONTROL ALTERNATIVES, Mg/year (Tons/year) Modet Control Alternatives Plant Plant Capacity Number Plant. Configuration ig day (rons aay) 1 2 3 1 Nonfluidized bed tover, 182 0 28.2 41.6 Agricultural grade (200) (0) (310) (45.8) 2 Nonfluidized bed tower, 726 0 112.7 166.4 Agricultural grade (800) (0) (124.0) (183.0) 3 Nonfluidized bed tower, 1090 0 169.1 249.6 Agricultural grade (1200) (0) (186.0) (274.5) 4 Fluidized bed tower, 182 0 74 29.6 Agricultural grade (200) (0) (8.1) (32.5) 5 Fluidized bed tower, 726 0 29.6 118.2 Agricultural grade (800) (0) (325) (130.0) 6 Fluidized bed tower, 1090 oO 44.3 177.3 Agricultural grade (1200) (0) (48.8) (195.0) 7 Feed Grade Tower 182 0 29.0 41.9 (200) (0) (31.9) (46.1) 8 Granulator, 363 0 Agricultural grade (400) (0) 9 Granulator, 726 0 Agricultural grade (800) (0) 10 Granulator, 1040 0 Agricultural grade (1200) (0) gased on an operating schedule of 301 days/year,6.1.2 Secondary Air Quality Impacts Secondary air pollutants are pollutants generated as a result of applying control equipment. There are no air pollutants generated directly by the control equipment required for each control level. However, the increased need for steam and electrical power to support the emission control systems will cause an increase in utility power plant emissions. Table 6-3 presents the emission reductions and corresponding increased power plant emissions for model urea plants and associated control alternatives. Also presented (as a percentage) is the increased power plant emission compared to the corresponding amount of urea plant- wide emission reduction. Increased power plant emissions range from 1 to 3 percent of the amount of plant-wide emission reductions. 6.1.3 Summary of Air Quality Impacts The primary air pollutant emissions from affected facilities in the urea industry are particulates. The major benefit of implementing control alternatives is the reduction of these particulate emissions. A range of particulate reductions is possible, depending upon the control alternative chosen. Alternative 3 has the greatest particulate reduction for prilling operations. For Model Plant 3 (1090 Mg/day nonfluidized bed prill tower), the primary air quality impact would be an annual reduction of 249.6 Mg/year (274.5 ton/year) of particulate, with a corresponding secondary air quality impact due to increased power plant emissions of 5.0 Ng/year (5.5 ton/year) (2 percent of the plant-wide reduction). Hence, the net reduction in particulates from Model Plant 3 would be 244.5 Mg/year (269.0 ton/year), or 98 percent of the plant-wide reduction. Similarly, the impact of secondary pollutants would be small for the other model plants and their respective control alternatives relative to plant-wide particulate emission reductions. 6.2 WATER POLLUTION IMPACT There would be no adverse water pollution impact due to the control alternatives, since the liquor used in the wet scrubbers controlling particulate emissions is typically recycled to the solution concentration 6-4s-9 TABLE 6-3, SECONDARY AIR POLLUTION IMPACTS ASSOCIATED WITH THE APPLICATION OF CONTROL ALTERNATIVES TO TYPICAL UREA PLANTS [rower Plant fnission aga (b/eon roe tanto Aitarmatiag = e/ton)p Impact Percent® | —tontrat ftternat ve — vission Reduction over ELOC._ka/Mg (1b/ton! cont roY_Aiternat fe Plant ~ Type 3 1 2 3 1 2 a Yon Flutdized Bed Pril} Tower, - 0.515 0.762 70,0042 o,ona1” - 1 2 Agricultural grade (1.030) (1.528) (0.0083) (0.0243) Flutdized Ged Pritt Tower, - 0.135 0.538 -*o,008s —o.on4a? - 3 3 Agricultural Grade (0.270) (1.076) (0.0089) (0.0279) PriN) Tower 0.530 764 "0.0028, 0.0076 - 1 1 Feed Grade - (1060) (1.528) (0.0056) (0.0151) (a) There are no additional energy requirements attributable to control devices corresponding to the ELOC. (b) These entsstons are averages for the various plant capacities. (c) Impact percent “( Power Plant, Emission ratsston ReductTon Sera) * 100.concentration process or used for fertilizer solutions. The amount of excess water discharged, already present in urea plants since it is produced as a byproduct of the carbamate decomposition reaction, will be reduced because of the large amount of water entrained in the exhaust of a wet scrubber. 6.3 SOLID WASTE IMPACT There would be no solid waste impact due to implementation of the control alternatives. Liquor from scrubbers is recycled to the solution concentration process or sold as fertilizer solution. 6.4 ENERGY IMPACT Emission control equipment for the urea industry uses electricity and, indirectly, steam. The primary electrical demand is from the control equipment fans used in conjunction with normal operating equipment to generate sufficient airflow rates and pressure drops across the control equipment. Pumps which circulate the scrubber liquor also require electrical energy. Steam is used to concentrate the scrubber liquor to a level where it can either be recycled to the solution concentration process or sold as fertilizer solution. Table 6-4 presents the total annual energy requirements of the control alternatives, assuming maximum steam requirements. The relative amounts of each type of energy (steam or electricity) vary by model plant. For prilling plants, 20-50 percent of the control equipment energy demand is represented by steam (assuming a scrubber Tiquor urea concentration of 20 percent by weight). Similarly, steam requirements can comprise more than 95 percent of the control equipment energy required for granulation plants. This high percentage is due primarily to the high uncontrolled emission rates from granulators which necessitate a greater scrubber liquor recycle rate. Also presented in Table 6-4 are the incremental energy requirements over the ELOC. The greatest increase in energy consumption occurs for Control Alternative 3, Model Plant 6, a 1091 Mg/day (1200 ton/day) fluidized bed prill tower producing agricultural grade prills. The 6-6TABLE 6-4. ANNUAL ENERGY REQUIREMENTS FOR UREA MODEL PLANT CONTROL ALTERNATIVES A - Model Control Increase over Alternative 1 Plant Alternative 10°stu Ty 10°Btu 1 1 Nonfluidized Bed 1 10.4 11.0 - - Prill Tower, a 15.0 15.9 4.6 4.9 Agricultural Grade, 3 24.3 25.6 4.6 4.9 182 Mg/day (200 TPD) 2 Nonfluidized Bed 1 33.8 35.7 - - Prill Tower, 2 46.3 48.9 12.5 13.2 Agricultural Grade 3 70.2 74.0 36.4 38.3 728 Mg/day (800 TPD) 3 Nonfluidized Bed 1 49.5 52.2 - = Prill Tower, 2 67.2 70.9 17.7 18.7 Agricultural Grade, E 100.8 106.3 49.3 51.4 1091 Mg/day (1200 TPD) 4 Fluidized Bed Prill 1 7.8 8.3 - - Tower, Agricultural 2 12.3 12.9 4.5 4.6 Grade, 182 Mg/day 3 22.0 23.2 14.2 14.9 (200 TPD) 5 Fluidized Bed Pril1 1 26.5 - - Tower, Agricultural 2 40.6 14.1 14.8 Grade, 728 Mg/day, 3 71.5 45.0 47.4 (800 TPD) 6 Fluidized Bed Pril] - 39.0 41.1 - - Tower, Agricultural Grade 2 59.5 62.8 20.6 21.7 1091 Mg/day (1200 TPO) 3 104.5 110.2 65.5 69.1 7 Prill Tower, Feed Grade 1 3.3 3.5 - - 182 Mg/day "(200 TPD) 2 5.6 5.9 2.3 2.4 3 9.8 10.4 6.5 6.9 8 Granulator 362 Mg/day (400 TPD) 1 275.9 291.0 9 Granulator 728 Mg/day (800 TPD) 1 551.8 582.0 10 Granulator 1091 Mg/day (1200 1 828.8 873.3 TPD) 6-7control equipment energy requirement increase over ELOC for this case is 69.1 Td/year (65.5 x 10° Btu/year), or 63 percent. The total annual energy requirement of the control equipment for this plant is 110.2 Td (104.5 x 109 Btu). The control equipment energy requirements of Model Plant 6 with Control Alternative 3 represent less than 7 percent of the total plant energy demand. 6.5 OTHER IMPACTS There would be no significant noise impact due to implementation of any of the control alternatives in the urea industry. The increase in noise from properly designed control equipment would be insignificant compared to the noise associated with production process equipment. 6-86.6 REFERENCES 1, Memo from Stelling, J., Radian Corporation, to file. June 30, 1980. 22 p. Increased power plant emissions. 2. Memo from Stelling, J., Radian Corporation, to file. June 30, 1980. 5p. Net consumption of water - test results and model plants.7.0 COST ANALYSIS A cost analysis of the control alternatives described in Chapter 5 ‘is presented in this chapter. This chapter is divided into two major sections. Section 7,1 presents the costs associated with the various control alternatives, including an analysis of capital and annualized costs. Both new facilities and existing facilities are considered. Other costs that may result from the application of control equipment are considered in Section 7.2, including costs imposed by water pollution control regulations and solid waste disposal requirements. 7.1 COST ANALYSIS OF CONTROL ALTERNATIVES 7.1.1 Introduetion The costs of implementing control alternatives in the urea industry are presented in this section. The cost analysis is based upon the model urea plants and the control alternatives presented in Table 7-1 and discussed in Chapter 5. Three sources were considered in the model plant matrix. These were prill towers, rotary drum coolers, and granulators. Control options are identified for each source and were used as the basis for the formulation of the control alternatives. The cost of purchasing, installing, and operating the various control devices are presented in the following sections. The purchased costs for the control equipment were obtained from vendor quotes. !~4 Cost estimating manuals and published reports were used to determine costs for auxiliary equipment, (fans, pumps, motors, starters, downcomers, and stacks).°"!0 Equipment costs were scaled up to first quarter 1980 dollars (abbreviated 1980) using either the Marshall and Swift Equipment Cost Indices or Chemical Engineering Plant Cost Indicies,!0>11 Total capital cost for installation of the various control devices was determined by applying component factors to the basic equipment 7-1ek TABLE 7-1. SUMMARY OF UREA MODEL PLANTS AND CONTROL ALTERNATIVES Model Plant Size Emission Cont rot No. Mg/D Configuration Sources Alternatives (tons /D) 2 3 1 181(200) Nonfluidized pril1 tower plant Prill’ Tower + + producing agricultural grade 2 726(800) prills. Supplementary Cooler 0 0 cooling required. 3 1090(1200) 4 181(200) Fluidized bed pril] tower plant producing Pril] Tower + a7 5 726(800) agricultural grade prills. No supplementary cooling 6 1090(1200) required. Pril] Tower + + 7 181(200) Pril] Tower Plant producing feed grade prills. 8 363(400) 9 726(800) Granulation Plant Granulator 10 1091 (1200) Legend: 0 - ELOC + = Option 1 +4 = Option 2costs. These component factors take into account direct costs (piping, electrical, instrumentation, structural costs, construction labor, etc.), indirect costs (engineering, contractor's fee, taxes, etc.), and contingencies. The capital component factors were obtained from a survey of industry and a cost estimating manual,0»12 The annual cost of operating and maintaining the control devices includes direct operating expenses (utilities, labor, maintenance) and capital charges. Capital charges include insurance, administrative overhead, taxes, and capital recovery (the annual cost for the payoff of the control devices).13+!4515 any credits or gains obtained from application of the control equipment is subtracted from the annual operating costs in order to obtain the net annual cost of the control alternatives. Credits are obtained from recovering urea captured by the control equipment. Net annual costs are divided by the quantity of pollutant removed by the control equipment to determine the cost effectiveness of the control alternatives. Cost effectiveness is used as a means of comparing the various alternatives. The costs associated with controlling emissions from new facilities are discussed in Section 7.1.2. Cost considerations for existing facilities are discussed in Section 7.1.3. 7.1.2 New Facilities The capital and annualized costs of applying control alternatives to new urea solids production, finishing, and handling facilities are presented in this section. Costs associated with the control alter- natives are presented in six subsections. Section 7.1.2.1 discusses important considerations used in the determination of control equipment costs. Section 7.1.2.2 presents the capital costs of the control alternatives, and Section 7.1.2.3 presents the annual cost of the control alternatives. The effect of the control alternatives on the cost of urea product is presented in Section 7.1.2.4. Section 7.1.2.5 compares the annual costs and cost effectiveness of the control options to Alternative 1 [existing level of control (ELOC)]. The base cost of a urea plant is discussed in Section 7.1.2.6. 7-37.1.2.1 Basis for Equipment Costs. This section presents important considerations in determining the costs of the control equipment. All the equipment, except for motors and starters, is made of stainless steel because of the corrosiveness of urea. Table 7-2 presents control equipment operating parameters which were obtained from vendors and are typical of industrial operation. The control devices and auxiliary equipment were sized to handle the airflows and emissions specified for the model plants in Tables 5-7 through 5-16. An example of the major equipment needed to control emissions from the sources in the model plants are presented in Tables 7-3a through 7-3c. For each emission source considered, the equipment for one plant size is presented as an example. Due to differences in plant design, scrubbers selected for prill towers are of various sizes while only one scrubber size was selected for each granulator processing train. For prilling operations, the prill tower and finishing equipment are constructed and sized to handle whatever capacity was chosen for design production. Likewise, the scrubbers and auxiliary equipment used to control emissions from these facilities were sized to handle the entire airflow from the facility. The airflow through the solids production equipment varies with plant size, therefore, all of the control equipment had to be resized for each plant size. Granulation plants, on the other hand, employ processing trains of specific sizes, as discussed in Chapter 5. The model granulation plants chosen were 363 Mg/day (400 TPD), 726 Mg/day (800 TPD), and 1089 Ng/day (1200 TPD). A 363 Mg/day (400 TPD) granulator was used as a base, representing a single processing train. Control equipment was sized to handle the emissions from a single 363 Mg/day (400 TPD) plant. For 726 Ng/day (800 TPD) and 1089 Mg/day (1200 TPD) plants additional granulator processing trains were added, and the total equipment cost for controlling emissions was obtained by doubling or tripling the cost of controlling a single processing train, 7-4TABLE 7-2. SPECIFICATIONS FOR PARTICULATE CONTROL SYSTEMS I, Spray Tower (For prill_ towers) A. "Pressure Drop: .77 kPa (3 in. .)? A B. Liquid to Gas Ratio: 0.40 wm (3.0 gal/1000 acf) C. Construction Material: 304 SS D. Fan Location: At scrubber inlet E. Scrubber Location: On top of prill tower II, Entrainment Scrubber (For prill towers) A. Pressure Drop: 1.3 kPa (5 in. 4W.G.) B. Liquid to Gas Ratio: 0.87 wm? (6.5 gal/1000 acf)* C. Construction Material: 304 SS D, Fan Location: At Scrubber inlet E. Scrubber Location: At grade level IIl, Wetted Fibrous Filter (For prill towers) A. Pressure Drop: 3.1 kPa (12 in,il.G.) i B. Liquid to Gas Ratio: 0.27 afm? (2 gal/1000 acf)** C. Construction Material: 304 SS 0, Fan Location: At scrubber outlet E, Scrubber Location: At grade Tevel IV. Plate Impingement (Tray type) Scrubber (For coolers) A. Pressure Drop: 1.3 kPa (5 in..W.G.) A 8. Liquid to Gas Ratio: 0.40 a/m> (3 gal/1000 act) 32° C. Construction Material: 304 Ss D. Fan Location: At scrubber inlet V. Entrainment Scrubber (For granulators) A. Pressure Drop: 4.1 kPa (16 ing W.G.) B. Liquid to Gas Ratio: 0.87 w/m> (6.5 gal/1000 acf)a C. Construction Material: 304 SS D. Fan Location: At scrubber inlet a. Reference 12 b. Reference 5 c. Reference 2 d. Reference 3 7-5TABLE 7-3a. EXAMPLE OF MAJOR EQUIPMENT REQUIREMENTS FOR CONTROL OF PRILL TOWERS . (726 Ma/day (800 TPD), Fluidized bed/Aqricultural grade configuration) Existing Level of Control Control Device Spray tower 304 SS, L/G = 3.0 gal/1000 ACF, ap = 3"NG Ducting 304 SS ductwork (4.0 feet diameter) Fan (each) 60740 ACFM @ 111°F, 275 rpm, 50 hp. Recirculation pump 1800 gpm, 30 ft TDH, 125 hp. Control Option 1 Control Device Entrainment scrubber, 304 SS construction, L/G = 6.5 gal/1000 ACF, 4p = 5"WG Duct ing 304 SS ductwork (7.0 feet diameter), including ducting from top of prill tower to grade level. Fan (each) 60740 ACFM @ 111 °F, 600 rpm, 125 hp. Stack 7.0 feet diameter, 85 ft high, CS Recirculation pump 2400 gpm, 100 ft TDH, 150 hp. Control Option 2 Control Device Wetted Fibrous Filter, 304 SS construction, L/G = 2 gal/1000 ACF, Sp = 12"WG Duct ing 304 SS ductwork (10.5 feet diameter), including ducting from top of prill tower to grade level. Fan 364,400 ACFM @ 111 °F, 1400 hp. Stack 10.5 feet diameter, 120 ft high, CS Recirculation pump 730 gpm, 100 ft TOH, 50 hp. Preconditioning system Recirculation pump 550 gpm, 220 ft TDH, 75 hp. Piping CS and SS, as required 7-6TABLE 7-3b. EXAMPLE OF MAJOR EQUIPMENT REQUIREMENT FOR CONTROL OF COOLERS 726 Mg/day (800 TPD) Control Device Ducting Fan Stack Recirculation pump Existing Level of Control Plate Impingement (Tray Type) Scrubber, 304 SS construction L/G = 3.0 gal/1000 ACF, Ap WG 304 SS ductwork (3.0 feet_diameter: 33800 ACFM @ 90°F, 1600 rpm, 200 hp 3.0 feet diameter, 40 ft high, CS 220 gpm, 100 ft TDH, 15 hpTABLE 7-3c. EXAMPLE OF MAJOR EQUIPMENT REQUIREMENTS FOR CONTROL OF GRANULATORS 363 Mg/day (400 TPD) Control device Ducting Fan Stack Recirculation pump Existing Level of Control Entrainment scrubber, 304 SS construction, L/G = 6.5 gal/1000 ACF, Ap - 16" WG 304 SS ductwork, 5.0 feet_diameter 64000 ACFM @ 190°F, 1250 rpm, 250 hp 5.0 feet diameter, 85 ft high, CS 400 gpm, 100 ft TDH, 15 hp 1-8The cost of purchasing the control equipment is shown in Table 7-4. This table presents the cost of the control device and the cost for all ‘the major equipment items associated with the control options. Table 7-5 presents an example cost breakdown of the major equipment items needed to control Model Plant 1 to the ELOC (Control Alternative 1). The same procedure shown in this example was used to derive the purchased equipment cost of the control alternatives for all model plants. 7.1.2.2 Capital Costs. Capital costs represent the total investment required for purchase and installation of the basic control equipment and associated auxiliaries. Capital cost estimates for each control system were developed with cost component factors.®*!2 These factors were applied to the control option costs presented in Table 7-6 to give installed capital costs. The capital costs for the control options were then combined to give the control alternative cost presented in Table 7-7. Costs for research and development and costs for possible production losses during equipment installation and start-up were not included. The costs are presented in first quarter 1980 dollars. In computing the total installed cost of the wetted fibrous filter for Control Option 2 on prill towers, actual installation costs provided by the vendor were substituted for generalized installation costs in the component factor.4 Therefore, the installation cost element of the component factor was deleted during these calculations. 7.1.2.3 Annualized Costs. Annualized costs represent the yearly cost of operating and maintaining the pollution control system. The basis of the annualized cost estimates are presented in Table 7-8. All annualized costs were based on 7224 hr/yr of operation. Electricity costs were based on the power required to run the electric motors used to operate fans and pumps. Brake horsepower for the motors was determined by using power curves from cost estimating manuals.® The annual cost of electricity was based upon an electricity cost of $.04/kwh. Annual labor cost for operation of the control equipment is the product of the total labor rate’ ($17.45/hr), operating hours perOre TABLE 7-4, PURCHASED EQUIPMENT COSTS ASSOCIATED WITH CONTROL OPTIONS (1980) Control Device Cost, 1000s Total Control Equipment Cost, 10005 Model Control Plant Option Prill Tower Cooler Granulator Pril] Tower Cooler Granulator 1 ELoc? 203.4 1.6 - 334.0 33.2 - 1 135.8 - - 362.2 - - 2 468 - - 622.2 - 2 ELOC 449.8 37.6 : 762.8 82.7 - 1 372.3 - - 864.3 - - 2 B15 - - 1140.0 - - 3 ELOC 532.0 47.0 - 976.1 121.1 1 467.7 - - 1185.4 S 2 1110 - - 1507.6 - 4 ELOC 206.5 - - - - 1 140.0 - - - - 2 478 - - - - § ELOC 506.6 - - - - 1 433.5 - - - 2 1025 : . = 6 ELOC 749.5, - - 1 620.5 - - 2 1350 - - - - 7 ELOC 159.5, : - - - 1 62.2 - - - - 2 325 - - - - 8 ELOC = - 69.4 - - 180.5 9 ELoc - - 138.8 - - 361.0 10 ELOc - - 208.2 - : 541.5 9 ELOC = Existing level of controlTABLE 7-5. EXAMPLE OF PURCHASED EQUIPMENT COST BREAKDOWN OF MAJOR EQUIPMENT FOR ALTERNATIVE 1 ON A 181 Mg/day (200 TPD) NONFLUIDIZED BED PRILL TOWER, AGRICULTURAL GRADE (MODEL PLANT 1), $1000 (1980). Item Cooler Pri] Tower Total Control Device 11.6 203.4 215.0 Fans, Motors, Starters 6.3 73.4 79.7 Pumps, Motors, Starters 7.5 19.7 27.2 Ducting 2.9 30.5 33.4 Stack 14 7.0 8.4 Tank 3,6 3.5 33,2 334.0 367.2TABLE 7-6. COMPONENT CAPITAL COST FACTORS FOR A WET SCRUBBER AS A FUNCTION EQUIPMENT COST, Q 6,12 oF Major Equipment Ductwork Instrumentation Electrical Founda tions Structural Direct costs Material 1.00 q 0.11 a 0.08 @ 0.06 9 0.03 9 0.06 g 0.02 g 0.008 0.09 @ 1.40 1.90.) Labor 0.09. @ 0.09 9 0.07 Q 0.12 9 0.05 9 0.03 9 0.02 9 0.02 q 0.08 0.580 @ Indi rect costs Conponent Measure of costs Factor Engineering ho percent caterial and labor 0.19.9 Contractor's fee 5 percent material and Tabor 0.29 Q Shakedown 5 percent material and Tabor . 0.10 9 Spares 1 percent mtertal 0.019 Freignt 3 percent material 0.08 Q Tues 3 percent mterial 0.08 Q Total indirect costs 0.67.0 Contingencies = 20 percent of direct and indirect costs 0.51 Q Total capital costs 3.08 9TABLE 7-7. CAPITAL COSTS OF CONTROL ALTERNATIVES FOR MODEL PLANTS, $1000 (1980) Total contrat Total Difference in ; Mode contro Eauipent Installed Total Installed Cost case! Plant Alternative cost cost (arternative = Existing Level) 1 1 1 367 na 5 1 2 335, ie 8 rb 3 555 1695 56a Bt 2 1 Bs 2604 . 22 2 367 215 32 2 5 wee 33 569 31 3 A 1097 379 5 22 2 1305 ‘soe 6a Fa 5 sz a2 353 41 4 1 30 1087 “2 2 368 1133 a 3 ez 1618 5 1 903 279 - 2 1092 5363 520 5 1385 3515 ee et 5 1 13 108 : 62 z 1661 siz 1003 53 3 1955 5007 398 Te 1 1 229 108 S ra 2 182 562 az) v3 5 a6 18 a2 br 8 1 180 555 5 ot 9 1 36 ur o 10-1 10 1 se 1688 : Airst number 4 model plant numbers second umber 1s contro? alternative nunber. Note: Yalues in parentheses represent net credits or gains.TABLE 7-8. BASES FOR SCRUBBER ANNUALIZED COST ESTIMATES (1980) Direct operating costs Utilities Water Condensate from solution formation processes assumed available free of charge’ Electricity $.04/kWh Operating Tabor Direct wage rate $7.66/hour Fringe benefits 25 percent of direct rate Supervision 15 percent of direct rate Total $10.72/hr . Operating hours Process equipment 7,224 hours/year Scrubbers 7,224 hours/year Each unit requires one eighth of an operator 4® Maintenance 5.5 percent of capital investment Capital charges Capital recovery factor 0.1669? Taxes and insurance 5.Opercent of capital investment Administrative overhead 2.5 percent of capital investment. Recovery credit $50 in solution” his condensate would contribute to a plant's water pollution loading, if not used by scrubbers and mist eliminators. Since costs of treatment and disposal are avoided, the assumption that it is available free of charge is conservative. bincludes wages plus 40 percent for labor-related administrative and overhead costs. Cost (4077) updated using Hourly Wage Index: 260.4: 212.8. Sgased on a 15-year equipment 1ife! and a 10.0 percent interest rate.!? Recovery credit is taken as cost of urea (f.0,b. plant, $120/ton), Tess the steam cost of removing, the scrubber water (12 MNBTU/ton at $6.25/M Tbs steam, $70/ton urea). 16 Reference 13. freference 15. SReference 6. 7-14year of the control process (7224 hr/yr), and number of operators required to run to control equipment (1/8 operator/unit). The annual labor cost to operate a single control device is estimated to be $15,760/yr. A conservative net credit of $55 per Mg ($50/ton) of urea was calculated for urea recovered in wet scrubbers. This recovery credit included the cost of removing all water from a 20 percent by weight urea solution in a single stage evaporator.!© the total credit allowed for each control option was dependent upon the uncontrolled emissions, control device efficiency, and the assumed hours of operation. An example of an annualized cost breakdown for Control Alternative 2 on Model Plant 2 is given in Table 7-9. The procedure shown in this example was used to determine the net annualized costs presented in Table 7-10 for the control options considered in this study. These costs were combined to give the net annualized costs of the control alternatives, which are presented in Table 7-11. 7.1.2.4 Effect of Control Alternatives on Product Cost. The impact of applying control alternatives on the price of the product was also determined and is presented in Table 7-11. This cost impact indicates the additional or credit cost per unit of urea produced. It was calculated by dividing net annual cost of the control alternative by annual model plant production. 7.1.2.5 Cost Effectiveness. Cost effectiveness is used as a means of comparing control alternatives, and is defined as the total annualized cost of the pollution control system divided by the quantity of pollutant removed by the system. The cost effectiveness of the control alternatives can be compared directly to the ELOC by using the following equation. Cost Effectiveness = Sx" ©, Poo Pe C, = Net annualized cost to remove a quantity of pollutant (P,) by alternative x. C_ = Net annualized cost to remove a quantity of pollutant (P_) to meet a specified ELOC.TABLE 7-9. COMPONENT ANNUALIZED COSTS FOR ALTERNATIVE 2 MODEL PLANT 4. Prill Tower, Control Option 2 Component Cost, $1000 per year (1980) Direct Costs Operating labor and supervision Maintenance labor and materials Utilities Electricity Total Direct Costs Administrative overhead Capital recovery charges Taxes and insurance Total Capital Charges and Overhead 234.1 Total Annualized Costs (without product recovery) 370.3 Credit for particulate recovery 7.9 Entrainment Scrubber Total Credit Net Annualized Costs 362TABLE 7-10, NET ANNUALIZED COSTS FOR CONTROL OPTIONS, 1000$ (1980) Model Control Plant Option = Prill Tower Cooler Granulator 1 ELOC 312 34 i 358 2 529 2 ELOC 697 51 1 834 2 1028 3 ELOC 894 66 1 141 2 1375 4 ELOC 314 1 362 2 537 5 ELOC 815 i 1045 2 1255 6 ELOC 1194 1 1571 2 1780 7 ELOC 2u1 1 183 2 345 8 ELOC (649) 9 ELOC (1314 10 ELOc (1979) Note: Values in parentheses represent net credits or gains. 7-17Bl-Z TABLE 7-11. NET ANNUALIZED COST AND COST EFFECTIVENESS OF CONTROL ALTERNATIVES FOR MODEL UREA FACILITIES (1980) (METRIC UNITS) Cast Effectiveness Cost Effectiveness Effect on cost® Effect on cost of product: Model Control et Arua per unit urea recovered Relative to FLOC* of product. ‘increase over LOC case Plant Alternative Cost 1000$ 50 Sims Sima s/t M1 1 1 376 1299 . 6.35 - 12 2 3932 1355 2007 720 0.85 13 3 645 1859) 5023 10.31 3.96 za 2 1 18.1 699 9 a2 : 22 2 236.0 163 ve 06 oe 2h 3 1079.1 a9 2296 «33 v1 a1 3 1 961.0 599 : : 32 2 1207.8 633 1810 os ua 3 wan 792 2226 146 4 1 362.4 2658 : 6.62 : 2 562.6 2520 23 6.64 0.02 3 503.0 ers oat 9.92 330 541 5 1 815.9 494 - 323 : 52 2 1046.8 1819 7800 a 10s 53 3 1252.1 1902 3795 S70 208 6-1 6 1 194.1 4487 E 2.64 : 2 2 1573.3 1822 asi9 aan uu 63 3 788.0 1796 3370 545, var ray 7 1 aa 2066 A 3.06 - 12 2 ie" 2202 (237) 30 C49) 73 3 382.0, 5854 376 eas 257 at 8 1 (649.1) (57.9) : 5.93) : 91 9 1 «aay (87.9) - (6.00) - 0-1 10 1 (1979) : (6.03) : FELOC = Eelsting Level oF Contvat ote: Values in parentheses represent net gains or credits. aased on a product price of $132/Mg of urea product ($120/ton).eink TABLE 7-11, NET ANNUALIZED COST AND COST EFFECTIVENESS OF CONTROL ALTERNATIVES FOR MODEL UREA FACILITIES (1980) (ENGLISH UNITS) case Note Contrat Wet Annual Cost Effectiveness Cost Ffectiveness Effect on cort® Effect on cost of rofucts® Plant Mternath Cost 10005 per'untt ures recovered Relative to ELOC: of product ‘Increase over LOC Sten! S/t0n ‘fon! ‘/ton ui 1 1 307.6 wa 5 5.7 ie 2 eh 1229 1839 653 76 rs 3 bees 1686 68 330 sie a4 2 1 149.1 6 P an e ped 2 ans:0 632 138 368 0.57 2a i 1079.1 206 203 cae 1a ay 3 1 961.0 503 2.66 42 z 1207'8 629 162 a4 0.68 xa 3 man's na 219 339 LB at 4 1 362.4 2409 : 6.02 - 42 2 36206 2aa 3s 8.03, oor a 3 543.0 2969 ss Siz 3°00 5 1 15.9 3.39 b 2 1046-8 nas as 9% a 262.1 aa 528 185 6 1 1198.1 tage eS a1 e 2 73.3 ess rai ery 105 i vaao 1629 ans a5 164 tm 7 1 aut 2507 a a8 12 2 184-0 1088 (250) 3.06 Cs rs 5 38200 3315 5063 aes an aa ® 1 (649.1) (82.2) (5.39) : oy ° 1 (138) (52.5) o (5.48) - mot 0 1 (1999) (52.8) : (5.48) - Hote: Values in parentheses represent net gains oF credits e10c = Existing Level of control Saased on 2 product price of $132/My of ures product ($120/ton).7.1.2.6 Base Cost of Urea Plants. Capital costs of control alternatives may be compared with the total capital costs of new urea manufacturing plants. Table 7-12 presents ranges of average capital costs for complete urea production plants, including solution synthesis, solution concentration, and solids formation processes. These values may be compared with the total capital costs and the capital cost relative to ELOC of each control alternative presented in Table 7-7. The capital cost relative to ELOC of control alternatives range from 3 to 7 percent of the total plant costs. The cost of producing urea has been estimated at 128 $/Mg (116 $/ton) for small plants and 101 $/Mg (92 $/ton) for large plants. 17>18 The major cost component of urea is the cost of natural gas used in manufacturing the ammonia feed to the urea synthesis process. 7.1.3 Existing Facilities The cost for installing a control system in an existing plant is generally greater than the cost of installing a control system in a new facility with the same exhaust gas parameters because special design modifications are often required. Cost components that may increase because of space restrictions and plant configuration are contractor and engineering fees, additional ducting and structural reinforcement. These costs vary from place to place and job to job depending on the difficulty of the job, the risks involved, and current economic conditions. Estimating this additional installation cost or retrofit penalty is difficult because of these plant-specific factors and additional engineer- ing requirements. However, these additional costs are not expected to be large or to preclude the application of control equipment. 7.2 OTHER COST CONSIDERATIONS 7.2.1 Cost Imposed by Water Pollution Control Regulations The costs of wastewater treatment at plants in the nitrogen fertilizer industry have been researched by previous investigators.2@>23 These costs are related to effluent limitations placed on the fertilizer industry and are not associated with air pollution control. Effluents 7-20TABLE 7-12. CAPITAL COSTS OF UNCONTROLLED UREA PLanTs’>1® (1980) Plant Size Relevant Model Cost Range Average Cost Ma/yr_(TPD) Plant Number $ mil lions $ millions 181 (200) 14,7 2.2 9.2 8.2 363 (400) 8 17 = 17.2 M4 726 (800) 2,59 19.2 - 27.4 23.3 1089 (1200) 3, 6, 10 25.8 - 33.2 29.5 7-21from air pollution control equipment are recycled to the solution process for economic reasons. Therefore, no additional wastewater treatment costs are expected due to air pollution control equipment. 7.2.2 Costs Imposed by Solid Waste Disposal Requirements Due to the high solubility of urea, any solid wastes can be dissolved and used as liquid fertilizer or recycled to the solution process to produce solid urea, Thus, no additional solid waste is anticipated due to air pollution control equipment. 7-227.3 REFERENCES 5 10. i. 12, 13, felecon. Brown, P., GCA Corporation, with Podhorski, J., Joy Manufacturing. March 12, 1979. Costs and other topics concerning scrubbers. Letter from Pilcher, L.Y., GCA Corporation, to Hosler, R., W.W. Sly. duly 24, 1979. Wet scrubber costs. Telecon. Stelling, J., Radian Corporation, with Hosler, R., W.W. Sly. May 2, 1980. Cost of plate impingement scrubbers. letter and attachment from Brady, J., Anderson 2000 Incorporated, to Jennings, M.S., Radian Corporation. May 9, 1980. Cost estimates of wetted fibrous filter systems. Peters, M.S. and K.D. Timmerhaus. Plant Design and Economics for Chemical Engineers, Second Edition. New York, McGraw-Hill Book Company, 1958. 850 p. Neveril, R.B. (GARD, Incorporated). Capital and Operating Costs of Selected Air Pollution Control Systems. (Prepared for the U.S. Environmental Protection Agency.) Research Triangle Park, N.C. Pub lication No. EPA-450/5-80-002, Decenber 1978. pp. 3-1 through 5-78. Guthrie, K.M. Process Plant Estimating, Evaluation and Control. Solana Beach, California, Craftsman Book Company of America, 1974, 603 p. Richardson Engineering Services, Inc. Process Plant Construction Estimating Standards, Volume Three. Solana Beach, California, 1980. Section 16-52. Richardson Engineering Services, Inc. Process Plant Construction Estimating Standards, Volume Four. Solana Beach, California, 1980, Sections 100 - 123, 100 - 281 through 100 - 283,'100 - 360, 100 - 360, 100 - 373, 100 - 652. Memo fron Stelling J., Radian Corporation, to file. July 9, 1980, Cost analysis computations for regulatory alternatives. Memo from Stelling, J., Radian Corporation, to file. duly 2, 1980. Compilation of cost indices used in updating costs. emo from Battye, W., GCA Corporation, to file. Septenber 7, 1979, Sunmary of Section 114 responses. Internal Revenue Service. Tax Information on Depreciation, 1979 Edition. Publication No. 534. Washington, 0.C. U.S. Government Printing Office, 1978. 38 p. 7-2314, 15. 16. 7, 18. 19. 20. PEDCo Environmental Incorporated. Cost Analysis Manual for Standards Support Document. (Prepared for U.S. Environmental Protection Agency.) Research Triangle Park, N.C. April 1979. 82 p. U.S. Environmental Protection Agency and Manufacturing Chemists Association. EPA-MCA Chemical Industry Cost Estimating Conference, February 1977: Notebook. Washington, D.C., EPA:EPAD. January 1979. 107 p. Memo from Stelling, J., Radian Corporation, to file. May 28, 1980 Determination of credit for recovered urea. Memo from Stelling, J., Radian Corporation, to file. July 6, 1980. Cost of manufacturing urea. Chemical Engineering (ed.). Sources and Production Economics of Chemical Products, Second Edition. New York, McGraw-Hill Publishing Company, 1979. pp. 119-121, 277-279. Martin, E. Development Document for Effluent Limitations Guidelines and New Source Performance Standards for the Basic Fertilizer Chemical Segment of the Fertilizer Manufacturing Point Source Category. (Prepared for U.S. Environmental Protection Agency.) Washington, D.C. Publication No. EPA-440/1-74-O1la. March 1974, p. 170, David, M.L., J.M. Matk and C.C, Jones (Development Planning and Research Associates, Inc.). Economic Impact of Costs of Proposed Effluent Limitation Guidelines for the Fertilizer Industry. (Prepared for U, S. Environmental Protection Agency.) Washington, D.C. Publication No. EPA-230/1-73-010. October 1973. p. 213. 7-24APPENDIX A = EMISSION SOURCE TEST DATA A.1 PLANT DESCRIPTIONS AND TEST RESULTS A.1.1 Introduction Available EPA data on particulate emissions and visible emissions from five different urea plants are presented in this Appendix. Results of formaldehyde and ammonia emission measurements are also presented. The uncontrolled and controlled emissions data included in this Appendix are analyzed and discussed in Chapters 3 and 4 respectively. The five plants where tests were performed are identified as Plants A, B, C, D, and E. The sources tested at each plant are presented in Tables A-1 and A-2. Mass emission measurements were determined by methods designated by EPA to provide consistent data and are similar or identical to the modified Method 5 presented in Appendix 8. Visible emission measure- ments were performed according to EPA Method 9 by a certified visible emission evaluator. Particle size distributions were determined using a cascade impaction collector, All standard units are for 293 K (68°F) and 29.92 in. Hg. of pressure. A brief description of each facility is presented followed by results of the testing. References for EPA emission tests are presented in Section A.2.TABLE A-1. SUMMARY OF MASS EMISSION TESTING PRILL TORER ‘GRANULATOR EMISSION SPECIES TESTED = TEU ORR RI TTTCTS Scrubber Scrubber | Urea Plant [ype Scrubter Inlet Scrubber Outlet | Scrubter Inlet Scrubber Outtet | Inlet duttet | Particulates Anmonta formalde a x x x x x 6 x x x x x oes x x x x Deco x x x x x x x owe x x x x ® heb = yon-flutdized ed FB = Fluidized bed © tuss aaission testing was also perforsed on 2 rotary drun cooler scrubber inletey TABLE A-2, SUMMARY VISIBLE BAtsSiOnS, OF VISIBLE EMISSIONS AND PARTICLE SIZE DISTRIBUTION TESTS PARTICLE SIZE DISTRIBUTIONS FanaTatO BrITT tower Tooter |——tranatatar Serabbar ower Scrubber [Cooler Scrubber [scrubber Scrubber Scrubber Pant putter outlet outlet Inlet ovtdet Inlet outlet Inlet outtet A x x x 8 x x c x x x x o x x x £ x x x 2 Vistble entsstons were also deternined for the outlet of a baghouse control ing bagging operations © vistote antssions not tested during test on Untt *C*4.1.2 Plant ale? Testing at Plant A was performed to gather urea particulate, anmonia, and formaldehyde enission data for the "A" and "C" granulators. Urea and anmonia emission measurements were also performed on the main vent for the urea solution synthesis and concentration process. The granulators operate on a 24 hr/day, 7 days/week basis at a production rate of approximately 363 Mg/day (400 tons/day) for each. Each granulator exhaust is ducted through a wet entrainment scrubber and fan before being discharged froma stack. The urea synthesis and concentration process operates on a continuous basis to provide urea solution for the entire urea plant. The exhaust from this process is vented from four locations which are combined and discharged through a conmon stack. Testing was performed at the outlet of this common stack. Mass emission tests and particle size distributions were conducted on the gas entering and exiting the "A" granulator scrubber. Visible emissions were determined for the exhaust exiting the 26 meter (85 foot) vertical stack from the "A" granulator scrubber. Mass emission tests were also conducted on gases exiting the granulator scrubber. Two different tests were performed to examine and evaluate factors affecting the accuracy of urea sampling and analytical techniques. Objectives of this study included establishment of a reference and analysis method quantification of possible sample degradation during storage (conversion of urea to other components) determination of the accuracy and con- sistency of analytical methods, and evaluation of the interfering effects of ammonia in the sample. This study concluded that urea particulate measurements for both granulator tests are representative of emissions. The results reported for the urea and formaldehyde measurements were determined for the samples using the colorimetric method of analysis. Ammonia concentrations were determined by direct nesslerization, for granulator "A" and nesslerization with preliminary distillation for granulator “C". Qutlet emission data for test run 9 on October 11, 1978 was discredited because a portion of the sample was lost. AndTABLE A-3. SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING THE "A" GRANULATOR SCRUBBER AT PLANT A’ (English Units) A 2 2 Ave. Senora) outa date 10-10-78 10-10-78 10-11-78 Teokinetie (5) 97.2 36.5 33.0, 97.2 Proguction Rate ¢ ron/day) 395 383 360 a7 Jnbitent Teno. (hve. Bry bul) > ® . > Relative Humidity 5 8 5 5 Exhaust Characteristics Floeraze inet: 0020 50670 43890 (aset/min) outlet 52090 5420 53500, Terperature ‘inlet: 163 1s 1s2 . outlet: 98 103 100 Motsture (t Vol.) “inlet 24 28 2.2 outlet ay 3a 39 Control Device characteristics Device Type Entrainment Scrubber Pressure Drop (in. ¥.S-) 9 7.3 18.3 we Lguid/Gas Ratio" (ga1/1000 ft) 3 ° > Etquor pi Ave.) 3.4 3.5, 9.3 Liquor Urea Cone. (1b/gal) inlet: 0.180 9.323 rae outlet! 8220 sao 4957 Urea Enisstons particulate Cones 1.75 : oos12 02a Crorke) 3.733 extaston Factor 38.7 Gorton} 0.238 Collection eftictency (2) 933 Ammonia Enissions Semonta. Cone. Gridser) Enission Rate Gove) Entasion Factor ‘ (o/ton) outtet siiat Coltection Efficiency (2) <0 Formaldehyde Enissions Formaldehyde Con. > > » > (cridser) 5 5 » 5 Enveston Rate > 5 5 5 Corre) 5 5 5 8 Extesion Factor 5 5 5 6 Goyton) 5 8 > 3 Cal lection efficiency 5 3 5 8 b= not avaitable ASTABLE A-4, SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING THE "A" GRANULATOR SCRUBBER AT PLANT A. (Metric Units) Test No. 1 2 3 vee ‘Seneral Dasa vate 10-10-78 wo1-7e = w-tl-78 Teokinette (x) o7.2 75.6 8.0 12 Production Rate (Mg/day) 359 383 ne M3 ‘Ambient Temp (K) (Ory Bulb) ’ b 5 e Relative humiaity (3) 3 . 5 5 Exhaust Characteristics Floxgate niet: 1385 1816 us un {ase" in) outlet: ware 1502 168 isle Senperature ‘let! 388 386 mS 5 te) outlet EH 508 Es no mossture (4 vot.) “inlet 10 2 2.8 22 outlet a3 a1 aa 33 control Device Characteristics Device Type Entrainment Serubber Pressure Srop (ke) 4.075 pic 3.650 4.075 Uigutayeas Recto (ave) t > 3 b Liquor ph (ive. Ey 38 3.3 3.4 Uiauor bree tone. tng/s )_ inte need 29720 5360 500 oat s0e200 1012000 0000 62940 rea Extssions Parzicylate Cone. 25.761 3.278 25.046 25.009 {fans} 0.0129 9.0250 0.0181, g.018s Extssion Rate 38760 42327 asset sears (g/hr) 19.091 35.737 29.911 28.246 Easton Factor 143.08 1360 12.85 169%68, (g/kg) 0.077 0.145 0.136 0.119 EB iReeton ertterency (2) 38:3 9 33 39 Aamonie Exisstons, Aanontg, Cone. 0.213 0.208 (Srasn') oad 03598 Extaston Rate 206 58 25.68 (nr) 55.08 370.08 Seetton Factor Tes {399 (o/s) 2681 Koa EEtTEdeton eettctency (2) a @ formaldenyce Enisstons Formalgehyde Con. . 5 > > (sresn’) 5 5 > 5 Enisston Rate 8 5 5 5 Grr) 5 5 3 8 Exieston Fector 5 3 8 b 10" (a/ks) eutlet! 5 3 5 S correction Efficiency (3) 5 8 8 ° De Wot avatlablzTABLE A-S. SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ‘ON GASES EXITING THE "A" GRANULATOR SCRUBBER AT PLANT A (English Units) Test ". 1 2 3 Seneral_ Oats fate 0-11-78 10-11-78 Teokinetic (2) 100.7 01.2 ot Broduceion Rate (Tons/dy) MT ag 305) Jmbient. Temp. (°F }(Ave.. Ory bul) > > > Relative Hunidtty 5 > 6 Exhaust istics Flowrate inlet: sza10 50860 51080 (aset in) cutter: 55359, 53550 5ua30 Temperature ‘let: er 168 tee *y outlet: 39 105 102 Moisture (# Vol.) ‘tet: 16 21 2.3 outlet 5 a3 ae Control Device Characteristics device Type Entrainment Scrubber Praseure Drop (ine M-Ge) 3 18.5 18.9 wa wa Uiquta/Gas’ fatto. (gat/1000 f°) ® ® > > quer pit (ave) 3 5 b 5 Uguor Grea Cone. (b/ga) énlet 5 8 5 5 outlet 5 > 5 > Urea Enisstons Particulate Cone. not 10.740 10.19 10.620 (ariaser) i} 0.0126 000342 000800 Entssion Rate 4946 4629) 4396) 9550) Cloyne) > 5.860 Lem 3.720 Entesion Factor 233 aa 250) 276 (G/ton) > 0.352 9.0908 8.221 Coltection & 5 99.3 100 93:3 smonta Entssions Smnonia, cone. inlet: 0.0936 0.0614 0.0685 0.0732 (gr/dset) cutlet: ° 0.138 91168 0.151 Emission Rate ‘alee: 42.080 252450 27380 zo (bite) outlet: > 621480 F720 Enisstion Factor ‘inlet: 2 1588 Veo Gorton} outlet ® 3780 oa29 Cab feet ton ertictency (2) 8 a @ yma} dehyde Eni ssions Fomal dehyde Con. inlet, 9.000225, 00000808 0.900427 9.000278 Gor/dset} outlet: b 9:0001285 000715, :0000368 Emission Rat ‘ntet 0.166 0.0350 0185 o:1z2, Gorhe) outlet: 8 0:0588 oo3z8 0.0858 Enission Factor ‘let: 9.00883 1.00209 90106, 0.00737 Gojton) couttet: > 000353 o:o0189 Dover Collection Efficiency 5 @ a b= noe avaiTableTABLE A-6, SUMMARY OF UREA, FORMALDEHYDE, AND AMMONIA TESTS ON GASES ENTERING AND EXITING THE "A" GRANULATOR SCRUBBERS AT PLANT A. (Metric Units) SSE ee a eS Test Ho. 1 2 3 ave. Genera tata tate 0-11-78 wens etre Teokineete (5) 00 iat ia am Production Rate, (Mg/éay) 360 33 EF Es Ambient Temp (K) (Ory Bu1b) . B R t Relative huntat ty 5 5 3 t Exhaust Charactertsties Flowgate inter: se ue use usr {sent in) ute 1587 1580 isi ise Tenpereture ‘net: He ue 39 et 1) HO Ee ne a Hoisture (x Vol.) 1s 3 27 2.3 ae at i a3 Control Device ch device Type Entrainment Serubber Pressure Drop (xP0), 3.88 3 3.83 3.70 Uiquta/tas patio. (17a) 3 3 3 ° Liquor pt (ave, 5 5 3 8 Liquor Grea Cone. (Mo/2) inlet: 5 5 3 b outlet 5 8 3 5 Urea Entss one Pareicylate Conc, inlet: 25.189 2.372 2.07 {ayésn’) outlet: ° 00288 900783 Extaston Rave 7025 0028 aul (g/m) 3 a6.34 11.08 Emission Factor 48.5 13 us (ars) outlet: ° 0.276 0.0882 Cat teetion Erticteney (4) 5 3.9 10 smwonta Enisstons Aneonig, Cone. : 0.214 o.140 0.147 0.186 (Green) outlet: 5 0: 308 388 oa Extaston Rate 318.10 200"18 aver 23. (g/m) outlet: ° a 52278 2778 Easton Factor han 0.795 0:80, 0385 (aves) outlet: » ers 2s 088 EErieeeion ertictency (2) 3 @ é 4 Formal senyce Enisstons Formalgehyde Con. : 9.000744 .00018 9.000976 (aresas) outlet 8 roooess ——‘Bronoied Extsston Rate 1.105 0:255 1388 (afm) outlet: 3 octas 0:248 Exgssion Factor Conese 8:00105 0083 10" (orks) cute ° o:00u72 00085 Gatveetion Eetiesency (2) . 3 @ b= Not available A-8TABLE A-7, SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES EXITING THE "C" GRANULATOR SCRUBBER AT PLANT A. (English Units) arcane Test No. 1 2 2 General Data tate 12-18-78 12-19-78 12-19-78 Isokinetic (2) 107.2 105.7 108.2 Production Rate an 370 370) Anbient Temp. 6 > > ‘Ambient Motsture (3) 5 5 > Exhaust Characteristics Flowrate fiolet: > > > (ascfm) outlet: 55180 54220 51130 Temperature ‘inlet > > > ry outlet: 2 102 108 Notsture (x Yor.) “inlet: ° 3 > outlet: 6.0 28 Ba Control Oevice Characteristice Device Type Entrainment Scrubber Pressure Brop (ins WoC.) 6 > b Liguid/Gas Ratio” (gat/1000 #1. 6 5 5 Liquor pl (Ave) 5 5 > Liquor Urea Cone. (2) ave.) 8 b B Urea Emissions Particulate Cone. inlet: » > b (ar/aset) outle 9.0278 0.0831 9.170 Enission Rate ‘inlet: > 5 b (oshe} outlet ad 20.19 7.438 Enission Factor ‘inlet > > b (1b/ton} outlet: 0.850 1.339 9.493 Collection Efficiency (4) » ° o Armonia Entssions Ammonia. Cone. b > > (orvaser) 9.106 0.148 0.279 Enission Rate b b b (lomre} e772 68.02 122.36 Extssion Factor » ° 2 Corton} 5,674 asi 8.118 Collection Erficiency (3), e D b Formaldehyde Emissions Fomaldenyde Con, > > > (srvaser, outlet: 8.00172 8.00210 9.00186 Enission Rate b 5 5 (orne) outlet: 8.813 0.986 0.683 Entasion Factor ® 6 > Corton} outlet: 0.0526 8.0654 9.0083 Col lection E*ficiency (3) b b 8TABLE A-8, SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES EXITING THE "C" GRANULATOR SCRUBBER AT PLANT A. (Metric Units) Test No. 1 2 3 General Data Date 12-18-78 12-19-78 12.18.78 Isokinetic (3) 107.2 108.7 108.2 Proauetion Rate (Mo/day) 338 336 336 Ambient Tenp. (X) . > > Anbient Woisture (2) 5 5 5 Exhaust Characteristics Flowrate > . > {asn3/min) 1563 1550 1448 Tenoerature > > : i 208 ma 13 Noisture (x Yo.) 3 5 ° 5.0 3 81 Device Type Entrainment Scrubber Pressure Brop_ (KP) > > > Ueuid/Gas atta. 3 5 . Liquor pi (Ave. ) 3 3 5 Uiguor Grea Cone. (3) (Ave.) 5 8 5 res Enisstons Pareicplate Cone. inter: ° ° ° (g/asm) outlet: 0.0636 0.0985, 0.0388 Entsston Rate ‘inlet 5 5 5 @/hr) outlet: 59580 91590, 33740 Entsston Factor ‘inlet: ° ° $ fa/ks) outlet 0.825 0,669 0.287 Gotlection EFfictency (3) $ 3 ° somonia_Enisstone Jenonta, conc. inet: ° > ° ( g/dsm3) outlet: 0.424 0.332 0.639 Enveston Rate ‘nlet : ° : (Cg/hr) outlet: 39790 30869 55500 Ealstion Factor ‘let! : ° ° Coat utter 2.es7 Bass 087 Eahieeson Erftetency (8) : e 3 formal enyse Eatsstons Formal gehyde Con. ‘ntet > . 2 (sresrs) outlet: 8.00098 B.oo4e. ——_—8.00856 EASeTon rate ‘ne 5 5 5 (amr) cutet: 8 “7 no EtEHon rector ‘net: ° ° ° Cts) cutlet: danse 0.0327 8.0226 cal ieeeton Efttetency 6 b 8 A-10TABLE A-9, SUMMARY OF RESULTS OF UREA, AMMONIA AND FORMALDEHYDE TESTS ON THE GASES EXITING THE "C" GRANULATOR SCRUBBER AT PLANT A, (English Units) Test to. * 5 6 aves Date a Tyokinette (3) 2 2 106.8 Preduet ion fate_(Tons/day) 30 noi ent * > Anoient Yorsture (2) > Exhaust Characteristics inet: 5 > > outlet s2s10 51730 53780 ‘let: 4 5 > outtet 1s 108 ot niet. b > 3 outlet 49 hd Be Congrol device characteristics Device Type Entratnment Serubber Pressure Drop (ine WaGe) 5 > > > 6 Liguia vias atto(gat/ic00 #23) 5 5 > 8 Uiguer pe (ave. 5 3 > 8 Uiauer Sires fone. (2) (Ave) 5 5 > 5 ees Emssions articulate Conc. > > > > (ariéset) 0.0239 3.0188 3.0230 9.0281 Eniasion Pate > > > 5 Tovar) 10.25, 5.492 6 11a Zeissian Factor > 8 > ° Goyeen) 0.720 0.491 9.708 0.787 Collection EFFicrency (3) 2 5 > ° Ancona cone. > > > > out 5.161 0.152 0.139 9.07 b 3 2 8 outlet 72.95 8 6.08 2.57 ‘let: > ° > 2 outlet rey ‘8.20 Bas 5.222 {alteetion efficiency (3) > 3 2 > Formaldehyde Snizsions Formaldehyde Con. > ® > > (arvaser} 0.00197 0.00087 8.00188 8.0016 Emission Rate » > 2 > oie) 0.893 0.432 0.683 8.786 Emission Factor a > > > 8 Co/ton) outlet: 0.0582 3.0236 8.0839 9.0893 ol fection EFftetency b+ noe avattable ALLLTABLE A-10, ON GASES EXITING THE "C" GRANULATOR SCRUBBER AT PLANT A. (Metric Units) SUMMARY OF RESULTS OF UREA, ANMONIA, AND FORMALDEHYDE TESTS Test vo. 7 General tata ate Teokinetic (+) Produet‘on Rate (g/day) Ambient Temp (8) Anpient potsture (2) shavst_ characteristics Flomate fase'yntn} on inlet: Control Device charac: Device Type Bressure bro (kPa) Liquia/fas’ Ratio Liquor of (Ave.) Liquor tea’ Urea Enissions inet: outlet: inlet: Parcicylate Conc. > a 4. outlet aszt ° ()as0 Enission Rave (aire) Entssion Factor ‘inles {aya} cutie! Ebtteceion EFeictency (3) ose? 360 noni Enissions Aanonty Cone. (grasn’) Steen nace fe) shan Factor oe autie 1 dlcion erricieney (3 368 Fomaldenyse tnissions Formal gehyée Con. ‘arasn’| ssion Rate {a/hr} Entssion Factor (ovis) Collection Efficiency inlet: nas Encrainnent: Scruboer 0338 2.240 00223 ous 0527 352 318 128 00329 cats Ave. 108. a0 > 1508 a2 9. 5 378 408 561 003 oot? b= Not avattabieTABLE A-11, SUMMARY OF RESULTS OF UREA AND AMMONIA TESTS ON GASES EXITING THE SOLUTION SYNTHESIS TOWER VENT AT PLANT A. (English Units) Test ¥o p 3 ve. General Data Date 0-13-78 tein7e 1013-78 Tsokinetic (2) > > ° > Proauetion ate Tons/cay > 3 5 3 Ecotent Terp.F | 3 3 5 5 fmojert sotsture ( 3 5 3 3 Exhaust Characteristics Flowrate inlet: 5 5 > > {asefa) outlet v8 ed 990.9 1 Tenperature inlet: : A : ° 4) cutlet 8 188 18 18 inlet 3 3 3 ° outlet sh.97 8.37 90.56 28.37 characteristics pe Hone Prasaure drgp (in. 4.6.) > ° : > Uieutayar acta” ‘gat 1000 3 3 3 3 Ulauor a (Ave) 3 3 3 5 Liauor Grea fone. (4) Ave.) 3 3 5 3 ate Cone inlet: 2 5 3 2 (geisser} cuties 3.0061 S.onizs 9.0152 8.oorse Exrsston Rate inte 3 5 : 3 itera) outer 8.085 O3 a2 3. Entsston Factor inte > 3 > 2 (iay/tan) outlet: > 3 2 : Collection eFfictency (2) 3 3 3 5 Ammonia Cone. inet 2 5 b 5 Gariaser} outiet: 13.2 vhs ul? 128.8 EMsston Rate ‘inlet 3 > 2 i iTerae) cuties vst 148 us eet Entaaion Factor ‘ntet: 3 > > ° ileyton) gutlet 3 3 3 3 atteetion efficiency (8) 3 3 3 3 Fomaldenyde Eatssiont Formal denyée Con > ° 5 > ariseetl 3 3 5 3 EMaaion Rate 3 3 8 3 ilsran) 3 > 3 5 Entasion Factor 3 > 3 3 {Teton} 3 5 3 3 Eatlectton e¢ftesency 3 5 3 3 = Noe aattale Ae13TABLE A-12, SUMMARY OF RESULTS OF UREA AND AMMONIA TESTS ON GASES EXITING THE SOLUTION SYNTHESIS TOWER VENT AT PLANT A. (Metric Units) Test No. 1 2 3 Wve. General Osta Date 10-13-78 8 Teoxsnetsc (=) > > > Production Fate (4a/day) b 6 5 Ambient Tenp. ( 5 5 5 Ambient Notsture (4) b 8 5 Exhaust Characteristics Flomgate inlets > > b > (asm fain) outlet: 35.32 38.02 28.08 32.46 Tanperature inlet: ° > . 3 i outlet: 382 358 388 388 Yelsture (4 Yo.) “intet: . > > 5 outlet: 3.97 8.5; 90.56 32.97 ntro}_ Device Charactertstics device Type one Pressure Srop (473) 5 5 . > > Liquia/éas’ tact (1/0°) 8 8 5 5 Liquor ai (Ave. 8 5 8 3 Liquor Urea Cone. (5) (Ave) 5 5 D 5 > > ® > a 0. 00268 5.0388 9.0172 Enission Rate > 5 ® . (ayne) 2 58.02 58.02 33.98 Eniesion Factor ° > 3 ° 3) b . 5 ° HTaction E*ttctency (3) 5 > 2 5 Ansonia. Eatestons ‘Ammonia. Cone. » > 2 (yas) sas 301. 28.7 Eniseion Rate > > . ns ees77 538266 582938 si1on Factor 5 . 3 42) 5 5 ® CSiTaceion eFticteney (4) 5 8 8 Fomaldenvae exissions Formaldenyde Con. inter. > ° > > (syased) outlet: 5 5 > 5 Enssion fate ‘alet! 5 5 5 3 Coyne outlet: 5 b 5 5 EAlsston Factor ‘inlet: > b 5 8 (9/49) outlet: 5 5 5 > cRteeeton efficiency (2) 5 b 5 2 b= tot wattable Aeldst-y TABLE A~13. ON "A" GRANULATOR SCRUBBER AT PLANT A\ SUMMARY OF INLET AND OUTLET PARTICLE SIZING TEST RESULTS Inlet Sampling Test Test Aerodynam' Mass in Location Date Time Size Range, Size Range, % Scrubber 10/12/78 0919-0929 >2.2 100 Inlet Scrubber 10/12/78 1109-1509 98.7 Outlet 0.00 0.14 0.21 0.39 0.39 Scrubber 10/12/78 1629-2029 89,35 Outlet 0,00 7.32 1.31 1,05 0.97 Scrubber 10/13/78 0855-1255 65.34 Outlet 0.00 4.54 0.67 23.89 5.56 Scrubber 10/13/78 1316-1317 100 Inlet Scrubber 10/13/78 1508-1509 >2.4 100TABLE A-14, SIX MINUTE ARITHMETIC AVERAGES OF OCTOBER 10, 1978 OPACITY READINGS ON "A" GRANULATOR SCRUBBER STACK AT UREA PLANT A Average Opacity Test Date Time for 6 minutes 10-10-78 RSEBRLRS A-16TABLE A-15. READINGS ON * AT UREA PLANT A SIX MINUTE ARITHMETIC AVERAGES OF OPACITY GRANULATOR SCRUBBER STACK Test Date Avg. Opacity Time for 6 min. Test Date Time Avg. Opacity for 6 min, 10-11-78 10-11-78 cof aaaaaaaaan ree wot wu bowen 125-14: 231-145 137-14: 243-14: 249-14: 155-152 301-15: 207-15: 210-16: 216-16: 122-16: 128-16: 134-16: 240-1 246-16 152-1 258-1 204-1 210-1 216-1 222-1 228-1 134-1 240-1 246-1 152-1 258-1 19-14: *Averaging time less than 6 minutes. AnLTTABLE A-16. SIX MINUTE ARITHMETIC AVERAGES OF OPACITY READINGS ON "A" GRANULATOR SCRUBBER STACK AT UREA PLANT A Test Date Time 10-12-78 4:45 1 1 1 ul i 7 1 1 “Averaging time less than 6 minutes. A-18 Average Opacity for 6 MinutesA.1.3 Plant _B? Testing at Plant B was performed to gather urea particulate, ammonia, and formaldehyde emission data for the "3" granulator as well as urea and ammonia emission data from the urea solution synthesis and concentration process. The granulator operates on a 24 hrs/day, 7 day/ week schedule. Exhaust from the granulator is ducted to a wet entrain- ment scrubber and then a fan prior to being discharged through a 24 meter (80 foot) vertical stack. Testing was performed on the gases entering the granulator scrubber. The urea synthesis and concentration Process operates continuously to provide urea solution for the entire urea plant. The exhaust from this process is vented from four locations which are combined and discharged through a common stack. Testing was Performed at the outlet of this common stack. Particle size distributions were determined for the granulator exhaust entering the granulator scrubber. Visible emissions tests were conducted on the emissions exiting the vertical exhaust stack from the scrubber, The urea concentration in the samples was determined by the Kjeldal method of analysis and is corrected for possible urea loss during analysis. Anmonia concentrations were determined by direct nesslerization and corrected for possible conversion of urea to anmonia. Formaldehyde data was determined by the chromotropic acid method of analysis. The inlet percent moisture values reported for the stack gas were based on separate moisture test runs. Moisture contents is typically measured concurrently with each test run. A-19TABLE A-17. SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING THE "B" GRANULATOR SCRUBBER AT PLANT B (English Units) Test fe. 1 2 3 ve. General ata Date 01-17-79 o1-t779 1-18.79 sokinetic (2) 102 103 102 102 Production Rate_ fTons/day) 2 a 2 3 Ambient Temp, “F (Ave. Ory buTb) 8 8 a 6 Relative Mumidtty (3) % ® Es 2 Exhaust Characteristics Flow ate intee: a1ai0 41760 suu7 (asctay cutter: $5550 45760 45870 Temperature ‘niet? 130 139 ) outlet: to. 10t Notsture (x vol.) “inlet: 2.253 2.257 2.286 outlet: 73 5.608 621 Control Device characteristics Device Type Presse Brap (in. V0.) 20.9 20.5 20.7 Liquia/Gas fatto” (ga1/io00 Ft?) > ° > Léquor pi (ave.) a6 37, 8.7 Liquor brea Cone.(18/¢81) inves 00000101, 000000898 ~oao000e6 outle ooasa Tooa7a ‘00863 22 Emissions Particulate Cone, 5.516 5.511 5.425 (ar/ésef) ovale 9:0101 0.210 Emission Rate 2312 2330) 2268 (orne} $.500 3.931 4.042 Enfssfon Factor 32815 aie wrz (ovton) out 0.285 9.226 0.248 Collection eftictency (3) 99.8 93.8 99.8 Sononia Eniesions anonta, cone. 9.0989 (Gorvases) 0878 Enission Rate sucka (arm) ielea Entasion Factor 30 Go/ton} outlet Le Catlection ertieteney (3) wt Forms) ehyée Enissions forataepyae Con. tntat o.oo ridse outlet: 9000 ‘i ign tate ciniet: pen Emission Factor nee: o.otis (in7eem) outiet 0:0063 Collection Efficiency (%) a7 9.1073 a + considered confidential by manufacturer 3 not avattable A-20TABLE A-18. SUNMARY OF RESULTS OF UREA ANMONTA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING THE "B" GRANULATOR SCRUBBER AT PLANT 8 (Metric Units) Tet 5 2 3 Wen Sage bate oar aiirzy —o1-ieag Baking (2) ie Fi re ws Prana sate (H/ > (eset) 308 33 38 Tencer 5 > > Fe 3 wi 190 votsture (& 1.) ° ° : * ea. 8.1 Control device device Type None Pressure Drop (in. HG.) 5 > > > > Ufquia/tas Ratio" (gal /t000 #t3) 5 5 5 ° Uiauor pi (ave) 5 3 8 5 Uiauer Urea one. (3) Ave.) 5 3 5 ° rea Eoisstons Particulate Cone. iter: > > > > Goriaser} outlet 8.518 O61 6.593, 6.619 Emission Rate ‘niet 3 > 3 5 hare} cutter se ba i ise Emission Factor ‘ntet 5 ® 5 3 ofan) qutlet: cert 0.0377 9.0306 5.0317 Colfection efficiency (3) 3 5 8 5 smmonia Enisstons Jmenia cone. fntet: b > > > (grvdset) outlet 115.2 Mia 178.5 43.5 ‘ntet » 2 3 > outlet 312.0 498.6 508.8 al Enisston Factor inlet: 3 3 ° ® Garton) utiee 5.09 58 3.00 5.22 atfection Efficiency (2) Formaldehyde Con. intet > > > > (sriaset), cuttet 5 5 5 5 Exission Rate intet 8 8 5 5 Grae) uciee: 8 3 5 5 Entazton Factor ‘te: 3 3 5 5 Garton) outtee: 5 5 5 5 Colfection Efficiency 5 5 5 5 a + Considered confidential by manufacturer 5S not avattabieTABLE A-20, SUMMARY OF RESULTS OF UREA AND AMMONIA TESTS ON GASES EXITING THE SOLUTION SYNTHESIS TOWER VENT AT PLANT 8. (Metric Units) Test. 1 2 3 aves Senerat Date 9 91-19-78 1-19-79 TSoksnetic (=) 8.2 58.5 73 16.2 Prequetton Rate (Ng/day) 3 2 a @ Acbtent. Ten. > 5 > 3 Anprent HoTsture (5 5 2 5 ° Exhaust characteristics Flowgate > > > {asm in) 3.660 37 9.598 3.77 Temperature 2 2 8 > ts) 362 388 381 380 Norseure (% vole} = 5 ° . si s.9 Bis 6a 0 kPa > > > > Liquid/Gas’ acto (103) 5 5 8 5 Liquor’ pi (Ave, 8 > 5 5 Uhavor Grea tone. > 3 5 8 Ue articglate Conc. inlet: > > > (gies cutie 408 1.407 I Las Emission Race ‘ale b 3 > ginny outlet 962.5 780.9 328.3 Emission Factor ‘inlet: 3 > 3 (gia) outlet: 0136 0.0189 3.0183 9.0159 Col eet ton EFFiciency (3) ° b > sononia Enissiont ‘ewonig Cone. inlets > > > > fa/dse’) outlet 288.3 303.3 383.3 ma. sston Rate inet? 3 > 3 > (sine outlet 137108 220018 230087 reser ERSston Factor inlets 5 > 8 Gy outlet 2.885 4.320 4.500 3.010 cBlVaeeion Efficiency (4) 5 » y > omaldehyde Sniss ong Formal detyde Con. inter > » > > sn) outtet: 5 5 5 ERisston Rate ‘alee 5 % > 5 (o/h outlet: 3 5 5 5 Enission Factor ‘alee: 5 5 8 5 (ayia) ou ® > 5 5 Collection eftictency 72) 3 ° 5 5 Teter nai haole A-23ve-v TABLE A-21. SUMMARY OF INLET PARTICLE SIZE TEST RESULTS ON 'B' GRANULATOR SCRUBBER INLET ON JANUARY 18, 1979, AT UREA PLANT B Sampling Test Test Aerodynamic Mass in Location Date Time Size Range, wm Size Range, % Scrubber Inlet 1/18/79 15:28-15:43 > 6.0 99+ Scrubber Inlet 1/19/79 10:00-10:15 > 5.7 99+ Scrubber Inlet 1/19/79 11:25-11::40, > 5.8 99+TABLE A-22, SIX MINUTE ARITHMETIC AVERAGE OPACITY READINGS ON "B" GRANULATOR SCRUBBER STACK AT PLANT B 6 Minute Avg. Opacity Date Time Period for 6 min, 1-17-79 COON OIVDCON CVO OOAAR MEA © CORO ROW EW BOW UL baw wWaNwowaw *Less than 6 min. average. An25TABLE A~23. SIX MINUTE ARITHMETIC AVERAGE OPACITY READINGS ON "B" GRANULATOR SCRUBBER STACK AT PLANT B 6 Minute Avg. Opacity Date Time Period for 6 min, 1-19-79 - 9.3 - 8.5 - 74 - 8.3 - 14 - 8.1 - val - 13 - 8.5 - 8.5 - 9.0 - 8.2 - 9.6 - 8.3 - 8.8 - 8.5 - 8.8 - 7.9 - 7 - 7:3 1-18-79 12:44 - 12:49 9.0 12:50 = 12:55 6.0 12:56 - 13:01 5.0 13:02 - 13:07 5.0 13:08 - 13:13 5.0 13:14 = 13:19 5.0 13:20 - 13:25 5.0 13:26 - 13:31 5.0 13:32 - 13:37 5.0 13:38 - 13:43 5.0 A-26Ad Plant ch Testing was conducted at Plant C to determine the urea and ammonia emissions in gases exiting one of four scrubbers on a prill tower and at the inlet of a rotary drum cooler scrubber. The testing was performed during the production of fertilizer grade urea. The prill tower operates at approximately 336 Mg/day (370 tons/day) on a 24 hr/day, 7 days/week basis. The prill tower exhaust is controlled by four packed bed wet scrubbers of in-house design. The exhaust from the rotary drum cooler is controlled by a mechanically aided scrubber. Particle size distributions were determined for the gases entering the prill tower scrubber and the rotary drum cooler scrubber. Tests performed on April 2nd and 3rd, 1979 were conducted during agricultural grade urea production. Tests performed on April 4th and 6th, 1979 were conducted during feed grade urea production. Visible emissions were conducted during both agricultural and feed grade urea productions for the outlet stacks from the prill tower scrubbers, and during agricultural grade urea production on the outlet of the rotary drum cooler scrubber. Also presented are flowrates through all four prill tower scrubbers. This is presented to verify that conditions in the single tested scrubber are representative for all four scrubbers. Mass emission samples (April 1980) were analyzed for urea content with the p-dimethylaminobenzadehyde analysis method. Mass emission samples were analyzed for ammonia content with the specific ion electrode analysis method. A-27TABLE A-24, SUMMARY OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING PRILL COOLER SCRUBBER DURING FERTILIZER GRADE PRODUCTION AT PLANT C. (English Units) Test fo. L 2 3 Ave. General Data oats oer. o2e09 ——obetse0 teatinette (3) ips ioe ios 1s Panties Aide (Tosider a a a a febtent Yap? : i : Baa feutiee Retatey 3 3 3 3 shave Oaracerstics Flowate niet: 186 noe ms 70 (ren) ltt z : : $ $S5tPlure et ue ug 122 "3 ot $ $ i voiture (2 vol) “iat Lan Sine 3s outta > > > foveral Oavice character tics device tpt Entratnant Sernber Faire ap Gn 45.) 2 : 2 > Cigtdae Balen Sal 2000 #2) 3 : ; $ IS eh ses 3 : : : THauor Ore foie 5/a) tates : : : : outlet: 3 3 $ Parcicalate ne, Last asa Las lar) dscf} b SUSE ce Bar Be das dee aoe > > ° > Beer Factor Sam Baca 3.006 3.480 fan) : : : : EXieton rttctecy > 3 : 3 Amant con. o.on661 o.o10s a o.oo. ire) : : ; $ Sisto nate Bas Boe ’. Bae uy : : ; $ Srten Factor Bone B.oses Bose Bose ° : ° ; nes ° : 3 $ fareldandstntastons Foralaaryde Con. inlet: . ° : > rach ati : : ; 3 Erion nace alee 3 $ ; 3 goin) atte : : : ; Enesin Factor wale: : : : $ (inven) icle 3 : ; $ GHLlon ertcseny (2 5 3 3 3 a= Considered confidential by ranufacturer b= not avaitabie A28TABLE A-25. SUMMARY OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING PRILL COOLER SCRUBBER DURING FERTILIZER GRADE PRODUCTION AT PLANT C. (Metric Units) tate Trokiner Production Rate (Mg/say) Asbient Teno. («) Relative mumidity st Characteristice Flowgate (ase"/min) Temperature (8 Moisture (3 Hot.) inlets outlet: ‘inlet: outlet ‘niet outtat ntrol_Oer Device T Bressure Broo {4a} Ulauia/Gas @atso (1/m?) Liguori ve.) iguor Urea oe. Ma/ chara eristics niet outta Urea Entestions pargteulate Cone, gyasn") Exteston Rate s/he) Enission Factor sovonte Entssions moni (eresn’) aiseon tte ne Exton ton Factor Gis) autiet Callection E¢Fictency (3) Formaldehyde Emissions mal genyde Con. (gfase') seston tate Gimnt Enisston Fa (arta) Collection efficiency (2] inet: outtet: inlet: outlet ‘net? Aves 04-28-00 108 106 255 04-23-80 105 288 08-28-20 > . > oes 0256 2 = Considered confidential by sanufacturers 8 fot avattapie A-29TABLE A-26, SUMMARY OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES EXITING PRILL TOWER NORTHEAST SCRUBBER DURING AGRICULTURAL GRADE PRODUCTION AT PLANT C. (English Units) Tet te 7 2 3 Tee General tata cate ot-28-20 ot-25-29 04-25-80 Teotinetse (2) ive ioe io 106 Prowuction Rate (tons/day) 288 200 8 235 fnotent Terp (=F) (ory Bulb) 3 8 2% 8 Relative nuniatty (3) 3 s 2 a Exhaust Ouaractersettes sone > : cutee: 13078 ert 1ass9 ‘niet ° f 3 outlet: " 7 3 foisture (1 vol.) “inlets 3 3 $ cut tet: tang 3.676 $000 control Device Characteristics Device Type In House Design Wet Scrubber Pressure Drop ° > > 2 Uiaeta/tas atte 2 : 2 3 ° Liguer ph ve.) : a > : Ugior bran Cone. (.) _ intets 23.9 E09 ul 12. whs autiet: ° ° 3 3 res Enissions Barticslate cone. 2 ° ‘arveser) 6.0127 3.00806 9.00893 EXGSNon ate ° > 2 Gain) {za Lot 1.154 Entadlon Factor 2 5 : ‘ievton) 8.9 3.0896 8.0838 cailedtlon ertictency ( ° 3 3 sononta_Entsstons fonania cone. » 2 : Boze 3.0561 Sones ° 3 : Geine) 3.189 5668 faa fateston Factor 2 3 3 Giorean) 8.262 see fate EaMteetlon eetictency (2) 3 3 5 Fomatsenyae Entsstons Formal enyde con. : > > 2 iarraset) : 3 3 5 eateston fate 5 3 5 ° (ib/he) 5 3 5 : fat sston Factor ° 3 3 2 {ieyton) cuttet: ° 3 3 5 daheetion erttcteney (2 3 3 5 . De tor waTtatle henTABLE A-27, SUMMARY OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES EXITING PRILL TOWER NORTHEAST SCRUBBER AT PLANT C DURING AGRICULTURAL GRADE PRODUCTION (Metric Units) eee a iz 3 a Production Rate {Ma/day) a z 38 26 Relative Humidity (2) > 35 82 5a wae ee era as ection ononia Exissioes Arwonig, Cone. > 5 + > (373503) 0.0683 9.0709 ° 9.0879 Envsston Race 2 2 2 5 afar) 1830 185d we} zo? Emission Factor 2 ° 3 ° fava) 8.135 3.146 3.271 9.160 cat teceton eFtictency (2) > b ° 5 Fama dehvde Eatssions Format gehyae Con. fintets > » > 2 (grasas} outtet 5 5 ° 8 Enisston ate inlet? > 6 3 5 (gimin} couttet: B 3 5 3 Emission Factor ‘niet? 5 8 5 8 (a/ka) outtet: 5 5 8 5 Collection efficiency (2) 5 5 5 8 4 Considered confidential oy manufacturer > = Not avaitapie A381ze-v TABLE A-28, SUMMARY OF PARTICLE SIZE TESTS ON THE PRILL TOWER SCRUBBER OUTLET DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT C Sampling Test Test Cut Diameter Mass in Size Location Date Time Size Range, um Range % Pril] Tower 4-2-79 1541-1556 >16.2 16.0 Scrubber Inlet 10.7-16.2 2.6 4.66-10.1 43 3.0-4,66 45 50-3.0 15.8 69-1.50 20.6 <0.69 36.2 Pril]. Tower 4-2-79 1806-1813 >17.79 5.2 Scrubber Inlet 11.11-17.79 0.0 5.14-11.11 3.3 3.31-5.14 41 1,66-3.31 18.7 0.77-1.66 50.5 <0.77 18.2 Prill Tower 4-3-79 0952-959 719.88 1.5 Scrubber Inlet 12.42719.88 1.9 §.75-12.42 2.9 3.70-5.75 3.6 1.87-3.70 49 0.87-1.87 44.5 <0.877 30.7 Pril]. Tower 4-3-79 1145-1149 >15.01 4.9 Scrubber Inlet 9.37-15.01 3.8 4,33-9.37 3.0 2.78-4.33 3.1 1.39-2.78 30.0 0.63-1.39 26.5 <0.63 28.7ce-v TABLE A-29, SUMMARY OF PARTICLE SIZE TESTS ON THE COOLER OUTLET AT PLANT C Sampling Test Test Cut Diameter Mass in Size Location Date Time Size Range, um Range % Cooler 4-3-79 1637-1652 217.49 98.50 Outlet 10.92-17.49 0.82 10.92-5.05 0.54 3.24-5.05 0.10 < 3.24 0.0 Cooler 4-4-79 1237-1247 >16.57 99.18 Outlet 10.35-16.57 0.44 4. 78-10.35 0.31 3.07-4.78 0.03 1.54-3.07 0.01 0.70-1.54 0.02 < 0.70 0.01 Cooler 4-4-79 1745-1755 >16.87 99.17 Outlet 10.53-16.87 0.38 4,87-10.53 0.39 3.13-4.87 0.04 1,57-3.13 0.00 0.72-1.57 0.00 < 0.72 0.02vey TABLE A-30. SUMMARY OF PARTICLE SIZE TESTS ON THE PRILL TOWER SCRUBBER OUTLET DURING FEED GRADE UREA PRODUCTION AT PLANT C Sampling Test Test Cut Diameter Mass in Size Location Date Time Size Range, um Range % Scrubber 4-6-79 1009-1015 > 16.61 11.9 Outlet 10.37-16.61 2.4 4,79-10.37 7.0 3.07-4.79 24.8 1.53-3.07 45.6 9.69-1.53 8.3 < 0.69 4.5 Scrubber Outlet 4-6-79 1530-1534 5 2 6 e2 2 6 Scrubber 4-6-79 1931-1934 12.7 Outlet 7.6 10.6 20.0 28.6 11.2 9.3TABLE A-31. SUMMARY OF VISIBLE EMISSIONS FROM THE PRILL TOWER SCRUBBER DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT C 6-Minute Average Date Time Period Opacity 04-02-79 1400 1406 20 14061412 26 1121418 16 14181424 B 1241430 16 1430 1436 20 1436 1442 23 14421448 21 14a8 1454 21 14541500 29 1530 1536 23.5 1536 1542 19 Y 1542 1548 23.5 1548 1584 26 04-03-79 0820 0826 12 9826 0832 8 0832 0838 6.5 0838 0844 5.5 08440850 6.5 0850 0856 6.0 0856 0904 9 0904 0910 7 0910 0916 6 0916 0922 8 0945 0951 7 0951 0957 15 0957 1003 18 1003 1009 24 1009 1015 16.5 1015 1021 10.5 1021 1027 13:5 1027 1033 13:5 1033 1039 15:5 1039 1045 14:0 ilo 1116 22 ls 1122 32 y 1122 1128 34 AA35TABLE A-32. SUMMARY OF VISIBLE EMISSIONS FROM THE PRILL TONER SCRUBBER DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT C 6-Minute Average Date Time Period Opacity 04-03-79 1128 1134 29 1134-1140 24 1425 1431 5 1431-1437 5.5 1437-1443 4.5 14431450 7 1450 1457 4 1457. 1503 6 1545 1551 5 1551 1557 3 1557 1603 6 1603 1609 8.5 AW36TABLE A~33, SUMMARY OF VISIBLE EMISSIONS FROM THE PRILL TOWER SCRUBBER DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT C 6-Minute Average Date Time Period Opacity 04-03-79 1609 1615 5.5 1615 1621 5 1621 1627 5 1627 1633 5 } 1633 1639 5.5 04-04-79 0845 0851 1.5 0851 0857 9.5 08570903, 9 0303 9909 13 0909 0915 7.8 0915 0921 2 0921 =0927 2.5 0927 0933 7 0933 0939 5 0939 ©0945 4.5 0946 ©0952 4 0952 0958 4 0988 1004 4 1004 1010 5 1010 1016 3.5 101s 1022 3.5 1022 1028 6 1028 1034 6.5 1034 1080 5.5 1040 1046 6.0 1046 = 1052 5 1052 1058 4.5 1058 = 1104 2.5 1104 1110 3 110 1116 4.5 161122 10 1122-1128 16 1128 1134 26 11341140 ul 11go 1146 6 1146 1152 45 11521158 10:5 ' 1158 1204 12:0 1204 1210 7.0 A-37TABLE A-34, SUMMARY OF VISIBLE EMISSIONS FROM THE PRILL TOWER SCRUBBER DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT C 6-Minute Average Date Time Period Opacity 04-04-79 1210 1216 8.0 1430 1436 16 1436 1442 u 1442 1448 2 14981454 11.5 1454 1500 10 1500 1506 10.5 1506 1512 10 1512 1518 12.5 1518 1524 12 1524 1530 1.5 1600 1606 0 1606 1612 0 1212 1218 0.5 1218 1224 215 1224 1230 015 1230 1236 1.9 1236 1242 2.5 1242 1248 1.0 ' 1248 1254 0.0 12541300 10 04-05-79 0930 0936 6 0936 ©0942 6 0942 0948 6 0948 0954 6 0954 1000 6 1000 1006 5 1006 1012 4 1012 1018 5 1018 1024 6 1024 1030 6.5 1030 1036 5 1036 1042 7 1042 1048 6 1048 1054 4 | 1054 1100 7 A-38TABLE A-35, SUMMARY OF VISIBLE EMISSIONS FROM THE PRILL TOWER SCRUBBER DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT C 6-Minute Average Date Time Period Opacity 04-05-79 1100 1106 9.5 106 1112 5 M2 1118 6 11181124 0 1241130 15 1220 1226 0.5 1226 1232 9 1232 1238 4 12381244 4 1244 1250 4 1250 1256 6 1256 1302 13 1305 1311 6 13111317 4 1317 1323 6 1323. 1329 8 1329 1335 9 1335-1342 8 1302 1348 16 13481354 30 1354 1400 37 1400 1406 34 A-39TABLE A-36. SUMMARY OF VISIBLE EMISSIONS FROM PRILL TOWER SCRUBBER EXIT DURING FEED GRADE UREA PRODUCTION AT PLANT C 6-Minute Average Date Time Period Opacity 04-05-79 1642 1648 5 1648 1654 5 1654 1700 5 1700 1706 5 1706 1712 5 21718 5 17181724 5 1724 1730 5 17301736 5 1 17361742 5 04-06-79 11001196 4 11061112 14.5 11121118 18 11g 1124 54 11241130 38 11301136 32 11361142 31 1421148 32 11481154 29 11541200 31 1206 1212 29 1242 1248 33 1248 1254 25 1254 1300 22 1300 1306 33 13061312 38 1312 1318 28 1318 1324 26 1324 1330 25 1330 1336 23 1622 1628 ctf 1629 1635 45 1636 1642 12 1643 1649 16 1650 1656 3 1657 1703 123.5 1704 1710 15 wi 1717 13.5 17g 1724 14.5 y 1725 1731 15 40TABLE A-37, VISIBLE EMISSIONS FROM ROTARY DRUM COOLER SCRUBBER OUTLET AT PLANT 'C'. 6-Minute Average Date Time Period Opacity 04-02-79 1400 1406 20 1406 1412 25 1g12 1418 15 1418 1424 20 1424 1430 23 1430 1436 7 1436-1442 30 1530 1536 25 1536 1542 20 1sa2 1548 25 1548 1554 27 An4lTABLE A-38. PRILL TOWER SCRUBBER OUTLET FLOW RATES* AT PLANT C Scrubber Outlet Time Run 1 Run 2 Run 3 Average Northeast During 13070 13730 13870 13560 Southeast Before® 11258 11808 12609 11892 After? + 12609 12150 12379 Average 11288 12208 12379 12135 Southwest Before 10496 12645 12888 12010 After a 12888 12798 12843 Average 10496 12766 12843 12426 Northwest Before 11814 12076 12902 12264 After aa 12902 12497 12699 Average 11814 12489 12699 12481 Total Flow” 46600 51200 51800 49900 aFiow rates calculated from velocity traverses performed before the indicated runs. below rates calculated from velocity traverses performed after the indicated runs. sum of during and average flow rates, rounded to the nearest 100 DSCFM. *Dry standard cubic feet per minute @ 68°F, 29.92 inches Hg. ‘*eVelocity traverse data invalid due to shut down of the prill tower. AnA2TABLE A~39. PRILL TOWER SCRUBBER OUTLET FLOW RATES* AT PLANT C Scrubber Outlet Time Run 1 Run 2 Run 3 Average Northeast During 370 389 393, 384 Southeast Before 319 334 387 337 after? * 357 308 351 Average 319 346 381 344 Southwest Before 207 358 365 340 After ” 365 363 364 Average 297 362 364 352 Northwest Before 335 342 366 348 After ” 366 384 360 Average 335 354 360 354 Total Flow 1321 1451 1468 1414 Flow rates calculated from velocity traverses performed before the indicated runs. Seiow rates calculated from velocity traverses performed after the indicated runs. “sum of during and average flow rates. *Dry standard cubic meters per minute @ 293 K, 29.92 inches Hg. **Velocity traverse data invalid due to shut down of the prill tower. AR43A1.5 Plant 0® Testing at Plant D was conducted to determine the urea, ammonia and formaldehyde emissions from a fluidized bed prill tower during feed and agricultural grade urea production. Urea and ammonia emissions were also determined for the urea solution synthesis and concentration process main exhaust vent. The prill tower is operated 24 hrs/day, 7 days/week producing approximately 1000 Mg/day (1100 tons/day) of urea prills. Exhaust from the fluidized prill tower is ducted to eight entrainment scrubbers located on the roof of the prill tower. Two (scrubbers "A" and "C") of the eight scrubbers were tested for mass emission in the inlet and outlet gas streams. Volumetric flow rates for all eight scrubbers are presented to verify that conditions in the two tested scrubbers are representative of all the scrubbers. The urea solution synthesis and concentration process operates on a continuous basis providing urea solution for the entire urea plant. Emissions from this process are vented through a single stack and then exhausted to the atmosphere. Mass emissions were determined at the outlet of this stack. Particle size distributions were determined for the inlet gas streams of prill tower scrubbers "A" and "C". Visible emissions were evaluated for the individual stacks of scrubbers "A" and "C" as well as for all eight scrubber stacks combined. Visible emissions were also evaluated for the outlet stack of the baghouse controlling emissions from bagging operations. Urea emission data was determined from samples using the p-dimethylaminobenzaldehyde colorimetric method of analysis. Ammonia emissions were determined using the direct nesslerization method of sample analysis. Formaldehyde emissions were determined with the chromotropic acid analysis method. It should be noted that preliminary velocity traverses on the scrubber inlet duct indicated that cyclonic flow existed due to axial flow fans. To account for this condition, the cyclonic flow angles were measured and the sampling probe rotated in accordance with the angle measured at each point. This is considered to be state-of-the-art procedure for these conditions. An4dTABLE A-40, DURING AGRICULTURAL GRADE PRODUCTION AT PLANT D. SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING PRILL TOWER SCRUBBER (English Units) Test to. 1 Wve. Generet Data tate 08-1579 08-16-79 08-17-19 Teokinetic (2) 13 102 Ed 101 roduetion Rate (tons/day) 104s 1088, 1038 id febient Texps tf (Aves Ory bulb) 79 73 31 30 Relative fumidtty(3) 8 % a 5 Exhaust Flowrate 8660 2800 7010 8250 (eset /nin) e210 0510 80830 61073 Temperature 1 i iis ua ca % %0 8 %0 Motsture (3 Vol.) 2.382 1.882 eae 2.038 588 556 Ben 3638 Control Device Characteristics Gevice Type Entrainment Sersboer Pressure Brop (ine 4.5.) a : a a Liguta/Sas Ratio” (561/000 Fe3) : 5 8 5 Uiauor pt ve.) : : : : Ltauor Ure Cone. (t8/gat) inlet : : a a outlet: i : i i Ures Emissions articulate Cone. 0.0922 0.0807 0.0811 (ar/eser} cuties: ori 900868 ovooes3 Exisston Rate ‘let sea way Govke} outlet 6.81 oa Envssion Factor ‘inlet: ee 156 Gojton} outlet: one o:099 Cobteet ton eFtetency (2) 8.2 08 somonta Enisstons fanonte, conc. sinter: 0.0283 0.0387 Gerisser) outtet: 0.105 9:108, Enleston Rate ‘let: 1193 ne ier} outtet: 5.01 sea? Entzston Factor 0.306 0.852 areon) outlet 127 1258 Collection Efficiency (3) o a mal denyee Emissions Formaldehyde Con. 9.000204 o.o0285 0.000281 0.000257 (arvaset} ‘ooocoae 0000102 ‘ooo0ni 0000008 Enlsston fate ons, o68, ates 01180 apne} 9.00837 900828 9.00876 000613 Emission Factor ‘nase 00368 “037i 00338 Go/ton) “pooto1 c00i1s “oootes cote Col ection EFFicteney( ost 96:3 6:8 96:8 ot avatteple Considered confidential by manufacturer AR45TABLE A-41, SUMMARY OF MASS EMISSION RESULTS FOR UREA, AMMONIA, AND FORMALOENYDE TESTS ON GASES ENTERING AND, EXITING PRILL "A" DURING AGRICULTURAL GRADE PRO AT PLANT D. (Metric Units) a Test Ne, t 2 a ve Senerat data vate Teakinetic (2 wt Broguerion Rate (Na/éay) Ed Anbient Tenp (K) i Relative Humidity ry aud) Exhaust Characteristics Flowgate inte ‘ese mia) utist: erperature ‘ntet te) outlet: wotsture (3 Vol.) “inlet outlet: nero] Device tharactertstics Gevice Tye Pressure Orop (4Pa) Ufauteyéas facig (17m?) quor pf (Ave. Tiguor Urea Cone.(¥y/2) inlet: outlet: Urea Entssions Parvicylate Conc. sntet: igitsn) outlet: Ehvssion Rate inte: arn) outet: extaston Factor ‘niet (area) uclet: Collaceion E¢#ietency (2) somonta Emissions farang, ores ayeene) eutlet Emission Rate s/t) out Bisson Gig) outlet £21 eekion eFtictency (4) Fomaldenyée Evissions Fomalgenyée Con. ‘arasn) out Entsston Rate (site) outlet: Emssion Factor (a/ia) utlet: Sbltection Efficiency (+) 1950 Vet ng ‘us 2.302 88 31855 577 Entrainment Scrubber 5 5 6 3 5.185, 0.186 0.0198 310135 zzo0g asd 2003, 2030 3.533 0.528 ol0zes 019495 094g 93.7 ar a8 9.0858 0.0865 0.9796 0.272 1227 0288 1030 10360 208 7340 23368 25263 0.282 9.251 0. 0.673 01586 a 0 a 3 cones 0.200888 0. 8 9;s000168 ‘:0000254 83900221 52.07 75:61 eae 1.38 281 233, otoose o:o0nes 00067 3°9000803, 9:000633 0000571, 96.2 96.6 96.6 foe avatlabie Considered confidential by manufacturer An46TABLE A-42, SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING THE PRILL TOWER SCRUBBER "C" DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT D. (English Units) Test No. 2 2 3 Ave. Generat_tai Date 08-15-78 08-16-79 08-17-79 Esokinettc (x) 110 is 108 a Production Rate_(Ton/day yo4e 1099 1092 1078 Anblent Temp. ry bulb) ry 79 at 8 Relative Humidity. (3) 6 6 a a Exhaust Characteristics Flowrate inlet: 62360 53660 59050 (asct/min) outlet: 56220 56450 62610 Tenperature ‘inlet: 13 nL ns ee} outlet 86 3 8 Moisture (x Yor.) inlet: 2.03 1.39 La? 58 outlet? 3:31 37 eo as Control Uevice characteris Device Type Inpingenent Scrubber Pressure Drop (in. MoS.) a 2 a 2 Liquia/Gas Ratio” (ga1/1000 #2°) > 5 5 5 Liquor 5 (Ave.) a a a 3 Liquor Urea Cone. (ib/ga} Dinter: 3 a a a outiet: a a a e cea Emissions Particulate Cone. tinier: 0.0806 0.0486 (arveser} outte 900965 Eatasion Rate inlet: 24.600 bin) outlet: % 51270 Entesion Factor inlet: 91408 0.541 (inten) outle oc 9.116 Collection Efficiency (2) 728 786 Anmonie Emissions Amonta, Cone. inlel 0.0237 0.0306 riase#) outle 0:0800 010385 fission Rate ‘nie: 10:90 1.49 Tb/ae) outle 18:37 18.45, Envsston Factor 0238 0349 01823, 0.805 a o Formaldehyde Emissions Formaldehyde Con inet 0.000128 0.000109 9.000338 0.000126 (ar/aset) 9000003, O.o0000%07 0: 0000088 000000086 Emission Rate 0.0683 0.0500 0.0707 0.0632 (io/hr) 0.00305 0:00517 000471 0:00829 Entssion Factor 0°00388 0.00108 0.00155 090181 (p/ton} 9.000070 9.000113 9.900103 000086 Collection e*Fictency (2) 95.6 8.7 93.3 93.2 2 = Considered confidential by manufacturer Not ovat Table An47TABLE A-43, SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING PRILL TOWER SCRUBBER NC" DURING AGRICULTURAL GRADE PRODUCTION AT PLANT 0, (Metric Units) Test No. 1 2 3 Aves General_oata, ate 08-15-79 08-16-79 08-17-79 Tsokinetic (2) 110 ne 108 a Production fate (1No/day) 3a 996 980 ot Aapient Tero. (K) 239 239 300 239 Anotent Notsture (3) 63 6 rH 3 Exhaust Characteristics, Flowate 178 1si9 1972 1552 (4sm°/min) 1592 60t 67 1653 Tenperature a8 ay ‘320 419 (¢) 403 400 301 401 Moisture ( ol.) 2.029 1.395 Lam 1.598 aa x71 so 35566 ontro}_ evi Device Type. Inptngenent Scrubber Pressure Orop (kPa) a a 2 a Licuid/Gas fatio (t/a) > > b > Efquor oH (Ave.) a a 2 a Liquor Urea conc. Ma/e inlet: a a a 2 cutee: a 2 2 a Urea Enisstons Particylate Conc. 0.106 9.0929 9.111 0.106 (a/dsm) outlet: 0.0131 o:028 9.0225 010299 Emission Rate zag) ages nga 10280 (g/he) outiet: 1247 2308 2380 1980 Extseton Factor 0.285 9.208 0.271 0.253 (o/s) outlet: oLo3i6 0.0856 919579 0.0486 Collection Eftictency (2) 38.3 728 78.5 20.8 Ammonia Enissions Jrmontg, Cone. inlet: 0.0871 0.0542 0.0700 9.0567 (g/ésn") outlet: 90590 00816 Loree 0.0768 Enission Rate ‘inet 4994 3998 7026 5629 (ayne) outlet: 3629 3785 3363 7578 Entsston Factor ‘al 0.127 ‘O.119 9.170 0.198 (o/s). outlet: 0183 oraz 0.203 0186 Cotlectton Effictency (2) Formaldehyde Emissions Formal ganyde Con. 0.000297, o.c0v2e9 9.000319 0.000288 Goyasn") outlet: o:o9001ee aloooczes 99000201 9:9000197, Entsston Rete aide 22183 32108 73.58. (aynr) outlet 138 238 21135 1345 Entsston Factor 0000736 D:000847 0.000777 ‘aoo7e1 (a/ea) outlet: 9:9000350 0.000080 00000520 ‘oao0180 Callectton efficiency (3) 95.8 88.9 28 2.8 == Considered confidential by manufacturer 5 tot avatiabie A-48TABLE A-44, SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING PRILL TOWER SCRUBBER "A" DURING FEED GRADE PRODUCTION AT PLANT OD, (English Units) Test Yo. General data ate 08-22-73 Trokinetic (+) ge 102 Production Rate (Tons/dey) 1138 1102 Ambient Teno, °F (Ave. Ory bulb) 3 32 Relative munidity (2) 6 65 Exhaust 0 Flowate 51720 s2010 52159 (asefal 49750 50390, anar0 Teaperature 189 ies tF} autiet: 105 02 Moisture (+ Yol.) inlet 2.385 utter 51360 Controt Device characters device Type Entrainnent Scrubber tssure Drop (ins Wee) 5 3 a a a Usquie/sas”aetio "(98171000 #23) 3 5 . 3 Ciauor pH ven! a a a a Liquor Urea Cone. (b/gat) inlet: a a 3 a outlet 3 a a a ea Enisstons Particulate Cone. inlet: o.u7 8 9.1002 outlet: 0.180 ° o.o1s2 ‘alee: 51.96 3 a7 Clove) outlet 5.03) 8 578 Eqission Factor ‘inte 1.036 1 0.957 Gayton} cutlet o.1as 0 oltze calieetion Eftieteney (2) 38.7 2 art (loyton) Col feet ion Giojton) Collection ertictency ( 900833 9.000123 0.000127 .0001120, Srogoner? :0009360 °0000¢88 9: oo02e8, 1.0359, 2.0879 2.0501 oroases 0.0131, a o.0nt7 ‘uo0a7e2 0.00315 o:00107 outlet s.000208 1.000275 9.000250 2) 38 78.1 78:5 78:5 4+ Considered confidential by manufacturer vot avarlapie An49TABLE A-45. SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING PRILL TOWER SCRUBBER "A" DURING FEED GRADE UREA PRODUCTION AT PLANT D. (Metric Units) Test to. i 2 7 nas General oaea Dace 08-20-79 08-22-79 Isokinetic (2 108 ioe 3 Proauction Rate (He/day) 1028 1000 1123 fanpient Teno. (x) 308 "01 302 fataetie tusialey (2) co 3 to Exnaust_ Characteristics Flowgate inlers ass 1855 1501 6 (45m ymin) outlet 1409 ns7 142 iad Temperature 355 360.2 355 387 te ie 33) 30 32 Moisture (+ Yol.) ‘ 2.7 3.416 2.508 2.085 par 51472 52st 50377 51380 trot Device characters sti Gevice 1 Entrainment Scrubber Uiquid/Gas Ration (i/m) 3 5 5 5 Ciguor pf (ve) 2 2 3 : Ligier Sea Conc. Ma/ 3 a 2 a outiet a a a a cea Enis tons Darticylate Cone. inet: 0.177, 0.229 (grasa) outlet: 9.0173 9.0528 ERtsston Rate ‘inlet 15566) 20s? aire) outlet: 1009 2823 ehisston Factor ‘let 0,268 0.878 sie) outlet: 9°0235 oosia cetlaction extictency (8) oat ant Anmonig, Conc. inl 0.238 9.289 263 Gres) cutter o.lie a5 5 esion fat ‘inlet 20970) 22162 (yar) outlet: 91675 ori2 Enistion factor inlet: 0.490, 9.529 $28 (ey). utter: 91225 01826 256 Calieetion Efficiency (3) 533 sr > Fomaldenyde Enissions inlet: 0.000283 9.000258 (/4sn*) outlet: 9ago08e4 99000856 Enfesion Rate ‘niet: 24a win (aire) utter: 5.82 531 Ereston Factor ‘inlet 9000877 07000535, (a7) outlet: 9.000138, 900125 Cotlaction efficiency (2) re) 78:5 Ton tdered cont Taanttal by aaRUaCEITE Rot avattable A-50TABLE A-46, SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING PRILL TOWER SCRUBBER “C” OURING FEED GRADE PRODUCTION AT PLANT D. (English Units .) Test to. 1 2 3 Ive. Seneral_ Data | ate 08-20-79 08-21-79 08-22-79, Tsokinetic (x) 103 104 108 108 Procuction Rate ( Ton/dey) 13 U3 101 23 Aabient Texp, °F (Ave. Ory bulb) 37 6 32 ra Relative Humiaity 6 6 65 6 Exhaust Charactersstfes Flowrate imet: 44180 48880 46920 85650 ) outiet: 46270 35160 50870 #7300 ‘inlet: 188 173 14 13 108 103 39 102 3.08 2.86 2.49 2.80 Baa 638 bi9 en Control Device Characteristics Device Type Entrainment Scrubber Pressure Drop (in, WS.) a @ a a Liquid/Bas Ratio" (21/1000 #%) > 5 ® 5 Lauer pi (Ave. 2 2 a a Liquor Ures Conc. tb/gat inlet: i a a 2 Urea Exissions Parcievlate Cone. 0.0983 0.0942 0.0987 fer/aset) 0062 9:0361 0.9138 Emission Rate 37:20 33.47 3.47 Corne) 270 6.23, 5.43 Entasion Factor 0.789, 0833 0.843 (iojton) outet 0.0873 0132 on116 Collection E#Fietency (2) 927 42 86.2 fenonia Emssions Armonia. Cone. inlet o.229 0.108 0.14 (grieser) outte 90893 0°9533, 00807 Emission Rate inlet? 28.96 44158 35.70 (io/he), outlet: 23152 20.63 24.61 Emission Factor inet 1035 9.382 378 Cayton} outiet ovag8 91495 526 Collection E¢tictency (3) 519 53.8 4621 Fomadehyde Emissions al dehyde Con. inlet 0.000100, 9.090325 0.0909886 0.000108 (orvaser) cutiet: ‘9.000028 (.0900337 ‘0090283 °.co00287 Emission a ‘inlet 9.0375. 0.0823 o.0387 o.o4z2| (ibyhr) outlet: 0.00987, 10130 oor oon Entssion Factor inlet: ‘s.000803 aoolo 000864 .0098ea (b/ton) outiet: ‘s.ooczez ooaers a.000266 a.n0024s Collection Efticiency (3) mar 75.0 63.2 13h Considered confidentia} by manufacturer hot available, ASS]TABLE A-47, SUMMARY OF RESULTS OF UREA, AMMONIA, AND FORMALDEHYDE TESTS ON GASES ENTERING AND EXITING THE PRILL TOWER SCRUBBER "C" DURING FEED GRADE PRODUCTION AT PLANT D. (Metric Units) 1 rare 2 3 we ieaal HE rete (2 ins oy was FEROS (he aye ie & ie es Jere antes oa ey ae s as ame Reidy es Bs x e 23 2 2 2 tant Guractertstics aa) EH B Fe ‘Seah 2 2 S 2 a Ed EF Es 3 rotate (2 .) foe fas m7 a aevce tee oie Sener FU Lae a. : : : ; rear at) : : : ; ESS BEE wre ine : : ; : vss isin (a/dsm") outlet: 0.0156 9.0368 9.038 199-9307 ee at hs sea oat od "2 (g/hr) outlet: 1226 2827 9299 2864 Ween ractr oe on 2.8 aus EB Retn errtcany (85 a a a a amon est seen oe ite: o.age oa oan ois ee mate mat a ant agg Br BO (g/hr) outlet: 10669 9350 13463 11163 (g/kg) outlet: 0.249 0.218 0,324 0.263 Collection Effictency (1) 81.9 $3.8 31.8 46.1 famulanede nissan faragene Gn aaese 2.iene oie? (ores at 1, ea 3a, (iar) outlet: ‘4361 ‘5.917 3.278 20 anrctr ates oles ee 3 ices (g/kg). out let: 0.000101, 0.000138 9.000133 0.000125. Collection Effictency () ‘14.7 175.0 69.2 m3 ‘Fo Conetcaree coatTaeneral By panuractarer 5 = wot availapie An52TABLE A-48. SUMMARY OF RESULTS OF UREA AND AMMONIA TESTS ON GASES EXITING THE SOLUTION SYNTHESIS TOWER MAIN VENT AT PLANT D. (Metric Units) Test to. 1 z 3 Aves ‘Seneral aaa date 08-22-78 08-22-79 08-22-79, Fokinettc (3) 124.4 132.6 130.7 129.2 Broguction Rate (%o/day) ose 1088 7 1088) Anblent Temp. (©) > > > ° Sapient Notsture’ (2) 5 5 5 5 Exhaust Characteristics Flowgate inlet: > > » > {dsm iin) out] 22.28 2.01 20.95 21.41 Temperature net: 2 > > ° © outlet: 355 Me we 38 Notsture (5 Vole} niet. in > > 2 outtet wa 78.1 3.9 79 Controt Device Characteristics . Device Type one Pressure Brap (42) 5 > > > > Uiqutdyeas facia (iva) 5 b 3 8 Liquor of (Ave, > 3 2 5 savor Grea onc. (2) (ave.) > 5 3 2 Urea Eatestonst efcutate Cone. inter: ° > ° > Caraiat) <.196 «0.169 3.168 «0.176 Enisston Rate inlet: ° > ° . Caine} outtet «262.1 ane 208.1 208.5 Enission Factor islet! > > 3 2 (ia) qutlet: 8.00605 «0.00880 -0,00855, «0.00815 Gol tection EFfictency (2) > > > > nmonta Conc. inter: > » » ‘ye outlet 483.5 00.4 498.1 Enission Pate ‘inlet: % 2 Coin) outlet: 646496 536962 actor ‘niet > (rio) gutter ae Col lection EFfictency (2) ° senvee Entssions Formal cehyde Con ° ° b > (fem) outlet: > 5 > > Emission Rate ° ° 5 > Comin outte ° 8 3 5 Emission Factor 8 5 5 8 (arg) ou ° 5 8 5 Col lection Efficiency (2) 5 5 3 . D> roe avattaple c * Concentrations were at the thresnold of detection A-53TABLE A-49, SUMMARY OF RESULTS OF UREA, AND AMMONIA TESTS ON GASES EXITING THE SOLUTION SYNTHESIS TOWER MAIN VENT AT PLANT D, (English Units) Test te. p z 3 sve. General Oata a 8-22-79 Kinetic (2) a u Bracuction Rate_(Tons/éay) a0 ue npient Tenge oF Give. Ory bulb) 5 Relative humidity i Exhaust Carectertsties Flowrate inlet . (suet). outlet: 706. 156 ‘tenpersture ‘net 3 outlet: 120 tet vovsture (2 v01.) “Inlet: . oui nd n nero} Device Characteristics device Tyge tore Pressure Brp (ine MS) 4 5 $ 5 Uiguteras sbes0"” (gat 1000 #¢3) 3 3 A Uiduor at (ives) 3 > 3 Chauor Grea Cosd. (x) fate 3 > 5 outle 3 3 3 Particulate Cone. . > ° (rreser} <0.tes8 2 2 Entsstons owonia, Cone» 2 : Gerace ute m3 293 26. reson nate 3 3 Govan} aie 42s 1382 03 Extaaton Factor 8 5 (ejeon) autiet ais 2.89 22. Eallectton efficiency (3) 3 Fomaldehyde Eatestons Forealdahyae Con. inlet > ° > {grvaser) outle: 8 3 3 Sesion ate ‘nee 3 : 3 (iar) cutter: 3 3 : Extaaton Factor ‘let 3 3 : Tin/ton) ou 3 : 3 Gailection eFttctency 3 : 3 2 o770 99 103 dential oy manufacturer Concentrations were at the thresrold of detection An54TABLE A-50. SUMMARY OF INLET PARTICLE SIZING TEST RESULTS ON SCRUBBER 'A‘ DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT D Sampling Test Test Aerodynamic Mass In Cumulative Location Date Time Size Range, um Size Range, % Percent A Inlet 08-14-79 1252 213.3 48.5 9,17-13.3 3.4 35 6.22-9.17 3.2 7 4,24-6.22 23) 9 2.72-4.24 13.8 6 1,36-2.72 4.0 8. 0.84-1.36 7.3 8 0.57-0.84 6.6 35 <0-57 10.9 9 A Inlet 08-15-79 0955 >14.5 24.9 10.0-14.5 9.2 75.1 1.8 65.9 12.8 64.1 6.4 51.3 10.7 44.9 0 34.2 20.9 34.2 13.3 13.3 A Inlet 08-15-79 1126 W RNEBOLwUON9os-v TABLE A-51. SUMMARY OF INLET PARTICLE SIZING TEST RESULTS ON SCRUBBER 'C* DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT D ‘Sampling Test Test Aerodynamic Mass In Cumulative Location Date Time Size Range, um Size Range, % Percent Cc Imet 08-16-79 122 >12.3 0 8.5-12.3 0 51.0 5.76-8.5 4 51.0 3.92-5.76 +0 50.6 2.52-3.92 22 47.6 1.26-2.52 +5 46.4 0.78-1.26 7 34.9 0,53-0.78 8 15.1 <0.53 +6 1.6 ¢ Inlet 08-16-79 1543 >12.2 21.2 8.4-12.2 0 78.8 5.69-8.4 1.4 78.8 3.88-5.69 2.5 77.4 2.49-3.88 3.1 74.9 1.25-2.49 22.8 71.8 0.77-1.25 22.2 49.0 0.52-0.77 21.2 26.8 < 0.52 5.6 5.6 € Inlet 08-17-79 1430 713.0 66.5 8.93-13.0 3.7 33.5 6.05-8.93 0.0 29.8 4.13-6.05 0.0 2.65-4.13 1.5 29.8 1.33-2.65 0.0 28.3 0.82-1.33 15.0 28.3 0.56-0.82 6.0 13.3 < 0.56 7.3 7.3is-v TABLE A-52. SUMMARY OF INLET PARTICLE SIZING TEST RESULTS ON SCRUBBER DURING FEED GRADE UREA PRODUCTION AT PLANT D Sampling Test Test Aerodynamic Mass In Cumulative Location Date Time Size Range, um Size Range, % Percent C Inlet 08-20-79 1555 215.10 71.8 10.4-15.1 0 28.2 6.0 28.2 3.5 22.2 49 18.7 3.4 13.8 5.0 10.4 0 5.4 5.4 5.4 C Inlet 08-21-79 1018 36.0 5.4 <0 11.9 6 8.7 7 5.7 0. 7.2 3 7.6 1 7.0 5 10.5 5 C Imet 08-22-79 0935 78.3 0.0 21.7 0.7 21.7 9.4 21.0 0.0 11.6 6.8 11.6 0.0 4.8 0.0 4.8 4.8 4.88s-y TABLE A-53. SUMMARY OF INLET PARTICLE SIZING TEST RESULTS ON SCRUBBER 'A' DURING FEED GRADE UREA PRODUCTION AT PLANT D Sampling Test Test Aerodynamic Mass In Cumulative Location Date Time Size Range, ym Size Range, % Percent A Inlet 08-21-79 1605 88.6 0.0 11.4 0.0 0.0 0.0 5.3 4 6.1 6.1 0.0 0.0 0.0 0.0 A Inlet 08-22-79 0935 84.4 0.1 15.6 0.0 15.5 7.9 15.5 0.0 7.6 1.9 7.6 0.0 5.7 0.0 5.7 5.7 5.7 A Inlet 08-22-79 1430 98.1 0.0 9.1 0.0 0.0 0.0 0.0 0.0 0.0 1.9 1.9TABLE A-54. VISIBLE EMISSIONS FROM SCRUBBER 'C' OUTLET DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT D (CONT) 6-Minute Average Date Time Period Opacity 08-17-79 1126 1131 11.5 11321137 16.3 11381143 15.4 11441147 10.3, 1215 1220 25.6 1221 1226 28.3 1227 1232 20.6 12331238 27.9 1239 1242 28.3 12491254 14.0 1255 1300 23:1 1301 1306 31.0 1311 1316 25.3 1317 1322 30.7 1329 1334 32.0 13351340 40.1 1341 1346 38.2 13471352 38.0 1353 1358 36.6 1359 1405 36.7 1408 = 1413 39.1 14141419 34.0 1420 1424 37.9 AH59TABLE A-55, VISIBLE EMISSIONS FROM SCRUBBER 'C' OUTLET DURING AGRICULTURAL GRADE UREA PRODUCTION AT PLANT D 6-Minute Average Date Time Period Opacity 08-16-79 1030 1035 25.5 1036 1041 7 1042 1047 16.7 1048 1053 20.2 1054 1059 17.9 1100 1105 19.4 1061111 22.1 42117 20.6 iia 1123 19.4 1241129 18.8 1215 1220 9 1221 1226 7 1227 1232 9 1233 1238 29 1239 1244 4 1245 1250 21 1251 1256 : 1257 1302 26.7 1303 1308 24.2 13091314 35.0 1315 1320 32:9 13211326 33.8 1327 1332 31.2 1333 1338 23.3 1339 1344 23.8 1415 1420 30.8 14211426 21:8 1427 1432 28.9 1433 1438 30.6 14391444 34.8 1445 1450 28.3 14511456 37.5 14571502 29.8 1503 1508 21.3 15091514 34.4 1515 1520 35.0 1521 1526 34.4 1527 1532 30.4 15331538 31.5 y 1539 1544 34.4 A-60TABLE A-56. VISIBLE EMISSIONS FROM SCRUBBER ‘A’ OUTLET DURING FEED GRADE UREA PRODUCTION AT PLANT D 6-Minute Average Date Time Period Opacity 08-20-79 1055 1100 19.0 11011106 16.5 1071112 15.9 11131118 17.9 11191124 12.7 11251130 13.5 11311136 10.9 11371142 7.5 11431148 5.0 11491154 3:3 11551200 5.8 1201 1206 9.6 1207 1212 71 12131218 ai { 12191224 13.1 08-21-79 0845 0850 22.3 0851 0856 26.3 0857 0902 29:6 0903 0908 24:0 0909 0914 22.3 0915 0920 20.2 0921 0926 e 0927 0932 20.8 0933 0938 21.7 09390944 26.5 0950 0955 15.4 0956 1001 16.4 10021007 20:0 1008 1013 215 1014 1019 21.7 1020 1025 21.9 1026 1031 28.1 1032 1037 24.6 { 1038 1043 26.7 1044 1049 22.7 A-61TABLE A-57, VISIBLE EMISSIONS FROM SCRUBBER ‘A’ OUTLET DURING FEED GRADE UREA PRODUCTION AT PLANT D (cont) 6-Minute Average Date Time Period Opacity 08-21-79 wz 1117 13.8 11181123 17.7 11241129 28.8 1130 1135 28.3 11361141 25.0 1421147 22.1 1481153 20.0 11541159 20.4 1200 1205 25.2 1206 1211 23.3 122-1217 23.1 1218 1223 23.1 1224 1229 20.0 1230 1235 24.0 12361241 21.9 A-62TABLE A-58, VISIBLE EMISSIONS FROM SCRUBBER 'A' OUTLET DURING FEED GRADE UREA PRODUCTION AT PLANT D 6-Mi nute Average Date Time Period Opacity 08-20-79 14471452 6.0 1453 1458 8.0 1459 1504 7.0 1505 1510 10.0 1511 1516 5.0 08-22-79 0950 0955 14.8 0956 = 1001 12.9 1002 1007 23.5 1008 1013, 26.3 1014 1019 28.3 1020 1025 29.8 1026 1031 33.1 1032 1037, 30.2 1038 1043, 32.3 1044 1049 31.9 1050 1055, 31.3 1056 1101 31.9 1102-1107 29.6 11081113, 27.3 11141119 26.5 1120 1125 26.5 1126 1131 28.8 11321137 25.0 1138 1143 24.4 11441149 23.8 A-63TABLE A-59. VISIBLE EMISSIONS FROM SCRUBBER 'A' OUTLET DURING FEED GRADE UREA PRODUCTION AT PLANT D (CONT) 6-Hinute Average Date Time Period Opacity 08-23-79 0930 0935, 14.0 0936 0943, 13.1 0942 0947 15.4 0948 ©0953 9.0 0954 0959 9.6 1000 1005 18:5 1006 1011 13:8 10121017 11:0 1018 1023 12.9 1024 = 1029 19.0 1010 1015 24.0 1016 1021 24.0 1022 «1027 . 1028 = 1033 27.5 1034 1039 25.6 1040 1045 29:8 1046 1081 30.0 1052 1057 28.8 1088 1103 29.4 11041109 26.3 110-1115 24.8 mis 1121 25.0 11221127 24.2 11281133 25.0 1134 1139 25.8 11401145 25.4 1461151 24.8 1521157 26.3 1158 1203 : 1204 1209 A-64TABLE A-60, SUMMARY OF VISIBLE EMISSIONS FROM BAGGING OPERATIONS AT PLANT D Test 6-Minute Average Location Date Time Period Opacity Baghouse 12-18-79 0845 0850 0 0851 0856 0 0857 0902 0 0903 0908 0 0909 0914 0 0915 0920 0 0921 0926 0 0927 0932 0 0933 0938 0 0939 ©0941 0 12-18-79 1030 1035, 0 1036 1041 0 1042 1047 0 1048 1053, 0.2 1054 1059 0 1100 1105, 1.0 11061111 0 eee 0 11181123 0 11241129 0 12-18-79 1135-1140 0 11411146 0 11471152 0 1153-1158 0 11591204 0 1205 1210 0 i211 1216 0 1217 1222 0 1223 1228 0 1229 1234 0 Baghouse 12-18-79 1340 1345, 0 1346 1351 0 13521357 0 1358 1403 0 14041409 0.6 A-659o-v * During 6 Berorg® Arter Average c During . efore Byter average © before Riter nverage F Before iter Average 6 Before Arter Average 4 Before After erage Toai® FHlowrates colculated fram velocity 0150. 1009 Spo 62360 57910 S720 ETbi0 65500 ‘0010 e870 60950 59480 0215 20 2450 78 soa10 210 Beato ‘92000 sasi0 eon su s7a000 “sum of averages, rounded to the nearast 1000 OSCFM. ory standard cuble feet per minute @ 60°F, 23.92 Inches ta. voi 44760 50720, aio 9050 si0 510 Biss ‘0900 S010, 0788 e210 siei0 59070, 59209 61020 61025 540 55530 55938 arz000 ‘8000 17000 (raverses perforsed before the Indicated runs at scrubbers A and C. Yrlourates calculated from velocity traverses performed after the Inticated runs at scrupbers A and C. 182000 185000 182000Daring 8 Sefore® 1676 426 1268 57 sa 9 10 wo? Atter fe Isi2 M37 Ie 969 oA uz Toy erage ier 1 1353 1436 o Toe nie oar ¢ uring ve 1520 154 nest ons 29 21 ° Before 1603 1370 M429 1393 1395 1296 101 iter ez wal ry 1295 re 13 1293 Breroge Be 1455 0 ne « tes) 1622 arse V0 1725 136 Average 8 1699 res F Before ua wea 1760 var After 386 170 es 160 iverage 1706 iat 1689 me ‘before 2046 1983 1962 1997 ter ig 1968, 1837 22 Average 1330 Toa, 1699, 1389 “ before 1599 186 1659 1639 After gaz oa 1368, ez Average 1385 ie tos 1630 Totat® 13500 13500 1400 13600 000 AFlonrates calculated fron velocity troverses performed tetore the indicated runt a scrubbers A and €, *rionrates catculated frou velocity traverses performed after the indicated runs at scrubbers A and C sun of averages, rounded to nearest 100 USCH. sory standard cuble meters per sinute @ 203°, 29.92 Inch Hs.A1.6 Plant €7 Testing was conducted at Plant E to determine the urea and ammonia emissions in gases entering and exiting a nonfluidized bed pril] tower scrubber. The testing was done during agricultural grade urea production. (Plant £ produces agricultural grade only). The prill tower operates at a production rate of approximately 272 Mg/day (300 tons/day) on a 24 hr/day, 7 day/week basis. The prill tower exhaust is ducted through a downconmer and then passed into a wetted fiber filter scrubber before being exhausted to the atmosphere. A preconditioning liquor spray is located in the downconmer prior to the entrance of the scrubber. Testing on April 15, 16, and 17 on the gases exiting the scrubber as well as the simultaneous inlet and outlet testing were performed with the preconditioning spray partially on, Testing on April 18, 21, and 22 on the gases exiting the scrubber was performed with the preconditioning spray fully on. Particle size distributions were determined for the prill tower exhaust entering the scrubber with the preconditioning sprays partially off. Visible emissions were determined for the gases exiting the prill tower scrubber stack and rotary drum cooler scrubber stack. Samples were analyzed for their urea content by the p-dimethylaminobenzaldehyde colorimetric (with preliminary distillation) analysis method. Samples were analyzed for their ammonia content by the specific ion electrode analysis method. Because of the relatively short (320 min.) sampling time and low emissions in the first exiting test on April 15, 1980, the amount of urea collected was below the direction limit for the analytic method. In order to detect the urea in this sample, a larger aliquot was concentrated during the preliminary ammonia removal step. For the Subsequent test runs, the sampling time was extended to 400 minutes. A-68TABLE A~63. SUMMARY OF RESULTS OF UREA AND AMMONIA TESTS. ON GASES ENTERING AND EXITING THE PRILL TOWER SCRUBBER AT PLANT E (English Units) Test to. 1 2 3 ive. bate, 4-18-20 ot-t6<89 04-17-20 Tsokinetic (3) it ioe 102 2 Production tate (Tons/day) 293 250 230 ai fsptent Teno.) 82 a 8 6 etattve Huniaity (4) 2 8 5 a shaust Charact Flowrate 2010 75100 8170 74830 asefn} 72310 4200 25590 20903, fercerature 108: 101 8 101 at cutter 3 7 % 7 Hoisture (2 Yol.) “inlet 3.58 1.86 tar 1 cutter oa 38 5108 a Control device Characteristics Device Tyee et Scrutber Pressure Drop (in 4.3.) 147 it wee 13.0 Ufauta/éas fatto, (gai/2000 143) 5 2 3 > Ltauor oh Gye.) 8.7 3.8 ba 8. Lauer brea Cone.) (Ave.)® ws 108 wa wt Urea Enissions Particulate Conc. 9.072 9.0466, 0.9 Griaset) 2 A:c00829 © 40 Enission Race 42.03 0.42 67 To/nr) ° 0.808 0:8 Evasion Factor 38 Fa fa ‘To/eon) o itso olovee eatiecvion efttciency b 97.3 9a. somonia_ Enissions Aemonta Cone. 9.0181 (gryases) 9:0686, EaTiston nate 1: woe ino} 6 6:49 EniSEfan Factor 8 ‘a8 {esa z 3:50 at feation Efficiency 0 a Formaldehyde Enissvons Fora genyde Con. a a a a (orraset cutiee: a a a 2 saission tate 2 a 3 2 (ibid outle: i a a 2 Emission Faczor A H 5 . o/tan) cutee: i i a a Gai feetion EFFictency (3) a a a a a+ nat available a= This average is for the scrubber Iiquor sun. A-69TABLE A-64, SUMMARY OF RESULTS OF UREA AND AMMONIA TESTS ON GASES ENTERING AND EXITING THE PRILL TOWER SCRUBBER AT PLANT E (Metric Units) Test No. 1 2 a ave. ate cect5-80 ot-16220 04-17-80 Tokinette (2) iat 102 rc 102 Production Rate (Hy/day) 256 268 288 288 Ambient Teno (K) (Gry Bulb) 330 287 269 288 Relative tuntatey (33 a2 38 5 8 Exhaust Characteristics Flongate inlet: 1926 215s 2242 2108 (esm'/nin) outlet: 2068 28s ist zoe Temperature ‘tet 313 ae at a2 te ito a Es) ios Moisture (x Yor.) 9.58 Has Lt Lor hat 38 2a as Control Device Type et Scrubber Pressure Srop (kPa) 37 Xo at aa UTautd/éas eacio,” (1/0?) > ° > 3 Liquor pf (Ave. > 33 88 38 38 Liquor brea one. ) m6 108 28 7 Uren Enisstons Particylate Cone. inlet: 9.165 0.218 0.168 (ciasn') outlet: ° 0.00892 0: 002¢7 eaisston fate 19081 29296 aus0a {ere} > 9.429 0.352 Emission Factor 178 2 si3e (c/a) outlet: > 0.0390 0: 320 Gattection efficiency (8) 5 98:8 98:3 fononta Emissions ‘smontg. Cone. o.ot10 0.0382 0.03485 0.0387 (craze) 0:236 0.256 9.208, 9.233 Emission Rate a 2933 ass aed (sine) 23250 36840 0160 32030 Enisston Factor 0.435 2.435 0.320 0.435 (Ges) Bs a3 27 29 Gattection Efficiency (x) o a a a Formaldehyde Ex\ssions Formaldehyde Con. a 2 a (ayasetl 2 : a Enssion Rate a : a a (re) a : a i Exisien factor a a a & i a a 2 a conection efftctency (2) a a 2 1 a= Not available bs Thisave. Is for the scrubber 1iquor sump A-70TABLE A-65, SUMMARY OF RESULTS OF UREA AND AMMONIA TESTS ON GASES EXITING THE PRILL TOWER SCRUBBER AT PLANT E. (English Units) Tet We. T 7 7 Te General Daca, tote ot-28-80 otzi-99 06-20-80 Tokinette (5) iat %¢ 2 99 Proauetion Rate, (Tons/d2y) 25 26 300 a fnpient Terp. (°F) ° ° ° 3 Sehative diy Ce S 3 5 $ Exhaust characteristics Fowrate inet > . 5 seta) outtet exei0 25408 e129 axo7a Fanperature ‘Het ° : : ° Cr cathe: 8 % ° * Moisture (4 vol.) “inte ° ° ° 3 outlet 3u6 fae in tne Control device Characteristics Bevice Tipe Met Scrubber Pressure Brop ( im ¥.6.) Ro 3 20 12.0 Hester Bate (asic ry ° ° ° ° Lauer pi Ove eee 7 77 agnor Brea tone (2) (Ave ae ae 28 af res entsstons Ferticulate Conc. 2 > > 8 eset 9.000732 0.000730 0.000775 9.000745 "tion Rate t 5 : ® Guin) 0.518 O36 2.560 8.537 salto restr $ 5 5 5 ay 8.c8a Q.ots0 Sones 0.0440 ERNEon erticteney (35 5 5 5 5 famonia Emissions enon Conc. niet: 2 > . ° Grreser) outlet: Butz 8.094 0839 o.02 Exisston Rate ‘te 5 : t 5 ia/ne) outlet: o.61 72.03 60.2 B86 Emission Factor ‘ne 3 : > itb/eon) cutlets ine 6.05 fee 8.00 Coitection E#tictency (2) 5 : 3 : Forme \dehyde Enissions Formaldehyde con. . » ° 5 (crise?) cutie 5 5 5 8 Esteston fate $ . 2 8 (lorie) cate : 5 . $ Eatetlan Factor : : : 5 Goiton) outlet 5 5 $ 5 Goiteetlon errictency (3) 3 2 5 3 b = not available € = This average ts for serubber quar sump. A-71TABLE A-66. SUMMARY OF RESULTS OF UREA AND AMMONIA TESTS ON GASES EXITING THE PRILL TOWER SCRUBBER AT PLANT E. (Metric Units) 1 2 3 Ae ot-t2-20 oe-2t-t0 nati (3) ‘ot 38 99 Proguetion 2ate (Ng/say) 28 260 28 febient Teno. () 3 5 3 Relative tuniaiey (3) 3 3 5 Exhaust Ouracteristics Fowrate inter: b 3 > o {ssn nin) outlet 230 ans 2388 zest Tanperatare ‘niet 3 > 3 3 i outst a 308 303 2 votsture (5 fot.) $ 3 3 3 5.6 a an hr device Type Pressure Srop (478) 5 30 3.0 Craute/ tas tette, (2) 3 3 lator’ pH Gve.J® a7 8.52 Trauor See’ Cone (2 (ave Be 33 Parvicylate Conc. inet: 3 3 3 5 iavésn’) cutter 9.00167 docs? 8.00172 Exisston Race ‘inte > 3 $ 5 ig/tr) ute abs 2 253. 283.5 EXanton Factor ‘nee ° ° 3 (ss) ule 8.0209 6. S.ozes CBr eeeton eetterency (3) 3 3 5 ievonia Enlssions 4emonig cone. ‘ater: 2 : 5 3 sidan) outtee 5.283 3 8.292 Oza zatsston Race Smit: 3 3 3 3 svar) cutter: sor 287 a8 23 Edson Factor ‘ate 3 : 3 3 ‘Gish qutlee: 3s Bo 2s 3.0 Catteceton efricseney (2) > a 3 5 Formal densde Enissions Formal genyce Con. tet 2 5 ° ° {greeny outlet 3 3 3 3 Enisston aace 5 3 3 3 res 3 3 3 3 ENS5'0n Factor 3 3 3 3 (ais) 3 3 3 3 collection Efttesency (=) 3 3 5 3 De Wer available ¢ = Trisaverage 1s for the scrusber liquor sump. An72TABLE A-67. SUMMARY OF RESULTS OF THE DOWNCOMER PARTICLE SIZE TESTS AT PLANT E el-¥ Test Test Aerodynamic Size Mass In Cumulative Date Time Range, (um) Size Range (%) (Percent) 04-16-80 1513-1520 80.0 5.8 20.0 8.8 44.2 4 5.4 1.3 1.3 04-17-80 0844-0904 67.1 2.3 32.9 8.2 30.6 10.2 22.4 12.2 12.2 04-17-80 1357-1411 66.8 0.8 33.2 2.3 32.4 5.6 30.1 9.2 24.5 15.3 15.3TABLE A-68. SUMMARY OF OPACITY READINGS ON THE SCRUBBER OUTLET AT PLANT E 6-Minute Average Date Time Period Opacity 04-15-80 1543 1548 0 1549 1554 0.4 1585 1600 0.2 1601 1606 0.4 1607 1612 0 1613 1618 0 1619 1624 0.2 1625 1630 0.2 1631 1636 0.2 1637 1642 0.2 1685 1650 0 1651 1656 Q 1657 1702 0 1703 1708 0 17091714 0 17181720 0.2 17211726 0.2 17271732 0 1733 1738 0.2 ’ 17391744 0 04-16-80 0925 0930 9.2 0931 ©0936 11.7 0937 0942 11.9 0943 0948 6.3 0949 = (9954 5.6 0955 1000 5.0 1001 1006 4.4 1007 1012 2.5 1013. (1018 3.8 1019 1024 4.0 1034 1039 3.5 1080 1045 3.3 1086 1081 71 1052-1057 8.5 1058 1103 6.5 11041109 7.5 An74TABLE A-69. SUMMARY OF OPACITY READINGS ON THE SCRUBBER OUTLET AT PLANT E 6-Minute Average Date Time Period Opacity 04-16-80 11101115 77 11161121 44 11221127 7.9 1128 1133 9.2 1158 1203 3.8 1204 1209 3.5 i210 1215 6.0 1216 1220 5.3 1248 1253, 4.0 1254 1259 5.0 1300 1305 4.6 1306 1311 17 1312 1317 6.7 1318 1323 8.8 1324 1325, 6.4 04-17-80 1104 1109 9.6 io 1115 9.2 11161121 8.1 11221127 6.9 1128 1133 7.5 1134-1139 6.3 11401145 6.3 1146 1151 4.6 1152 1157 7.9 1158 1203 8.8 1205 1210 7A 12111216 71 1217, 1223 9.6 1253 1258 11.5 1259 1304 9.0 1305 1310 6.5 1311 1316 9.2 1317 1322 11.7 1323 1328 13.8 y 1329 1334 15.5 A-75TABLE A-70. SUMMARY OF OPACITY READINGS ON THE SCRUBBER OUTLET AT PLANT E 6-Minute Average Date Time Period Opacity 04-17-80 1340 1345 14.4 1346 1351 13621357 1358 1403 1404 1409 14101415 14161421 1422 1427 1428 1433 1434 1439 1541 1546 15471552 1553 1558 1559 1604 1605 1609 1610 1613 1617 1622 1623 1628 Y 1629 1631 04-18-80 0900 9905 0906 0911 0912 0917 0918 0923 0924 0929 0930 ©0935 0936 0941 = 0942 0947 12.8 0948 0953 1.5 0954 0959 14.0 1010 1015 13.1 1016 1021 10.8 1022 1027 17.9 1028 1033 12.7 1034 1039 11.9 1040 1045 12.3 1046 1051 19.4 1052 1057 11.3 1058 1103 11.0 v 1104 1109 12.9 A-76TABLE A-71. SUMMARY OF OPACITY READINGS ON THE SCRUBBER OUTLET AT PLANT E 6-Minute Average Date Time Period Opacity 04-18-80 1115 1120 10.8 11211126 9.0 1127-1132 9.2 1133 1138 12.1 1139 1144 11.7 1145 1150 9.6 11511156 10.0 1157 1202 13.8 1203 1208 13.3 1209 1214 14.8 y 04-21-80 1005 1010 13.5 1011 1016 12.5 1017 1022 17.9 1023 1028 16.5 1029 1034 12.3 1035 1040 13.8 1041 1046 8.8 1047 1052 13.8 1053 1058 12.7 10591104 9.6 1120 1125 11.5 1126 1131 91 1132 1137 15.4 1138 1143 10.0 11441149 8.8 1150 1155 14.0 1156 1201 9.8 1202 1207 12.5 1208 1213 12.1 v 1214 1219 14.6 04-22-80 0845 0850 23.7 0851 0856 26.9 0857 0902 21.9 0903 0908 19.0 090g 0914 23.3 0915 0920 13.5 t 0921 0926 14.2 0927 0932 12.1 A-77TABLE A-72. SUMMARY OF OPACITY READINGS ON THE SCRUBBER OUTLET AT PLANT E 6-Minute Average Date Time Period Opacity 04-22-80 0933 0938 13.3 0939 ©0944 14:2 1010 1015 12:3 1016 1021 ase 1022 1027 5.8 1028 1033 10.0 1034 1039 15.0 1040 1045 12.3 1046 1051 10.0 1052 1087 8.0 1058 1103 4.2 1104 1109 5.7 1101115 7.6 1128 1133 95 11341139 wil 1140-1145 9.8 11461151 5.0 11521157 6.1 1158 1203 5.2 1204 1209 6.1 1210 1215 2.5 1216 1221 5.3 1222 1227 10:0 1300 1305 5.0 1306 1311 13:1 1312 1317 9.0 1318 1323 6.5 1324 1329 4.4 1330 1335 48 1426 1431 6.1 1432-1437 5.0 14381443 7.7 14aa 1847 12.0 04-23-80 0920 0925 B41 0926 ©0931 19:0 | 0932 0937 18.1 A-78TABLE A-73, SUMMARY OF OPACITY READINGS ON THE SCRUBBER OUTLET AT PLANT E 6-Minute Average Date Time Period Opacity 04-23-80 0938 0943 26.0 0944 0949 27.9 0955 1000 31.3 1001 1006 29.8 1007 1012 18.8 1013 1018 16.9 1019 1024 19.0 1036 1041 14.8 1042 1047 11.3 1048 1053 11.7 1054 1059 11.0 1100 1105 15.8 ilo 1115 23.1 iis = 1121 23.2 31221127 24.4 11281133 18:8 11341139 12:3 AR79TABLE A-74, VISIBLE EMISSIONS FROM ROTARY DRUM COOLER SCRUBBER OUTLET AT PLANT E. Test Date Time Avg. Opacity for 6 min. 10-16-80 900 - 905 3.1 10-16-80 906 - 911 2.5 10-16-80 912 - 917 3.8 10-16-80 918 - 923 4.0 10-16-80 924 - 929 3.5 10-16-80 930 - 935 4.4 10-16-80 936 - 941 3.1 10-16-80 942 ~ 947 2.9 10-16-80 948 - 953 1.5 10-16-80 954 - 959 1.0 A~80A.2 REFERENCES a Urea Manufacture: Agrico Chemical Co pany Emission Test Report, EMB Report 79-NHF-13a, September 1980, Urea Manufacture: Agrico Chemical Company Emission Test Report, EMB Report 78-NHF~4, April 1979, Urea Manufacture: CF Industries Emission Test Report, EMB Report 78-NHF=8, May 1979. Urea Manufacture: Union 011 of California Emission Test Report, EMB Report 78-NHF-7, October 1979. Urea Manufacture: Union Oi1 of California Emission Test Report, EMB Report 80-NHF-15, September 1980, Urea Manufacture: W.R. Grace & Company Emission Test Report, EMB Report 78-NHF-3, December 1979, Urea Manufacture: Reichhold Chemicals Emission Test Report, EMB Report 80-NHF-14, August 1980. A-81APPENDIX B UREA EMISSION MEASUREMENT AND CONTINUGUS MONITORING 8.1 Emission Measurement Methods 8.1.1 Background The standard method for determining particulate emissions form stationary sources is EPA Mehtod 5, whereby a particulate sample is extracted isokinetically from a source and is collected on a heated filter. The particulate mass is then determined gravimetrically. Initial evaluations by EPA and others of the applicability of Method 5 for urea sampling indicated that the standard procedures of Method 5 would not be practical.’ Factors that affected the sampling and analysis Procedures of Method 5 included the following: * High water-solubility of urea (greater than 1 gram per ml water); * Relatively high vapor pressure and volatility of urea melts at 133° C and will begin to decompose at lower temperatures). As a result of these factors, major modifications to Method 5 were adopted. A summary of these modifications and the reasons for each are presented in the remainder of Section B.1. 8.1.2 Brief Summary of Urea Method Development Industry and EPA assessments in 1977 of the applicability of Method 5 for urea sampling and analysis determined that the following modifications were necessary: * Use of water-filled impingers as the primary urea collection devices. Use of a urea-specific analytical procedure for measurement of urea collected in the impinger water. Emission tests to develop new source performance standards (NSPS) for the urea industry were begun in October 1978, using sampling and analytical procedures incorporating these Method 5 modifications. The sampling train for the first program included five impingers in series, B-1with water in the first impinger (for urea capture), sulfuric acid in the second and third impingers (for ammonia capture), the fourth impinger empty, and the fifth impinger containing silica gel. A heated filter (Filter temperature not exceeding 71°C was positioned in front of the first impinger. The filter catch was weighed and then dissolved in the water impinger contents which was then analyzed for urea with the p- dimethylaminobenzaldehyde (PDAB) procedure. Water and acid impinger contents were analyzed for ammonia by direct nesslerization. The urea sampling and analytical procedures were further modified during the seven EPA emission tests conducted from October 1978 through April 1980. Each of these tests provided information that helped to clarify and simplify the procedures, The modifications were the result of pertinent questions raised by industry as well as investigations performed by EPA and its contractors, The in-train filter was eliminated, and urea analyses were performed by the Kjeldahl procedure, the Kjeldahl and PDAB procedures together, and eventually just the PDAB procedure. Urea, ammonia, and formaldehyde emissions were measured during these testing programs at urea solution formation process units, solid urea formation process units (prill towers and granulators), and solid urea coolers. The results of these programs demonstrated the applicability of the recommended Method 28 for urea sampling and analysis. The modifications to Method 5 that are incorporated into the recommended method are as follows: Sampling Five impingers in series with the following sequence: impingers 1, 2 and 3 each contain 100 ml water, impinger 4 contains 100 m1 IN sulfuric acid, and impinger 5 contains silica gel. © Elimination of an in-train filter. Analysis ‘© Sample Recovery: Combine the probe washes with contents of impingers 1, 2 and 3. Measure the volume of the contents of impinger 4 for moisture gain and then discard. Weigh the contents of impinger 5 for moisture gain. B-2* Sample Analysis: Dilute a 100 ml aliquot of the combined probe wash and water impinger contents to 500 ml, adjust the pH to greater than 9.5, then boil this solution down to about 75 ml. Dilute up to 100 ml, add PDAB cooler reagent to a 10 m1 aliquot and measure color intensity in a spectrophotometer. The POAB procedure was determined to be the simplest and most direct procedure for urea analysis. The interfering effects of ammonia on the PDAB analysis procedure are eliminated through the boiling step, whereby ammonia is removed from the sample. Ammonia and formaldehyde sampling and analytical procedures are not included in the reconmended method. The acid impinger is used to protect sampling train equipment from ammonia corrosion. B.1.3 Detailed Development of Urea Sampling and Analysis Method B.1.3.1. Initial Method Development Urea sampling modifications to Method 5 were needed because of the following source conditions: © Urea has a substantial vapor pressure even as a solid, and if a sampling is heated in a probe or on a filter for extended periods of time it would tend to decompose. '4s15 * The high water-solubility of urea implied that a water medium in the sampling train would be an efficient urea particualte collector. © Anmonia and formaldehyde are additional pollutants emitted from urea manufacturing processes, and both were considered secondary pollutants in the NSPS work plan. Anmonia cannot be efficiently collected in the Method 5 sampling train water impingers, so additional impingers containing acid would be required. Formaldehyde can be efficiently collected in water. Factors that would affect urea analysis procedures were the following: © With water impingers as the primary particulate collector, the sampling train water would have to be analyzed for urea. * The volatility of urea would preclude rapid heating of the samples 8-3to dryness in order to do a gravimetric analysis. At the same time, evaporating large quantities of water without heating would be inefficient and tedious. © The insoluble fraction of particulate emitted form urea sources was considered to be insignificant. Through literature surveys and discussions with industry sources, EPA determined that only three procedures were routinely used for urea analysis. These were the urease procedure, the Kjeldahl procedure, the POAB procedure. With the urease procedure, urea is hydrolyzed to ammonium carbonate, the sample solution is acidified and then back- titrated with standard base. This procedure is applicable only for high concentration urea analyses, such as for scrubber liquors. The Kjeldahl procedure can be applied for urea measurement in one of two ways: direct or indirect. With the direct procedure, ammonia is boiled from a sample and the urea in the residue is converted to anmonia; this converted ammonia is distilled off and the distillate is analyzed for ammonia by either nesslerization or titration. With the indirect procedure, one sample portion is distilled and the distillate is then analyzed for ammonia. The urea in a second sample portion is converted to anmonia and this solution is distilled. This distillate is then analyzed for ammonia and the armonia measured in the first portion is subtracted from the ammonia measured in the second portion. For both the indirect and the direct Kjeldahl procedures, urea is calculated by applying a stoichiometric 5 and conversion factor to the final ammonia measurement. With the PDAB procedure, color reagent is added to a sample aliquot and color intensity is then related to urea concentration. This analytical procedure was chosen for the initial urea sampling and analysis method because it is simple and easy to use in the field and it will measure Jow levels of urea (less than 10 mg/1). The interfering affect of ammonia on the POAS procedure? was of concern, and for this reason the Kjeldahl procedure was evaluated with the PDAB procedure in the early stages of the EPA emission test program.The urea sampling and analytical procedures recommended for the EPA Program included the following specific Method 5 modifications: Sampling © Use of five impingers in series: impinger 1 contains 100 ml deionized, distilled waters impingers 2 and 3 each contain 100 ml IN sulfuric acids impinger 4 is empty, and impinger 5 contains silica gel. The heated filter was retained in its position just before the first impinger. * The probe and filter temperatures should not exceed 160°F. Sample Recovery Place filter in jar 1; the distilled water wash of the probe, nozzle and front half of the filter holder in jar 2; the silica gel in jar 3; and contents of impinger 1 and its distilled water wash in jar 4; the contents of impingers 2, 3, and 4 and their acid wash in jar 5, Sample Analysis Jar 1 - Desccate and weigh the filter, then place in 50 m1 water in an ultrasonic bath. Combine this solution with jar 4. dar 2 - Measure the volume, then evaporate the liquid and weigh the residue. Redissolve in water and combine with jar 4. Jar 3 - Weigh for moisture gain. Jar 4 - Analyze for urea with the PDAB procedure. Jar § - Analyze for ammonia by direct nesslerization. Two acid impingers were employed to ensure capture of anmonia and to protect downstream sampling train equipment form the corrosive effects of ammonia. Direct nessleration® is a widely used ammonia colorimetric analytical procedure. B.1.3.2 First EPA Emission Tests At the first urea plant tested in October 1978, urea, anmonia, and formaldehyde were measured at a rotary drum granulator scrubber inlet and outlet and at a solution tower vent.* Formaldehyde sampling and analysis results are discussed in Section 8.1.6. Sampling at the solution 8-5tower was performed with a train modified to handle the high moisture and anmonia levels in the off-gas vent: eight impingers were used (four with water, one empty, two with 10N sulfuric acid, and one with silica gel) along with an in-stack orifice. Urea analyses were performed by the contractor using the PDA procedure within 15 days of sample collection, and by plant personnel using the Kjeldahl indirect procedure within 24 hours of sample collection. The analyses by plant personnel were performed in order to address questions of sample stability raised by industry. The results of this testing program show the POAB urea results exceeding the Kjeldahl results by about 8% at the granulator scrubber inlet (where ammonia concentrations were much less than the urea concentrations). At the outlet however, (where ammonia concentrations greatly exceeded the urea concentrations) , the POAB results were 48% Tower than the Kjeldahl results. Urea audit sample analyses of urea standards containing varying concentrations of anmonia did indicate approximately a 2% positive interference with a 17.6 anmonia-to-urea molar ratio. The positive interference increased as the molar ratio increased. These results generally corroborated earlier evaluations of the interfering effect of ammonia on the PDAB method. 2 10 days showed no urea degradation with time. The large difference between the PDAB and Kjeldahl analyses of the outlet samples is considered to be due to analytical errors and limitations in the Kjeldah1 analysis. The POAB procedure is more accurate than the Kjeldahl procedures at low urea concentrations. Other analyses of urea standards performed periodically over The sampling train water impingers were purged with ambient air for 15-20 minutes at the end of each test run to flush most anmonia into the acid impingers where it would not interfere with the PDAB urea analyses. A significant amount of ammonia still remained in the water impingers, 0 this flushing technique was discontinued after these tests. The results of this emission testing program showed that the amount of urea collected in the acid impingers was insignificant, indicating that nearly al] sampled urea was caught in the water impinger and on thefilter. Solution tower vent emissions were shown to consist primarily of water vapor and ammonia and very little urea. B.1.3.3 Method Modifications As a result of these first tests, EPA further modified the sampling and analytical procedures. An additional water impinger was added to ‘insure the complete capture of urea. The filter was deleted to simplify the method (since any filter catch had been merely added to the water ‘impinger contents) and to eliminate the possibility of an accidently overheated filter decomposing any collected area. The urea analytical procedure was changed to the Kjeldahl direct procedure (with preliminary distillation to remove ammonia). This was due to the high levels of ammonia collected in the controlled granulator emissions and the susceptibility of the POAB procedure to ammonia interference. These modifications are summarized as follows: Sampling © Six impingers in series: impingers 1 and 2 each contain 100 ml water, impingers 3 and 4 each contain 100 ml IN sulfuric acid, impinger 5 is empty, and impinger 6 contains silica gel. © Elimination of the in-train filter. Analysis © Sample Recovery: Combine the probe and nozzle washes with the contents of impingers 1 and 2. Combine the contents of impingers 3, 4, and 5 in a separate container. Weigh the silica gel for moisture gain. * Sample Analysis: Add buffering agents to the samples and distill into a boric acid solution. Analyze this distillate solution for ammonia by nesslerization. Add digestion reagents to the residue, converting organic nitrogen (urea) to ammonia. Distill and analyze this distillate solution for ammonia by nesslerization; calculate urea concentration stoichiometrically from the indicated ammonia concentration. B-7Several of these modifications to the urea sampling and analytical procedures were applied during the next three emission tests in December 1978, January 1979, and April 1979, 8.1.3.4 Second EPA Emission Tests The December 1978 tests were performed at the same facility as the October 1978 tests, and consisted of emission measurements at the outlet of one granulator scrubber.® A major purpose of this test was to provide field samples for time stability evaluations and to attempt to establish the validity of the Kjeldahl indirect procedure analyses performed during the previous test program. The December 1978 samples were analyzed on-site for urea by the contractor using the Kjeldahl direct procedure and by plant personnel using the Kjeldahl indirect procedure. The sampling train used in the Decenber 1978 tests contained only § impingers (impingers 1, 2, and 3 contained water, impinger 4 was empty, and impinger 5 contained silica ge1) because the primary concern was with urea; ammonia capture was of secondary importance. The combined contents of impingers 1 through 4 were filtered for insoluble particulate, and the filtrate was then analyzed for urea by the Kjeldahl procedures as described above and for formaldehyde as described in Section B.1.6. Anmonia analyses were also performed, by direct nesslerization and by nesslerization with preliminary distillation. The insoluble particulate catch averaged about 1.4% of the total particulate catch. The Kjeldahl indirect analyses performed by plant personnel on the scrubber outlet samples yielded results that averaged 30% higher than the kjeldahl direct analysis results. Audit sample analyses by the contractor using the Kjeldahl direct procedure (ending with a final anmonia analysis by nesslerization) agreed within 6% of the actual urea sample weights. These same audit samples were analyzed by plant personnel using the Kjeldahi direct method ending with titration, and the results averaged 93% higher than the actual sample weights. EPA concluded that the Kjeldahl analyses performed by plant personnel during this testing program and the previous program (October 1978) were unreliable. 8-8A complication with the use of the Kjeldahl procedures for urea analyses is the need to correct indicated urea and anmonia concentrations in order to account for the conversion of some urea to anmonia during the preliminary distillation step.’ the standard correction factor is: Té of the sample urea content is converted to anmonia during distillation. There is evidence, however, that this correction factor is not constant, but may vary with absolute urea concentration or with the ratio of urea to ammonia concentrations in the sample.°*7 Use of the 7% factor produces difficulties with samples containing relatively high urea concentrations compared to ammonia concentrations; for example, with uncontrolled emission samples or scrubber liquor samples, negative corrected ammonia concentrations can be calculated. The granulator scrubber outlet samples from the December 1978 program contained relatively small amounts or urea, and no problems were encountered. The granulator outlet samples from the December 1978 emission testing program and specially prepared urea laboratory samples were periodically analyzed over a 20-day period subsequent to the completion of the program, The purpose of these time analyses was to determine if the urea content of samples deteriorated over tine. These analyses were performed with the Kjeldahl direct procedure, and the results showed no detectable change in urea content of the samples.° EPA concluded that there would be no problems with the stability of urea samples analyzed up to 20 days after the sample collection. B.1.3.5 Third and Fourth EPA Emission Tests The January 1979 and April 1979 emission tests were performed with the modified urea sampling and analytical procedures (6 impingers, Kjeldahl direct analysis procedure). Emissions at a granulator scrubber inlet and outlet and at a solution tower vent were measured during the January 1979 tests.” Ammonia analyses were performed by both the direct nesslerization and the nesslerization with preliminary distillation procedures. The granulator inlet samples contained relatively large urea concentrations and, as a result of the urea to ammonia conversion, ‘the two ammonia analytical procedures differed greatly in indicatedammonia concentrations. No significant difference occurred with the outlet samples. Formaldehyde analyses were also performed, as described in Section 8.1.6. The urea content of the acid impingers was less than 0.2% of the total urea caught at the granulator scrubber inlet and near the threshold of detection at the outlet. During the April 1979 tests, prill tower uncontrolled and controlled emission samples were analyzed by both the Kjeldahl and the PDAB procedures.® Procedural difficulties during the analyses, however, prevented any reliable evaluation of the results. B.1.3.6 Method Modification At this time, the urea sampling and analytical procedures were further modified, based on the results of these three recent emission tests. Ammonia and formaldehyde sampling was discontinued because no immediate need for emission standards for these pollutants was foreseen. The PDAB procedure was retained for the urea analyses because of its advantages (its simplicity and the fact that it measures urea directly) and because of the disadvantages of the Kjeldahl procedure. Its complexity, the number of reagents and amount of apparatus needed, as well as the problems associated with the urea to ammonia conversion during distillation, were the major reasons for deleting the Kjeldahl procedure. The problem of ammonia interference in the PDAB procedure was investigated in more detail at this time.? The interfering effects of anmonia as initially described in the literature? and as discussed above were corroborated. Investigations with prepared laboratory standard solutions showed a slight (less than 2%) positive interference for approximately a 20:1 ammonia to urea molar ratio. Higher molar ratios increased the interference. To eliminate the ammonia interference during the field sample analyses, a preliminary distillation step was included in the PDAB procedure, whereby ammonia is boiled off prior to the actual analysis. The modifications to the urea sampling and analytical procedures made at this time are summarized as follows: B-10Sampling © Five impingers in series: impingers 1 and 2 each contain 100 ml water, impinger 3 contains 100 ml IN sulfuric acid, impinger 4 is empty, and impinger 5 contains silica gel. Analysis © Sample Recovery: Combine the nozzle and probe washes with the contents of impingers 1 and 2, Mesaure the voluje of the contents of impingers 3 and 4, then discard. Weigh the silica gel for moisture gain, * Sample Analysis: Dilute a 100 m1 aliquot to 500 ml, adjust the pH to greater than 9.5, then boil sample down to about 75 ml. Dilute up to 100 ml, and add PDAB color reagent to a 10 ml aliquot and measure color intensity in a spectrophotometer. The one acid impinger was retained to protect the downstream sampling train equipment from the corrosive effects of ammonia. During the final three emissions tests, ammonia sampling and analysis was continued in order to accumulate background data for potential future use. An dditional acid impinger was therefore added to the train immediately preceding the empty impinger, making a total of six impingers in the train. B.1.3.7 Fifth, Sixth, and Seventh EPA Emission Tests The fifth emission tests were performed in August 1979 on uncontrolled and controlled prill tower emissions and on emissions from a solution tower vent. 1! The sampling and analytical procedures described inmediately above were used (6 impingers, PDAB procedure). The water impinger contents were filtered prior to urea, ammonia, and formaldehyde analyses to retain any insoluble particulate. Water and acid impingers were analyzed for urea by PDAB procedure with preliminary distillation, and for ammonia by both direct nesslerization and specific ion electrode. The two anmonia analysis results agreed with each other within 10%. The urea content of the acid impingers was negligible (at the threshold of detection) for both controlled and uncontrolled prill tower emissions samples.During the urea analysis of the first series of the first series of water and acid impinger samples, the analyst determined that sulfuric acid was acting as a negative interference to the PDAB urea analysis. In order to compensate for this interference, urea standards used to establish absorbance vs. concentration calibration curves were prepared with the same acid concentrations as the samples being analyzed.9!! The effect of the preliminary distillation step (boiling ammonia off) was investigated during this field program as part of the audit sample analyses. The investigation results indicate that the extent of urea loss during the distillation step is 12 to 14 percent. This urea loss can be compensated for as long as both samples and standards are handled in the same way (both undergo distillation)? Prior to the sixth emission tests, the absolute threshold of detection for the PDAB urea analysis procedure was investigated with laboratory standard solutions and was determined to be 5 to 7 mg/1.9 The sixth set of emission tests were performed in April 1980 on uncontrolled and controlled prill tower emissions.'2 It was known beforehand that the controtled prill tower emissions would be very Tow, so an unusually long sampling time was planned (320 minutes), Even with this extended sampl ing time the first samples yielded urea concentrations near the threshold of detection. Consequently, sampling times were further extended (400 minutes) to collect more urea, and the PDAB analysis procedure was modified in order to assess the low concentrations. Instead of diluting a 100 ml sample aliquot to 500 mI and then boiling down to 75 ml, larger sample aliquots (500 to 700 ml) were taken and boiled down without dilution. In this way, 5 to 7 times as much urea was concentrated in the same volume and the sensitivity of the analysis method was effectively increased. 912 The urea content of the acid impingers was about 2.5% (less than 10 mg) of the total urea catch at the scrubber inlet (uncontro’ led emissions) and at or below the threshold of detection at the scrubber outlet (controlled emissions). Annonia analyses were performed with the direct nesslerization procedure and with the specific ion electrode (SIE) procedure. The results of both analysis procedures agreed within 6 percent. B-12The stability of urea field samples was further documented during this emission testing program. The urea analyses of the scrubber outlet samples were performed in the field within 24 hours of sample collection and at the contractor's laboratory within 16 days of sample collection. No significant difference existed between the results. The last EPA emission tests were conducted in April 1980 on the outlet of a prill tower scrubber and on the inlet of a prill cooler scrubber.'3 One purpose of this program was to document the urea collection efficiency of the sampling train (six impingers: 1 and 2 water, 3 and 4 acid, 5 empty, and 6 silica gel). The nozzle and probe wash, the contents of impinger 1, the contents of impinger 1, the contents of impinger 2, and the combined contents of impingers 3, 4, and 5 were analyzed separately for urea (PDAB procedure with ammonia removal) and ammonia (SIE procedure). The analysis results showed that 70% of the urea is caught by the probe and first water impinger, and the remaining 30% is caught by the second water impinger. The urea content of the acid impingers was below the threshold of detection. The ammonia analysis results showed that about half the ammonia is caught in the water impingers and half in the acid impingers.9*13 During several of the EPA emission tests, acid impinger samples turned turbid when the PDAB color reagent was added, due perhaps to the amount of sodium hydroxide added to adjust the sample pl. In all cases, however, the turbidity was removed with the addition of a small amount (1 or 2 m1) of concentrated hydrochloric acid, !!»12513 8.1.3.8 Recommended Method The recommended urea sampling and analysis method (Method 28) incorporates an additional modification to the sampling train, based on the results from all the EPA emission tests and the urea method development investigations. These final method modifications are surmarized as follows:Sampling © Five impingers in series with the following sequence: impingers 1, 2, and 3 each contain 100 ml water, impinger 4 contains 100 ml IN sulfuric acid, and impinger 5 contains silica gel. Analysis © Combine the probe washes and the contents of the three water impingers and analyze for urea by the PDAB procedure with anmonia removal. “Measure the volume of the contents of impinger 4 for water gain and discard. Weigh the contents of impinger 5 for water gain, The third water impinger is included to ensure capture of all sampled urea and to eliminate the need to make up separate urea standards for acid impinger sample analysis. (As discussed above, the acid content of samples and standards should be the same.) The acid impinger is included to protect the sampling train equipment from ammonia. In situations where ammonia sampling is desired, an additional impinger (containing 100 ml 1N sulfuric acid) can be added to the train directly preceding the silica gel impinger. In this case, the combined contents of the first three impingers are analyzed for urea (by POAB) and ammonia (by SIE or direct nesslerization), and the combined contents of impingers 4 and 5 are analyzed for ammonia only. The results of the urea EPA emission tests have demonstrated the utility and economy of the recommended method. The urea (and ammonia) analytical procedures require a minimum amount of equipment and field laboratory space. All analyses can be performed on-site and immediately after each individual test run. The ability to perform sampling analyses quickly in the field allows for rapid evaluation of emission values and sampling technique. 8.1.4 Potential Problems with the Recommended Method Two difficulties may be encountered with the use of the recommended Method 28: © Decomposition of urea in the probe at elevated temperatures; © Incomparability of Method 28 data and Method 5 data.By maintaining probe temperatures at about 6°C above stack temperature, sample decomposition and moisture condensation in the probe can be avoided. Most emission control devices operate at or near saturation and with outlet gas stream temperatures less than 49°C. Solid urea melts at about 133°C but will decompose at temperatures below the melting point. Routine analyses of the urea product by industry that Show the presence of biuret indicate that decomposition does take place. Industry procedures for drying solid urea specify heating at 70°C overnight. Therefore, to ensure the integrity of a sample, a reasonable upper limit on probe temperature is approximately 71°C. A source subject to particulate emission regulations normally undergoes periodic compliance tests. Method 28 would be used specifically to verify the particulate emission compliance of a new urea source. The results of a Method 28 urea compliance test could not be directly compared to the results of a Method 5 urea compliance test because of the factors discussed in Section B.1.3. In addition, the relationship between Method 5 urea collection and Method 28 urea collection is not established. This relationship would depend on the type of emission source, the operating conditions of that source, and the amount of urea particulate caught. Small amounts of particulate (for example, less than 10 mg) can be analyzed accurately by Method 28, but are difficult to assess with a Method 5 gravimetric analysis. 8.1.5 Relationship of Data Gathered Under EPA Tests to Data Gathered with the Recommended Test Method The majority of the EPA emission tests were conducted using the same sampling and analytical procedures as contained in the recommended Method 28. The first tests and the last three tests (all utilizing the POAB analysis method) differed from each other in that the first tests did not include an ammonia removal step and did utilize an in-train filter. Since the ammonia-to-urea molar ratio in the samples of the First tests did not exceed 20, the urea analysis results of these samples would not be in error by more than approximately 2 percent due to ammonia interference. The in-train filter catch was redissolved in the waterimpinger contents and so was included in the urea analysis results. The second, third, and fourth emission tests utilized the Kjeldahl urea analytical procedure. The data gathered during these three tests may be in error by approximately 7 percent due to the urea-to-anmonia conversion that occurs during distillation. Urea audit sample analyses performed during these tests showed that the Kjeldahl procedure produced results within the required accuracy (+10 percent). The results of the fourth tests are not considered valid, as noted in Section 8.1.3. B.1.6 Formaldehyde Sampling and Analysis During the EPA Tests A formaldehyde-based additive is often used in urea production processes to coat solid urea prills. Formaldehyde emissions were sampled and analyzed through the August 1979 EPA emission tests. Formaldehyde emissions were very low, and subsequently formaldehyde sampling was discontinued. During the first EPA test, formaldehyde at the granulator scrubber inlet and outlet was sampled with a sampling train separate from the train used for the urea and ammonia samp] ing.® The procedure in the EPA document "Tentative Method for Isokinetic Determination of Pollutant Levels in the Effluent of Formaldehyde Manufacturing Facilities" was followed, which utilized the impinger sequence of Method 5 but without a filter. Formaldehyde analysis was performed with the chromotropic acid procedure. The analytical results showed that formaldehyde emissions at the granulator scrubber inlet and outlet were about 0.045 and 0.023 kg per hour, respectively. During the second EPA test in December 1978, the urea sampling train impinger contents were analyzed for formaldehyde with the chromotropic acid procedure.® Formaldehyde emissions from this granulator outlet averaged less than 0.36 kg per hour. The same formaldehyde sampling and analysis procedures that were followed in the second EPA test were followed in the third and fifth tests (an aliquot from the urea sampling train impingers were analyzed). During the third test, formaldehyde emissions in the granulator scrubber B-16inlet and outlet were approximately 0.09 and 0.045 kg per hour, respectively.” During the fifth test, prill tower scrubber inlet and outlet formaldehyde emissions averaged less than 0.045 and 0.0045 kg per hour, respectively.)TABLE 1 MANTIME BREAKDOWN FOR COMPLIANCE TEST Number. Number Total Task of People of Days Man-Days Site Visit 1 1 1 Field Work 2 1 2 Preparation and Cleanup 1 2 2 Lab Analysis 1 2 2 Data Reduction 1 0.5 0.5 Report Preparation 1 2 2 Management 1 0.5 0.5 Total 10REFERENCES lL 5. 6. 10. i. 12. Grove, J. D., "Prill Tower Sampling Approahces: Urea and Ammonia Nitrate Processes". Entropy Envirormentalists, Inc., Prepared for U.S. EPA under Contract 68-01-4148, Task No. 32. October 1977. Watt, G. W. and J. D. Chrisp, “Spectrophotometric Method for Determination of Urea", Analytical Chemistry, Volume 26, 1974, pp. 452-453. Standard Methods of Water and Wastewater Analysis, APHA, AWWA, WPCF, T4th Edition, 1975, p. 412. EPA Report 78-NHF-4, "Emission Test Report, Agrico Chemical Company, Blytheville, Arkansas". Prepared by TRC-Environmental Consultants, Inc. under EPA Contract 68-02-2820, Work Assignment 6. Standard Methods of Water and Wastewater Analysis, APHA, AWWA, WFC, Wath Edition, 1975, p. 437 EPA Report 79-NHF-13a, "Emission Test Report, Agrico Chemical Company, Blytheville, Arkansas". Prepared by TRC-Environmental Consultants, Inc., under EPA Contract 68-02-2820, Work Assignment Ve EPA Report 78-NHF-8, “Emission Test Report, CF Industries Inc., Donaldsonville, Louisiana". Prepared by TRC-Environmental Consultants, Inc., under EPA Contract 68-02-2820, Work Assignment 10, EPA Report 78-NHF-7, "Emission Test Report, Union Oi] Company, Brea, California". Prepared by Engineering-Science under EPA Contract 68-02-2915, Work Assignment 26, EPA Report 79-NHF-13, "Development of Analytical Procedures for the Determination of Urea from Urea Manufacturing Facilities". Prepared by TRC-Environmental Consultants, Inc., under EPA Contract 68-02- 2820, Work Assignment 11. Standard Methods of Water and Wastewater Analysis, APHA, AWMA, WCF, Tath Edition, 1975, p. 408. a EPA Report 79-NHF-3, “Emission Test Report, W. R. Grace and Co., Memphis, Tennessee". Prepared by TRC-Environmental Consultants, Inc., under EPA Contract 68-02-2820, Work Assignment 9. EPA Report 80-NHF-14, "Emission Test Report Reichhold Chemicals, Inc., St. Helens, Oregon". Prepared by TRC-Environmental Consultants, Inc., under EPA Contract 68-02-2820, Work Assignment 19.13, EPA Report 80-NHF-15, “Emission Test Report, Union 011 Company, Brea, Californi Prepared by TRC-Environmental Consultants, Inc., under EPA Contract 68-02-2820, Work Assignment 20. 14, The Condensed Chemical Dictionary, 9th Edition, Van Nostrand Company, 1977 p. S08 15. Cramer, J. H., "Urea Prill Tower Control - Meeting 20% Opacity". Presented at the Fertilizer Institute Environmental Symposium, New Orleans, Louisiana, April 1980. 8-20TECHNICAL REPORT DATA ad Inttrucrions on the reserse before completing) RECIPIENTS ACCESSION NO. Please TTREFORT NG. z EPA-450/3-81-001 I a TIrce ANG SUBTITLE Technical Document for the Urea Industry REPORT DATE Janaury 1981 Is. PERFORMING ORGANIZATION CODE FROTRERTST [PERF SAWING GRGARTERTION REPORT NS | [a PERFORMING ORGANIZATION NAME AND ADDRESS 16. PROGRAM ELEMENT NG Radian Corporation 3024 Pickett Road Ourham, North Carolina 27705 68-02-3058 PEN eae” PS RRTER Fil 13, TYPE OF REPORT ANO PERIOD COVERED Office of Air Quality Planning and Standards final Emission Standards and Engineering Division 1* SPONSORING AGENCY Cope Industrial Studies Branch Research Triangle Park, North Carolina 27711 78. SUPPLEMENTARY NOTES TSRESTRACT This report presents information on the emission levels, control techniques and costs associated with the control of particulate emission sources and facilities in the urea solids producing industry. Sources of emissions include prill towers, granulators and coolers. Alternative control techniques and supporting data are described and discussed, and an analysis of environmental and economic impacts of control techniques are presented. KEY WORDS AND COCUMENT ANALYSIS: 3 DesCRPTORS 1OeNTIFIERS/OPEN ENDED TERNS [e. GOEATI TUICIONp Air Pollution Air Pollution Control Control Technology Particulate Control Urea Wet Scrubbers Particulate Emissions Netted Fibrous Filter Te OSTRRUT ON STRTEENT TS BREDADY COREE TH Reson FT NSO PSE Unlimited Unclassified 7°" | “e [RO SECURIT? Coase Tar paReT—— PC LUnelass: Pa FormU.S. Environmental Protection Agency* Region 5, Library (PL.12)) 77 West Jackson Boulevard, 12th Fy, Chicago, IL” 60604-3590 cor