vxEPA United States Environmental Protection Agency Industrial Environmental Research Laboratory Research Triangle Park NC 27711 Research and Development EPA-600/S7-81 -169 August 1982 Project Summary Emissions from Refinery Process Heaters Equipped with Low-NOx Burners R. J. Tidona, H. J. Buening, and J. R. Hart This report summarizes results of an investigation of the performance of commercial Iow-N0x burners in refin- ery process heaters. Refineries in Southern and Central California were surveyed to determine the number of low-NOx burner installations existing or planned. Ten process heaters, equip- ped with low- N Ox burners, were tested to measure gaseous emissions, partic- ulates, and efficiences over a normal range of operating conditions. The as- found NOX emission increased from 58 to 245 ng/J as the fuel-bound nitrogen increased from 0 to 0.81 percent. The NOX concentrations were strongly influenced by excess air levels in most cases. Reducing excess air to about 3-4 percent reduced NOxto 34- 200 ng/J, depending on fuel nitrogen. Comparisons of present emissions data with past field test data for refinery heaters equipped with stan- dard burners showed that for mechan- ical-draft gas-fired heaters, low-NOx burners may reduce the NOX emission factor by 32-77 percent below the mean emission factor for standard burners. Three heaters (one firing gas, another firing distillate oil, and the third firing residual oil) were selected as suitable candidates for 30-day continuous monitoring of gaseous emissions. This Project Summary was devel- oped by EPA's Industrial Environmen- tal Research Laboratory, Research Triangle Park, NC, to announce key findings of the research project that is fully documented in a separate report of the same title (see Project Report ordering information at back). Introduction Approximately 6000 refinery process heaters are in operation in the United States (Ref. 1). Of these, about 5400 have natural draft burners; the remain- der have forced or balanced draft burners. Both types emit a total of about 121.5 Gg/y (134,000 tons/y) of NO,, making process heaters one of the largest industrial emitters of this pollu- tant. Several field test programs have been conducted to characterize NO* emis- sions from process heaters over a wide range of operating conditions. However, all the previous field test work was conducted on heaters equipped with conventional burners. Only recently have refinery heater burner manufac- turers begun to market low-NOx burners for the petroleum industry, and new commercial installations are somewhat sparse. It was the purpose of this program to (1) locate refineries in Southern Cali- fornia which have installed or ordered low-NOx burners, (2) test nine gas- or oil-fired process heaters in which low- NOx burners have been installed (NO, and other gaseous emissions were to be measured over a normal range of oper- ating conditions), and (3) assess the potential for long term (30-day) tests of ------- the heaters with the goal of finding three suitable sitesone firing gas, a second firing distillate oil, and the third firing residual oil. Southern California was selected for the heater survey because of its strict NOx emission regulations; hence, a high likelihood of finding installations with low-NOx burners. Results of the survey are summarized in Table 1. At least 29 heaters were found with low-NOx burners installed or planned for instal- lation. Heaters which were tested in this program were assigned the site num- bers shown in Table 1. Results of the tests conducted at these 10 sites are reported here. (Site 10 was added to the required nine because it was near Sites 6 through 9 and the refinery was willing to allow it to be tested.) Summary of Program Results and Comparison with Past Results This program required that baseline tests be conducted for nine refinery process heaters equipped with low-NOx burners. Tests were completed on 10 process heaters, since one extra could be tested for minimal additional cost. At each test site, the gaseous emis- sions and stack gas temperature were measured. Samples of fuel were taken and submitted to an independent labora- tory for analysis. Unit operational data such as flow rate, pressure, and tem- perature were recorded periodically. After testing the unit in the as-found condition, burner registers and/or stack dampers were adjusted to determine the effect of excess oxygen on unit operation, gaseous emissions, and heater efficiency. The heaters tested included four natural draft and six mechanical draft units firing gas, distillate, and residual oils. Four of the mechanical draft units had preheated combustion air. Emission Test Methods and Instrumentation All emission measurement instru- mentation was carried in a 2.4 x 12.8 m (8 x 42 ft) mobile laboratory trailer. Thi trailer was used at all test sites. Gaseou species were measured with analyzer in the trailer. The emission measure ment instrumentation used is listed ii Table 2. Results Figure 1 shows NOx emissions as i function of stack oxygen for the 11 process heaters tested. These data shov that for all heaters, lowering the oxygei resulted in lower N0» emissions. Th< gas-fired heaters had lower NO, emis sions than did the distillate-oil-firei heater, and the residual-oil-fired heater: had the highest NOx emissions. Thi gas-fired heaters with combustion ai preheat had higher NO, emissions thai gas-fired heaters with ambient combus tion air. The points at which significan CO emissions or visible smoke occurrei are marked in the figure as "CO limits.' Gas-fired heaters with ambient com bustion air showed NOx emissions a< 40-50 ng/J (80-100 ppm at 3 percent Table 1. Heater Low NOK Burner Test Site Survey Location A B C D E F G/i G/2 H 1 J/1 J/2 K L/1 L/2 M N 0/1 O/2 P Site No. 2 - - - _ - 6, 7, 8,9 10 - - - - 7 4 3 5 - - - - Air Quality Control Region 31 31 24 24 24 31 30 30 24 31 24 24 24 31 31 31 24 30 30 31 No. Heaters 1 1 - - 1 11 4 1 1 1 1 1 1 1 1 1 - 1 1 - Heat Input Rate 106 MW Btu/Hr 4.7 16.4 - - 7.5 - 93.8° 8.8 7.7 10.3 - - 6.1-6.7 6.7 11.7 11.7 - 21.4 6.4 - ~ 16 56 - - 25.7 - 320 30 26.4 35 - - 21-23 22.9 39.9 40 - 73 22 - Burners 1 8 - - 4 - 32 1 8 3 - - 3 3 12 10 - - - - Operational Yes Yes - - Dec. 8O- Jan. 81 - Sept. 80 Sept. 80 Jan. 81 March 81 Jan. 81 Jan. 81 Yes Yes Yes Yes - Yes Yes - Fuel LO-S resid 6 Gas, 2 Oil - - Gas - Gas Gas - #6 Oil Gas Gas Gas #5 Oil, gas #5 Oil, gas #£+/?G - Gas Gas - Comments Natural Draft Old heater; leaks above firebox Natural Draft Preheat, Balanced Draft Ambient, Forced Draft Natural Draft Natural Draft Natural Draft Forced Draft Natural Draft Natural Draft Natural Draft Natural Draft Natural Draft aTotal for four heaters. ------- 02, dry*) in the as-found condition. By lowering the excess oxygen to a CO limit or to the minimum acceptable condition as determined by the plant, NO* emis- sions were reduced to about 25 ng/J (50 ppm). Heaters with preheated com- bustion air showed as-found NO* emis- sions as 58-76 ng/J (115-150 ppm). Lowering excess oxygen resulted in NO, reduction to 25-43 ng/J (50-85 ppm). As-found NO emissions from a natural- draft distillate-oil-fired process heater were 92-117 ng/J (168-212 ppm) and were reduced to approximately 58 ng/J (105 ppm) by lowering excess air to the single burner. Emissions in the as-found condition were considerably higher for the three units firing residual oil. All were natural draft units, and two were found to be emitting 207-281 ng/J (370-500 ppm) of NO. The unit at Site 5 fired 15-20 percent refinery gas simultaneously with residual oil. This unit was found to be emitting 194-205 ng/J (350-370 ppm) of NO and was the only unit firing residual oil which showed a significant tendency toward reduced emissions when excess air levels were lowered. At reduced excess air settings, this unit produced about 166 ng/J (300 ppm) of NO. Table 3 is a summary of gaseous emissions and efficiency for each site tested. Emissions at as-found conditions and at optimum low-NO* conditions (i.e., lowest NOx emission without adverse effects on flame stability or unit effi- ciency) are compared in this table. At most sites, significant reductions in NO* emission below as-found levels could be achieved along with small increases or, at worst, no change, in efficiency. With respect to flame stability, product flows and temperatures, and emissions of CO and unburned hydrocarbons, unit operations at the optimum Iow-N0x conditions were generally unchanged from the as-found conditions. Figure 2 is a plot of the effect of fuel- bound nitrogen content on the NO, emission factor. The three circles plotted in this figure represent data from the oil-fired units tested. Site 2 burned 100 percent No. 2 oil; Sites 3 and 4, No. 6 oil (both units burned fuel having the same composition); Site 5, a combination of 80 percent No. 6 oil (by heat input rate) and 20 percent refinery gas. The "zero *AII emission concentrations given as ppm are corrected to 3 percent 02, dry basis. Table 2. Emission Measurement Instrumentation Species Manufacturer Measurement Method Model No, Hydrocarbon Carbon Monoxide Oxygen Carbon Dioxide Nitrogen Oxides Particulates Sulfur Dioxide Beckman Instruments Beckman Instruments Teledyne Beckman Instruments Thermo Electron Co. Joy Manufacturing Co. DuPont Instruments Flame lonization IR Spectrometer Polarographic IR Spectrometer Chemiluminescent EPA Method 5 Train UV Spectrometer 402 865 326A 864 10A EPA 400 500 400 £300 I d ^ 200 100 I-Site 4 'Residual Oil Natural Draft Site 3 Residual Oil Natural Draft -Site 5 fS Residual Oil/RG xv/' Natural Draft 0 Figure 1. 468 Stack Oxygen, % Dry Site 2 Distillate Oil Natural Draft Site 6-9 Refinery Gas Balanced Draft Site 10 Refinery Gas Forced Draft Site 1 NG/RG Forced Draft RG = Refinery Gas NG = Natural Gas = Denotes as-found cond. 10 12 14 NOX emissions as a function of stack oxygen for Iow-N0x burners in process heaters. 3 ------- Table 3. Optimum Low-N0x and As-Found Gaseous Emissions and Efficiencies at 10 Heaters Equipped withLow-NOxBurners As-Found Optimum Low-N0x Site 1 2 3 4 5 6 7 8 9 10 Heater' Test Nos. Config. (A.F.:Low-NOx) 121 1-13; 1-24 211 2/1-1e;2/1-3a 311 3-1:3-3 311 4-1:4-4 411 5-3:5-6 132 6/5-3:6/5-1 132 7/1-1:7/2-2 132 8/1-1:8/1-3 132 9/1-1:9/1-3 121 10/1-1; 10/1-4 3Heater configuration designations 1st Digit Fuel Burned 1 = gas 2 = dist. oil 3 = residual oil 4 = combined oil &gas NO (ng/J) 39.0 92.4 222.0 268.0 194.0 57.9 65.8 74.9 60.2 51.6 are: NO 02 (ppm) (%) 77 6.2 168 5. 1 396 5.2 477 8. 1 352 7.7 114 4.8 130 5.3 148 7.7 119 8.4 102 6.7 2nd Digit Draft Type 1 = natural 2 = forced 3 = balanced CO (ppm)b 0 11 11 0 11 11 11 14 0 13 Htr. Eff. (%) 79.9 64.0 63.4 67.9 69.8 68.0 65.4 66.4 73.1 NO (ng/J) 24.0 80.4 203.0 264.0 167.0 38.1 35.4 41.0 32.9 32.9 NO (ppm) 48 145 361 471 303 75 70 81 65 65 02 (%) 3.0 4.0 4.3 7.1 4.4 3.5 2.3 3.8 3.3 2.8 CO lppm)b 20 11 11 0 22 11 33 15 10 10 Htr. Eff. (%) 83.0 64.0 64.5 68.6 71.3 68.5 69.2 68.8 74.4 %NO Reduction 38.5 13.0 8.5 1.5 13.9 34.2 46.2 45.3 45.3 36.2 3rd Digit Air Temp. 1 = 2 = ambient preheater bDry, corrected to 3%02. Fuel Bound Nitrogen, Percent by Weight Figure 2. Fuel-bound nitrogen effect on NOX emission factor. level" or base HO* emission factor for gaseous fuel (containing no fuel-bound nitrogen) was computed by taking the average as-found NOX emission factor for all sites firing gas fuel only. This value was 58 ng/J. The increase. A, in emission factor for No. 2 oil was 34 ng/J (as found). For No. 6/gas fuel, the increase over the gas fuel emission factor was 136 ng/J (as found). For No. 6 oil, A was 187 ng/J (as found). Comparing these points to the curve fit of previous laboratory data (Figure 2), shows that in all cases the emissions from low-NOx burners fall below that curve. It is interesting to note that operation at the optimum low-NOx mode (represented by squares in Figure 2) did not significantly alter the fuel-bound nitrogen conversion to NO,. Also, except for Site 2, in the as-found condition all points plotted in Figure 2 fall within the bounda ry curves for the laboratory data. The effect of fuel nitrogen on NOx emissions is thus responsible for the large differences in emissions between gas- and oil-fired heaters shown in Figure 1. Table 4 summarizes NO emissions from the heater configurations studied over the tested range of operating variables. The values reported in paren- theses are ppm corrected to 3 percent Oz, dry. (NOz emissions were not in- cluded due to lack of data at some sites. The data available from Sites 6-10, however, show that the NOz emissions average only 3.6 percent of the total NO, emissions.) ------- Past Results - Comparison to Present Reference 2 summarizes past NOx emissions data on refinery process heaters firing oil and gas. These data are presented in Table 5. Unfortunately, the paucity of baseline data from oil-fired process heaters with standard burners makes it difficult to compare them with present Iow-N0x burner data. Mechan- ically drafted heaters with low-NOx burners firing gas, however, appear to have significantly lower NO, emissions relative to the standard burners. Tables 4 and 5 reveal that low-NO. burners produced NO* emissions (neglecting NOz) which were 32-77 percent less than the average emission factor for all standard burners on mechanical-draft gas-fired heaters. The variations in NO* emissions from low-NOx burners were due to operating differences between units (e.g., excess air level and air register settings). It is important to note that there is no NO, data available on a ny of the units tested for operation with standard burners. Thus, NO, emissions with and without low-NO, burners cannot be compared directly for any heater. Because of a lack of availability, no natural-draft gas-firing low-NOx burners were tested; hence, they cannot be compared with their standard counter- parts. Conclusions and Recommendations for Future Testing The following conclusions may be drawn from the study: 1. The effectiveness of low-NOx burn- ers depends on operating tech- niques, especially excess air level and air register settings. 2. By changing excess air and air register settings, NO, emissions were reduced at every site with improved overall heater perform- ance. 3. Commercially available low-NOx burners may reduce the NO, emis- sion factor by 32-77 percent below the average emission factor for standard burners on mechanical- draft gas-fired heaters, depending on operating techniques. 4. Because of the lack of a good data base for oil-fired heaters with standard burners, a meaningful comparison of the performance of low-NOx oil-fired burners versus Table 4. NO Emissions from Process Heaters with Low-NOx Burners (1) Fuel Type Unit Type Gas (Ref. or NG) Distillate Oil Residual Oil + Gas Residual Oil Natural Draft ,"9*, 58_j 77 21(105-212 ppm) 167-202.(303-366ppm) 207-284^(369-506ppm) J J J Forced Draft, co^/c* m? nnm> AmbientAir 26-52-2(51-102 ppm) Balanced Draft, _ -,.,n9 ,.-n , ,0 , AirPreheat 25-75^(50-148 ppm) aRatio of heat input from oil to heat input from gas = 83/ J 7. Table 5. Mean NOX Emissions from Process Heaters with Standard Burners Fuel Type Unit Type Gas (Ref. or NG) Distillate Oil Residual Oil + Gas Residual Oil Natural 60.6 (25 tests) Mechanical 111 "J! (6tests) J 89.6 22 (5 tests) 138 ^-(3 tests) \J Not Reported (Bay Area A.P.C.D.) 70.022 (64 tests) \J Not Reported (AP-42) 95 (8 tests) ng 198 (4 tests) - oil - firing only; type unspecified 'Ratio of heat input from oil to heat input from gas 60/40. ------- standard oit-fired burners cannot be made at this time. 5. It is recommended that further testing include mechanical-draft oil-fired heaters with and without air preheat and equipped with low-No, burners as well as natural- draft gas-fired heaters equipped with low-NO, burners. 6. Existing data on heaters with standard burners have a number of gaps (see Table 5); further testing is required on standard burners firing distillate oil (both natural- and mechanical-draft) and standard burners firing resid- ual oil (natural-draft) for a more valid comparison with Iow-N0x burners. 7. Each test site in the program was evaluated as to its potential for a 30-day test. Considerations in- cluded: Fuel type fired. Ability to fire this fuel independ- ently of other fuels. Availability of the heater and the desired fuel for 30-40 days. Based on these considerations, the following sites are proposed as desirable for 30-day tests: Site 1 - gas-fired, forced-draft Site 2 - distillate-oil-fired, natural-draft. Site 4 - residual-oil-fired, natural-draft. Although Sites 6 through 10 are all gas-fired mechanical-draft heaters, they are not desirable sites for a 30-day test because they are still experiencing start-up problems; i.e., availability prob- lems. Site 5, firing a combination of gas and oil, does not satisfy the fuel specific- ity requirement. Sites 3 and 4 are both residual-oil-fired natural-draft heaters, but Site 4 is a vertical cylindrical heater similar to the other recommended sites. Because of this similarity in design. Site 4 is the better choice. The influence of heater design on NO, emission between the three candidate sites will thus be minimal. Since none of the residual oil tests provided low NO* levels, it may be desirable to try to find a site which can achieve lowr NO, levels. References 1. Hunter, S. C. et al., "Application of Advanced Combustion Modifica- 2. tions to Industrial Process Equip- ment: Subscale Test Results," USEPA, Industrial Environmental Research Laboratory, Research Triangle Park, NC (Draft No. IERL- RTP-1271). Hunter, S. C. and Cherry, S. S., NOjL Emission from Petroleum Industry Operations. API Publica- tion No. 4311, American Petroleum Institute, Washington, DC, October 1979. R. J. Tidona, H. J. Buening, andJ. R. Hart are with KVB, Inc., Irvine, CA 92714. Robert E. Hall is the EPA Project Officer (see below). The complete report, entitled "Emissions from Refinery Process Heaters Equipped with Low-NO* Burners,"(Order No. PB 82-231 838; Cost: $16.50, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Industrial Environmental Research Laboratory U.S. Environmental Protection Agency Research Triangle Park, NC 27711 ------- United States Environmental Protection Agency Center for Environmental Research Information Cincinnati OH 45268 Postage and Fees Paid Environmental Protection Agency EPA 335 Official Business Penalty for Private Use $300 USS MM329 ------- |