CHAPTER 14
NITROGEN OXIDE EMISSIONS AND THEIR DISTRIBUTORS
INTRODUCTION
Nitrogen Compounds
Nitrogen, the most abundant gas in the atmosphere, is found in a variety of gaseous and particulate forms. The overwhelming amount in air (79 percent of air by volume or 4.6×1012 tons) is present as relatively inert nitrogen gas, N2. However, the oxidation of nitrogen by bacteria, lightning, organic protein decay, and high temperature combustion and chemical processing causes the appearance of nitrogen in a variety of compounds. The most important, because of health effects and reactivity, are: NO (nitric oxide), NO2 (nitrogen dioxide), NH3 (ammonia) and to a lesser extent N2O (nitrous oxide).
Nitrous Oxide (N2O)
Nitrous oxide, a colorless and odorless gas, has been used as an anesthetic (laughing gas). It is present in the atmosphere in concentrations averaging about 0.25 ppm (Junge 1963, Bates and Hays 1967). There are no direct pollutant sources of N2O, although it may be an indirect and minor product of NO2 photolysis with sunlight and hydrocarbons. The in-
terest in N2O is not in the troposphere (ground level to 8–15 km) where it is practically inert but in photodissociation reactions in the stratosphere. Bates and Hays (1967) indicate that the dissociation of N2O into NO and atomic nitrogen accounts for about 20 percent of the dissociation in the stratosphere. The NO thus formed provides an important sink reaction for ozone.
Ammonia (NH3)
As an industrial emission, ammonia is produced mainly from coal and oil combustion but natural production from biological generation over land and ocean is many times greater than that from anthropogenic sources (250 to 1). NH3’s importance is the significant role it plays in atmospheric reactions in both the nitrogen and sulfur cycles. Nearly three-fourths of the NH3 is converted to ammonium ion condens ed in droplets or particles. These aerosols are then subject to the physical removal mechan isms of coagulation, washout, rainout and dry deposition.
In general, ambient background concentrations of NH3 vary directly with the intensity of biological activity. The highest concentrations occur in the summer and in the tropical latitudes. Concentrations, as reported by many investigators, range from 1 to 10 ppb (Strauss 1972).
Nitric Oxide-Nitrogen Dioxide
NO, a colorless, odorless gas, is formed naturally from the nitrates in various materials by bacteria and then is oxidized to NO2 (Peterson 1956).
Altschuller (1958) and others have reported very hazardous conditions for farm workers near closed silos where NO→NO2 bacterial production has resulted in toxic concentrations of several hundred ppm of NO2.
Organic nitrogen compounds are found in
coal and oil in concentrations of a few tenths to a few percent by weight. Bituminous coal contains 1–2 percent nitrogen, and United States crude oil approximately 0.05–0.5 percent nitrogen (Demski et al. 1973).
Natural gas, while containing up to 4 percent nitrogen gas, does not contain any significant organic nitrogen (Perry et al. 1963). Because organic nitrogen compounds have relatively high molecular weights, they tend to be concentrated in the residual and heavy oil fractions during distillation.
Nitrogen oxides are produced during combustion by the oxidation of organic nitrogen compounds in fossil fuels and by the thermal fixation of atmospheric nitrogen gas, N2.
The primary sources of NO and NO2 as pollutants are combustion processes in which temperatures are high enough to fix N in the air and fuel, and in which the quenching of combustion is rapid enough to reduce decomposition back to N2 and O2. The predominant product of this high temperature combustion is NO. During combustion, approximately 5–40 percent of the nitrogen in coal, and 20 to 100 percent of the nitrogen in oil is oxidized to nitrogen oxides (see Chapter 15).
NO is subsequently oxidized to NO2 either in the stack gas or, to a lesser extent, in the diluted plume. Once the NO has been diluted to 1 ppm (1230 μg/m3) or less, the direct reactions with O2 do not contribute significantly to NO2 formation (USEPA 1971). However, the reaction of NO with tropospheric ambient concentrations of O3 (ozone) to form NO2 is rapid. It is believed that the almost everpresent background concentrations of O3 will yield NO2 predominance over NO, although some researchers have reported higher NO than NO2 concentrations in remote areas (Lodge and Pate 1960, Ripperton et al. 1970).
NO2 is removed from the atmosphere either by further O3 oxidation to a nitrous salt or by the more favored conversion to HNO3 in the presence of water vapor. The HNO3 is then rapidly removed by reactions with NH3 and absorption by hygroscopic particles (Strauss 1972)
Nitrogen Oxides
Global Emissions
To estimate the total annual emission of oxides of nitrogen (NOx), emission factors have been applied to the fuel usage of several sources (Robinson and Robbins 1968). According to Robinson and Robbins, the annual production in 1967 was about 53×106 tons with coal combustion contributing the majority, 51 percent, followed by petroleum production and combustion contributing 41 percent. Natural gas combustion on a world-wide basis is comparatively less important (4 percent). However, it should be noted that on a local or regional basis, it could be the major source of NOx. (It should also be noted that these figures include combustion sources only since careful surveys of industrial process losses had not been undertaken at the time these estimates were prepared.) See Table 14–1.
National Emissions
Anthropogenic sources in the United States produce nearly 50 percent of the world’s NOx emissions (USEPA 1971). While emissions from human activities amount to far less than the estimated 50×107 tons of NOx emitted annually from natural sources (USEPA 1971), the spatial concentration of emissions in urban areas leads to concentrations of NOx 10 to 100 times higher than those in non-urban atmospheres.
Fuel combustion is the major cause of anth ropogenic NOx emissions in the United States (See Figure 14–12). In 1972, coal, oil, natural gas and motor-vehicle fuel combustion contributed over 86 percent of the estimated 24.6 million tons of NOx emitted in the United States. Stationary area and point sources account for approximately 64 percent of all the NOx. Direct stationary fuel combustion is the largest source
TABLE 14–1
World-Wide Urban Emissions of Nitrogen Dioxide
|
Fuel |
Source |
NO2 Emissions (106 Tons) |
% Total |
Sub % |
|
TOTAL |
|
52.9 |
100 |
|
|
COAL |
51 |
100 |
||
|
|
Power Generation |
12.2 |
23 |
47 |
|
Industrial |
13.7 |
26 |
52 |
|
|
Domestic/Commercial |
1.0 |
2 |
3 |
|
|
PETROLEUM |
41 |
100 |
||
|
|
Refinery Production |
0.7 |
1 |
3 |
|
Gasoline |
7.5 |
14 |
34 |
|
|
Kerosene |
1.3 |
2 |
6 |
|
|
Fuel Oil |
3.6 |
7 |
16 |
|
|
Residual Oil |
9.2 |
17 |
41 |
|
|
NATURAL GAS |
4 |
100 |
||
|
|
Power Generation |
0.6 |
1 |
25 |
|
Industrial |
1.1 |
2 |
50 |
|
|
Domestic/Commercial |
0.4 |
<1 |
25 |
|
|
OTHER |
||||
|
|
Incineration |
0.5 |
<1 |
|
|
Wood |
0.3 |
<1 |
||
|
Forest Fires |
0.5 |
1 |
||
|
Source: Modified from Robinson and Robbins (1968). |
||||
category (49.7 percent) with coal as the single largest contributor of NOx in this group.
Gasoline powered vehicles are the overwhelming source of transportation-related NOx contributing 32 percent of all NOx and 82 percent of the transportation NOx.
Significant quantities of NOx are emitted from industrial processes, primarily the manufacturing and use of nitric acid and refining of petroleum. On a local scale, electroplating, engraving, welding, and metal cleaning are responsible for industrial NOx emissions which may also be significant. In 1972, industrial process losses accounted for 2.9 million tons of NOx, or 11.7 percent of total nationwide emissions.
Overall, about 77 percent of the total NOx emissions occur in highly populated areas. Eighty percent of stationary source emissions occur in populated areas as do 71 percent of motor vehicle emissions.
The 1972 NOx emissions for the Nation will be defined in greater detail in the following section.
There has been a steady growth of NOx nationwide emissions. The decade of the sixties witnessed a greater increase in emissions than the previous two decades (see Table 14–2).
TABLE 14–2
NITROGEN OXIDES: Estimated Total Nationwide Emissions (106 tons) (USEPA 1974)
|
1940 |
1950 |
1960 |
1970 |
|
6.5 |
8.8 |
11.4 |
22.1 |
Over the past three decades, total nationwide emissions are estimated to have quadrupled. During this period, emissions from motor vehicles have increased at a steady rate of 4.6 to 4.9 percent per year. Emissions from stationary sources, however, have contributed progressively increasing proportions. Total NOx
emissions from power plants have increased at an annual rate of 6.9 to 7.4 percent.
Nitrogen Oxide Concentrations
Controversy and uncertainty about NOx measurement methods have made reliable urban NO and NO2 concentration data almost as scarce as the remote area data. Global remote measurements show NO2 variations with growing season, latitude, and altitude. Lodge and Pate (1966) reported average dry season values of 0.9 ppb and wet season values at 3.6 ppb NO2 in Panama. Junge reported in 1966 measured NO2 concentrations averaging 0.9 ppb in Florida and 1.3 ppb at 10,000 foot high Mauna Kea, Hawaii (Junge 1956). In the continental U.S., several investigators found NO2 values in the 4 ppb range and NO concentrations about 50 percent lower at 2 ppb (Hamilton et al. 1968, Ripperton et al. 1970).
Based upon the estimated global background levels and the annual emissions rates, the average residence time of NO2 in the atmosphere is about 3 days and that of NO is about 4 days. Residence times of atmospheric pollutants reflect the action of natural scavenging processes including photochemical reactions. Figure 14–1 provides a flow diagram summary of the atmospheric NO-NO2 cycle.
The spatial and temporal variations in ambient NO2 concentrations are great. Not surprisingly, the highest concentrations are found in urban regions. Measurements of NO2 have been taken since 1961 through the CAMP and SCAN programs of EPA (formerly NAPCA). However, there is now considerable uncertainty regarding ambient levels and trends for NOx. Although an EPA study of CAMP data for 5 cities reported slight increases in ambient NOx for 1962–1971, the data were obtained by the Jacobs-Hochheiser method for NOx analysis, which has been shown to overestimate ambient NOx levels at low concentrations (CEQ 1975). Thus, these NOx data must be viewed with caution. (See Appendix 14-B)
The National Air Surveillance Network (NASN) program for measuring NOx by a modified Jacobs-Hochheiser method was started in 1972 and it will be several years before the program can provide useful trend data on ambient NOx levels.
EPA has suspended all Air Quality Control Regions priority classifications based on NOx. However, regions where the standards for NO2 are believed to be exceeded are Los Angeles, Chicago, and Baltimore, and possible, New York-New Jersey-Connecticut, Salt Lake City, and Denver.
In spite of the uncertainties in the absolute values, there are NOx data available to indicate the general yearly trends in a few areas. The trends in CAMP observations of nitrogen dioxide are provided in Table 14–3. With the exception of Chicago, annual avenge nitric oxide concentrations of the period 1967–1971 are consistently higher than those of the period 1962–1966.
Annual nitrogen dioxide (NO2) averages show a greater variability among cities than do the nitric oxide averages, and the trends for nitrogen dioxide do not parallel those of nitric oxide. It is not clear whether this deviation is a result of instrument variation, or whether it can be attributed to differences in atmospheric conversion rates in various cities (NAS 1974).
Monitoring data for NO and NO2 in New Jersey cities presented in Figure 14–2 suggest that a pattern of change similar to that observed in the CAMP cities has occurred at these locations. Maximum monthly averages of nitric oxide appear to have increased after 1971, while the levels of NO2 have remained essentially constant over the same period.
In Los Angeles County there has been a direct relationship between increasing NOx emissions and the reported annual average one hour concentrations of NO2. Between 1965 and 1972, NOx emissions have increased in L.A. County at an annual average rate of 3.8 percent per year. The annual average of the maximum hourly average total NOx concentrations has increased approximately 3.2 percent per year from 1965 to
TABLE 14–3
Nitrogen Dioxide Trends in CAMP Cities 1962–71 (National Academy of Sciences 1974)
|
|
Average NO concentration, ppm |
Average NO2 concentration, ppm |
||
|
Station |
1962–66 |
1967–71 |
1962–66 |
1967–71 |
|
Chicago |
0.10 |
0.10 |
0.04 |
0.05 |
|
Cincinnati |
0.03 |
0.04 |
0.03 |
0.03 |
|
Denver |
0.03 |
0.04 |
0.04 |
0.04 |
|
Philadelphia |
0.04 |
0.53* |
0.04 |
0.04 |
|
St. Louis |
0.03 |
0.04 |
0.03 |
0.03 |
|
*The unusually high NO concentration for Philadelphia in 1967–71 is discussed in the section of this chapter on U.S. Nitrogen Oxide Emissions. |
||||
1972 at Burbank and downtown L.A. However, Anaheim and Azusa experienced increases in these annual averages of close to 11 percent per year.
There is more detailed discussion of NOx, hydrocarbon, and oxidant trends and relationships presented in Vol. 3, “The Relationship of Emissions to Ambient Air Quality” of the National Academy of Sciences Report, Air Quality and Automobile Emission Control, September 1974.
Diurnal Nitrogen Oxide Concentrations
The concentrations of NO and NOx in the ambient air change during a typical day. While all aspects of the typical diurnal cycle are not completely understood (Comm. of Mass, 1973, 1974; Coffey and Stasink 1975), it is possible to trace the concentrations of NO and NO2 throughout the day.
On a normal day in a city, ambient NOx levels follow a regular pattern with the sun and traffic. Nocturnal levels of NO and NO2 are relatively stable and are usually higher than the minimum daily values. During the dawn hours of 6 to 8 a.m., NO begins to increase as a result of automobile emissions. With increasing ultraviolet sunlight to drive the conversion reactions, NO2 concentrations increase until most of the NO is converted to secondary NO2. Photochemical oxidants accumulate as NO decreases to low levels (<0.1 ppm) and they reach a peak about midday. Secondary and tertiary reactions involving NO and NO2 occur leading to complex formations of eye irritants such as PAN (peroxyacylnitrates).
Later in the afternoon, there is decreased mixing, and increased atmospheric stability. Evening rush hour traffic (4 to 7 p.m.) produces another build-up of NO which is not readily converted to NO2 or more oxidants. The absence of sunlight does not completely halt NO2 formation, however. The principal oxidant present, ozone (O3), continues to react rapidly with NO to form NO2, until the O3 supply is
exhausted, and thus the early evening NO2 concentration may continue to rise. This condition may be ascribed to meteorological factors and, on cold evenings, to increased emissions from stationary sources.
A smoothed concentration profile for NO, NO2, carbon monoxide (CO), and O3 in Los Angeles on July 19, 1965, is displayed in Figure 14–3. The figure shows concentration on only a single day but it graphically displays the diurnal phenomenon described above. The NO2 peak lags the NO peak. The build-up of O3 is coincidental with the decrease of NO. The evening increase of NO2 apparently did not occur on this day.
The classical diurnal trend is also apparent in Figure 14–4, which illustrates the diurnal variation in monthly mean 1-hour average NO2 concentrations from St. Louis, Philadelphia, and Bayonne, N.J.
Figure 14–5 compares the diurnal patterns of NO for weekend days and weekdays (for Chicago CAMP stations). The Sunday 8 a.m. peak is about one-third of the weekday peak concentration. On some weekends, some locations have peak values, but these occur between 9 and 11 a.m.
The effect of stagnation conditions on the nitrogen oxide- photo-oxidant relationship is graphically exemplified in the diurnal concentrations of NO, NO2 and oxidants during a stagnating air mass over Washington, D.C. (USDOC 1967). Figures 14–6 through 14–8 compare the diurnal concentration pattern for NO, NO2 and oxidants that occurred during a four day stagnation in Washington, D.C., October 15 to 19, 1963, with a composite of the normal concentrations without stagnation and inversions. The efficiency of the NO-NO2 conversion to photo-oxidants is obvious. During midday when solar energy is most intense, the photo-oxidation of NO and NO2 in the presence of hydrocarbons is so efficient that their concentrations differ little from the norm. Details of these photochemical reactions are discussed in Appendix 14-A.
Seasonal Patterns
The concentration of NO displays a seasonal variation while that of NO2 does not. For NO, higher mean values are observed during late fall and winter months coinciding with decreased atmospheric mixing and generally less untraviolet energy available for the formation secondary products.
Figure 14–9 shows the seasonal patterns in NO by presenting the mean values by month of the year. The higher winter levels are quite distinctive.
The seasonal pattern for NO2 is less consistent, as seen in Figure 14–10. Thus there appears to be little monthly variation for Denver (1966) and Bayonne, N.J. (1967). Even though a greater amount of NO is converted to NO2 during the summer months, the actual concentrations of NO2 are governed by the rate of conversion to oxidant. This can lead to apparently contradictory comparisons between Los Angeles and Chicago monthly trends.
NO2 is highest during summer months in Chicago and during winter months in L.A. On closer examination of the L.A. aerometric data, one finds an inverse relationship between oxidant levels and NO2 levels. The presence of oxidants during summer months, when the synoptic weather condition superimposes a persistent inversion lid over the L.A. basin, and when there is higher solar radiation intensity and higher temperature, effectively scavenges the NO molecules. Figure 14–11.
Indoor Concentrations of Nitrogen Oxides
There is now a very active interest in determining pollutant levels in the home environment. However, very little of the research performed to date has been concerned with NOx indoor levels (USEPA 1973). One of the studies which has examined nitrogen oxide concentrations both indoors and outdoors was performed in Tokyo and found that although particulate matter and sulfur dioxide concentrations in indoor air
were lower than concentrations in outdoor air, there was no such difference for nitrogen dioxide (Miura et al. 1975). The results for NO2 confirmed the findings of an earlier study in Dushambe, Russia (Berdyer et al. 1967). It has been found that nitrogen oxide concentrations in private homes with gas stoves can exceed 100 μg/m3 (Wade et al. 1974). The existence of such pollutant levels indicates that there is clearly a need for further investigation of the sources of nitrogen oxides within the home and of the factors affecting the indoor/outdoor concentration relationship.
U.S NITROGEN OXIDE EMISSIONS
Description of Methodology of Emissions Estimation
Emission data from the Environmental Protection Agency’s National Emission Data System (NEDS) can be used to describe the present pattern of nitrogen oxide (NOx) emissions across the United States (USEPA 1972b). NEDS, a computerized emission data system, summarizes emissions by source type, e.g., transportation, electric generation, and by fuel use, e.g., coal, oil, and natural gas for every county, Air Quality Control Region (AQCR) and State in the U.S. Emissions are further summarized by point and area source classification for each source type. Point sources are all industrial process emission sources and stationary fuel combustion sources in urban areas emitting more than 100 tons per year of any one contaminant, and stationary fuel combustion sources in rural areas emitting more than 25 tons per year of any one pollutant. Area sources include all sources other than those defined as point sources.
The 1972 National Emissions Report from NEDS provides the basis for all of the current emission descriptions in this section. The 1972 report contains emission data representative of either 1970, 1971, or 1972 for each political
jurisdiction, depending on the base year inventory used in state implementation plans required by the Clean Air Act. Because the 1972 report represents the first such comprehensive emissions summary and because the sophistication of state and local jurisdictions in gathering emissions data varies widely, the accuracy of the data varies from state to state. Such variability can result in apparent data anomalies. For example, according to the 1972 report, the Metropolitan Philadelphia Interstate AQCR accounts for 75 percent of the nationwide industrial process emissions of nitrogen oxides. This is presumed to occur because this AQCR contains the only comprehensive survey of petroleum refinery emissions in the U.S. and seems to indicate that nationwide industrial process emissions are significantly underestimated. However, the emission data for stationary fuel combustion and transportation sources which account for over 85 percent of the nationwide NOx emissions are much more accurate and consistent. In fact, the national emission totals obtained from the 1972 report agree quite closely with independent national emission estimates performed with national fuel use and vehicle usage data (Mason and Shimizu 1974, Cortese 1975)
As new and more accurate emission factors become available, emission estimates and projections will be improved. Currently there is variability in emission estimates resulting from use of different emission factors. Because of recent changes in transportation source emission factors, the 1972 NEDS data overestimates diesel emissions by 590,000 tons per year and underestimates gasoline emissions by 320,000 tons per year (USEPA 1973b, 1974b). Thus total transportation emissions are overestimated by approximately 270,000 tons per year.
Nationwide Nitrogen Oxide Emission Patterns
Nationwide NOx emissions for 1972 are summarized in Table 14–4 and Figure 14–12. In 1972, 24.64 million tons of NOx were emitted into the nation’s atmosphere; of this total 49.7 percent
TABLE 14–4
Summary of Nationwide NOx Emissions, 1972
|
|
Emissions (106 tons/year) |
% of Total Emissions |
||||
|
Source Type |
Total |
Point |
Area |
Total |
Point |
Area |
|
Stationary Fuel Combustion |
12.27 |
10.78 |
1.49 |
49.7 |
43.8 |
5.9 |
|
Transportation |
8.72 |
- |
8.72 |
35.4 |
- |
35.4 |
|
Industrial Process Lossess |
2.88 |
2.88 |
- |
11.7 |
11.7 |
- |
|
Solid Waste Disposal |
0.18 |
0.03 |
0.15 |
0.7 |
0.1 |
0.6 |
|
Miscellaneous |
0.17 |
- |
0.17 |
0.7 |
- |
0.7 |
|
New York Point Sourcesa |
0.42 |
0.42 |
- |
1.8 |
1.8 |
- |
|
TOTAL |
24.64 |
14.11 |
10.53 |
100.0 |
57.3 |
42.7 |
|
aNew York point sources were separately treated since a further dissection of these data by source type was not possible. |
||||||
or 12.27 million tons were produced by stationary fuel combustion sources and 35.4 percent or 8.72 million tons were produced by transportation sources. All fuel combustion accounted for over 85 percent of the national total. Dissecting the data in a slightly different manner, point sources accounted for 57.3 percent (14.11 million tons) while area sources accounted for 42.7 percent (10.53 million tons) of the total national emissions. The major portion of the point source emissions was contributed by stationary fuel combustion and industrial process losses (97 percent) and the major portion of area source emissions was contributed by transportation sources (83 percent).
Since stationary fuel combustion sources accounted about half of the total national emissions in 1972, it is appropriate to examine the types of stationary sources and fuels which make the major contribution to stationary fuel combustion emissions. As shown in Table 14–5 and Figure 14–13, electric power generation (48 percent) and industrial fuel combustion (44 percent) shared almost equally in the production of over 92 percent of the stationary fuel combustion emissions. However, a look at the type of fuel consumed in electrical power generation and industrial heating discloses differences. Coal accounted for 67 percent of the NOx emissions from electric power generation, but only 14 percent of the industrial fuel combustion emissions. The major contributor to industrial fuel combustion NOx emissions was industrial process gas, such as coke oven gas and refinery gas, which accounts for 48 percent of the industrial fuel combustion emissions. Natural gas was the second largest contributor for each source type, accounting for 30 percent of industrial fuel combustion emissions and 19 percent of electric power generation emissions.
The predominance of coal-related NOx electric power generation emissions is a result of two factors. First, coal is the major fuel used in electric power generation. Secondly, coal has a higher NOx emission rate than oil or natural gas on an equivalent heat production basis as indicated in Table 14–6.
TABLE 14–5
Summary of Nationwide NOx Emissions by Source Type and Fuel Use, 1972
|
Source Type |
Emissions (106 tons/year) |
% of Total Emissions |
||||
|
Stationary Fuel Combustion |
12.27 |
|
49.7 |
|
||
|
Electric Generation |
|
5.94 |
|
|
24.1 |
|
|
Coal |
|
3.95 |
|
16.0 |
||
|
Oil |
0.85 |
3.4 |
||||
|
Natural Gas |
1.14 |
4.7 |
||||
|
Industrial Fuel COmbustion |
|
5.39 |
|
|
21.8 |
|
|
Coal |
|
0.76 |
|
3.0 |
||
|
Oil |
0.41 |
1.6 |
||||
|
Process Gas |
2.58 |
10.5 |
||||
|
Natural Gas |
1.64 |
6.7 |
||||
|
Commercial-Institutional |
|
0.65 |
|
|
2.6 |
|
|
Residential |
0.29 |
1.2 |
||||
|
Transportation |
8.72 |
|
35.4 |
|
||
|
Gasoline |
|
6.62 |
|
|
26.9 |
|
|
Diesel |
1.90 |
7.7 |
||||
|
Other |
0.20 |
0.8 |
||||
|
Industrial Process Losses |
2.88 |
|
11.7 |
|
||
|
Solid Waste Disposal |
0.18 |
0.7 |
||||
|
Miscellaneous |
0.17 |
0.7 |
||||
|
New York Point Sources |
0.42 |
1.8 |
||||
|
TOTAL |
24.64 |
100.0 |
||||
Projected increases in energy demand coupled a desire to convert existing electric utilities from oil and natural gas to coal combustion and to require new electric utilities to burn coal could result in a substantial increase in NOx emissions from electric power generation in the future.
Geographical Nitrogen Oxide Emission Patterns
General Patterns
Geographical nitrogen oxide emission patterns were determined by summarizing NOx emission data for all the AQCR’s and states located within the jurisdictional boundaries of each of the Environmental Protection Agency’s (EPA) 10 regional offices. A complete listing of the states located within the jurisdictional boundaries of each EPA regional office is given in Table 14–7.
Table 14–8 displays the geographical distribution of 1972 NOx emissions for the United States. The North-east includes all states east of the Mississippi River and north of the Mason-Dixon line, the South includes all states south of the Mason-Dixon line and east of Arizona, and the West includes all of the area west of the Mississippi River and north of Oklahoma and Arkansas. The North-east was responsible for 56 percent of total U.S. NOx emissions with the South and West being nearly equal contributors to the remaining 44 percent. Distribution of stationary fuel combustion emissions followed a similar pattern. Transporation emissions followed a somewhat different pattern more closely following population patterns. The North-east accounted for 42 percent of U.S. NOx emissions from transportation while the South and West contributed 32 percent and 26 percent respectively.
A closer look at the contribution of EPA Regions to total emissions reveals that 27 percent of U.S. total NOx emissions and 39 percent of U.S. stationary fuel combustion emissions
TABLE 14–7
Listing of States Located Within Jurisdictional Boundaries of EPA Regional Offices (EPA 1972)
|
EPA Region |
States and Territories |
|
I |
Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, Vermont |
|
II |
New Jersey, New York, Puerto Rico, Virgin Islands |
|
III |
Delaware, Maryland, Pennsylvania, Virginia, West Virginia |
|
IV |
Alabama, Florida, Georgia, Kentucky, Mississippi, North Carolina, South Carolina, Tennessee |
|
V |
Illinois, Indiana, Michigan, Minnesota, Ohio, Wisconsin |
|
VI |
Arkansas, Louisiana, New Mexico, Oklahoma, Texas |
|
VII |
Iowa, Kansas, Missouri, Nebraska |
|
VIII |
Colorado, Montana, North Dakota, South Dakota, Utah, Wyoming |
|
IX |
Arizona, California, Hawaii, Nevada, American Samoa, Guam |
|
X |
Alaska, Idaho, Oregon, Washington |
TABLE 14–8
Geographical Distribution of Nitrogen Oxide Emissions, 1972a
|
Geographical Area (EPA Region)b |
% Total U.S. Emissions |
% Total U.S. Stationary Fuel Combustion Emissions |
% Total U.S. Transportation Emissions |
||||
|
EAST |
|
56 |
|
59 |
|
42 |
|
|
|
V |
|
27 |
|
38 |
|
19 |
|
III |
19 |
12 |
10 |
||||
|
II |
7 |
6 |
9 |
||||
|
I |
3 |
3 |
4 |
||||
|
SOUTH |
|
24 |
|
23 |
|
32 |
|
|
|
IV |
|
14 |
|
14 |
|
19 |
|
VI |
10 |
9 |
13 |
||||
|
WEST |
|
20 |
|
18 |
|
26 |
|
|
|
IV |
|
10 |
|
10 |
|
12 |
|
VII |
5 |
5 |
7 |
||||
|
VIII |
3 |
2 |
4 |
||||
|
X |
2 |
1 |
3 |
||||
|
aPercentages have been rounded off bRegions listed in decreasing order of contribution to Area emissions. |
|||||||
were produced in the six Eastern states located in Region V. Regional transportation emissions were consistent with population distribution. Stationary fuel combustion emissions on the other hand, reflected geographical differences in industrialization and electric power requirements.
NOx Emission Patterns in EPA Regions
Tables 14–9 through 14–20 compare the distribution of NOx emissions in each of the ten EPA Regions. The Regional NOx emissions are dissected according to the degree of urbanization is determined by the largest Standard Metropolitan Statistical Area (SMSA) population within an AQCR’s are grouped in the following manner: grouped in the following manner:
-
large urban AQCR’s: AQCR’s with an SMSA population greater than 1 million;
-
medium-sized urban AQCR’s: AQCR’s with an SMSA population of 250,000–1,000,000
-
small urban AQCR’s: AQCR’s with an SMSA population of 50,000–250,000; and
-
rural AQCR’s: AQCR’s not containing an SMSA.
Although it reflects the obvious distribution of population, electric power generation and industrialization, Table 14–9 reveals some interesting points:
-
Region VIII had high rural emissions due to electric power generation. Electric generation emissions were 62 percent of rural emissions and 25 percent of the Region’s total emissions. This reflects the presence of the Four Corners power plant in Region VIII.
-
In Region X the rural emissions were also high. However, 66 percent of the emissions
TABLE 14–9
Effect of Urbanization on Nitrogen Oxide Emissions for Different Geographical Areasa,b
TABLE 14–10
1972 NOx Emissions
EPA Region I
|
Air Quality Control Regions |
||||||||||||||||||
|
Source Category |
Rural (7)a |
Medium-Sized, Urban (3)a |
Total Urban |
Total Region |
||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Ton |
% |
Tons |
% |
Tons |
% |
|||||||
|
Stationary Fuel Combustion |
20632 |
33 |
100 |
89116 |
48 |
100 |
158335 |
51 |
100 |
98664 |
59 |
100 |
346115 |
52 |
100 |
366747 |
50 |
100 |
|
Residential |
2790 |
|
13 |
5657 |
|
6 |
8000 |
|
5 |
5918 |
|
6 |
19575 |
|
6 |
22365 |
|
6 |
|
Electric Generation |
3359 |
16 |
48298 |
54 |
90825 |
57 |
47617 |
48 |
186740 |
54 |
190099 |
52 |
||||||
|
Industrial |
10785 |
52 |
20407 |
23 |
19271 |
12 |
6581 |
7 |
46259 |
13 |
57044 |
16 |
||||||
|
Commercial-Institutional |
3695 |
18 |
14754 |
17 |
40239 |
25 |
38548 |
39 |
93541 |
27 |
97236 |
26 |
||||||
|
Industrial Process Losses |
23 |
0 |
|
1041 |
1 |
|
184 |
0 |
|
0 |
0 |
|
1225 |
0 |
|
1248 |
0 |
|
|
Solid Waste Disposal |
1254 |
2 |
3816 |
2 |
4105 |
1 |
1058 |
1 |
8979 |
1 |
10233 |
1 |
||||||
|
Transportation |
41654 |
66 |
100 |
90528 |
49 |
100 |
148409 |
48 |
100 |
68830 |
41 |
100 |
307767 |
46 |
100 |
349421 |
48 |
100 |
|
Light-duty gas vehicles |
27778 |
|
67 |
64178 |
|
71 |
102637 |
|
69 |
56075 |
|
81 |
222890 |
|
72 |
250668 |
|
72 |
|
Other |
13876 |
33 |
26350 |
29 |
45772 |
31 |
12755 |
19 |
84877 |
28 |
98753 |
28 |
||||||
|
Miscellaneous |
0 |
0 |
|
0 |
0 |
|
0 |
0 |
|
0 |
0 |
|
0 |
0 |
|
0 |
0 |
|
|
Total |
63564 |
100 |
184500 |
100 |
311032 |
100 |
168553 |
100 |
664085 |
100 |
727649 |
100 |
||||||
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA population within AQCR c Total urban=emissions from all AQCR’s except Rural AQCR’s |
||||||||||||||||||
TABLE 14–11
1972 NOx Emissions
EPA Region IId
|
Air Quality Control Regions |
||||||||||||||||||
|
Source Category |
Rural (2)a |
Total Urbanc |
Total Region |
|||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
|||||||
|
Stationary Fuel Combustion |
9516 |
18 |
100 |
31985 |
58 |
100 |
35120 |
18 |
100 |
317296 |
36 |
100 |
384401 |
34 |
100 |
393917 |
33 |
100 |
|
Residential |
1797 |
|
19 |
819 |
|
3 |
7208 |
|
21 |
39896 |
|
13 |
47923 |
|
12 |
49720 |
|
13 |
|
Electric Generation |
824 |
9 |
22884 |
72 |
0 |
0 |
139924 |
44 |
162808 |
42 |
163632 |
42 |
||||||
|
Industrial |
713 |
7 |
2851 |
9 |
0 |
0 |
57939 |
18 |
60790 |
16 |
61503 |
16 |
||||||
|
Commercial-Institutional |
6181 |
65 |
5431 |
17 |
27913 |
79 |
79538 |
25 |
112882 |
29 |
119063 |
30 |
||||||
|
Industrial Process Losses |
0 |
0 |
|
402 |
0.7 |
|
0 |
0 |
|
23335 |
3 |
|
23737 |
2 |
|
23737 |
2 |
|
|
Solid Waste Disposal |
706 |
1 |
240 |
0.4 |
3261 |
2 |
15523 |
2 |
19024 |
2 |
19730 |
2 |
||||||
|
Transportation |
41843 |
80 |
100 |
22512 |
41 |
100 |
161947 |
81 |
100 |
511706 |
59 |
100 |
696165 |
62 |
100 |
738008 |
63 |
100 |
|
Light Duty gas vehicles |
25126 |
|
60 |
11325 |
|
50 |
96189 |
|
59 |
341373 |
|
67 |
448887 |
|
64 |
474013 |
|
64 |
|
Other |
16717 |
40 |
11187 |
50 |
65758 |
41 |
170333 |
33 |
247278 |
36 |
263995 |
36 |
||||||
|
Miscellaneous |
0 |
0 |
|
0 |
0 |
|
0 |
0 |
|
2145 |
0 |
|
2145 |
0 |
|
2145 |
0 |
|
|
Total |
52064 |
100 |
55139 |
100 |
200329 |
100 |
870005 |
100 |
1125473 |
100 |
1177537 |
100 |
||||||
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA population within AQCR c Total urban=emissions from all AQCR’s except Rural AQCR’s d Does not include 421,000 tons from New York City Point Sources. |
||||||||||||||||||
TABLE 14–12
1972 NOx Emissions
EPA Region III
|
Air Quality Control Regions |
|||||||||||||||||||
|
Source Category |
Rural (12)a |
Total Urbanc |
Total Region |
||||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
||||||||
|
Stationary Fuel Combustion |
200902 |
65 |
100 |
219192 |
73 |
100 |
422045 |
61 |
100 |
723466 |
22 |
100 |
1364703 |
31 |
100 |
1565605 |
34 |
100 |
|
|
Residential |
4515 |
|
2 |
3389 |
|
1 |
11185 |
|
3 |
19096 |
|
3 |
33670 |
|
2 |
38185 |
|
2 |
|
|
Electric Generation |
156591 |
78 |
162769 |
74 |
301993 |
72 |
496032 |
69 |
960794 |
70 |
1117385 |
71 |
|||||||
|
Industrial |
35347 |
18 |
49738 |
23 |
85460 |
20 |
146640 |
20 |
281838 |
21 |
317185 |
20 |
|||||||
|
Commercial-Institutional |
4559 |
2 |
3926 |
1.7 |
23407 |
6 |
61697 |
9 |
89030 |
7 |
93589 |
6 |
|||||||
|
Industrial Process Losses |
2025 |
0.6 |
|
6210 |
2 |
|
4524 |
0.6 |
|
2175857 |
65 |
|
2186591 |
50 |
|
2188616 |
47 |
|
|
|
Solid Waste Disposal |
1749 |
0.5 |
1259 |
0.4 |
1341 |
0.2 |
6312 |
0.1 |
8912 |
0.2 |
10661 |
0.2 |
|||||||
|
Transportation |
103413 |
34 |
100 |
75748 |
25 |
100 |
265678 |
38 |
100 |
439907 |
13 |
100 |
781333 |
18 |
100 |
884746 |
19 |
100 |
|
|
Light-duty gas vehicles |
59474 |
|
58 |
43950 |
|
58 |
160580 |
|
60 |
266356 |
|
61 |
470886 |
|
60 |
530360 |
|
60 |
|
|
Other |
43939 |
42 |
31798 |
42 |
105098 |
40 |
173551 |
39 |
310447 |
40 |
354386 |
40 |
|||||||
|
Miscellaneous |
56 |
.01 |
|
0 |
0 |
|
0 |
0 |
|
4 |
0 |
|
4 |
0 |
|
60 |
0 |
|
|
|
Total |
308147 |
100 |
302410 |
100 |
693,587 |
100 |
3345546 |
100 |
4341543 |
100 |
4649690 |
100 |
|||||||
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA Population within AQCR c Total urban=emissions from all AQCR’s except Rural AQCR’s |
|||||||||||||||||||
TABLE 14–13
1972 NOx Emissions
EPA Region IV
|
Air Quality Control Regions |
||||||||||||||||||
|
Source Category |
Rural (16)a |
Total Urbanc |
Total Region |
|||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
|||||||
|
Stationary Fuel Combustion |
156522 |
31 |
100 |
376839 |
52 |
100 |
950265 |
52 |
100 |
237407 |
49 |
100 |
1564511 |
51 |
100 |
1721033 |
48 |
100 |
|
Residential |
6672 |
|
4 |
8115 |
|
2 |
11984 |
|
1 |
1646 |
|
1 |
21739 |
|
1 |
28411 |
|
2 |
|
Electric Generation |
97421 |
62 |
247034 |
66 |
689749 |
73 |
184154 |
77 |
1102937 |
70 |
1218358 |
71 |
||||||
|
Industrial |
43360 |
28 |
111834 |
30 |
220141 |
23 |
45886 |
19 |
377861 |
24 |
421221 |
24 |
||||||
|
Commercial-Institutional |
9066 |
6 |
9858 |
3 |
28393 |
3 |
5726 |
2 |
43977 |
3 |
53043 |
3 |
||||||
|
Industrial Process Losses |
9072 |
2 |
|
9180 |
1 |
|
67525 |
4 |
|
5539 |
1 |
|
82244 |
3 |
|
91316 |
3 |
|
|
Solid Waste Disposal |
6359 |
1 |
8169 |
1 |
12670 |
1 |
2332 |
1 |
23171 |
1 |
29530 |
1 |
||||||
|
Transportation |
338750 |
66 |
100 |
329640 |
46 |
100 |
804008 |
44 |
100 |
234761 |
49 |
100 |
1368409 |
45 |
100 |
1707159 |
48 |
100 |
|
Light-duty gas vehicles |
243307 |
|
72 |
199124 |
|
60 |
589073 |
|
73 |
146702 |
|
62 |
934899 |
|
68 |
1178206 |
|
69 |
|
Other |
95443 |
28 |
130156 |
40 |
214935 |
27 |
88059 |
38 |
433510 |
32 |
528953 |
31 |
||||||
|
Miscellaneous |
1076 |
0 |
|
0 |
0 |
|
2394 |
0 |
|
1974 |
0 |
|
4368 |
0 |
|
5444 |
0 |
|
|
Total |
511776 |
100 |
723834 |
100 |
1837044 |
100 |
482016 |
100 |
3042894 |
100 |
3554670 |
100 |
||||||
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA population within AQCR’s except Rural AQCR’s c Total urban=emissions from all AQCR’s except Rural AQCR’s |
||||||||||||||||||
TABLE 14–14
1972 NOx Emissions
EPA Region V
|
Air Quality Control Regions |
||||||||||||||||||
|
Source Category |
Rural (13)a |
Total Urbanc |
Total Region |
|||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
|||||||
|
Stationary Fuel Combustion |
550994 |
66 |
100 |
136182 |
46 |
100 |
754891 |
55 |
100 |
3375474 |
80 |
100 |
4266552 |
73 |
100 |
4817546 |
72 |
100 |
|
Residential |
9695 |
|
2 |
5694 |
|
4 |
22719 |
|
3 |
37865 |
|
1 |
66278 |
|
2 |
72973 |
|
2 |
|
Electric Generation |
441305 |
80 |
53019 |
39 |
445422 |
59 |
574867 |
17 |
1073308 |
25 |
1514613 |
31 |
||||||
|
Industrial |
84982 |
15 |
63809 |
47 |
234826 |
31 |
2693250 |
80 |
2991885 |
70 |
3076867 |
64 |
||||||
|
Commercial-Institutional |
15048 |
3 |
13657 |
10 |
51925 |
7 |
69498 |
2 |
135080 |
3 |
150128 |
3 |
||||||
|
Industrial Process Losses |
47450 |
6 |
|
2361 |
1 |
|
35693 |
3 |
|
51781 |
1 |
|
89835 |
2 |
|
137285 |
2 |
|
|
Solid Waste Disposal |
3399 |
0.4 |
3004 |
1 |
9946 |
1 |
21970 |
1 |
34920 |
1 |
38319 |
1 |
||||||
|
Transportation |
231182 |
28 |
100 |
151566 |
52 |
100 |
562855 |
|
100 |
727989 |
17 |
100 |
1442410 |
25 |
100 |
1673592 |
25 |
100 |
|
Light-duty gas vehicles |
121864 |
|
53 |
79263 |
|
52 |
294859 |
|
52 |
438331 |
|
60 |
812453 |
|
56 |
934317 |
|
56 |
|
Other |
109318 |
47 |
72303 |
48 |
267996 |
48 |
289658 |
40 |
629957 |
44 |
739275 |
44 |
||||||
|
Miscellaneous |
2 |
0 |
|
0 |
0 |
|
0 |
0 |
|
1 |
0 |
|
1 |
|
|
3 |
0 |
|
|
Total |
832307 |
100 |
293103 |
100 |
1363388 |
100 |
4197602 |
100 |
5854093 |
100 |
6686400 |
100 |
||||||
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA population within AQCR c Total urban=emissions from all AQCR’s except Rural AQCR’s |
||||||||||||||||||
TABLE 14–15
1972 NOx Emissions
EPA Region VI
|
Air Quality Control Regions |
||||||||||||||||||
|
Source Category |
Rural (9)a |
Total Urbanc |
Total Region |
|||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
|||||||
|
Stationary Fuel Combustion |
72887 |
38 |
100 |
157265 |
40 |
100 |
417050 |
50 |
100 |
488012 |
48 |
100 |
1062327 |
47 |
100 |
1135214 |
47 |
100 |
|
Residential |
2510 |
|
3 |
7711 |
|
5 |
6385 |
|
2 |
5527 |
|
1 |
19623 |
|
2 |
22133 |
|
2 |
|
Electric Generation |
38846 |
53 |
81637 |
52 |
201899 |
48 |
301193 |
62 |
584729 |
55 |
623575 |
55 |
||||||
|
Industrial |
28789 |
39 |
62068 |
39 |
194014 |
47 |
163377 |
33 |
419459 |
39 |
448248 |
39 |
||||||
|
Commercial-Institutional |
2744 |
4 |
5838 |
4 |
14751 |
4 |
17915 |
4 |
38504 |
4 |
41248 |
4 |
||||||
|
Industrial Process Losses |
4736 |
2 |
|
28701 |
7 |
|
25989 |
3 |
|
123458 |
12 |
|
178148 |
8 |
|
182884 |
7 |
|
|
Solid Waste Disposal |
2127 |
1 |
3396 |
1 |
6610 |
1 |
8825 |
1 |
18831 |
1 |
20958 |
1 |
||||||
|
Transportation |
113393 |
59 |
100 |
207023 |
52 |
100 |
387972 |
46 |
100 |
386572 |
38 |
100 |
981567 |
44 |
100 |
1094960 |
45 |
100 |
|
Light-duty gas vehicles |
55143 |
|
49 |
90624 |
|
44 |
198071 |
|
51 |
205388 |
|
53 |
494083 |
|
51 |
549226 |
|
50 |
|
Other |
58250 |
51 |
116399 |
56 |
189901 |
49 |
181184 |
47 |
487484 |
49 |
545734 |
50 |
||||||
|
Miscellaneous |
0 |
0 |
|
0 |
0 |
|
0 |
0 |
|
0 |
0 |
|
0 |
0 |
|
0 |
0 |
|
|
Total |
193143 |
100 |
396375 |
100 |
837620 |
100 |
1012866 |
100 |
2246861 |
100 |
2440004 |
100 |
||||||
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA population within AQCR c Total urban=emissions from all AQCR’s except Rural AQCR’s |
||||||||||||||||||
TABLE 14–16
1972 NOx Emissions
EPA Region VII
|
Air Quality Control Regions |
||||||||||||||||||
|
Source Category |
Rural (10)a |
Total Urbanc |
Total Region |
|||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
|||||||
|
Stationary Fuel Combustion |
90687 |
32 |
100 |
195354 |
46 |
100 |
16667 |
41 |
100 |
375130 |
67 |
100 |
587151 |
57 |
100 |
677838 |
51 |
100 |
|
Residential |
4731 |
|
5 |
6176 |
|
3 |
616 |
|
4 |
4916 |
|
1 |
11708 |
|
2 |
16439 |
|
2 |
|
Electric Generation |
54077 |
60 |
146372 |
75 |
12879 |
77 |
327690 |
87 |
486941 |
83 |
541018 |
80 |
||||||
|
Industrial |
24567 |
27 |
30833 |
16 |
1539 |
9 |
35877 |
10 |
68249 |
12 |
92816 |
14 |
||||||
|
Commercial-Institutional |
7311 |
8 |
11975 |
6 |
1633 |
10 |
6647 |
2 |
20255 |
3 |
27566 |
4 |
||||||
|
Industrial Process Losses |
8530 |
3 |
|
11695 |
3 |
|
1354 |
3 |
|
15619 |
3 |
|
28668 |
3 |
|
37198 |
3 |
|
|
Solid Waste Disposal |
2926 |
1 |
3991 |
1 |
216 |
0.5 |
2470 |
0.4 |
6677 |
0.6 |
9603 |
1 |
||||||
|
Transportation |
184239 |
64 |
100 |
214704 |
50 |
100 |
21855 |
54 |
100 |
170659 |
30 |
100 |
407218 |
40 |
100 |
591457 |
45 |
100 |
|
Light-duty gas vehicles |
84109 |
|
46 |
111437 |
|
52 |
13191 |
|
60 |
96782 |
|
57 |
221410 |
|
54 |
305519 |
|
52 |
|
Other |
100130 |
54 |
103267 |
48 |
8664 |
40 |
73877 |
43 |
185808 |
46 |
285938 |
48 |
||||||
|
Miscellaneous |
745 |
0.2 |
|
356 |
0.1 |
|
80 |
0.1 |
|
0 |
0 |
|
436 |
.04 |
|
1181 |
0.1 |
|
|
Total |
287126 |
100 |
426102 |
100 |
40172 |
100 |
563877 |
100 |
1030151 |
100 |
1317277 |
100 |
||||||
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA population within AQCR c Total urban=emissions from all AQCR’s except Rural AQCR’s |
||||||||||||||||||
TABLE 14–17
1972 NOx Emissions
EPA Region VIII
|
Air Quality Control Regions |
||||||||||||||||||
|
Source Category |
Rural (16)a |
Total Urbanc |
Total Region |
|||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
|||||||
|
Stationary Fuel Combustion |
229297 |
51 |
100 |
31325 |
18 |
100 |
21900 |
38 |
100 |
30194 |
37 |
100 |
83419 |
26 |
100 |
312716 |
41 |
100 |
|
Residential |
5390 |
|
2 |
1394 |
|
4 |
1288 |
|
6 |
1441 |
|
5 |
4123 |
|
5 |
9513 |
|
3 |
|
Electric Generation |
187860 |
82 |
11847 |
38 |
6298 |
29 |
18192 |
60 |
36337 |
44 |
224197 |
72 |
||||||
|
Industrial |
27406 |
12 |
14400 |
46 |
10647 |
49 |
6696 |
22 |
31743 |
38 |
59149 |
19 |
||||||
|
Commercial-Institutional |
8704 |
4 |
3685 |
12 |
3667 |
17 |
3863 |
13 |
11215 |
13 |
19919 |
6 |
||||||
|
Industrial Process Losses |
7976 |
2 |
|
91591 |
51 |
|
5095 |
9 |
|
4726 |
6 |
|
101412 |
32 |
|
109388 |
14 |
|
|
Solid Waste Disposal |
2864 |
1 |
691 |
0 |
329 |
1 |
188 |
0 |
1208 |
0 |
4072 |
1 |
||||||
|
Transportation |
203745 |
46 |
100 |
54018 |
30 |
100 |
30581 |
53 |
100 |
47469 |
57 |
100 |
132068 |
41 |
100 |
335813 |
44 |
100 |
|
Light-duty gas vehicles |
92125 |
|
45 |
27671 |
|
51 |
18981 |
|
62 |
29631 |
|
62 |
76283 |
|
58 |
168408 |
|
50 |
|
Other |
111620 |
55 |
26347 |
49 |
11600 |
38 |
17838 |
38 |
55785 |
42 |
167405 |
50 |
||||||
|
Miscellaneous |
3780 |
1 |
|
434 |
0 |
|
10 |
0 |
|
0 |
0 |
|
444 |
0 |
|
4224 |
1 |
|
|
Total |
447700 |
100 |
178059 |
100 |
57916 |
100 |
82578 |
100 |
318553 |
100 |
766253 |
100 |
||||||
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA population within AQCR c Total urban=emissions from all AQCR’s except Rural AQCR’s |
||||||||||||||||||
TABLE 14–18
1972 NOx Emissions
EPA Region IX
|
Air Quality Control Regions |
||||||||||||||||||
|
Source Category |
Rural (12)a |
Total Urbanc |
Total Region |
|||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
|||||||
|
Stationary Fuel Combustion |
156693 |
55 |
100 |
5177 |
42 |
100 |
155213 |
33 |
100 |
882293 |
57 |
100 |
1042683 |
51 |
100 |
1199376 |
52 |
100 |
|
Residential |
1915 |
|
1 |
255 |
|
5 |
4164 |
|
3 |
12172 |
|
1 |
16591 |
|
2 |
18506 |
|
2 |
|
Electric Generation |
112924 |
72 |
4259 |
82 |
121888 |
78 |
182969 |
21 |
309116 |
30 |
422040 |
35 |
||||||
|
Industrial |
39254 |
25 |
0 |
0 |
19381 |
12 |
660586 |
75 |
679967 |
65 |
719221 |
60 |
||||||
|
Commercial-Institutional |
2604 |
2 |
662 |
13 |
9781 |
7 |
26568 |
4 |
37011 |
4 |
39615 |
3 |
||||||
|
Industrial Process Losses |
13167 |
5 |
|
1 |
0 |
|
3804 |
0.5 |
|
62140 |
4 |
|
65945 |
3 |
|
79112 |
3 |
|
|
Solid Waste Disposal |
2703 |
1 |
5 |
0 |
7368 |
2 |
17189 |
1 |
24562 |
1 |
27265 |
1 |
||||||
|
Transportation |
110240 |
39 |
100 |
7049 |
|
100 |
304801 |
64 |
100 |
589829 |
38 |
100 |
901679 |
44 |
100 |
1011919 |
44 |
100 |
|
Light-duty gas vehicles |
51967 |
|
47 |
4342 |
|
62 |
163512 |
|
54 |
381100 |
|
65 |
548954 |
|
61 |
600921 |
|
59 |
|
Other |
58273 |
53 |
2707 |
38 |
141289 |
46 |
208729 |
35 |
352725 |
39 |
410998 |
41 |
||||||
|
Miscellaneous |
0 |
0 |
|
0 |
0 |
|
2611 |
.5 |
|
0 |
0 |
|
2611 |
0 |
|
2611 |
0 |
|
|
Total |
282820 |
100 |
12232 |
100 |
473794 |
100 |
1551453 |
100 |
2037479 |
100 |
2320299 |
100 |
||||||
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA population within AQCR c Total urban=emissions from all AQCR’s except Rural AQCR’s |
||||||||||||||||||
TABLE 14–19
1972 NOx Emissions
EPA Region X
|
Air Quality Control Regions (Washington, Oregon, Idaho) |
||||||||||||||||||
|
Source Category |
Rural (8)a |
Total Urbanc |
Total Region |
|||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
|||||||
|
Stationary Fuel Combustion |
32346 |
21 |
100 |
5192 |
21 |
100 |
4673 |
47 |
100 |
37512 |
21 |
100 |
47377 |
22 |
100 |
79723 |
22 |
100 |
|
Residential |
3283 |
|
10 |
500 |
|
10 |
277 |
|
6 |
4386 |
|
12 |
5163 |
|
11 |
8446 |
|
11 |
|
Electric Generation |
0 |
0 |
0 |
0 |
0 |
0 |
313 |
1 |
313 |
1 |
313 |
0 |
||||||
|
Industrial |
27772 |
86 |
4328 |
83 |
4239 |
91 |
28285 |
75 |
36852 |
78 |
64624 |
81 |
||||||
|
Commercial-Institutional |
1286 |
4 |
363 |
7 |
156 |
3 |
4519 |
12 |
5038 |
11 |
6324 |
8 |
||||||
|
Industrial Process Losses |
13337 |
9 |
|
0 |
0 |
|
0 |
0 |
|
1509 |
1 |
|
1509 |
1 |
|
14846 |
4 |
|
|
Solid Waste Disposal |
2193 |
1 |
329 |
1 |
214 |
2 |
3325 |
2 |
3868 |
2 |
6061 |
2 |
||||||
|
Transportation |
100035 |
66 |
100 |
19179 |
76 |
100 |
5027 |
51 |
100 |
137280 |
76 |
100 |
161486 |
74 |
100 |
261521 |
71 |
100 |
|
Light-duty gas vehicles |
48796 |
|
49 |
8816 |
|
46 |
2777 |
|
55 |
77714 |
|
57 |
89307 |
|
55 |
138103 |
|
53 |
|
Other |
51239 |
51 |
10363 |
54 |
2250 |
45 |
59566 |
43 |
72179 |
45 |
123418 |
47 |
||||||
|
Miscellaneous |
2858 |
2 |
|
483 |
2 |
|
21 |
0 |
|
2071 |
1 |
|
2575 |
1 |
|
5433 |
1 |
|
|
Total |
150770 |
100 |
25184 |
100 |
9934 |
100 |
181697 |
100 |
216815 |
100 |
367585 |
100 |
||||||
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA population within AQCR c Total urban=emissions from all AQCR’s except Rural AQCR’s |
||||||||||||||||||
TABLE 14–20
1972 NOx Emissions
EPA Region X
|
Air Quality Control Regions (Alaska) |
||||||||||||||||||
|
Source Category |
Rural (1)a |
Total Urbanc |
Total Region |
|||||||||||||||
|
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
Tons |
% |
|||||||
|
Stationary Fuel Combustion |
20,873 |
58 |
100 |
|
|
100 |
|
|
100 |
|
|
100 |
|
|
100 |
20873 |
58 |
100 |
|
Residential |
1095 |
|
5 |
|
1095 |
|
53 |
|||||||||||
|
Electric Generation |
7022 |
34 |
7022 |
34 |
||||||||||||||
|
Industrial |
9513 |
46 |
9513 |
46 |
||||||||||||||
|
Commercial-Institutional |
3243 |
16 |
3243 |
16 |
||||||||||||||
|
Industrial Process Losses |
0 |
0 |
|
0 |
0 |
|
||||||||||||
|
Solid Waste Disposal |
776 |
2 |
776 |
2 |
||||||||||||||
|
Transportation |
14459 |
40 |
100 |
|
|
100 |
|
|
100 |
|
|
100 |
|
|
100 |
14459 |
40 |
100 |
|
Light-duty gas vehicles |
3879 |
|
27 |
|
3879 |
|
27 |
|||||||||||
|
Other |
10580 |
73 |
10580 |
73 |
||||||||||||||
|
Miscellaneous |
0 |
0 |
|
0 |
0 |
|
||||||||||||
|
Total |
36108 |
100 |
|
|
100 |
|
|
100 |
|
|
100 |
|
|
100 |
|
36108 |
100 |
|
|
a Number in parentheses indicates the number of AQCR’s in that category b Largest SMSA population within AQCR c Total urban=emissions from all AQCR’s except Rural AQCR’s |
||||||||||||||||||
-
were transporation related. Surprisingly, this represented 27 percent of the total Region’s emissions. Further, rural industrial fuel combustion emissions equaled urban industrial fuel combustion emissions in this Region.
-
In Region II 81 percent of emissions arose in three large urban AQCR’s. The transportation emissions in just these large urban AQCR’s represented 43 percent of the Region’s total emissions. Stationary fuel combustion in these largest urban AQCR’s generated 27 percent of the Region’s emissions with only 12 percent coming from electric power generation.
-
In Region III the urban emissions were dominated by industrial processing. Forty-six percent of the Region’s emissions arose from industrial processing in a single AQCR, the Metropolitan Philadelphia Interstate AQCR. (This is due, at least in part, to the fact that petroleum refinery emissions have been carefully surveyed in the Philadelphia area, as discussed earlier in this chapter.)
-
In Region I transportation emissions are dominated by light duty gasoline vehicles (72 percent). For all other Regions, light duty gas line vehicles contribute an average of 57 percent of the transportation emissions.
There are some similarities among Regions in the distribution of emissions which reflect the degree of urbanization and industrialization of the Regions:
-
Regions I and IV (New England and the Southeast) had a similar distribution of emissions between rural and urban AQCR’s. The majority of the Regional emissions were produced in medium-sized urban AQCR’s. However, Southern states had a greater proportion of electric power emissions in rural areas.
-
Regions II, III, and V represent major population centers. Sixty-three to 80 percent of the Regions’ emissions were produced in large urban areas.
Effects of Degree of Urbanization on Nationwide Nitrogen Oxide Emissions
Figure 14–14 displays the effects of the degree of urbanization on nationwide NOx emissions in 1972. Urban AQCR’s accounted for 87 percent of the total NOx emissions. Large urban AQCR’s alone accounted for 53 percent of total emissions while rural AQCR’s contributed only 13 percent of total emissions. Thus, the major portion of the NOx problem as characterized by NOx emissions was in urban areas.
Analysis of the 10 AQCR’s with the largest SMSA populations indicates that these AQCR’s accounted for 39 percent of the total U.S. NOx emissions. The emissions from these ten AQCR’s are shown in Table 14–21. The majority of the emissions in these AQCR’s were produced by stationary fuel combustion and industrial process loss sources with transportation sources contributing only 20 percent of the total. This is in marked contrast to the remainder of the country where transportation contributed 45 percent of total emissions.
Nitrogen Oxide Emissions by States
Table 14–22 ranks the states and the District of Columbia according to their total statewide NOx emissions. Table 14–22 also provides another basis for characterizing NOx emissions; it compares emissions per capita and emissions per unit area for each state to the NOx emission density for the contiguous United States. The U.S. NOx emission density is approximately 8.15 tons per square mile and National emissions per capita are approximately 0.112 tons.
The column labeled Ratio Emission Densities (State/U.S.) in Table 14–22 presents a tabulation of each state’s NOx emissions per square mile divided by 8.15 tons per square mile, the nationally averaged emission density. The ratios range from 104 for the District of Columbia to .008 for Alaska. The states of Pennsylvania, Michigan, Indiana, New Jersey, Massachusetts, Maryland and Rhode Island had NOx emission den-
TABLE 14–21
1972 NOx Emissions from 10 Largest Urban Air Quality Control Regionsa
|
AQCRb (Name of Largest City) |
Total Emissions (106 tons/year) |
Stationary Fuel Combustion Emissions (106 tons/year) |
Transportation Emissions (106 tons/year) |
Other Sources |
|
New York-New Jersey-Connecticut Interstate |
1.15 |
0.69 |
0.43 |
0.11 |
|
Metropolitan Los Angeles Intrastate |
1.20 |
0.79 |
0.36 |
0.05 |
|
Metropolitan Chicago Interstate |
1.33 |
1.07 |
0.23 |
0.03 |
|
Metropolitan Philadelphia Interstate |
2.58 |
0.24 |
0.18 |
2.16c |
|
San Francisco Bay Area Intrastate |
0.28 |
0.06 |
0.19 |
0.03 |
|
Metropolitan Detroit-Port Huron Intrastate |
2.01 |
1.85 |
0.15 |
0.01 |
|
Greater Metropolitan Cleveland Intrastate |
0.29 |
0.15 |
0.13 |
0.01 |
|
Metropolitan Boston Intrastate |
0.17 |
0.09 |
0.07 |
0.01 |
|
National Capitol Interstate (Washington, D.C.) |
0.18 |
0.09 |
0.09 |
0 |
|
Metropolitan St. Louis Interstate |
0.43 |
0.31 |
0.11 |
0.01 |
|
TOTAL |
9.62 |
5.26 |
1.94 |
2.42 |
|
aAQCR’s chosen by SMSA population bAQCR’s listed in descending order of SMSA population cDue to industrial process emissions |
||||
TABLE 14–22
NOx Emissions by States and the District of Columbia
|
Rank |
State |
State’s Emissions |
Running % U.S. Total |
Ratio Emission Densities (State/US) |
Ratio Emission Capita (State/US) |
|
1 |
Pennsylvania |
3,326,053 |
13 |
9.1 |
2.5 |
|
2 |
Michigan |
2,449,819 |
22.9 |
5.3 |
2.5 |
|
3 |
California |
1,833,297 |
30.9 |
1.4 |
.82 |
|
4 |
New York |
1,659,019 |
37.6 |
2.7 |
.52 |
|
5 |
Indiana |
1,511,526 |
43.7 |
5.1 |
2.6 |
|
6 |
Texas |
1,427,195 |
49.6 |
.67 |
1.2 |
|
7 |
Ohio |
1,214,163 |
54.5 |
3.6 |
.98 |
|
8 |
Illinois |
1,074,061 |
58.9 |
2.4 |
.87 |
|
9 |
Florida |
710,764 |
61.8 |
1.6 |
.94 |
|
10 |
South Carolina |
574,904 |
64.1 |
2.3 |
1.9 |
|
11 |
New Jersey |
539,268 |
66.3 |
8.8 |
.67 |
|
12 |
Missouri |
494,166 |
68.3 |
.13 |
2.1 |
|
13 |
Tennessee |
470,085 |
70.2 |
1.4 |
1.1 |
|
14 |
Louisiana |
466,076 |
72.1 |
1.3 |
1.1 |
|
15 |
Kentucky |
462,025 |
73.9 |
1.4 |
1.3 |
|
16 |
North Carolina |
454,813 |
75.8 |
.11 |
.79 |
|
17 |
Wisconsin |
450,332 |
77.6 |
1.0 |
.09 |
|
18 |
Alabama |
437,693 |
79.4 |
.03 |
.04 |
|
19 |
Georgia |
407,653 |
81 |
.86 |
.79 |
|
20 |
Massachusetts |
368,590 |
82.5 |
5.8 |
.58 |
|
21 |
Virginia |
363,000 |
84 |
1.1 |
.69 |
|
22 |
Minnesota |
343,738 |
85.4 |
.53 |
.80 |
|
Rank |
State |
State Emissions |
Running % U.S. Total |
Ratio Emission Densities (State/US) |
Ratio Emission Capita (State/US) |
|
23 |
Maryland |
293,337 |
86.6 |
7.2 |
.68 |
|
24 |
Iowa |
267,337 |
87.7 |
.59 |
.85 |
|
25 |
Kansas |
257,926 |
88.7 |
.39 |
1.0 |
|
26 |
West Virginia |
253,088 |
89.8 |
1.3 |
1.3 |
|
27 |
Oklahoma |
245,470 |
90.8 |
.44 |
.86 |
|
28 |
New Mexico |
219,559 |
91.5 |
.22 |
1.9 |
|
29 |
Washington |
207,150 |
92.2 |
.38 |
.54 |
|
30 |
Mississippi |
190,170 |
.49 |
.76 |
|
|
31 |
Arkansas |
186,278 |
|
.03 |
.06 |
|
32 |
Connecticut |
171,775 |
.56 |
.06 |
|
|
33 |
Montana |
163,588 |
.13 |
2.1 |
|
|
34 |
Colorado |
163,586 |
.19 |
.66 |
|
|
35 |
Oregon |
149,637 |
.19 |
.64 |
|
|
36 |
Arizona |
136,544 |
.15 |
.69 |
|
|
37 |
Nebraska |
112,378 |
.18 |
.68 |
|
|
38 |
Nevada |
98,032 |
.11 |
1.8 |
|
|
39 |
North Dakota |
94,477 |
.17 |
1.3 |
|
|
40 |
Utah |
89,285 |
.13 |
.71 |
|
|
41 |
Maine |
84,592 |
.34 |
.76 |
|
|
42 |
Wyoming |
79,997 |
.1 |
2.1 |
|
|
43 |
New Hampshire |
74,195 |
1.0 |
.89 |
|
|
44 |
Delaware |
64,383 |
3.9 |
1.11 |
sities over 40 tons/sq. mile. (It should be noted, however, that the Pennsylvania data include a comprehensive survey of petroleum refinery emissions near Philadelphia. Since such surveys do not exist elsewhere, emissions from states other than Pennsylvania may be greater than shown here.)
A comparison of per capita emissions is given in the column of Table 14–22 labeled Ratio Emission Capita (State/U.S.). The per capita NOx emissions range from 2.6 for Indiana to 0.4 for Alabama. Pennsylvania, Michigan, Indiana, Missouri, Montana and Wyoming had per capita emissions which were two or more times the National average of 225 pounds per person per year. Montana and Wyoming contributed less than 1 percent each to the total U.S. emissions but had a high per capita emission ratio because of their small population.
Table 14–22 also compiles the running accumulating percent contribution of state’s emissions to the total U.S. NOx emissions (24.64 million tons/year). Interestingly, six states (Pennsylvania, Michigan, California, New York, Indiana and Texas) were responsible for about 50 percent of the nationwide NOx emissions. Pennsylvania, itself, contributed over 13 percent. Over 90 percent of the annual NOx emissions are accounted for in the first 27 states listed, with the remaining 23 states and the District of Columbia combining to contribute less than 10 percent.
Nationwide NOx Emission Trends
1940–1972
Nationwide NOx emission trends from 1940–1972 are displayed in Table 14–23 and Figure 14–15 (USEPA 1973a). Total nationwide emissions tripled in the three decades from 1940–1972. The stationary source fuel combustion category revealed the largest emission growth increasing from 3.5 million tons in 1940 to 12.3 million tons in 1972. In this category, emissions from electric power generation showed the greatest increase, climbing nearly 800 percent over the three decades. Transportation emissions tripled during this period.
TABLE 14–23
Nationwide NOx Emissions, 1940–1970 (EPA 1973b)
|
|
NOx Emissions (106 tons/year) |
|||||
|
Source Category |
1940 |
1950 |
1960 |
1968 |
1969 |
1970 |
|
Stationary Fuel Combustion |
3.5 |
4.3 |
5.2 |
9.7 |
10.2 |
10.0 |
|
Electric Generation |
0.6 |
1.2 |
2.3 |
4.2 |
4.3 |
4.7 |
|
Industrial |
1.9 |
2.0 |
1.8 |
3.7 |
4.6 |
4.5 |
|
Commercial-Institutional |
0.1 |
0.1 |
0.2 |
1.0 |
0.4 |
0.2 |
|
Residential |
0.9 |
1.0 |
0.9 |
0.8 |
0.9 |
0.6 |
|
Industrial Process Losses |
<0.1 |
0.1 |
0.1 |
0.2 |
0.2 |
0.2 |
|
Solid Waste Disposal |
0.1 |
0.2 |
0.2 |
0.4 |
0.4 |
0.4 |
|
Transportation |
1.7 |
2.9 |
4.3 |
7.5 |
8.3 |
8.8 |
|
Road Vehicles |
1.4 |
2.2 |
3.5 |
5.5 |
5.8 |
6.2 |
|
Gasoline |
1.4 |
2.15 |
3.2 |
4.8 |
5.1 |
5.4 |
|
Diesel |
<0.1 |
0.05 |
0.3 |
0.7 |
0.7 |
0.8 |
|
Other |
0.3 |
0.7 |
0.8 |
2.0 |
2.5 |
2.6 |
|
Miscellaneous |
0.8 |
0.4 |
0.2 |
0.2 |
0.2 |
0.1 |
|
Total |
6.1 |
7.9 |
10.0 |
18.2 |
19.3 |
19.5 |
Total Emissions: 1972–1990
In Appendix 14-C, total emissions of nitrogen oxide are projected to the year 1990 under various assumptions about the implementation of regulatory programs and about the growth rate of stationary and mobile sources.
Assuming the implementation of the present statutory program for NOx control; using emission projections for electric power generation based on forecasts by the Federal Energy Administration (USEPA 1975) and the National Academy of Engineering (1974); assuming a slowing of the growth rate of NOx emissions from industrial, commercial, and institutional fuel combustion; and using other assumptions presented in detail in Appendix 14-C, a conservative estimate yields a 36 percent increase in total NOx emissions from 1972–1990 (see Figure 14–16). The largest increases will result from industrial process losses, non-highway transportation, industrial fuel combustion, and electric power generation. Highway-related transportation emissions will decline under the assumptions of this projection as Federal controls on nitrogen oxide emissions from automobiles are implemented.
Appendix 14-C examines future NOx emissions from electric power generation and highway vehicles in some detail. Protections of power generation emissions are critically dependent upon the rate at which nuclear power generation increases. If for any reason no new nuclear power plants were built after 1975, NOx emissions from power generation facilities could more than double. Road vehicle emissions are projected under three options: (1) no control of NOx emissions, (2) implementation of a proposed five year delay of the 1977 statutory standards and (3) implementation of the present statutory program. The analysis shows a 25 percent reduction in road vehicle NOx emissions by 1990 (compared with 1972) under the present statutory program (if vehicle miles of travel increase by 4 percent per year). The five year delay leads to a 17 percent reduction in emissions by 1990. Emission reductions are somewhat smaller than might be expected from the
LITERATURE CITED
Altschuller, A.P. (1958) Tellus 10, 479.
Bates, D.R. and P.B.Hays (1967) Planetary Space Sci., 15, 189.
Berdyer, Kh.B, N.V.Pavlovich, and A.A. Tuzhilina (1967) Effect of Motor Vehicle Exhaust Gases on Atmospheric Pollution in Dwellings and in a Main Street, Hyg. and Sanitation (Moscow) 32:424–426, Ap-Jn.
Coffey, P.E. and W.N.Stasiuk (1975) Evidence of Atmospheric Transport of Ozone into Urban Areas, Env. Sci., and Tech., 9(1).
Commonwealth of Massachusetts (1974) Ozone Data, 1973.
Cortese, A.D. (1975) Harvard School of Public Health, Unpublished NOx Emission Estimates, January.
Council on Environmental Quality (1975) The Fifth Annual Report, 274.
Hamilton, H.L. et al. (1968) Am. Atmospheric Physics and Chemistry Study on Pikes Peak in Support of Pulmonary Edema Research RTI for Army Research Office, Research Triangle Park, N.C.
Junge, C.E. (1956) Recent Investigations in Air Chemistry, Tellus 8:127–139, May.
Junge, C.E. (1963) Air Chemistry and Radioactivity, Academic Press, New York.
Lodge, J.P., Jr. and J.B.Pate (1960) Science, 153, 408.
Lodge, J.P., Jr. and J.B.Pate (1966) Atmospheric Gases and Particulate in Panama, Science 153:408–410, July 22.
Mason, H.B. and A.B.Shimizu (1974) Briefing Document for the Maximum Stationary Source Technology Systems Program for NOx Control, EPA Contract No. 68–02–1318.
Miura, T., K.Kimura, K.Kimotsuki, M.O.Kusa, O.Tada, and T.Sawano (1975) Comparison of the Concentration of Suspended Particulate Matter and Gaseous Pollutants Between Indoor Air and Outdoor Air in Urban Areas, J. Sci. Labour (Tokyo) 41(10) 493–500.
National Academy of Engineering (1974) U.S. Energy Prospects: An Engineering Viewpoint, Task Force on Energy, October.
National Academy of Sciences (1974) The Relationship of Emission to Ambient Air Quality, Ser. NO. 93–24, Vol. 3, P. 65.
Perry, J.H., C.H.Chilton, and S.D.Kirkpatrick (1963) Chemical Engineer’s Handbook, 4th edition, McGraw-Hill, New York.
Peterson, W.H. (1956) Production of Toxic Gas (NO2) from Silage, presented at 130th National Meeting, Am. Chem. Soc., Atlantic City, N.J.
Ripperton, L.A. et al. (1970) J. Air Poll. Cont. Assoc. 20(9) 589–592.
Ripperton, L.A., L.Kornreich, and J.J.B.Worth (1970) Nitrogen Dioxide and Nitric Oxide in Non-Urban Air, J. Air Poll. Cont. Assoc. 20, 589–592, September.
Robison, E. and R.C.Robbins (1968) Sources, Abundance, and Fate of Gaseous Atmospheric Pollutants, Stanford Research Institute, Final Report.
Schuck, E.A., J.N.Pitts, Jr., and J.K.S.Wan (1966) Relationships Between Certain Meteorological Factors and Photochemical Smog, Air and Water Pollution 10:689–711.
Strauss, Werner (1972) Air Pollution Control, Wiley.
System Development Corporation (1970) Comprehensive Technical Report on All Atmospheric Contaminants Associated with Photochemical Air Pollution. Santa Monica, California, Report No. TM (L) 4411/002–01, June.
U.S. Department of Commerce (1967) The Automobile and Air Pollution, A Program for Progress, Part II, Washington, D.C., December.
U.S. Environmental Protection Agency (1971) Air Quality Criteria for Nitrogen Oxides, AP-84.
U.S. Environmental Protection Agency (1972a) Federal Air Quality Control Regions, AP-102, Rockville, Maryland, January.
U.S. Environmental Protection Agency (1972b) Guide for Compiling a Comprehensive Emission Inventory, APTD-1135, Research Triangle Park, North Carolina, June.
U.S. Environmental Protection Agency (1973) Indoor-Outdoor Air Pollution Relationships, Vol. II, AP-1126.
U.S. Environmental Protection Agency (1973a) Nationwide Air Pollutant Trends: 1940–1970, AP-115, Research Triangle Park, North Carolina, January.
U.S. Environmental Protection Agency (1973b) Compilation of Air Pollutant Emission Factors, AP-42, Research Triangle Park, North Carolina, April.
U.S. Environmental Protection Agency (1974) The National Air Monitoring Program, Air Quality and Emission Trends (1975) Vol. 1, Table 1–5.
U.S. Environmental Protection Agency (1974a) 1972 National Emissions Report, NEDS, AEROS. Report Number EPA-450/2–74–012, Research Triangle Park, North Carolina, June.
U.S. Environmental Protection Agency (1974b) Compilation of Air Pollutant Emission Factors, AP-42, Draft Supplement No. 5, Research Triangle Park, North Carolina, November.
U.S. Environmental Protection Agency (1975) Federal Energy Administration Electric Generation Projections, personal communication with James Speyer, Office of Planning and Management, January.
Wade, W.A., III, W.A.Cote, and J.E.Yocon (1974) A Study of Indoor Air Quality, presented at the 68th Annual Air Poll. Control Assoc. Conf., Denver, June.
APPENDIX 14-A
ATMOSPHERIC REACTIONS
Nitrogen oxides play a principal role in the atmospheric reactions which produce photochemical smog. The complex reactions and the increased toxic potential associated with these secondary pollutants complicate the evaluation of control strategies significantly.
Dr. J.G.Calvert (1973) and Dr. E.R.Stephens (1973) have discussed the atmospheric reactions in detail in the Proceeding of the Conference on Health Effects of Air Pollution of the National Academy of Sciences.
PRODUCTION OF NITROGEN OXIDES
Nitrogen oxides are produced during combustion by the oxidation of organic nitrogen compounds in fossil fuels and the thermal fixation of atmospheric nitrogen gas, N2.
Nitrogen gas (N2) is generally inert with respect to tropospheric reactions. The equilibrium of the reaction
(1)
is far to the left at normal atmospheric temperatures and NO concentrations are small. However at temperatures found in combustion chambers this equilibrium shifts to the right. Rapid cooling of the resultant N2, O2, and NO mixture quenches the reverse reaction, 2NO N2+O2. The return to equilibrium proceeds very slowly and the NO persists. Although NO2 may be an important intermediate during combustion, the
amount released is small. For temperatures greater than 2000 F (1093 C), only 0.5 percent of the nitrogen oxides emitted are NO2 (EPA 1970)
NO TO NO2 CONVERSIONS
The levels of NO, NO2 and O3 as a function of time for typical diurnal smoggy conditions were described in detail in Chapter 14 in the discussion of diurnal NOx concentrations. Nitric oxide emissions increase during the morning rush hour. However NO concentrations begin to fall rapidly as the NO2 concentrations climb toward their maximum. Ozone levels do not begin to increase until most of the NO has disappeared, reaching a maximum at midday and falling off in the afternoon. There is no peak in NO concentrations in the afternoon corresponding to the evening rush hour.
These observations imply that NO is rapidly being converted to NO2 in the atmosphere. Recent observations tracking power plant plumes have shown similar processes occurring (Davis et al. 1975). The time required for 50 percent conversion of the NO to NO2 in these plumes was estimated to be between 12 and 60 minutes.
During initial dilution in the atmosphere NO2 is produced by the reaction
(2)
Because this is a three body reaction requiring two NO molecules, the rate of production is proportional to the square of the nitric oxide concentration. Therefore, nitrogen dioxide production by this mechanism is only important for high NO concentrations (greater than about 100 ppm) and significant O2 concentrations. As the nitric oxide is diluted to concentrations below 1 ppm, these direct reactions with oxygen become unimportant (EPA 1970).
These conversions during the initial dilution can produce nitrogen dioxide equal to up to 25 percent of the total NOx (Calvert 1973). However, this reaction cannot account for the conversion of NO to NO2 in the atmosphere at ambient levels of NO which are typically less than 1 ppm.
Ozone (O3) oxidation of the NO does provide an explanation; nitric oxide is rapidly oxidized in the atmosphere by ozone,
(3)
Ozone is naturally found in the atmosphere in concentrations of 10 to 50 ppb (Com et al. 1975). The 50 percent conversion time for this reaction is 0.3 minutes for nitric oxide and ozone concentrations of 100 ppb (Stephens 1973). Therefore, in the presence of excess atmospheric ozone this reaction appears to be the primary link in NO to NO2 conversion.
The brown gas, nitrogen dioxide, is a strong absorber of sunlight, and is rapidly photolyzed by the ultraviolet radiation to produce the reactive ground state oxygen atom, O(3P), and an NO molecule (EPA 1970):
(4)
The highly reactive O(3P) reacts with an oxygen molecule and a third molecule, M, which removes excess vibrational energy, to form ozone:
(5)
The net effect of these two reactions is the reverse of the equation (3) above. Thus we have a dynamic balance between two very fast (2 to 3 minutes) reactions (Stephens 1973):
(6)
This mechanism does not yet explain the net conversion of NO to NO2 as observed in the atmosphere. Moreover, this reaction sequence does not explain the conversion of NO2 to ozone.
OXIDANT PRODUCTION
The key reactions in shifting this equilibrium to convert NO to NO2 are chains involving various transient molecules and reactive species such as carbon monoxide and hydrocarbons found in polluted air.
Carbon monoxide reacts with the hydroxyl radical HO through a chain of reactions to drive the equilibrium to the left:
(7)
(8)
(9)
(7)
Each hydroxyl radical produced by this method is able to recycle back and oxidize many NO molecules to NO2.
In the presence of ozone, the HO radical is produced by the reaction sequence (Levy 1971):
(10)
(11)
where O(1D) is the highly reactive ground state oxygen atom. Only 1 percent of the O(1D) atoms can be expected to react with water. The remainder are rapidly quenched by other molecules
(12)
The HO production rate is then linearly dependent on the absolute H2O concentration (Davis 1974).
Hydrocarbon reactions are major sources of the hydroperoxyl radical, HO2, as in the sequence (Calvert 1973):
(13)
The HO2 radical can then oxidize NO to NO2 also produce a hydroxyl radical:
(9)
The reaction chains of hydrocarbon radicals, R, are very complex, but the key reactions convert NO to NO2 analagously to reactions (8) and (9) (Stephens 1973):
(14)
(15)
The net effect of these radical reaction chains is to promote the conversion of NO to NO2:
(16)
This drives the dynamic equilibrium between NO and NO2 towards NO2 production:
(17)
By scavenging NO, these reactions permit O3 to accumulate (Stephens 1973).
REMOVAL PROCESSES
Ozone is lost by reacting with nitrogen dioxide:
(18)
The transient NO3 reacts with nitrogen dioxide:
(19)
N2O5 is also a transient molecule which rapidly dissociates or reacts with water to form nitric acid, HNO3:
(20)
(21)
In addition, nitric acid is formed by the OH removal process:
(22)
Davis et al. (1974) have calculated that this reaction for power plant plumes in summer daylight conditions should have a 50 percent conversion time of 2 to 3 hours. The nitric acid
formed presumably reacts with various other contaminants in the air to form nitrates. Ammonium for example, will react rapidly with nitric acid to form ammonium nitrate. In fact, composite samples in the Los Angeles basin have shown ammonium nitrate to comprise 10–15 percent of the total suspended particulates (Gordon and Bryan 1973).
The net effect of these reactions is to remove nitrogen dioxide from the air by loss reactions with the important reactive molecules to produce nitrate aerosols.
The other important loss mechanism for nitrogen dioxide is through the production of the peroxyl acyl nitrates, PAN’s. The PAN’s are a family of organic compounds with the general formula:
The PAN’s have a temporal variation which is very similar to that of ozone. Stephens (1973) suggests that PAN formation is delayed because NO and NO2 compete for the peroxyl acyl radicals:
(23)
(24)
The first reaction, (23), converts NO to NO2 and claims the major portion of the peroxyl acyl radical. As soon as the NO has been depleted however, the second reaction becomes important and PAN’s are formed. This is very similar to the reaction scheme for ozone.
LITERATURE CITED
Calvert, J.G. (1973) Interactions of air pollutants. Proceed. of the conf. on health effects of air pollution. National Academy of Sciences, 19–101. October 3–5.
Corn, M., R.W.Dunlap, L.A.Goldmuntz, and L.H.Rogers (1975) Photochemical oxidants: sources, sinks, and strategies. J. Air Poll. Con. Assoc., 25(1) 16–18.
Davis, D.D., G.Smith, and G.Klauber (1974) Trace gas analysis of power plant plumes via aircraft measurement: O3, NOx, and SO2 chemistry. Science 186:733–736.
EPA (1970) Air quality criteria for nitrogen oxides. AP-84.
Gordon, R.J. and R.J.Bryan (1973) Ammonium nitrate in airborne particles in Los Angeles. Environmental Science & Technology 7(7):645–647. July.
Levy, H. (1971) Normal atmosphere: large radical and formaldehyde concentrations predicted. Science 173 (3992):141–143. July 9.
Stephens, E.R. (1973) Photochemical formation of oxidants. Proc. of conf. on health effects of air poll. 465–487. October 3–5.
APPENDIX 14-B
MEASUREMENT OF AMBIENT LEVELS OF NO2
Ambient air concentrations of NO2 are not well known because of problems with the Federal Reference method of measurement which was employed prior to 1972. The reference method used to determine compliance with national air quality standards was the Jacobs-Hochheiser (J-H) method as modified for the National Air Surveillance Networks (NASN). This method, along with other NO2 measurement techniques, is summarized in Table App. 14-B1 (Saltzman 1973).
Jacobs-Hochheiser Method
Briefly, the Jacobs-Hochheiser (J-H) method involves bubbling ambient air through an aqueous solution for 24 hours. The nitrate ions formed in the reagent are treated to form an azo dye which is measured colorimetrically. The intensity of the absorbed light at a specific wave length is related to the integrated 24-hour NO2 concentration. The collection efficiency of the reference method for NO2 was previously determined as 35 percent, and a correction for this efficiency had been applied to all data.
In 1971–72 it became apparent that the J-H method has two inherent problems which affected its ability to accurately estimate air quality in the field. These are: (1) the collection efficiency varies with concentration of NO2 and (2) a small portion of the nitric oxide (NO) in the ambient air shows up as NO2. The collection efficiency of the J-H method at NO2 concentrations in excess of 110 μg/m3 has been known for some time and calibration curves developed. These
TABLE App. 14-B1
Manual Methods Using Alkaline Absorbing Reagents
|
Name |
Reference |
Volume and Composition of Absorbing Reagent |
Absorbing Apparatus |
Sampling Rate and Time |
Procedure for Analysis |
Remarks |
|
Jacobs and Hochheiser (J-H) |
2 |
30–35 ml: 0.1N NaOH, 0.2% v/v butyl alcohol |
Coarse Fritted glass bubbler in sequential sampler |
1.3 lpm, 40 min. |
Add 1 drop 1% H2O2, 10 ml of 2% sulfanilamide in 5% v/v H3PO4; 1 ml of 0.1% N-(1-naphthyl)-ethylene diamine dehydrochloride |
No validation or stoichiometry studies. Data shows inverse relationship between SO2 and NO2 analyses, suggesting interference. 90% collected in first of two bubblers. |
|
Perry and Tabor (NASN) |
15, 16 |
20 ml: Same as J-H |
Polypropylene test tube containing glass inlet tube with orifice 0.36 ±0.05 mm. THERMOSTATED AT 35°C. |
0.15–0.20 lpm, 24 hrs. |
Autoanalyzer, using J-H reagents. |
Absorption in first of 4 bubblers varied from 30–69%. Collected samples stable for 3 weeks. Assumed 50% overall conversion. Large mass of data available using this method. |
|
Meadows and Stalker (Alabama Study) |
14 |
35 ml: 0.1N NaOH (in each of 2 bubblers) |
Two NASN bubblers in series. |
0.35–0.50 lpm, 24 hrs. |
Same as NASN. |
Absorption in 1st of 6 bubblers was 19%, in 1st 2 was 30%. Overall conversion of 36% used, with assumption of 100% stoichiometry. |
|
Federal Reference Method |
1 |
50 ml: 0.1N NaOH |
Membrane filter leading to polypropylene test tube with fritted glass inlet of porosity B (70–100 μm max. pore diameter). |
0.20 lpm, 24 hrs. |
Similar to J-H. |
Overall conversion of 35% is used. |
|
Name |
Reference |
Volume and Composition of Absorbing Reagent |
Absorbing Apparatus |
Sampling Rate and Time |
Procedure for Analysis |
Remarks |
|
Christie, et al. |
17 |
5 ml: 0.025 N NaOH, 0.1% NaAsO2, 0.75% sulfanilic acid |
Glass bubbler with open tube inlet. |
0.12 lpm, 1 min. |
Add 3 ml of 0.02% N-(1 naphthyl)-ethylene diamine dihydrochloride, 6% oxalic acid. |
Overall conversion 94% at 6 ppm NO2. For industrial hygiene use. |
|
Nash |
18 |
4 ml: 0.1N NaOH, 0.05% guaiacol (2-methoxy phenol) |
Glass Arnold Tube (with open tube inlet) |
0.6 lpm, 2 hrs. |
Add 4 ml of 0.2N HCl 0.5% sulfanilic acid, 1.5% glycine, 0.002% N-(1-naphthyl)-ethylene diamine dihydrochloride |
Limited amount of H2O2 may be added to reduce SO2 interference. Stoichiometric factor 0.73, absorption efficiency 99%. |
|
Huygen and Steerman |
19 |
15 ml: 0.1N NaOH, 0.02% R-salt (2-naphthol-3, 6-disulfonic acid disodium salt), 0.1% triethanolamine (in each of two bubblers). |
Two bubblers in series, each with coarse glass frit (90–150 μm) of 6.5 cm2 area. |
1 lpm, 24 hrs. |
Add to each absorber 15 ml of 0.4% sulfanilamide, 0.03% Cleves acid (8-aminonaphthl-2-sulfonic acid), 0.12N HCl. A drop of H2O2 may be added first to eliminate SO2 interference. |
Stoichiometric factor varied from 96% to 80% in range 48–2200 μg/m3 NO2. Absorption efficiency averaged 96%. |
|
EPA Tentative Candidate Arsenite Method |
6 |
50 ml: 0.1N NaOH, 0.1% sodium arsenite |
Polypropylene tube containing glass inlet tube with orifice 0.6±0.2 mm i.d., 6 mm from bottom of tube. |
0.2 lpm, 24 hrs. |
To 10 ml of absorbing solution, add 1 ml 0.024% H2O2; 10 ml of 2% sulfanilamide, 5% v/v H3PO4; and 1.4 ml of N-(1-naphthyl) -ethylene diamine dihydrochloride. |
Overall conversion is 85% over range 50–750 μg/m3 NO2 concentration. |
|
Levaggi, Siu, and Feldstein (triethanolamine) |
20 |
50 ml: 1.5% triothanolamine, 0.3% v/v n-butanol. |
Polypropylene tube with fritted glass inlet of 70–100 μm max. pore diameter. |
0.15–0.20 lpm, 24 hrs. |
Similar to J-H. |
Overall conversion efficiency 76–92% over range 56–750 μg/m3 NO2, collection efficiency 95–99%. Constant 0.85 factor recommended. |
show an approximate 35 percent collection efficiency at NO2 concentrations of 110 to 180 μg/m3. This means that about one-third of the NO2 in the air that bubbles through the sampler is trapped and bubbles through the sampler is trapped and analyzed. Late in 1971 lab methods became available to generate reliable low level NO2 concentrations and it was possible to extend the calibration curve for NO2 concentrations below 100 μg/m3. The extended curve showed that collection efficiencies greatly increased as NO2 concentrations decreased below 100 μg/m3. For example, at 100 μg/m3 the collection efficiency is just below 40 percent, whereas at 30 μg/m3 the collection efficiency is 60 percent. This variable collection efficiency may have been a major source of error in the observations used for classifying regions for NO2.
The J-H method as published in the Federal Register uses a constant average correction factor on a collection efficiency of 35 percent based on the early efficiency curves. It is clear (see Figure App. 14-B1) that a 35 percent collection efficiency is too low for concentrations in the 30 to 60 μg/m3 range. Therefore, at lower concentrations where collection efficiencies are much higher than 35 percent, the calculated and reported concentrations of NO2 obtained by the J-H method are higher than actual ambient NO2 concentrations. Since the NO2 concentration in the air varies throughout the day, the collection efficiency may vary significantly throughout any one 24-hour sample, making accurate calculation of NO2 concentrations impossible.
The NO interference previously mentioned is a second factor leading to an additive effect and is particularly important in areas of low NO2 concentrations. Table B2, The Effect of NO on the Reference Method for NO2, displays this additive interference of NO (Hauser and Shy 1972).
Because of the questionable data base, EPA reviewed all 47 AQCR’s in 29 states that were classified as Priority I for NO2 control and thus required stationary source controls. In July 1973, 43 of the original Priority I AQCR’s were reclassified as Priority III for NO2. The Los Angeles and Chicago AQCR’s are the only ones
TABLE App. 14-B2
Effect of NO on the Reference Method for NO2
|
μg/m3 |
Ratio NO/NO: |
Expected NO: recovered, % |
Apparent NO: recovered, % |
|
|
NO2 |
NO |
|||
|
100 |
0 |
0.0 |
39 |
38 |
|
102 |
63 |
0.6 |
39 |
38 |
|
105 |
127 |
1.2 |
38 |
52 |
|
122 |
627 |
5.1 |
36 |
57 |
|
189 |
0 |
0.0 |
29 |
29 |
|
244 |
1205 |
4.9 |
24 |
45 |
|
248 |
1279 |
5.2 |
23 |
55 |
|
215 |
1242 |
5.8 |
26 |
50 |
|
311 |
0 |
0.0 |
20 |
17 |
|
316 |
111 |
0.4 |
20 |
30 |
|
318 |
332 |
1.1 |
20 |
33 |
|
356 |
1060 |
3.0 |
18 |
44 |
where the data indicated NO2 concentrations exceeding 110 μg/m3 annual average. Both of these AQCR’s remain Priority I. In the New York-New Jersey-Connecticut, Wasatch Front (Salt Lake City) and Denver AQCR’s (originally Priority III) arsenite data shows concentrations below the cutoff point for a Priority III classification, but chemiluminescence and/or Saltzman data show concentrations above it.
For more complete details of comparative sampling data performed in the original 47 Priority I AQCR’s, refer to Federal Register 38 FR 15181 (Federal Register 1973).
Proposed Reference Methods
Three methods of ambient NO2 monitoring have been proposed as the Federal Reference method: the modified sodium arsenite method, the continuous Saltzman method and the chemiluminescence detection method.
Each method has its own inherent limitations of either interferences, stability, complexity, toxicity, or cost. Each of the methods is briefly described.
Arsenite Method
Figure App. 14-B2 (Federal Register 1973) displays the apparatus used to determine NO2 (24-hour sample) by the Arsenite Method. Air is bubbled through a reagent of sodium hydroxide-sodium arsenite solution to form a stable solution of sodium nitrite (Christie et al. 1970). The nitrite ion produced during sampling is reacted with phosphoric acid, sulfanilamide, and N-1-naphthylethylenediamine dihydrochloride to form an azo dye. NO2 concentrations are infered by measuring the azo dye colorimetrically.
The method has an overall conversion efficiency of 85 percent over the range of 50–750 μg/m3 NO2. However, it has been reported that there is a 12 percent positive interference for NO which can be deducted (Saltzman 1973). The toxicity of the absorbing reagent is an obvious disadvantage of this method.
Continuous Saltzman Method
The principles of this method have been used for ambient air sampling for over 14 years (Saltzman 1960, Lyshkow 1965). By contacting an air flow with a liquid diazotizing-coupling reagent, the 
(nitrite ions) form a deeply colored azo dye which is measured colorimetrically. The absorbance of the azo dye is directly proportional to the number of NO2 molecules in the air stream that are converted to 
.
This continuous method is applicable for concentrations from 18.8 μg/m3 to 1880 μg/m3 (.01–1 ppm). Apparently there is some interference from ozone. This is a negative interference reported as 5.5 percent for NO3/NO2=1, 19 percent for O3/NO2=2 and 32 percent for O3/NO2=3 (Federal Register 1973).
This method has the drawback of requiring considerable skill and effort for calibration and maintenance, especially replacement of the liquid reagent.
Chemiluminescence Method
The chemiluminescence instrument represents a major break with absorbing wet chemistry methods of ambient air sampling. Figure App. 14-B3 schematically illustrates the NOx chemiluminescent analyzer (Federal Register 1973). In general, the analyzer draws an air sample into a light-tight reaction chamber and mixed with ozonated air. The chemiluminescence produced by the reaction of nitric oxide is measured by a photomultiplier tube, and the data is stored in a peak-holding amplifier.
The first generation chemiluminescence instruments alternately sampled NO and total NOx. The total NOx was determined by converting all NO2 to NO and measuring as described above. The difference between total NOx and NO was taken as NO2. This 1-minute cycling of samples often gave negative NO2 values when NO concentrations were high and rapidly varying, such as near traffic. The latest analyzers operate
with dual chambers or a rapidly cycling single chamber. This improvement has apparently removed spurious negative values.
Since the instrument measures the light emissions of individual molecules, its lower detectable values are dependent on photomultiplier sensitivity. The lower limit of this method is approximately 9.4 μg/m3 (0.005 ppm).
The direct measurement of NO has no interferences. However, the conversion of NOx to NO by the converter can decompose unstable nitrogen compounds to give a falsely high NOx reading. Fortunately interference is small (Federal Register 1973).
Selection of a New Federal Reference Method
An extensive field study is now being conducted by EPA to determine the proper priority classifications of numerous air quality control regions. The arithmetic average results for 42 regions by simultaneous analysis with the Tentative Candidate Chemiluminescent Method and the Tentative Candidate Arsenite Method have been tabulated in 38 FR 15174. Figure App. 14-B4 shows a plot derived from these data by Saltzman. Even though the methods were selected as best, after much testing, there is a disturbing scatter in the comparison plot. The straight line fit by the method of least squares has the equation:
The diffuculty and expense of validating analytical methodology has been grossly underestimated in the past.
Calibration Techniques
Standardized calibration techniques are essential to all ambient air quality monitoring. Precise procedures for calibration of each of the tentative NO2 methods are given in 38 FR 15174. There are two techniques described:
NO2 permeation tube and dilution, and NO2 gas-phase titration.
The latter, gas-phase titration, is diagrammed in Figure App. 14-B5 (Federal Register 1973). This apparatus can calibrate chemiluminescent NO, NO2, and O3 instruments.
LITERATURE CITED
Christie, A.A., R.G.Lidzey, and D.W.F.Radford (1970) Analyst 95, 519.
Federal Register (1973) 38(110), 15174–15191 June 8.
Hauser, T.R., and C.M.Shy (1972) Position paper: NOx measurement env. res. and tech. 6(10), October.
Lyshkow, N.A. (1965) A rapid sensitive colorimetric reagent for nitrogen dioxide in air. J. Air Poll. Cont. Assoc. 15(10):481.
Saltzman, B.E. (1973) Analytical methodologies for nitrogen oxides in perspective. Proceed of the conf. on health effects of air pollution. National Academy of Sciences, Washington, D.C. October 3–5.
Saltzman, B.E. (1960) Modified nitrogen dioxide reagent for recording air analyses. Anal. Chem. 32, 135.
APPENDIX 14-C
PROJECTIONS OF FUTURE NOx EMISSIONS
TOTAL EMISSIONS: 1972–1990
Total emissions of nitrogen oxides projected to 1990 assuming implementation of the present statutory program for NOx control are displayed in Table App. 14-C1 and Figure 14–16 in Chapter 14. This estimate projects a 36 percent increase in total emissions from 1972–1990. The largest increases in emissions will be produced by industrial process losses (98 percent), non-highway transportation (72 percent), industrial fuel combustion (54 percent) and electric power generation (49 percent). Highway-related transportation emissions will decline by 24 percent during this period assuming the statutory emission standards for automobiles are achieved. The estimate is considered to be conservative since it assumes that (1) most new electric power generation will be produced by nuclear reactors, (2) the statutory automotive emission standards will remain in effect and be achieved and (3) the 1940–1970 annual growth rate of NOx emissions from industrial, commercial and institutional fuel combustion will slow during the next two decades.
Emission projections for electric power generation are based on recent electric generation forecasts by the Federal Energy Administration (USEPA 1975) and the National Academy of Engineering (NAE 1974). These projections are discussed in greater detail below.
Industrial fuel combustion emissions of nitrogen oxides grew at the rate of 9.5 percent per year during the 1960–1972 period due to a
TABLE App. 14-C1
Nationwide NOx Emissions Projected to 1990 Assuming the Present Statutory Program
|
|
NOx Emissions (106 tons/year) |
|||
|
Source Category |
1972 |
1980 |
1985 |
1990 |
|
Stationary Fuel Combustion |
12.27 |
15.96 |
16.82 |
18.46 |
|
Electric Generation |
5.94 |
8.16 |
8.20 |
8.88 |
|
Industrial |
5.39 |
6.73 |
7.46 |
8.31 |
|
Commercial-Institutional |
0.65 |
0.76 |
0.84 |
0.93 |
|
Residential |
0.29 |
0.31 |
0.32 |
0.34 |
|
Industrial Process Losses |
2.88 |
3.91 |
4.72 |
5.71 |
|
Solid Waste Disposal |
0.18 |
0.22 |
0.25 |
0.28 |
|
Transportationa |
8.45 |
8.47 |
7.49 |
7.60 |
|
Road Vehicles |
7.48 |
7.14 |
5.89 |
5.68 |
|
Gasoline |
6.59 |
5.97 |
4.30 |
3.95 |
|
Diesel |
0.89 |
1.17 |
1.59 |
1.73 |
|
Other |
0.97 |
1.33 |
1.60 |
1.92 |
|
Miscellaneousb |
0.59 |
0.74 |
0.87 |
1.02 |
|
TOTAL |
24.37c |
29.30 |
30.15 |
33.07 |
|
aAssumes a 4% annual VMT growth rate bIncludes New York City Point sources assumed to grow at 4% per year c1972 total emissions are 270,000 tons lower than those summarized in Table 2.2 due to the use of new road vehicle emission factors |
||||
large increase in natural gas and industrial process gas combustion. In this projection it is assumed that the growth rate will slow to approximately 2.5 percent per year through 1990 due (1) to a gradual shift to purchase of electricity from utility companies, and (2) to an assumed 3 percent annual growth rate in use of natural and industrial process gas, and in the use of coal. A 3 percent growth rate for industrial process gas use follows the growth rate for steel and petroleum production. No increase in oil consumption is assumed after 1975. Since there are no federal new source performance standards and few state regulations for NOx emissions from medium-sized and small boilers, all emissions are uncontrolled. Because historical trends show a much larger growth rate than as sumed here, these projections are considered conservative.
NOx emissions from commercial and institutional fuel combustion have grown at the rate of 4.5 percent per year over the last three decades. In this projection it is assumed that the growth rate will slow to approximately 2 percent per year through 1990 due to (1) a gradual shift to the purchase of power for space heating and electricity from utility companies, and (2) energy conservation measures. No controls were assumed for industrial fuel combustion sources. These projections are also quite conservative since they depart significantly from historical trends.
Residential fuel combustion emissions of NOx have declined at the rate of 2 percent per year over the last three decades. This is primarily due to a shift in fuel usage from coal to distillate oil and natural gas and to the purchase of power for space heating from electric utilities. In this projection it is assumed that residential emissions will grow 1 percent per year through 1990 since: (1) the benefits of a coal to oil or natural gas shift will all have been realized by 1975; (2) the number of housing units increase at a rate of 2 percent per year; and (3) approximately half of the new housing units will be heated by electricity.
Lack of knowledge of NOx emission factors from industrial process losses in the past make
historical trends for these sources difficult to interpret. It is assumed that refinery emissions which amount to 86 percent of all industrial process loss emissions, will increase at the rate of 4 percent per year, which is the growth rate for petroleum usage. Since there are no federal new source performance standards for NOx from refineries it is assumed that these sources will be uncontrolled. Nitric acid manufacturing emissions of NOx are assumed to grow at 2.6 percent per year which reflects the mix of old and new sources and the 90 percent control of emissions from new sources required by federal new source performance standards. These emission estimates are conservative since baseline 1972 petroleum refinery emissions are probably underestimated.
NOx emissions from solid waste disposal grew at a rate of 4 percent per year from 1940–1970 due to open burning and incineration. State implementation plans required by the Clean Air Act will eliminate most of the open burning emissions so that it is more realistic to assume only a 1 percent annual growth rate of such emissions through 1990. The scarcity of land for sanitary landfill, and the increase in per capita solid waste production will probably result in at least a 5 percent annual growth rate in incineration emissions. Overall, it is assumed that solid waste emissions will grow at a rate of 2.6 percent per year.
Non-highway transportation sources grew at the rate of 5 percent per year from 1940–1960 and 12.5 percent per year from 1960–1970. As reliance on petroleum products decreases with energy conservation measures, this projection assumes that the growth rate will slow to 4 percent per year through 1980 and 3.8 percent per year from 1980–1590. The slower increase in the 1980–1990 period is due to the effect of new aircraft emission standards effective on 1979 and later model year aircraft engines. Again, it is likely that the estimates are conservative since they depart significantly from historical trends.
ELECTRIC POWER GENERATION EMISSIONS: 1972–1990
Emission projections for electric power generation were based on recent electric generation forecasts by the Federal Energy Administration (USEPA 1975) and the National Academy of Engineering (NAE 1974). These forecasts include generating capacity and actual power output for 1980 and generating capacity projections for 1985. Actual power output for 1985 and both generating capacity and actual power output for 1990 were projected from 1972–1980 data which indicates a 6.5 percent annual increase in power usage. The ratio of actual power output to generating capacity, i.e., load factor, was assumed to remain constant from 1980 to 1990 for each power source. The projections of generating capacity, actual power output and emissions are summarized in Table App. 14-C2.
As seen in Table App. 14-C2, nuclear power generation will increase significantly from 1972–1990 under the assumptions of Project Independence. Coal generation will also increase and natural gas generation will decrease by 10 percent during that same period. All new coal, oil and natural gas generation is assumed to meet federal new source performance standards for NOx emissions. One-quarter of the existing plants are assumed to be retired and replaced by new facilities every 5 years. Resulting emissions from electric power generation will increase 49 percent from 1972–1990 due to the increase in emissions from coal fired plants.
These emission estimates reflect the projected predominance of nuclear power in electric generation. If for some reason a decision should be made not to build any new nuclear power plants after 1975, quite a different picture would emerge. As can be seen in Table App. 14-C3 and Figure App. 14-C1, such a policy decision would result in nearly a doubling of NOx emissions in 1990 if coal fired plants provided the capacity that was projected for nuclear power. If the projected growth of nuclear generation were slowed rather than halted then the resulting NOx emissions would fall somewhere between the projected extremes displayed in Table App. 14-C3 and Figure App. 14-C1.
TABLE App. 14-C2
NOx Emission Projections to 1990 from Electric Power Generation Assuming Project Independence
|
Energy Source |
Gen. Cap. (MW) |
Power Output (109 kwh) |
Emissions (106 tons) |
Gen. Cap. (MW) |
Power Output (109 kwh) |
Emissions (106 tons) |
Gen. Cap. (MW) |
Power Output (109 kwh) |
Emissions (106 tons) |
|
|
1980 |
1985 |
1990 |
||||||
|
Coal |
295,000 |
1608 |
7.21 |
330,000 |
1734 |
7.21 |
402,000 |
1972 |
7.89 |
|
Oil |
59,000 |
165 |
0.52 |
52,000 |
146 |
0.52 |
52,000 |
146 |
0.52 |
|
Natural gas |
86,000 |
145 |
0.48 |
78,000 |
178 |
0.44 |
78,000 |
178 |
0.44 |
|
Gas Turbine |
61,000 |
53 |
0.03 |
61,000 |
53 |
0.03 |
61,000 |
53 |
0.03 |
|
Hydroelectric |
88,000 |
325 |
- |
75,000 |
276 |
- |
75,000 |
276 |
- |
|
Nuclear |
73,000 |
394 |
- |
250,000 |
1314 |
- |
500,000 |
2,452 |
- |
|
Other |
- |
- |
- |
61,000 |
53 |
- |
75,000 |
66 |
- |
|
Total |
662,000 |
2,740 |
8.16 |
907,000 |
3,754 |
8.20 |
1,243,000 |
5,143 |
8.88 |
TABLE App. 14-C3
Nationwide Emissions of NOx from Electric Power Generation Projected to 1990 for Two Policy Options
|
|
NOx Emissions (106 tons/year) |
|||||||
|
YEAR |
Project Independence |
No New Nuclear Plants Built After 1975 |
||||||
|
Totala |
Coal |
Oil |
Natural Gas |
Totala |
Coal |
Oil |
Natural Gas |
|
|
1972 |
5.94 |
3.95 |
0.85 |
1.14 |
5.94 |
3.95 |
0.85 |
1.14 |
|
1980 |
8.24 |
7.21 |
0.52 |
0.48 |
9.32 |
8.29 |
0.52 |
0.48 |
|
1985 |
8.20 |
7.21 |
0.52 |
0.44 |
12.81 |
11.82 |
0.52 |
0.44 |
|
1990 |
8.88 |
7.89 |
0.52 |
0.44 |
17.56 |
16.57 |
0.52 |
0.44 |
|
aTotal contains 0.03×106 tons/year from gas turbines |
||||||||
ROAD VEHICLE PROJECTIONS
Past highway statistics (FHA 1972) on vehicle-miles-of-travel (VMT) are used to project NOx emissions to 1990 for road vehicles including light-duty passenger vehicles, light and heavy duty gasoline trucks and heavy duty diesel trucks and buses. Three options for control are considered: (1) no controls on NOx emissions with the statutory program for CO and HC in effect; (2) a 5 year delay of the 1977 statutory standards, i.e., maintenance of the interim standard for NOx emissions of 2 g/mile until 1983; and (3) the present statutory program (NOx emissions of 0.4 g/mile by 1977). Two different VMT growth rates are considered: (1) 2 percent annual increase and (2) 4 percent annual increase. The former assumes stringent gas conservation measures while the latter assumes no restraint. The resulting projections are displayed in Table App. 14-C4 and Figure App. 14-C2.
The present statutory program will result in a 25 percent reduction in 1972 road vehicle emissions by 1990 for a VMT growth rate of 4 percent. A 5 year delay of the 1977 automotive standards will result in only a 17 percent reduction in 1972 emissions by 1990 for the same VMT growth rate. In addition to growth in total VMT, growth in uncontrolled heavy duty gasoline and diesel truck usage will be responsible for the fact that emission reductions predicted in 1990 are considerably smaller than the percentate reduction in emissions per mile required by the standards for light-duty automobiles. These heavy duty sources become major contributors to total transportation emissions during the 1980’s.
TABLE App. 14-C4
Nationwide NOx Emissions From Road Vehicles Projected to 1990 for Three Policy Options
|
|
NOx Emissions (106 tons) |
|||||
|
2% Annual Growth in VMT |
4% Annual Growth in VMT |
|||||
|
YEAR |
No Control |
Present Statutory Program |
Proposed 5-year Standards Delaya |
No Control |
Present Statutory Program |
Proposed 5-year Standards Delaya |
|
1972 |
7.62 |
7.48 |
7.48 |
7.62 |
7.48 |
7.48 |
|
1975 |
8.55 |
7.82 |
7.82 |
9.08 |
8.31 |
8.31 |
|
1980 |
9.49 |
61.4 |
7.37 |
11.06 |
7.14 |
8.60 |
|
1985 |
10.35 |
4.44 |
5.60 |
13.48 |
5.89 |
7.38 |
|
1990 |
11.22 |
4.03 |
4.38 |
15.86 |
5.68 |
6.18 |
|
a1977 auto emission standards for NOx postponed until the 1982 model year. California interim standard of 2.0 g/mi applies to 1977–1982 model years. |
||||||
LITERATURE CITED
Federal Highway Administration 1972, “Highway Statistics,” Office of Highway Planning.
National Academy of Engineering 1974b, “U.S. Energy Prospects: An Engineering Viewpoint,” Task Force on Energy.
U.S. Environmental Protection Agency 1975, “Federal Energy Administration Electric Generation Projections,” Personal Communication with James Speyer, Office of Planning and Management.
CHAPTER 15
NITROGEN OXIDE CONTROL TECHNIQUES
OVERVIEW
This chapter makes substantial use of reports and information made available by a large number of government and private institutions. Special thanks are due to the Combustion Research Section of the EPA Control System Division.
Throughout this section, emissions have been converted to lbs/106 BTU expressed as NO2 in order to facilitate comparison of emissions from different sources. A table of conversion factors appears in Appendix 15-A.
Typically, nitrogen oxides are formed in localized, high-temperature regions in combustors by the oxidation of both atmospheric nitrogen (thermal NOx) and nitrogen that may be contained in the fuel (fuel NOx). The formation of NOx in combustion systems can be suppressed, with varying degrees of success, by reducing the oxygen content and the temperature in the localized regions in the combustor contributing to emissions, usually in the vicinity of the flame. Reductions in the oxygen content in the flame zone reduce the emissions of both fuel and thermal NOx; reductions in temperature, however, produce significant reductions in only the thermal NOx.
Injection of cooled combustion products, steam, or water into the flame volume; reduction of temperature to which combustion air is
preheated; and extraction of heat from the flame volume are methods that have been used to reduce the temperature in the combustor. The injection of cooled combustion products into the flame volume can be achieved externally by ducting relatively cool combustion products to the burner area (referred to as flue gas recycle for furnaces or exhaust gas recycle for engines) or internally by modifying the burner design so as to induce the entrainment of colder combustion products by the hot gases leaving a burner.
Methods for reducing the oxygen content in the flame zone involve lowering the volume of air supplied to the burners by reducing the overall air/fuel ratio to the combustor (referred to as low-excess-air firing) or by reducing the air/fuel ratio for some burners without reducing the overall air/fuel ratio (referred to as staged combustion).
Low excess-air firing can achieve modest reductions in NOx emissions. However, its application is limited by the ability to reduce the air/fuel ratio to burners without producing excessive emissions of carbon monoxide and particulates.
There are several variations on staged combustion, referred to as biased firing, off-stoichiometric firing, overfire air and two-stage combustion. In general, the air/fuel ratio to some or all of the burners is reduced, to as low as seventy percent of the value required for combustion. The combustion is then completed by the addition of the balance of the air requirement after an interval in which the combustion product is allowed to partially cool. In engines a form of staged combustion is achieved by stratifying the fuel charge to provide local regions of relatively high fuel concentration, facilitating combustion at overall air/fuel ratios considerably higher than those practical for a uniform charge.
Many of the results reported in this chapter are from tests performed in laboratories or on small-scale units. Some of the results point to promising techniques or methods for nitrogen oxide emission control. However, it should be borne in mind that the application of test results
to large scale, operational systems often introduces significant problems, many of which are indicated throughout the chapter. The success of a test does not necessarily imply easy transfer to a large, commercially practicable system and considerable effort is being, and will continue to be, directed toward the resolution of the problems which arise when such a transfer is attempted.
Utility boilers and gas turbine engines have received priority in the development of nitrogen oxide emission control technology. Other major sources, particularly stationary reciprocating engines, have received relatively little attention.
Utility Boilers
As noted in Chapter 14, utility boilers fired by coal, gas, and oil accounted for nearly half of the stationary emissions of NOx in 1972. Based upon a field test performed by Esso Research and Engineering and other data, EPA has developed emission factors which indicate that NOx emission rates for existing boilers range from 0.6 to 1.2 lbs/106 BTU for coal-fired units, from .27 to .7 lbs/106 BTU for oil-fired units, and from .12 to .7 lbs/106 BTU for tangentially fired boilers using gas. The lowest emissions were from tangentially-fired units and the highest from units with intense combustion. Emissions decreased substantially with decreased furnace load.
Low-excess-air firing, staged combustion, flue-gas recirculation, water injection and reduced air preheat are control techniques that have been successfully demonstrated on field units. The latter two methods, however, have an associated and usually unacceptable penalty in thermal efficiency of the boiler. Flue gas recirculation is most effective for gas-fired units but is relatively uneffective for oil-fired and coal-fired units in which the fuel NOx contribution is significant. By the utilization of a combination of control techniques an average reduction in emissions of 60 percent has been achieved for gas-fired units, 48 percent for oil,
and 37 percent for coal. In the case of coal, the locally fuel rich conditions may contribute to slagging or corrosion problems. Accelerated corrosion testing over 300 hours showed no adverse effects but longer range trials for a wide range of coals are needed before the control techniques can be commercially accepted.
The applicability of combustion process modification to existing furnaces must be evaluated on a case-by-case basis. Boilers can generally be adapted for low-excess-air firing and staged combustion without major modification. Flue-gas recirculation, however, may be impractical for retrofitting except on units which have had the ducting already installed for steam temperature control. The capital costs vary widely with specific installation size and design; they range from under $0.50/KW for staged combustion to $6.0/KW for flue gas recirculation on existing units and from negligible costs for staged combustion to $4.0/KW for flue gas recirculation on new units.
Equipment manufacturers can meet existing emission standards on new units but usually specify an upper limit on the nitrogen content of the fuel oils that can be burned without exceeding the standards. EPA has proposed emission levels that might be achieved in the future: 0.12 lb/106 BTU (100 ppm) for gas, 0.20 lb/106 BTU (150 ppm) for oil, and 0.28 lb/106 BTU (200 ppm) for coal by 1980; and 0.06 lb/106 BTU (50 ppm) for gas, 0.12 lb/106 BTU (90 ppm) for oil, and 0.14 lb/106 BTU (100 ppm) for coal by 1985. These are technological goals that can probably be met on individual units if adequate research is carried out on new control techniques, pilot-plant development and field-demonstration. It should be recognized that high-intensity combustion burners may be required to burn certain types of coals, and these will yield higher emissions than the projected goals. Further, there is a significant lag time between the development of new designs and installation of field-units by the utilities.
Industrial Boilers
Most boilers with 10 to 500 million BTU/hr
capacity can meet the new source performance standards for units greater than 250 million BTU/hr capacity with only minor modifications in operating conditions. Still lower emission levels are possible, but redesign of the boilers would be required to allow off-stoichiometric or staged-combustion in units burning heavy fuel oils or coal since very few existing units possess the necessary flexibility. Problems which must be considered in the design of new units, and particularly in the modification of existing units, include corrosion and deposits on boiler tubes, flame instability, and combustion noise. The level of control that is achievable on industrial boilers is close to but not as great as that attainable with utility boilers.
Commercial and Residential Space Heating
Small space heating equipment (less than 10 million BTU/hr) fueled with natural gas or distillate oil generally has the lowest specific NOx emissions of any class of combustion equipment (i.e., 0.05 to 0.20 lb NO2/106 BTU) and produces about seven percent of the stationary source NOx. Existing units have little flexibility for NOx control, and the emissions are unaffected by normal maintenance operations. Combustion improving devices generally increase NOx emissions by producing a more intense flame region, but optimized burners, based either on conventional designs or on new combustion concepts such as catalytic combustion, which emit from 0.015 to 0.05 lb NO2/106 BTU have been demonstrated in tests.
Stationary Engines
About 18.8 percent of the stationary source NOx is produced by reciprocating engines. Spark-ignition gas engines produce about 63 percent of the engine NOx, primarily in applications associated with the gas industry, i.e., pipelines and natural gas production and processing.
Moderate reduction in NOx emissions from these engines (20 to 40%) can be achieved, while reducing fuel consumption, by cooling the fuel/ air mixture. Further reductions in NOx emissions from existing, engines are possible, but have generally been accompanied by a significant increase in fuel consumption. New engine designs, such as those using stratified charge concepts, may produce substantial reductions in NOx without increasing fuel consumption; however, further development work is required.
Diesel engines produce about 34 percent of the engine NOx in agricultural and industrial applications. As in spark ignition engines, cooling the intake air, can reduce NOx emissions while decreasing the fuel consumption. Water injection can also reduce NOx emissions with little change in fuel consumption; however, corrosion may be a problem, particularly if the engine is used on a standby basis. NOx emissions have been reduced by as much as 50 percent with little penalty in fuel consumption, but lower emission levels may require substantial engine redesign, e.g., prechamber engines; or exhaust gas treatment using catalytic reduction of NO to N2; or may result in significant increases in fuel consumption. EPA projects that emission levels of 0.14 lb, NO2/106 BTU for spark ignition gas engines and 0.16 lb NO2/106 BTU for diesel engines may be reasonably achieved by 1980, but there may be some increase in fuel consumption.
Stationary gas turbine NOx emissions can be reduced to the levels of currently proposed emission standards through the use of steam or water injection into the combustion chamber. Further reductions are possible through combuster design modifications including use of out-of-line combustion chambers; improved fuel atomization and mixing; prevaporized, premixed combustion; internal recirculation of combustion products; and surface combustion. The use of heavy fuel oils, rather than natural gas or distillate oils, will probably result in higher emission levels due to fuel nitrogen and may require new measures to control NOx.
The efficiency of gas turbines varies from 24 percent for simple cycle gas turbines to more
than 40 percent for combined gas turbine—steam turbine cycles. Emission standards based upon thermal input (e.g., lb.NOx/106 BTU or ppm NOx at 15 percent oxygen) provide no credit for higher efficiencies and, in fact, may inhibit the development of more efficient systems.
Fluidized Bed Combustion
Fluidized bed combustion of coal provides a potential alternative to current utility boiler design that is competitive when a low-sulfur fuel or flue-gas treatment is required to meet the emission standards for sulfur oxides. Atmospheric pressure and pressurized combustors with dolomite injection for control of sulfur oxide emissions are at a pilot-plant stage. A demonstration of a 30 MW atmospheric pressure fluidized-bed combustor is scheduled for 1975 and it is projected that 200 MW and 800 MW units may be built as early as 1977 and 1980, respectively.
At the low temperatures of operation of fluidized bed combustors, typically 1500F to 1800F, most of the NOx is contributed by the oxidation of fuel nitrogen. Tests on laboratory and pilot-scale fluidized-bed combustors have yielded emissions that meet the current standards for new coal-fired units. Emissions as low as 0.11 lbs/106 BTU have been obtained without the benefit of staging. Tests on larger-scale units are needed to establish practical emission levels for commercial units.
Tall Stacks and Intermittent Control Strategies
The only intermittent control strategy that appears practical for NOx emission reduction is load switching of electric power generation. Load switching has limited applicability because of the variability in the contribution of electric power generation to local emissions. The advantages of tall stack release of sulfur dioxide to reduce ground level concentrations do not apply for NO. Tall stacks
potentially reduce ground level NO concentrations however, NO converts to nitric acid and nitrates faster than does SO2 to sulfuric acid and sulfates and since the reaction products precipate, there is a greater potential for local impact. There is considerable uncertainty about the effects of NOx release from tall stacks on the formation of photochemical oxidants and the ground level concentrations of oxidants and nitrogen dioxide.
Future Trends
A major shift from gas to coal is expected in fuel utilization for utility boilers, and to a lesser extent smaller combustors. Emissions would be expected to increase unless new levels of control are achievable for the combustion of coal. Shale and synthetic fuels from coal may have higher nitrogen contents than existing fuels and may yield high NOx emissions.
Although significant success has been achieved in reducing NOx emissions from stationary sources, the reductions that have been achieved, typically 50 percent, are much smaller than existing potential. New combustion systems, utilizing surface or catalytic combustion, have the potential to eliminate thermal NOx. Significant reduction in the emissions from coal-fired units have been achieved by burner modification on a pilot scale, and the potential exists of designing burners with emission levels for coal below 0.2 lb/106 BTU.
Flue gas treatment is more expensive than combustion process modification but at present can attain much higher reduction in NOx emissions. A number of flue gas treatment methods are at a pilot plant demonstration stage, primarily in Japan.
Institutional Constraints
The reduction of NOx from stationary sources is subject to a number of economic and logistic constraints. Capital availability may limit the
rate at which new, more expensive equipment with reduced NOx emissions can be brought on line and may severely constrain the rate at which existing equipment is brought into compliance with emission standards by retrofit. Increased operating costs due to new maintenance requirements, increased fuel consumption, decreased capacity, or increased labor to provide the closer control of combustion conditions required for NOx control must also be factored into the decision making process.
Equipment outages for modification of existing facilities present an additional constraint, particularly in the electric utilities. Some modifications may be prepared in advance and installed during scheduled outages; however, prolonged outages, which may be required for the modification of some units, may strain the system generating capacity and will, therefore, require careful scheduling in order to permit utilities to meet demand. The logistics of modifying a very large number of units of diverse types may inhibit the retrofit of smaller sources.
NOx control techniques have been developed for those applications for which emission regulations have been proposed (e.g., utility boilers and stationary gas turbines) providing an economic incentive for equipment manufacturers to develop NOx controls. Where funding has been available for research on NOx control, significant progress has been made toward developing control techniques. On the other hand, there is little incentive to develop burners or small package boilers with low-NOx emissions, so progress in development is slow for these systems. Reduction of NOx emissions from systems for which development incentives do not exist, (e.g., industrial process furnaces) has received little attention.
INTRODUCTION
Stationary Sources of Nitrogen Oxides
Stationary sources currently contribute more than half of the man-made emissions of NOx in the U.S. The NOx emissions are dominated by combustion sources burning coal, gas, and oil (Mason and Shimzu 1974). The major sources are utility boilers and stationary engines. Nitric acid and nitrogen fertilizer plants contribute a small fraction of the national total, although their emissions may be locally significant in some areas.
The nitrogen oxides emitted by combustion sources are predominantly in the form of nitric oxide (NO) with the residual, usually less than 5 percent, in the form of nitrogen dioxide (NO2). The oxides are formed either by the oxidation of atmospheric nitrogen at high temperatures (thermal NOx) or by the oxidation of nitrogen compounds in the fuel (fuel NOx). The relative contributions of thermal and fuel NOx depend on combustor design and operating conditions, as well as on the nitrogen content of the fuel. As much as half of the total stationary emissions may be contributed by the oxidation of the nitrogen in the fuel, primarily in units burning heavy-oils and coals.
Control Options
The technology for controlling the emission of nitrogen oxides from combustion sources is based on either the modification of the combustion process to prevent formation of the oxides or the treatment of the product gases to destroy or remove the oxides. The latter option is made difficult by the relative inertness and insolubility of NO; wherever combustion process modification has yielded the needed reduction in NOx emissions, it has proven to be the most economical method for doing so. In some cases, where stringent emission controls have been imposed on combustion sources, a combination of
combustion process modification and product gas treatment has been proposed (Tohata 1974). For NOx emissions from chemical plants, a variety of control techniques are available, including catalytic destruction, afterburning, absorption and adsorption.
MECHANISMS FOR NITRIC OXIDE FORMATION
Thermal Fixation of Atmospheric Nitrogen
The formation of thermal NOx is determined by highly temperature dependent chemical reactions the so-called Zeldovich (1946) reactions. The rate of formation is significant only at high temperatures (greater than 3300F) and doubles for every increase in flame temperature of about 70F. The rate of formation increases with increasing oxygen concentration (rate proportional to the square root of the oxygen concentration) except in a small region near the flame zone in which superequilibrium concentrations of oxygen atoms are found (Thompson et al. 1972, Sarofim and Pohl 1973, Livesey et al. 1971). Although the NOx emitted by most practical combustors is predominantly in the form of NO, there is evidence from laboratory studies that a significant fraction of the NOx may be present as NO2 in localized regions of a flame (Merryman and Levy 1974).
Formation of Fuel NOx
An upper bound for the potential contribution of fuel-nitrogen to NOx formation may be gauged from the nitrogen content of fuels. Natural gas has negligible amounts of organically bound nitrogen, U.S. crude oils have nitrogen concentrations that average about 0.15 percent by weight (Martin and Berkau, Ball et al. 1951) and U.S. coals have nitrogen concentrations that average about 1.4 percent by weight. The nitrogen content in the crude oil varies between oil fields (Table 15–1), and is high for California oils. During the refining of oils, the nitrogen
TABLE 15–1
Nitrogen Content in U.S. Crude Oils in Million Barrels (Ball et al. 1951, Ball and Wenger 1958, Vandaveer 1965, Aga et al. 1973)
|
|
Proved Reserves (1972) |
Production (1972) |
Average Weight % Nitrogen |
||
|
MM bbl |
% |
MM bbl |
% |
|
|
|
Texas |
12,144 |
33.4 |
1,258 |
38.3 |
0.074 |
|
Alaska |
10,096 |
27.8 |
73 |
2.2 |
— |
|
Louisiana |
5,028 |
13.8 |
780 |
23.8 |
0.056 |
|
California |
3,554 |
9.8 |
346 |
10.5 |
0.49 |
|
Oklahoma |
1,303 |
3.6 |
198 |
6.0 |
0.151 |
|
Wyoming |
950 |
2.6 |
139 |
4.2 |
0.183 |
|
New Mexico |
582 |
1.6 |
106 |
3.2 |
0.082 |
|
U.S. Total |
36,339 |
92.6 |
3,231 |
88.2 |
— |
concentrates in the heavy fractions so that, residual oils from a California crude may have as much as 1.1 percent nitrogen whereas light distillate oils generally have nitrogen concentrations under 0.1 percent. The spread in nitrogen concentrations for coals is shown in Figure 15–1. The concentration is reported as lbs-NO2 per million BTU assuming 100 percent conversion of fuel nitrogen to NO2. Also shown on the plot are the sulfur concentrations, reported as lbs-SO2/106 BTU, and the percentage of U.S. coal reserves that have nitrogen and sulfur concentrations below certain levels. There is no apparent correlation between coal nitrogen content and coal sulfur content.
A fraction of the fuel nitrogen is converted to nitric oxide in practical combustors. Tests on carefully controlled laboratory-scale units have typically shown that 15 to 100 percent of the fuel nitrogen is converted to NOx (Figure 15–2), with the higher conversion efficiencies obtained when the nitrogen content is low or when the combustor is operated lean.
Because it is not easy to separate the contribution of fuel NOx from thermal NOx, the conversion efficiency of fuel nitrogen to NOx in large scale units is not known with certainty, but estimates obtained from field tests show trends similar to those obtained in the laboratory. A plot of the total uncontrolled NOx emissions from large and small scale units, Figure 15–3, suggests that the emissions increase with increases in fuel nitrogen content for units ranging in size from laboratory units to utility boilers. Emissions from large gas-fired units account for the high emission data at zero nitrogen content.
The mechanism by which the fuel nitrogen is converted to nitrogen oxides is imperfectly understood. The conversion efficiency of fuel nitrogen to NOx is found to increase markedly with increased oxidizing conditions in the flame but is insensitive to changes in temperature (Fenimore 1972, Martin and Berkau).
FACTORS INFLUENCING NOx EMISSIONS
The rate of formation of nitric oxide in flames by the two mechanisms discussed above is a sensitive function of oxygen content and, for thermal NOx, temperature. Any factor which influences temperature and oxygen concentration profiles in combustors may therefore have an influence on NOx emission. This makes generalizations for practical systems difficult since fuel/air ratios, fuel/air mixing patterns, fuel type, interaction of different burners, and placement of heat transfer surfaces all influence emissions. Additional complications are introduced by the fact that NO may be reduced by reactions with hydrocarbons in flames and that hydrocarbons can react with atmospheric nitrogen to form nitrogen-containing compounds that are in turn oxidized to nitric oxide. The following sections indicate some of the factors that influence emissions.
Air Preheat
Increased air preheat increases flame temperature and the contribution of thermal NOx. The increases observed in practice for premixed flames are in agreement with predictions from theory (Lange 1972). Reducing air preheat has been shown to decrease emissions on full-scale utility boilers (Blakeslee and Burbach 1972) but it carries with it an associated penalty in thermal efficiency.
On the other hand, in reciprocating engines reducing the inlet air temperature increases efficiency and is thus a very attractive NOx control technique for these systems.
Water or Steam Injection
Flame temperatures can be reduced through the use of steam or water injection. A 90 percent reduction in NOx emissions has been achieved by the use of 15 percent (by weight) water in
the fuel air mixture in a gas-fired system (Halstead et al. 1972). The water may be introduced by forming a water-in-oil emulsion which has the additional benefit of providing better fuel atomization and lower soot and particulate emissions (Toussaint and Heap 1974). Water injection, as expected, has little effect in controlling the emission of fuel NOx since fuel NOx formation is insensitive to temperature changes. Although the effectiveness of water injection for reducing NOx emission has been demonstrated on a full-scale utility boiler (Blakeslee and Burbach 1972), the associated loss in thermal efficiency was high. The major application of water injection for NOx control is in engines where it provides substantial NOx reductions without incurring large losses in fuel economy.
Flue Gas Recirculation
The recirculation of flue gases reduces temperature and dilutes the oxygen in the air. It reduces thermal NOx by an amount equal to that of an equivalent thermal load of water injection, and has little influence on fuel NOx. To be effective the flue gas has to be injected into the flame zone. A recirculation of 10 percent (by weight) of the flue gases has achieved a 60 percent lowering of thermal NOx and 20 percent flue gas recycle a 75 percent reduction on a small scale furnace (Halstead et al. 1972). Flue gas recycle is sometimes used in furnaces for temperature control. In such cases, the cost of implementing NOx control is small; otherwise the cost of installation of ducting makes flue gas recycle impractical in retrofitting units. Flue gas recycle, unlike water injection, does not have an adverse effect on the thermal efficiency of a furnace, but does on an engine.
Furnace Load and Size
As the size of a furnace of a given type
increases the emissions per unit weight of fuel increase slightly (Woolrich 1961). This is consistent with the postulate that flames radiate less effectively to the walls in larger furnaces and, therefore, are somewhat hotter. The level of emissions per unit weight of fuel varies much more dramatically with furnace load, being a little more than proportional to load for gas-fired field units and a little less than proportional to load for coal and oil-fired units (Bartok et al. 1971). This difference between fuels can be explained by the influence of fuel NOx which is expected to be relatively load insensitive. Reduction of NOx emissions by use of reduced loads or the oversizing of combustion chambers carries with it a capital cost associated with underutilized equipment.
Excess Air
As the amount of excess air is increased in a combustor, the oxygen content in the flame zone generally increases and the temperature decreases. As a consequence of these two opposing effects, the emissions of thermal NOx pass through a maximum (Armento 1974) (Wasser et al. 1968). Most, but not all, furnaces operate in a range where NOx emissions decrease with a reduction in excess air. Fuel NOx, being temperature insensitive, increases monotonically with increased excess air (Fenimore 1972, Martin and Berkau, Turner et al. 1972). Low excess air firing can reduce NOx emissions, but care must be taken not to replace those emissions with CO and soot emissions. When operating a multi-burner system near stoichiometric, care must be taken to regulate the fuel/air ratio of each burner so as to prevent some burners from running fuel-rich. Low excess air firing of furnaces, although necessitating more control equipment and instrumentation to monitor emissions, provides fuel savings through increased thermal efficiency. In contrast, fuel lean combustion with high excess air can reduce the NOx
emissions from engines, but improved carburation, fuel injection, or stratification may be necessary for efficient operation.
Staged Combustion
Staged combustion takes advantage of both low temperatures and low oxygen concentration by operating some burners fuel rich, allowing the partially combusted products to cool between stages, and completing the combustion by addition of the residual air. Several variations of staged combustion can be used. In a multi-burner system the lower burners may be run fuel rich and the upper burners lean; this is known as biased or off-stoichiometric combustion. Alternatively, all of the fuel can be injected through the lower burners and the air introduced through the upper burners or higher in the furnace through ‘NO’ ports; this is referred to as overfire-air (Winship and Brodeur 1973) or two-stage combustion (Barnhart and Diehl 1959, Barnhart and Diehl 1960). Staged combustion reduces the emission of both thermal and fuel NOx. By operating the first stage with as little as 70 percent of stoichiometric air, the emission from a 0.25 percent nitrogen oil could be reduced to under 100 ppm (Siegmund and Turner 1974). As much as a 90 percent reduction in the fuel nitrogen has been observed by use of fuel rich operation in laboratory burners (Fenimore 1972).
The potential for NOx control in practical systems depends on many factors. In retrofitting units the number of burners, the ability to change fuel rates to individual burners, and the availability of ‘NO’ ports for air addition above the burners are physical constraints. In addition, in changing operation or design conditions to reduce NOx, care must be taken that the other pollutants are not substituted for the NOx. Low excess air operation increases the potential for emissions of unburned hydrocarbons, carbon monoxide and soot. Staged combustion may result in the formation of hydrogen cyanide or ammonia in the first stage,
which will be partially oxidized to NO in the second stage (Yamayishi et al. 1974). Changes of fuel/air ratio on burners may create problems of fuel stability and noise (Lockling et al. 1974). These complicating factors and the very wide variety of combustion systems in use make it difficult to state any generalization on the potential reduction in emissions that may be achieved by combustion process modification. Staged combustion, however, provides the most successful method currently in use for the control of NOx emission from combustion sources.
Burner Design
Variation in the momentum of the fuel and air streams, in the swirl or rotation of the air stream, in the shape of a burner, in the positioning of the fuel nozzle in a burner, in the atomization of liquid fuels, in the amount of air that is premixed with the fuel, or in the positioning of the flame stabilizer, if any, can make large differences in mixing and combustion patterns and hence in NOx emissions from different burners. The potential for major reductions in NOx emission by modification of burner design has been established on experimental burners (Heap et al. 1973). Although a complete characterization of the processes is not possible, some general rules seem to apply.
High intensity combustion favors the formation of thermal NOx. Evidence for this is provided by an inverse correlation of NOx emission with residence time in the combustion zone in a laboratory burner as the swirl was varied (Wasser and Berkau 1972). Other, less quantitative, evidence is provided by the observation that burners which entrain air gradually and produce relatively long flames, such as in a tangential boiler, produce low thermal NOx whereas high intensity burners for example cyclones, yield high emissions. Low emissions of thermal NOx also result when a burner produces entrainment of relatively cold combustion products into the flame, a form of
internal flue gas recirculation (Calvert 1973 Hemsath et al. 1972).
High emissions of fuel NOx are found when air is premixed into a burner or when a flame is lifted so that air is entrained before ignition (Heap et al. 1973). For oil and coals containing fuel nitrogen it is preferable that the fuel nitrogen be released into an oxygen deficient ambient atmosphere (Heap et al. 1973). Long diffusion flames therefore favor low fuel NOx in addition to low thermal NOx.
Through the use of multiple concentric fuel and air ports, apparently providing delayed mixing or a form of staging, emissions as low as 150 ppm have been reported for an experimental coal-fired burner (Heap et al. 1973).
These results suggest that with developmental effort, significant reduction in NOx emissions may be derived from changes in burner design.
Fuel Atomization
Under normal operating conditions for practical systems, the technique of fuel atomization and atomized pressure has little effect on either thermal NOx or fuel NOx. Normal atomization procedures produce fuel droplets which are sufficiently large and which have a velocity relative to the air flow sufficiently low that the droplets burn with an attached flame. This diffusion type flame results in near stoichiometric combustion of much of the fuel vapor and results in both thermal and fuel NOx levels which are relatively insensitive to the overall fuel/air ratio of the combustor (Flagan and Appleton 1974, Pompei and Heywood 1972). However, by using more efficient atomizers which produce much smaller droplets with a high velocity relative to the air flow, distillate fuels may evaporate and partially premix with air prior to burning. Thus, in a burner operating fuel lean, thermal NOx may be significantly reduced by increasing the atomizer efficiency (Pompei and Heywood
1972). Fuel NOx, on the other hand, tends to increase with increasing atomizer efficiency (Flagan and Appleton 1974).
Burner Interaction
Interaction between burners either on the same wall (Lowes et al. 1974) or on opposed walls (Bartok et al. 1971) leads to higher emissions, probably as a consequence of the reduced heat transfer from the central flames to the walls.
Reduction of NO by Reaction with Hydrocarbons
Laboratory experiments in which hydrocarbons and ammonia where injected into the post-flame zone have shown significant reduction in NOx levels (Wendt et al. 1973), and may suggest new methods for NOx control. Related experiments demonstrate that NO was destroyed with 30 to 95 percent efficiency by injection into the combustion air of a burner (Turner and Siegmund 1972), into a fluidized bed coal combustor (Hammons and Skopp 1971), and into a diffusion flame (Sarofim et al. 1973). In practical turbulent combustors, some NO may be destroyed by these processes as the turbulent eddies bring fuel rich pockets together with NO.
The factors that influence NO formation are many and are imperfectly understood. The results reported here suggest some directions, but what is practically achievable must be determined from field studies for the various classes of stationary sources, as described in subsequent sections. The large variation in peak temperatures, pressures, sizes and residence times in different combustion systems explains the wide range of emissions that will be reported, with residential heating units having the lowest emissions per unit fuel consumption and reciprocating engines the highest. These same differences necessitate different control methods for the different
classes of units and may limit the degree of control which can be achieved by combustion modification alone.
UTILITY BOILERS
Background
Utility boilers are a major source of NOx emissions. The NOx from utility boilers is discharged from a relatively small number of tall stacks. The boilers are fired by coal (54.3 percent of the energy supplied in 1972) (Mason and Shimzu 1974), gas (27 percent) and oil (18.6 percent), sometimes in combination. Environmental constraints on sulfur emissions had resulted in a substitution of low-sulfur oils and gas for coals, with a short term decrease in percentage use of coal, but the trend has been reversed and projections (NERC 1974) indicate a continued reliance on coal, a decrease in the rate of installation of oil-fired units, and a gradual phasing out of gas. Emissions are contributed by over 3000 boilers in use for steam-electricity generation, varying widely in size, design, and age. The most common boiler types, designated by the location of the burners in the combustion chamber, are tangential (T), horizontally opposed (HO), front wall (FW), cyclone (Cyc), vertical (V), turbo-fired (turbo) and all wall (AW) (Bartok et al. 1969).
Control Methods for Gas-Fired Units
The emissions from gas fired units are due entirely to thermal NOx and fall typically in the range of 0.30 to 1.31 lbs/106 BTU for uncontrolled units at full-load (Table 15–2 taken from an EPA-sponsored systematic field study) (Bartok et al. 1971). The lowest emissions are expected for units with delayed mixing of fuel and air and in which the flames can radiate effectively to the walls; the highest emissions
TABLE 15–2
Uncontrolled Emissions from Gas, Oil, and Coal fired Utility Boilers Operated at Full, Intermediate, and Low Load (Bartok et al. 1971)
|
Fuel |
Size (MW) |
Type of firing |
Full Load (MW) |
PPM @ 3% |
|
|
Gas |
180 |
FW |
180 |
390 |
0.51 |
|
Gas |
80 |
FW |
82 |
497 |
0.65 |
|
Gas |
315 |
FW |
315 |
992 |
1.29 |
|
Gas |
350 |
HO |
350 |
946 |
1.23 |
|
Gas |
480 |
HO |
480 |
736 |
0.96 |
|
Gas |
600 |
HO |
559 |
570 |
0.74 |
|
Gas |
220 |
AW |
220 |
675 |
0.88 |
|
Gas |
320 |
T |
320 |
340 |
0.44 |
|
Gas |
66 |
V |
66 |
155 |
0.20 |
|
Oil |
180 |
FW |
180 |
367 |
0.50 |
|
Oil |
80 |
FW |
80 |
580 |
0.79 |
|
Oil |
250 |
FW |
250 |
360 |
0.49 |
|
Oil |
350 |
HO |
350 |
457 |
0.62 |
|
Oil |
450 |
HO |
455 |
246 |
0.33 |
|
Oil |
220 |
AW |
220 |
291 |
0.39 |
|
Oil |
320 |
T |
320 |
215 |
0.29 |
|
Oil |
66 |
T |
66 |
203 |
0.27 |
|
Oil |
400 |
CY |
415 |
530 |
0.72 |
|
Coal |
175 |
FW |
- |
- |
- |
|
Coal |
315 |
FW |
275 |
1490 |
2.04 |
|
Coal |
600 |
HO |
563 |
838 |
1.15 |
|
Coal |
800 |
HO |
778 |
905 |
1.24 |
|
Coal |
575 |
T |
- |
- |
- |
|
Coal |
300 |
T |
300 |
568 |
0.78 |
|
Coal |
700 |
CY |
665 |
1170 |
1.60 |
|
Intermediate Load (MW) |
PPM @ 3% O2 |
|
Low Load (MW) |
PPM @ 3% O2 |
|
%N in Fuel |
|
120 |
230 |
0.30 |
70 |
116 |
0.15 |
- |
|
50 |
240 |
0.31 |
20 |
90 |
0.12 |
- |
|
223 |
768 |
1.00 |
186 |
515 |
0.67 |
- |
|
- |
- |
- |
150 |
341 |
0.44 |
- |
|
360 |
610 |
0.79 |
250 |
363 |
0.47 |
- |
|
410 |
335 |
0.44 |
325 |
253 |
0.33 |
- |
|
190 |
550 |
0.72 |
125 |
313 |
0.41 |
- |
|
240 |
230 |
0.30 |
- |
- |
- |
- |
|
- |
- |
- |
- |
- |
- |
- |
|
120 |
322 |
0.44 |
80 |
266 |
0.36 |
0.29 |
|
50 |
361 |
0.49 |
21 |
258 |
0.35 |
0.36 |
|
172 |
306 |
0.41 |
- |
- |
- |
0.31 |
|
- |
- |
- |
150 |
264 |
0.36 |
0.46 |
|
365 |
219 |
0.30 |
228 |
186 |
0.25 |
- |
|
170 |
267 |
0.36 |
120 |
324 |
0.44 |
0.42 |
|
220 |
220 |
0.30 |
- |
- |
- |
0.30 |
|
- |
- |
- |
- |
- |
- |
0.62 |
|
258 |
205 |
0.28 |
- |
- |
- |
0.53 |
|
140 |
660 |
0.90 |
- |
- |
- |
1.36 |
|
190 |
1280 |
1.75 |
160 |
1200 |
1.64 |
1.36 |
|
462 |
781 |
1.07 |
363 |
643 |
0.88 |
1.38 |
|
580 |
741 |
1.01 |
- |
- |
- |
1.17 |
|
470 |
405 |
0.55 |
310 |
264 |
0.36 |
- |
|
240 |
418 |
0.57 |
- |
- |
- |
- |
|
545 |
882 |
1.21 |
- |
- |
- |
1.19 |
are expected for large units with high combustion intensities. Generally, tangentially-fired boilers yield the lowest emissions. Reduction in load has a major influence on emissions (Table 15–2); to a first approximation there is a proportionality between emissions and furnace load.
The control techniques that have been successfully demonstrated on field units are low-excess-air firing, staged combustion, flue-gas recirculation, water injection, and reduced air preheat. The concept of staged combustion was pioneered on gas units in the late 1950’s and 1960’s by Babcock and Wilcox and Southern California Edison (Bagwell et al. 1970, Barnhart and Diehl 1959, Barnhart and Diehl 1960, Teixera and Breen 1973). A 33 to 75 percent reduction in the emissions level has been attained on 12 units of 175 MW to 480 MW by the use of staged combustion (two-stage and/or off-stoichiometric combustion). The off-stoichiometric combustion, in which the fuel from up to 25 percent of the burners was diverted to the remaining burners, yielded better results than the two-stage combustion, in which part of the air to all of the burners was diverted to ‘NO’ ports higher in the furnace.
Southern California Edison also tested the potential for control of NOx by flue-gas recirculation on tangentially-fired boilers designed with flue-gas recirculation to the windbox (the compartment confining the air to the burners). Recirculation of flue gases at a rate of 15 per cent by weight of the sum of the fuel and air reduced the emissions by 30 to 60 percent; recirculation of 30 percent yielded emission levels below 100 ppm (0.131 lb NO2/106 BTU) (Teixera and Breen 1973).
The EPA funded systematic field study of NOx emission control methods determined the control achievable on representative front wall horizontally opposed, and tangentially fired boilers at three load levels, by use of various combinations of low-excess-air firing, staging, and flue gas recirculation (see Table 15–3).
TABLE 15–3
Percentage Reduction in NOx Emissions Achieved by Staging, Low Excess Air Firing, and Flue Gas Recirculation (Bartok et al. 1971)
|
Combustion Operating Modification and Furnace Load(1) |
||||||||||||||||
|
Fuel Fired |
Type (2) of Firing |
% Reduction in NOx Emission |
||||||||||||||
|
Low Exc. Air |
Staging |
LEA+Staging |
Flue Gas Rec. |
“Full”(3) |
||||||||||||
|
Full |
Int. |
Low |
Full |
Int. |
Low |
Full |
Int. |
Low |
Full |
Int. |
Low |
Full |
Int. |
Low |
||
|
GAS |
FW |
13 |
24 |
7 |
37 |
30 |
30 |
48 |
42 |
36 |
— |
— |
|
48 |
42 |
36 |
|
HO |
17 |
15 |
32 |
54 |
35 |
59 |
61 |
48 |
68 |
— |
— |
20 |
73 |
52 |
72 |
|
|
T |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
60 |
— |
66 |
65 |
— |
|
|
ALL (Average) |
16 |
19 |
26 |
45 |
31 |
52 |
54 |
44 |
52 |
— |
60 |
20 |
64 |
51 |
60 |
|
|
OIL |
FW |
27 |
20 |
28 |
29 |
20 |
20 |
39 |
32 |
21 |
46 |
31 |
— |
50 |
41 |
21 |
|
HO |
10 |
16 |
12 |
34 |
34 |
47 |
35 |
44 |
42 |
— |
— |
— |
38 |
35 |
55 |
|
|
T |
28 |
22 |
— |
— |
17 |
— |
— |
45 |
— |
10 |
13 |
— |
— |
59 |
— |
|
|
ALL (Average) |
19 |
19 |
18 |
30 |
22 |
34 |
38 |
37 |
32 |
28 |
23 |
— |
47 |
42 |
38 |
|
|
COAL |
FW |
— |
14 |
— |
— |
40 |
— |
— |
55 |
— |
— |
— |
— |
— |
60 |
— |
|
HO |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
— |
|
|
T |
27 |
18 |
— |
— |
39 |
— |
— |
50 |
42 |
— |
— |
— |
— |
50 |
42 |
|
|
ALL (Average) |
27 |
17 |
— |
— |
39 |
— |
52 |
52 |
42 |
— |
— |
— |
— |
55 |
42 |
|
|
(1) Furnace load: “Full”=85%–105%, “Intermediate”=60%–85%, Low=50%–60% of rating. (2) Type of firing: FW=Front Wall, HO=Horizontally Opposed, T=Tangential. (3) “Full control”: combination of techniques achievable on boilers tested. |
||||||||||||||||
Use of low excess air achieved modest reductions in emissions (13 to 32 percent). The minimum value of the excess air that could be used was determined by the problem of CO emissions which increased dramatically below a critical excess air level. The minimum practical level of excess air is a function of burner and furnace design; low-excess-air firing therefore requires careful control of the fuel and air rate to individual burners and careful monitoring of products of incomplete combustion in the stack. These disadvantages, however, are partially offset by the increase in the thermal efficiency of a boiler which results from a reduction in the excess air level. Staging and flue gas recirculation produced improvements in the systematic field study similar to those observed in Southern California. A combination of control techniques, as expected, yields a smaller effect than that computed from the gains observed when each technique is applied individually. The final three columns in Table 15–3 show that a 50 to 60 percent reduction in emissions was achievable when combinations of all the techniques available on each of the boilers in the field study were utilized to their fullest.
Injection of water, use of water-in-oil emulsions, and reduction of air preheat are all less desirable methods of NOx emissions control because they impose a penalty in thermal performance. Water injection tests carried out by Combustion Engineering on a 150 MW tangentially fired unit showed a maximum reduction of 50 percent in NOx emission at a water injection rate of 45 pounds per million BTU fired and an associated 5 percent decrease in furnace efficiency (Blakeslee and Burbach 1972). Significant reductions in NOx emission and furnace efficiency were also observed when the air preheat was reduced from 490F to 81F at three-quarter and half loads (Blakeslee and Burbach 1972).
In summary, the techniques most readily adapted to reducing emission from existing gas-fired units are low excess air firing and staged combustion, particularly in units where
the number of burners permits the changes to be made without any equipment modification. Flue gas recirculation provides a viable option only for those existing units already equipped with a recirculation system. Water injection or reduction in air preheat are less desirable control techniques in view of the penalty of increased fuel consumption entailed in their use. For gas units, a reduction in emissions of 50 to 60 percent can be expected with the use of low excess air and staged combustion alone, and greater reductions are possible when these can be combined with flue-gas recirculation. It should be noted, however, that units differ significantly in design and individual units may show much lower potential for emission control.
Control Methods for Oil-Fired Units
Emissions from oil-fired units are due to a combination of thermal and fuel NOx. Control methods which are based on a temperature reduction, such as flue gas recirculation, water injection, and load reduction, are effective in controlling the thermal NOx contribution but have little effect on the fuel NOx. Methods which reduce the oxygen availability in the primary conbustion zone, such as low excess air firing, staged combustion, and long flames with delayed oxygen-uptake are effective in controlling both sources of NOx. The emissions from nine oil fired utility boilers included in the systematic field survey (Bartok et al. 1971) range from 0.27 to 0.79 lbs/106 BTU at full load (Table 15–2). Tangential-fired units were found to yield the lowest NOx emissions for a given size unit, as had been previously found for gas fired boilers. It was estimated from results obtained using oils with different nitrogen contents that 30 percent of the fuel nitrogen in the nitrogen concentration range of 0.3 to 0.6 percent by weight was converted to NOx, corresponding approximately to an incremental emission of 0.06 lb/106 BTU per 0.1 percent increment in fuel nitrogen. Independent tests on tangentially-
fired boilers indicate a range of conversion of fuel nitrogen to NOx from 43 percent for 0.2 percent nitrogen oils to 30 percent for 1 percent nitrogen oils when the boilers where operated with 3 percent O2.
Reduction in load for oil-fired boilers resulted in a reduction in emissions but the dependence of emissions on load is less marked than for the case of gas-fired boilers (Table 15–2).
The control methods tested in the field study included low-excess-air firing, staging, flue gas recirculation and combinations thereof (Table 15–3). The trends are similar to those obtained for gas-fired units but the fractional reduction in NOx emissions is lower. The results show an average reduction in NOx for the best combination of control techniques ranging from 48 percent at full load to 38 percent at low load. Although the uncontrolled emissions of gas fired units were higher than the emissions from oil units of the same design and size, the emissions with the best control achieveable were generally lower for the gas-fired units. Part of the difference may be due to the fuel-nitrogen contribution and part due to the difficulty of controlling the very complicated processes that define the atomization and combustion of liquid fuels.
Control Methods for Coal-Fired Utility Boilers
Uncontrolled emissions from coal-fired boilers are higher than those for gas and oil (Table 15–2 and 15–4), ranging from 0.53 to 2.04 lbs/106 BTU at full load. Tangential-fired units again show the lowest emissions. Emissions from coal-fired units show a less than proportional dependence on load (Table 15–2). From the dependence of the emissions on load it is inferred that about 20 percent of the fuel nitrogen content, typically 1.3 percent by weight, is converted to NOx and that the fuel nitrogen contributes 50 percent of the emission at full-load (Crawford et al. 1974). The relative proportions of fuel NOx and thermal NOx depend
TABLE 15–4
Uncontrolled Emissions from Coal-Fired Utility Boilers and Percent Reduction in Emission Achievable by Staged Combustion (Crawford et al. 1974).
|
Fuel |
Size (MW) |
Type of Firing |
Load (MW) |
PPM @ 3% O2 |
|
%N in Fuel |
% Reduction at Full Load |
|
Coal |
105 |
FW |
101 |
454 |
0.60 |
— |
53 |
|
Coal |
125 |
FW |
125 |
634 |
0.84 |
1.35 |
40 |
|
Coal |
256 |
FW |
253 |
703 |
0.93 |
1.31 |
48 |
|
Coal |
320 |
FW |
350 |
832 |
1.11 |
1.38 |
34 |
|
Coal |
218 |
Ho |
219 |
569 |
0.76 |
— |
34 |
|
Coal |
480 |
Ho |
490 |
711 |
0.95 |
1.81 |
35 |
|
Coal |
800 |
Ho |
800 |
935 |
1.24 |
1.27 |
48 |
|
Coal |
250 |
T |
250 |
410 |
0.55 |
1.49 |
25 |
|
Coal |
330 |
T |
334 |
531 |
0.71 |
1.56 |
59 |
|
Coal |
350 |
T |
350 |
415 |
0.55 |
1.4 |
35 |
|
Coal |
348 |
T |
306 |
434 |
0.53 |
0.80 |
4 |
|
Coal |
850 |
Turbo |
370 |
600 |
0.80 |
1.34 |
34 |
upon many variables including coal moisture content, nitrogen content, burner design, furnace design and operating conditions.
More conclusive evidence on the contribution of fuel nitrogen in coal is provided by laboratory studies in which pulverized coal was burned in argon-oxygen mixtures (Pershing et al. 1973). In these tests 25 to 40 percent of the 1.2 percent nitrogen content of the coal was converted to NOx.
Based on the importance of both thermal NOx and fuel NOx it is expected that techniques that reduce temperature in the flame zone will be less effective in controlling NOx emission than those that reduce the oxygen content. Low-excess-air firing and staged-combustion have proven effective in reducing NOx emissions from coal-fired units by an average of 37 percent for the 12 units included in the EPA funded field tests (Crawford et al. 1974). The carbon content of the ash and the carbon monixide in the stack gas increased slightly during the tests (Crawford et al. 1974). Accelerated corrosion studies (Crawford et al. 1974) for the 300 hour duration of the trials with staged combustion showed no adverse effects but additional long range trials will be needed to confirm these preliminary findings.
Design Modifications and Costs
The success of low-excess-air and staged combustion has enabled all manufacturers to develop boilers that will meet the national NOx emission standards of 0.2 lbs/106 BTU for gas, 0.3 lbs/106 BTU for oil, and 0.7 lb/106 BTU for coal. Some manufacturers specify a maximum nitrogen content on the oil to be fired in new oil fired units. Further reduction in emissions can be achieved at considerably increased costs by use of flue gas recycle. Water injection may be used but with a penalty in thermal efficiency.
Severe constraints may be imposed on the ability to make modifications on existing units by the design of the particular unit and by the
space available for adding ducting. The modifications easiest to apply are low-excess-air firing and off-stoichiometric combustion. Addition of NO ports may be feasible in many units but the addition of flue-gas recycle would require a major effort.
Costs for modifying new and existing coal-fired tangentially-fired units have been estimated by Combustion Engineering (Blakeslee and Seler 1973) and are presented in Figures 15–4 and 15–5. (Additional data supplied by a utility and by a boiler manufacturer are consistent with the figures presented here). The costs were calculated for the introduction of 20 percent of the total combustion air over the fuel firing zone as overfire air, for the recirculation of 30 percent of the flue gas to the secondary air ducts and windbox, for a combination of overfire air and flue-gas recirculation, for the recirculation of 17 percent of the flue-gas through the coal pulverizers (mills), and for water injection into the fuel-firing zone at 5 percent of the steam rate for the boiler. The practical limit for single cell furnaces is about 600 MW; larger units use divided furnaces with two times as many burners which results in an increased cost for installation of ductwork for the various control methods (Blakeslee and Seler 1973). Cost figures are expected to vary considerably between units particularly for the case of retrofit. It should be noted that the costs in Figures 15–4 and 15–5 are only capital costs; in addition to capital costs, operating (Bartok et al. 1969, NAE 1972) and testing costs will be incurred which will be totally dependent on the specific situation.
INDUSTRIAL BOILERS
Background
Boilers with capacities from 10 to 500 million BTU per hour emit an estimated 18.1 percent of the U.S. stationary source nitrogen oxides. The industrial boiler population is made up of about 75000 units of various types, ages, and applications which burn gas (57 percent of the energy supplied in 1968), coal, (23 percent) and oil (20 percent) (Lockling et al. 1974). Residual oils and a small quantity of distillate oils are used. As a result of regulations on the emissions of sulfur oxides, recent trends have been toward the use of natural gas and low sulfur oils in all but the largest units. However, due to increased prices and reduced availability of these fuels, the industrial sector may be forced to rely increasingly upon coal and residual oil.
About a third of the industrial boilers, accounting for 10 percent of the industrial capacity, are small units (10 to 16 million BTU per hour capacity) (Lockling et al. 1974). These are primarily packaged firetube boilers which burn oil or gas. The 17 to 100 million BTU per hour units, which are primarily packaged watertube boilers, account for 53 percent of the units and 45 percent of the capacity. The larger units are watertube boilers, mostly field erected. Currently only 1 percent of the units and 9 percent of the capacity are greater than 250 million BTU per hour capacity, the size range for which new source performance standards have been established.
The data on emission levels and combustion modification effectiveness for NOx control are limited, mostly derived from a field study performed by KVB, Engineering ((Cato et al.) in which about 80 units were tested for baseline emission levels and minimum NOx achievable without hardware modifications. Due to the limited flexibility in combustion operating conditions
for these units, only moderate reductions were obtained in most cases. The results are presented in Table 15–5.
When combustion modification strategies are considered, it will be necessary to consider several problems which may arise such as corrosion and deposits on boiler tubes, flame instability, blow-off, flashback, combustion driven oscillations, and combustion noise or roar. The need for tuning individual units to minimize these problems will place constraints on retrofit programs due to the large number of units in existence and their diverse characteristics (Lockling et al. 1974).
Gas-Fired Boiler Emissions
Gas-fired units without air preheat, firetube boilers, and some of the watertube boilers had NOx emission levels less than 0.14 lb NO2 /106 BTU. (Cato et al.) The watertube units with air preheat had generally higher emissions, from 0.08 to 0.45 lb NO2/106 BTU. By comparing units with different degrees of air preheat, it was estimated that the preheat effects vary from 0.02 lb NO2/106 BTU per 100F for units of less than 30 million BTU per hour capacity to 0.15 lb NO2/106 BTU per 100F for much larger units. (Cato et al.) This appeared to be a function of the burner firing rate (BTU/hr/ burner). Reducing the oxygen level had only a small effect on emissions from units without air preheat, but did decrease the NOx levels from preheated boilers. Off-stoichiometric firing, achieved by taking one burner out of service, resulted in NOx reductions of 16 to 42 percent. By combustion modification, the percentage of units with emissions below 0.2 lb NO2/106 BTU was increased from 75 percent to 82 percent, (Cato et al.)
Oil-Fired Burner Emissions
The baseline emission levels for those units fired with No. 2 fuel oil were 0.13 to 0.26 lb
TABLE 15–5
Uncontrolled NOx and Low NOx Emissions from Industrial Boilers (Cato et al.)
|
|
Baseline |
Low NOx |
|
||||||
|
Fuel |
Size MBH |
Burners |
Test Load MBH |
PPM @ 3% O2 |
|
PPM @ 3% O2 |
|
% reduction |
|
|
Type |
Number |
||||||||
|
Gas |
25 |
Ring |
1 |
20 |
72 |
.086 |
65 |
.077 |
10 |
|
Gas |
29 |
Ring |
1 |
22 |
70 |
.083 |
- |
- |
- |
|
Gas |
29 |
Ring |
1 |
11 |
97 |
.115 |
68 |
.081 |
29.6 |
|
Gas |
59 |
Ring |
6 |
48 |
116 |
.138 |
- |
- |
- |
|
Gas |
60 |
Ring |
2 |
48 |
101 |
.124 |
86 |
.102 |
17.7 |
|
Gas |
60 |
Ring |
4 |
46 |
242 |
.288 |
138 |
.164 |
43.1 |
|
Gas |
160 |
Ring |
1 |
136 |
374 |
.445 |
355 |
.423 |
4.9 |
|
Gas |
158 |
Ring |
4 |
125 |
299 |
.343 |
110 |
.131 |
61.8 |
|
Gas |
300 |
Ring |
4 |
259 |
199 |
.237 |
169 |
.201 |
15.2 |
|
Gas |
10 |
Ring |
1 |
8 |
55 |
.065 |
38 |
.045 |
30.8 |
|
Gas |
20 |
Ring |
1 |
14 |
107 |
.127 |
75 |
.089 |
29.9 |
|
Gas |
11 |
Ring |
1 |
10 |
87 |
.104 |
83 |
.099 |
4.8 |
|
Gas |
7 |
Ring |
1 |
6 |
71 |
.085 |
67 |
.080 |
5.8 |
|
Gas |
10 |
Ring |
1 |
7 |
91 |
.108 |
90 |
.107 |
.9 |
|
Gas |
13 |
Ring |
1 |
12 |
57 |
.068 |
- |
.067 |
1.5 |
|
Gas |
18 |
Ring |
1 |
17 |
56 |
.067 |
53 |
.063 |
6.0 |
|
Gas |
20 |
Ring |
1 |
13 |
79 |
.094 |
67 |
.080 |
14.9 |
|
Gas |
8 |
Ring |
1 |
6 |
70 |
.083 |
45 |
.054 |
35.0 |
|
Gas |
20 |
Ring |
1 |
10 |
101 |
.120 |
- |
- |
- |
|
Gas |
33 |
Ring |
1 |
24 |
90 |
.107 |
81 |
.096 |
10.3 |
|
Gas |
65 |
Ring |
6 |
53 |
98 |
.117 |
- |
- |
- |
|
Gas |
110 |
Ring |
1 |
85 |
94 |
.112 |
- |
- |
- |
|
Gas |
225 |
Nozzle |
8 |
180 |
212 |
.252 |
147 |
.175 |
30.6 |
|
Gas |
325 |
Nozzle |
8 |
260 |
320 |
.381 |
230 |
.274 |
28.1 |
|
|
Baseline |
Low NOx |
|
||||||
|
Fuel |
Size MBH |
Burners |
Test Load MBH |
PPM @ 3% O2 |
|
PPM @ 3% O2 |
|
% reduction |
|
|
Type |
Number |
||||||||
|
#2 Oil |
110 |
Steam |
2 |
88 |
177 |
.231 |
- |
- |
- |
|
#2 Oil |
10 |
Air |
1 |
7 |
169 |
.220 |
149 |
.194 |
11.8 |
|
#2 Oil |
18 |
Steam |
1 |
14 |
65 |
.085 |
63 |
.082 |
3.5 |
|
#2 Oil |
18 |
Air |
1 |
14 |
97 |
.127 |
86 |
.112 |
11.8 |
|
#2 Oil |
18 |
Mech |
1 |
12 |
80 |
.104 |
80 |
.104 |
- |
|
#2 Oil |
11 |
Air |
1 |
11 |
128 |
.167 |
- |
- |
- |
|
#2 Oil |
18 |
Air |
1 |
16 |
116 |
.151 |
- |
- |
- |
|
#2 Oil |
18 |
Steam |
1 |
16 |
118 |
.154 |
- |
- |
- |
|
#2 Oil |
20 |
Air |
1 |
16 |
193 |
.252 |
141 |
.184 |
27.0 |
|
#2 Oil |
29 |
Steam |
1 |
10 |
103 |
.134 |
- |
- |
- |
|
#2 Oil |
7 |
Air |
1 |
7 |
127 |
.166 |
- |
- |
- |
|
#2 Oil |
158 |
Steam |
4 |
115 |
181 |
.236 |
120 |
.157 |
33.5 |
|
#2 Oil |
33 |
Steam |
1 |
23 |
123 |
.160 |
104 |
.136 |
15.0 |
|
#2 Oil |
13 |
Air |
1 |
11 |
84 |
.110 |
- |
- |
- |
|
#5 Oil |
19 |
Cup |
1 |
12 |
200 |
.261 |
139 |
.181 |
30.7 |
|
#5 Oil |
85 |
Steam |
4 |
59 |
329 |
.429 |
243 |
.317 |
26.1 |
|
#5 Oil |
11 |
Air |
1 |
12 |
183 |
.239 |
162 |
.211 |
11.7 |
|
#5 Oil |
18 |
Air |
1 |
18 |
177 |
.231 |
154 |
.201 |
13.0 |
|
#5 Oil |
18 |
Steam |
1 |
17 |
161 |
.210 |
152 |
.198 |
5.7 |
|
#5 Oil |
13 |
Air |
1 |
11 |
181 |
.236 |
171 |
.223 |
5.5 |
|
#5 Oil |
7 |
Air |
1 |
7 |
275 |
.359 |
- |
- |
- |
|
#5 Oil |
59 |
Steam |
6 |
46 |
619 |
.807 |
516 |
.673 |
16.6 |
|
#5 Oil |
65 |
Steam |
6 |
50 |
466 |
.608 |
431 |
.562 |
7.6 |
|
#5 Oil |
125 |
Steam |
1 |
100 |
337 |
.440 |
322 |
.420 |
4.5 |
|
#5 Oil +RG |
110 |
Steam |
4 |
88 |
172 |
.224 |
166 |
.217 |
3.1 |
|
#5 Oil +RG |
110 |
Steam |
4 |
90 |
215 |
.280 |
211 |
.275 |
1.8 |
|
|
Baseline |
Low NOx |
|
||||||
|
Fuel |
Size MBH |
Burners |
Test Load MBH |
PPM @ 3% O2 |
|
PPM @ 3% O2 |
|
% reduction |
|
|
Type |
Number |
||||||||
|
NSF oil |
17 |
Cup |
1 |
15 |
184 |
.240 |
- |
- |
- |
|
#6 oil |
18 |
Steam |
1 |
14 |
350 |
.457 |
331 |
.432 |
5.5 |
|
#6 oil |
18 |
Air |
1 |
15 |
334 |
.436 |
277 |
.361 |
17.2 |
|
#6 oil |
80 |
Steam |
1 |
51 |
305 |
.398 |
268 |
.350 |
12.1 |
|
#6 oil |
90 |
Steam |
3 |
71 |
246 |
.321 |
175 |
.228 |
29.0 |
|
#6 oil |
65 |
Steam |
2 |
54 |
186 |
.243 |
174 |
.227 |
6.6 |
|
#6 oil |
105 |
Steam |
4 |
80 |
251 |
.327 |
222 |
.290 |
11.3 |
|
#6 oil |
260 |
Steam |
4 |
130 |
240 |
.313 |
201 |
.262 |
16.3 |
|
#6 oil |
500 |
Steam |
9 |
400 |
267 |
.348 |
180 |
.235 |
32.5 |
|
#6 oil |
7 |
Air |
1 |
7 |
298 |
.389 |
249 |
.325 |
16.5 |
|
O/C |
513 |
Cyc |
2 |
320 |
716 |
.934 |
- |
- |
- |
|
O/C |
513 |
Cyc |
2 |
320 |
710 |
.926 |
- |
- |
- |
|
Coal |
60 |
UFS |
7 |
48 |
266 |
.327 |
188 |
.263 |
29.3 |
|
Coal |
60 |
UFS |
7 |
46 |
224 |
.314 |
198 |
.277 |
11.8 |
|
Coal |
135 |
Sprd |
2 |
110 |
370 |
.518 |
335 |
.469 |
9.5 |
|
Coal |
50 |
Sprd |
1 |
40 |
465 |
.651 |
330 |
.462 |
29.0 |
|
Coal |
75 |
Sprd |
1 |
63 |
465 |
.651 |
387 |
.542 |
16.7 |
|
Coal |
225 |
Pulv. |
8 |
181 |
378 |
.529 |
360 |
.504 |
4.7 |
|
Coal |
210 |
Sprd |
5 |
120 |
553 |
.774 |
471 |
.659 |
14.9 |
|
Coal |
230 |
Sprd |
6 |
162 |
547 |
.766 |
360 |
.504 |
34.2 |
|
Coal |
500 |
Pulv |
6 |
400 |
580 |
.812 |
- |
- |
- |
|
Coal |
513 |
Cyc |
2 |
320 |
800 |
1.120 |
742 |
1.039 |
7.2 |
|
Coal |
10 |
UFS |
1 |
8 |
273 |
.382 |
- |
- |
- |
|
Coal |
10 |
UFS |
1 |
8 |
346 |
.484 |
- |
- |
- |
|
Coal |
325 |
Pulv |
8 |
260 |
484 |
.678 |
- |
- |
- |
|
Abbreviations NSF=Navy Standard Fuel Cup=Rotary cup fuel atomizer Air=Air assist fuel atomizer Steam=Steam-fuel atomizer Cyc=Cyclone furnace coal combustor UFS=Underfed stoker coal burning equipment Sprd=Spreader stoker coal burning equipment Pulv=Pulverized coal burning equipment MBH=Million BTU per hour |
NO2/106 BTU (Cato et al.). Increasing combustion intensity resulted in increased NOx, but other changes in operating parameters had little effect.
The heavy fuel oil emissions were higher, 0.20 to 0.81 lb NO2/106 BTU, and more sensitive to the operating conditions. The nitrogen content of the fuel had a particularly strong effect since an estimated 44 percent of the fuel nitrogen was converted to NOx. NOx emissions decreased with decreasing oxygen levels, burner heat release rate (BTU/hr/burner), and combustion intensity. Emission reductions of 6 to 29 percent were obtained by using stoichiometric or staged combustion. The technique of fuel atomization did not strongly influence the emissions as long as good atomization was achieved. The two highest emission levels for No. 5 oil were the result of preheating the oil to only 130F, rather than 160 to 180F as in most other tests. Increasing the oil temperature improved the atomizer effectiveness and reduced the NOx emission levels.
Coal-Fired Boiler Emissions
Due largely to the high nitrogen content of coals (1.29 to 1.80 percent by weight in the KVB study), the emissions from coal-fired units were generally higher than those from oil or gas fired units, i.e., 0.31 to 1.12 lb NO2/106 BTU. (Cato et al.) The lowest emissions from coal-fired units were from underfed stoker coal burning equipment. In these units the air fed up through the grating is insufficient for complete combustion, so additional air must be introduced above the grating through overfire air ports. The combustion is, therefore, effectively staged, and the NOx emissions were quite low, .31 to .48 lb NO2/106 BTU.
Spreader stokers, in which the fuel is introduced with the air flow above the grate, had intermediate emission characteristics, 0.52 to 0.77 lb NO2/106 BTU. Some of the fuel is burned in the fuel spray; the remaining fuel is burned on the grate as in the underfed stoker. The
resultant combustion is only partially staged. The combustion intensities are also higher than for underfed stokers, possibly increasing thermal NO formation.
Pulverized coal units, in which all of the fuel is burned in suspension, had higher emissions, 0.53 to 0.81 lb NO2/106 BTU. A unit equipped with two cyclone-type coal combustors produced 1.12 lb NO2/106 BTU, the highest emissions of all the units tested due to the very high combustion intensity. (Cato et al.)
Reductions of about 0.07 lb NO2/106 BTU percent reduction in oxygen were obtained by reducing the amount of excess air. A decrease in underfire air-rate with a compensating introduction of air through openings higher in the furnace for a spreader stoker-fired boiler resulted in about a 46 percent NOx reduction, but the grate temperature increased beyond its allowable limits. The most acceptable operating conditions reduced NOx by 20 to 25 percent. Controlled NOx emissions were below 0.68 lb NO2/106 BTU for all but the cyclone fired boiler. (Cato et al.)
RESIDENTIAL AND COMMERCIAL SPACE HEATING
Background
Space heating equipment of less than 10 million BTU hr capacity emits about 7.1 percent of the U.S. stationary source NOx. The emissions are largely near ground level in areas of high population densities. (Hall et al. 1974). The problem of NOx emission from these sources may, in fact, be more severe than this number would indicate since the emissions are confined to the heating season.
Commercial units, which generally fall in the range of 0.3 to 10 million BTU hr capacity, include a mix of packaged firetube, cast iron, and watertube boilers. (Barrett et al. 1973). These units are equipped to burn natural gas (56 percent), number 2 oil (38 percent), number 4 and 5 oils (8 percent), and number 6 oil (4 percent). About 6 percent of the units have dual
fuel capabilities. Commercial boilers are designed to operate with a minimum of manual control and maintenance for a life of from 10 to 45 years depending upon the type of unit.
Residential units are much smaller; about 70 percent have capacities of less than 0.2 million BTU/hr, and 96 percent have capacities of less than 0.42 million BTU/hr. (Barrett et al. 1973). The units include warm air furnaces hot-water and steam boilers, and water heaters, and are fueled primarily by distillate oil (No. 2) or natural gas. Operation is generally controlled by a thermostat and maintenance is on an annual schedule, at most.
Emissions from Residential Space Heating Equipment
Specific NOx emission levels from residential units average about 0.14 lb NO2/106 BTU, (Barrett et al. 1973) which is low, primarily as a consequence of low combustion intensities in these units. About 90 percent of the residential units tested in a recent field survey (Barrett et al. 1973) of oil-fired space heating equipment had emissions below 0.2 lb NO2/ 106 BTU; the highest emission level measured in that survey was about 0.29 lb NO2/106 BTU. Normal service and adjustment practice had very little effect on the emission level of NOx or any pollutant other than smoke.
A recent EPA study (Hall et al. 1974) (USEPA 19) provides more detailed information on the effects of operating conditions and burner design on