20367
ECONOMIC IMPACT ASSESSMENT FOR THE
NATIONAL AMBIENT AIR QUALITY STANDARD FOR LEAD
Office of A1r Quality Planning and Standards
Office of Air and Waste Manaqement
U.S. Environmental Protection Aqency
November 22, 1977
-------
ECONOMIC IMPACT ASSESSMENT
Chapter 1 Summary and Introduction
1.1 Summary
1.2 Introduction
Chapter 2 Stationary Source Assessment
2.1 Primary Lead Smelting
2.1.1 Industry Structure
2.1.2 Model Plant Specifications and Dispersion Model Results
2.1.3 Control Costs
2.1/4 Model Plant Closure Assessment
2.2 Secondary Lead Smelting
2.3 Primary Copper Smelting
2.4 Grey Iron Foundry Casting
2.5 Gasoline Lead Additives Manufacturing
2.6 Lead-Acid Battery Manufacturing
Chapter 3 Other Affected Sectors
3.1 Mobile Source Assessment
3.2 State and Local Control Agency Assessment
-------
1.0 SUMMARY AND INTRODUCTION
1.1 Summary
The purpose of this analysis is to estimate the economic impact
which can be expected from a constant air quality for lead. This analysis
examines three levels which are deemed to fall within the likely range of
the final standard. It should be noted, however, that this economic assess-
ment is not a basis for selecting the standard 'since the Clean Air Act
requires that the standard be based solely on health and welfare criteria.
This assessment considers three possible standards: 1.0, 1.5, and 2.0
(monthly average). The economic impact assessment emphasizes the impacts
upon stationary source lead emitters but also discusses the impacts upon
mobile source lead emitters and state and local control agencies
charged with implementing the standard.
Dispersion models indicate that plants in at least six industries may
be required to install control devices to meet the alternative standards
under consideration. These control devices would be in addition to those
control systems required by typical state regulations for control of particulate
or other emissions. The six industries under consideration are primary lead
smelting, secondary lead smelting, primary copper smelting, grey iron foundries,
gasoline lead additive manufacturing, and lead-acid storage battery manufacturing.
The economic impact assessment is based primarily upon the use of model
plants and estimated emission factors. The model plant emissions were used in
a meteorological dispersion model that predicts maximum ambient lead concen-
trations. The dispersion modeling results were then used to estimate which
emission sources needed additional control to meet the alternative ambient
standards and the extent of control required. The resulting control
requirement then determined the type of control equipment needed and
the cost of the equipment. Control costs to meet alternative standards were
1
-------
then factored into a discounted cash flow model that served as the basis for
evaluating potential plant closures.
As expected, the results of the economic impact analysis indicate
that model plants of the six industries mentioned above will be affected
in varying degrees by the alternative standards under consideration. As
shown in Table 1-1, annual1zed compliance costs expressed as a percent
of model plant revenues range from zero to over 7 percent depending upon
the level of the standard and the type of model plant being analyzed.
It is possible that some plants, facing control costs of the magnitude
shown in Table 1-1, may choose to close rather than comply with emission
regulations required to achieve a given ambient lead standard. An
analysis of this issue shows that, for lead air quality standards of
either T.5 or 2.0 yg/ra3, potential plant closures may be possible in the
» *
primary lead industry, the secondary lead industry, and the primary
copper industry. In addition, at an ambient standard of 1.0 ug/m plant
closures may he possible for some grey iron foundries. Plant closures are
not projected for gasoline lead additive plants or lead-acid battery
plants for any of the alternative standards under consideration.
The economic impacts described above are based upon a number of factors
such as emission rates, plant profit margins, and terrain and weather conditions
in the vicinity of the source. Since these variables are difficult to quantify
with any reasonable degree of-precision, it must be borne in mind that impacts
at any one specific source may vary considerably from the impacts described here.
It should be noted that a rollback analysis (an analysis that assumes a linear
relationship between the percent of air quality improvement and emission
reductions) indicated that only two industries— primary lead smelting and
-------
Table 1-1. RELATIVE COSTS OF ALTERNATIVE AMBIENT LEAD STANDARDS - MODEL
PLANT ANNUALIZED COSTS AS A PERCENT OF ANNUAL REVENUE
Alternative Ambient Standards
(yg/m3 monthly average)
Model Plant Type
Primary Lead
Secondary Lead
Primary Copper
Grey Iron
Gasoline Lead Additives
Lead Add Batteries
2.0
0.9% to 7.7%*
7.4%**
0.0% to 3.3%
0.0% to 2.0%
0.2%
0.0% to 0.1%
1.5
1.2% to 7.7%*
7.4%**
0.0% to 4.3%
0.0% to 3.8%
0.2%
0.0% to 0.1%
1.0
1.8% to 7.
7%*
7.4%**
0.0% to 5.
0.0% to 6.
0.2%
0.0% to 0.
2%
2%
1%
*The control efficiency of the costed system 1s assumed to be 95%. At this level of efficiency the model primary
lead smelter with high fugitive emissions could only attain an air quality standard of 3.9 ng/m3.
**The model secondary lead smelter with low fugitive emissions could meet a standard of 2.0 ug/m3 with an expen-
diture of 7.4% using the 95% control system referred to above. A model secondary lead smelter with high
fugitive emissions, however, could only meet an air quality level of 3.7 ng/m3 with the same
-------
primary copper smelting—would need to install control devices to meet the
alternative standards. 'This result is at variance with the dispersion analysis
upon which the stationary source economic impact assessment was based that
indicated six industries would need to install control devices. This fact
emphasizes the effect of various modeling assumptions upon the results of
the economic impact analysis as well as the general problem of the variability
of the results depending upon various input assumptions.
An analysis of potential impacts to mobile source lead emitters for the
ambient lead standards under consideration has also been developed. This
analysis indicates that mobile sources in portions of one to four Air
Quality Control Regions, or a possible total of 58,000 to 1,300,000 vehicles,
may have to be controlled, assuming the ambient lead standard must be achieved
in 1982. However, because of lead phasedown regulations and the increased
use of lead free gasoline for catalyst equipped vehicles, the total number of
mobile source lead emitters is expected to decrease after 1982. One control
device that may be feasible for mobile sources is a lead trap muffler. Other
means of control include reducing vehicle miles traveled and further reducing
the lead content of gasoline.
State and local air pollution control agencies will also incur costs
to develop and implement plans to achieve an ambient air quality standard for
lead. Total first year costs for all state and local agencies are estimated
to range from $1.0 - $1.7 million, or 0.6 - 1.0 percent of current expenditures.
Recurring costs are estimated to be $1.4 - $2.8 million, or 0.9 - 1.8 percent
of current expenditures. Man-year requirements, both first-year and recurring,
are similarly estimated to be less than 2 percent of current levels.
-------
1.2 INTRODUCTION
This report assesses the cost and economic impact of alternative
ambient lead standards. The stationary sources covered in the assessment
are model primary lead smelters, secondary lead smelters, gasoline lead
additive manufacturing plants, lead-acid battery manufacturing plants,
primary copper smelters, and grey iron foundries. In addition, the
potential costs of mobile source emission control and of requisite state
and local control agency information, administration, and enforcement
activities for alternative standards are also estimated. The alternative
standards considered are 2.0, 1.5, and 1.0 ug/m , monthly average. The
detailed methodology and documentation of the analysis are provided in
the report entitled "Background Document Supporting the Economic Impact
Assessment of the Lead Ambient Air Quality Standard".
1.2.1 Reasons for Selection of Sources for Consideration
In 1975, about 142 thousand metric tons of lead were emitted nationwide.
Combustion of lead containing gasoline accounted for 90% of those emissions.
Combustion of waste crankcase oil, solid waste, oil, and coal accounted for
an additional 5% of national emissions in 1975. The remaining 5% came from
19 industrial stationary source types. As a result of phasedown of lead
in gasoline, lead emissions from gasoline combustion are expected to
2
decrease about 60% by 1985 from current levels. Although this is a large
relative decline, gasoline combustion emissions are still projected to
be the greatest emission source nationally in 1985. Because of this,
-------
gasoline combustion sources (mobile sources) are included in the economic
impact assessment.
Combustion of waste crankcase oil, solid waste, oil, and coal is not
o
considered in the economic impact assessment. Combustion of waste crankcase
oil is nqt considered because the phasedown of lead in gasoline will cause
combustion of waste crankcase oil to cease to be a major source. Combustion
of solid waste is not considered in the analysis since many solid waste
combustion facilities (i.e., municipal incinerators) are scheduled to be closed.
Therefore, lead emissions from this source category will be substantially
reduced. Finally, sources burning oil and coal are not considered since
dispersion modeling and ambient data analysis near these sources have indicated
that oil and coal combustion will probably not result in violations of any of
the alternative ambient standards under consideration.
Growth projections as well as analyses of estimated emissions and measured
ambient impacts were developed for the 19 industrial stationary source types
previously mentioned. As a result, 8 of the 19 were identified as sources
4
probably requiring additional emission control. These 8 industrial source
types are primary lead smelters, secondary lead smelters, primary copper
smelters, gasoline lead additives manufacturing plants, lead-acid battery
manufacturing plants, grey iron foundries, ferroalloy plants, and lead ore
crushing and grinding plants. The economic impact assessment includes the
first six of the aforementioned source types. Ferroalloy plants are not
considered in the economic impact assessment because dispersion modeling
indicated that emissions from a model ferroalloy plant in zero background-isolated
-------
source situations should not result in violations of any of the alternative
standards under consideration. Ore crushing and grinding plants are not
considered in the economic impact analysis because of the inability to
adequately measure fugitive emissions from these sources and also to predict
the ambient impact of these emissions. These fugitive emissions are a
function of particle density and wind speed. The density data is not
currently available nor is a dispersion model which handles windblown
emissions.
One assumption which tends to understate the economic impact of any
given ambient lead standard is that each model plant is an isolated source
with no background ambient levels of lead present. To the extent that
sources of'lead emissions may be clustered, the combined ambient lead
concentrations of the cluster result in higher control costs and impacts
than assumed.
In addition, even if the isolated source assumption were valid, there
are other factors that influence the results of the economic impact analysis.
These include topographical and meteorological conditions, stack characteristics,
particulate emission factors, lead content and particle size of particulate,
fugitive emissions factors, and baseline process economics. Values for many
of these factors are initially specified as range estimates with midpoints
(arithmetic means) of the ranges used in the model plant analysis. For other
factors, values used in the model plant analysis are derived from site specific
measurements. However, the degree of error in the measurement per se or in the
application of the measured value to the model plant is not well known.
-------
Variations in these data inputs (factors) among plants and source
'pes limit the accuracy of the economic impact assessment. However,
le assessment does explicitly address some variations via a reasonable
mge analysis for certain data inputs.
.2.2 Methodology
.2.2.1 Stationary Source Assessment Methodology
Plant, dispersion, control cost, and discounted cash flow models provide
fie bases for the economic impact assessment for the selected stationary sources
utputs of the plant models include emission, size, process, and location
haracteristies. These are inputs to the dispersion model. The dispersion
ode! provides maximum predicted concentrations and source contribution file
stimates. The latter relate point and fugitive emissions generated by the
lant to maximum predicted ambien-t concentrations. The outputs of the disper-
ion model are used together with control systems engineering and cost data
o produce estimates of investment and annualized control costs. These
stimates are the outputs of the cost model. They are used together with
rocess economic data in the discounted cash flow models to produce a numerical
stimate of the value of a plant after control. If these calculations show a
lant is worth more closed than it is open, closure is predicted.
In situations where all control costs could be passed on without any
ffect on production levels, closure would never be predicted using the
8
-------
forementioned methodology. However, the ability of an individual plant to
ass costs on and sustain pre-control production levels becomes less likely
s alternatives to accepting the cost pass on become available to his
ustomers or raw materials suppliers. The assumptions of the economic impact
ssessment and the competitive structure of the industries analyzed imply
any alternatives to accepting a cost pass on. Consequently, no cost pass-on
s considered in the closure analysis.
.2.2.2 Mobile Source Assessment Methodology
A 1975 source emissions inventory and projected mobile and stationary
ource growth rates are used to estimate a 1982 lead emissions inventory for
:ach Air Quality Control Region. In a similar manner, ambient concentrations
re rolled forward. If the alternative ambient standards are predicted to be
txceeded, the 1982 source emissions inventory and ambient concentrations are
tilled back so that the standard is achieved. Of course, several different
ombinations of mobile and stationary source control can achieve the same
oil back. In this analysis mobile source control is assumed used as a last
esort and then with emission reduction effectiveness limited to 75%.
The 1982 stationary source emissions inventory used in the mobile source
ssessment includes eleven types of process sources. These are primary lead
melting, secondary lead smelting, primary copper smelting, grey iron production,
asoline lead additives production, lead-acid storage battery production,
erroalloy production, coal-fired power generation, oil-fired power generation,
olid waste incineration, and iron and steel production. It is important to
ote that only primary lead smelting and primary copper smelting were projected
o require additional rollback in 1982. This runs counter to the dispersion
-------
model findings o'f the stationary source assessment which identify the first 6
of the 11 aforementioned sources as requiring further control, This apparent
inconsistency is a function of the different methodologies employed and serves
to underscore another source of variability. Moreover, like data variability,
this inconsistency limits our ability to make accurate judgements regarding impact.
1.2.2.3 State and Local Control Agency Assessment Methodology
The state and local control agency assessment builds on the stationary
and mobile source assessments. Control requirements for the stationary source
model plant assessments for the three alternative standards are aggregated to a
national level and used as inputs to EPA's Air Pollution Strategy Resource
Estimator (APSRE). APSRE then provides an estimate of additional stationary
source related state and local control agency needs. Additional resource
requirements resulting from mobile source control are developed as supplementary
calculations using the findings of the mobile source assessment.
10
-------
2.1 PRIMARY LEAD SMELTING
2.1.1 Industry Structure
Six primary smelters owned by four corporations comprise the domestic
industry. Company names, the plant capacities and locations are given
in Table 2-1.
The four corporations smelting lead are both vertically and horizontally
integrated. For example, they mine lead ore and refine smelted lead. In
addition, they mine, smelt, and/or refine zinc, silver, coal, molybdenum,
and copper. The dependence that these corporations have on lead varies,
ranging from 6.U of revenues for AMAX to 17.955 for Gulf Resources and Chemical.
Lead is an intermediate good. As such, the demand for lead is derived
from the demand from lead-using products. These include batteries, gasoline
lead additives, electrical cables and sheathing, paints, sheeting, plumbing,
and ammunition. Through the year 2000, U.S. lead demand is projected to grow
at 1.5% per year.
Lead is a small percent of end product value and has few close substitutes.
Therefore, its .demand is charactericterized as price inelastic. This means
that when lead prices increase, total revenue (price times quantity sold) increases.
Price inelastic demand notwithstanding, domestic primary smelters still
could face competitive pressures from other lead suppliers. For example,
unilateral price increases by domestic primary smelters could foster increased
competition from domestic secondary producers. Also, unilateral price increases
by domestic primary smelters could also foster increased competition from foreign
producers. As indicated in Table 2-2, foreign producers' prices
11
-------
Table 2-1. U. S. PRIMARY LEAD SMELTERS
Company
Amax
ASARCO
Gulf Resources
& Chemical
St. Joe Minerals
Corp.
Plant(s) Locations)
Boss, Missouri
East Helena, Montana
El Paso, Texas
Glover, Missouri
Kellogg, Idaho
Herculaneum, Missouri
1974 Capacity
(thousands of metric tons)
127
82
82
100
118
204
713
Source: U.S. EPA, October 1974. Background Information for New Source Performance Standards: Primary
Copper, Zinc, and Lead Smelters; Volume 1; Proposed Standards.Bureau of Mines Minerals Fact;
and Problems. 1975.
-------
Table z-z. • -LEAD PRICES
(Average annual price, cents per kilogram)
Year
1964
1965
1966
1967
_, 1968
CJ
1969
1970
1971
197Z
1973
1974
1975
1976
London Metals Exchange
27.8
31.8
26.3
22.7
24.1.
28.9
30.5
25.4
30.2
43.0
59.2
41.5
45.3
Mineral Facts and Problems, 1975.
Minerals and Materials, December, 1976.
New York
30.0
35.3
33.6
30.9
29.1
32.9
34.7
30.7
33.1
36.0
49.7
47.5
51.0
Bureau of Mines, Minerals Yearbook, 1970, 1972, 1974.
-------
(London Metals Exchange) have been consistently less than domestic
producers' prices (New York Exchange). Anticipated reaction by foreign
producers to domestic primary smelter price increases may determine
whether domestic smelters elect to absorb long run cost increases or
push them forward as price increases.
2.1.2 Model Plant Specifications and Dispersion Model Results
The model plant specifications used in the economic impact assessment
are primarily dependent on the assumptions and data requirements of the
dispersion model. The dispersion model employed in the assessment is
the Single Source (CRSTER) Model. It is a steady state Gaussian
plume dispersion model. CRSTER will be recommended to the states to
develop lead emission regulations for isolated stationary sources.
Flat terrain is a basic assumption inherent in the CRSTER dispersion
model, and hence is one of the model primary lead smelter's specifications.
The flat terrain assumption may be important in predicting the distance
of the maximum concentration from plant. The maximum concentration for
the model primary lead smelter is predicted to be 300 meters away from
the plant. With rugged terrain, the maximum concentration could be expected
to be closer to the plant. A review of topographical maps for all six
smelters indicates elevations greater than the smelter site within 300
meters of the* plant. Consequently, the maximum concentration may occur
closer to the plant than 300 meters.
14
-------
Data requirements of the dispersion model include meteorological
iditions such as wind speed, wind direction, and ambient temperature.
;h data is not always available for every plant location. St. Louis
teorological data is available and is used in the analysis, since three
!the six smelters are located near St. Louis.
Data requirements of the dispersion model also include emission
aracteristics such as stack gas exit velocity and temperature as well
5 point and fugitive emission rates and release heights. These emission
Aracteristics are sometimes related to plant size. The production rate
ed in the dispersion modeling is 62 thousand metric tons annual production.
is size is smaller than the current production rates of 4 of the 6
elters. However, this size was chosen because fugitive emission
asurements were available for a plant of that size. Furthermore, because
ission characteristics are dependent on factors other than size, scaling
the fugitive emission measurements of the model plant to be consistent
th an average plant size could provide atypical results.
The model smelter is assumed to have both stack and fugitive emissions
lead particulate. Stack emissions are assumed controlled to average
P particulate allowable process weight rates. The fugitive emissions
•e assumed uncontrolled. These emissions come from the sinter machine
hiding, blast furnace building, reverberatory furnace building, zinc
aning area, and the zinc furnace building.
Stack emissions from the model smelters have a negligible effect on
•edicted maximum ambient lead concentrations. Moreover, higher stack emission
ites within a range thought to be reasonable do not change this finding.
15
-------
gitive emissions have the predominent impact on predicted ambient
ncentrations, and this impact appears to vary significantly from
elter to smelter. For example, fugitive emission rates derived from
asurements at a Montana primary lead smelter result in a maximum
edicted concentration for the model smelter of 3.8 yg/m , monthly
erage. Fugitive emission rates derived from measurements at an Idaho
alter are higher and result in a maximum predicted concentration before
3
rttrol of 78.2 ug/m , monthly average.
1.3 Control Costs <
1.3.1 Model Plant
The alternative ambient standards considered here are 2.0, 1.5, and
D pg/tn , monthly average. The predicted ambient concentrations
om both sets of fugitive emission rates mentioned above are greater than
a alternative standards. Hence, control costs corresponding to reduced
gitive emission rates which achieve the alternative ambient standards should
developed for the model primary lead smelter. Control costs representing
» least costly means of achieving the three standards were developed for
» lower set of derived fugitive emission rate. For the higher set, control
;ts corresponding to 95% control efficiency were developed. The 95% estimate
an engineering judgement based on the best demonstrated control system
•rently available (building evacuation to a fabric filter). However, 95%
itrol applied to the higher set of fugitive emission rates still results in
•redicted maximum concentration of 3.9 ug/m3 which exceeds any of the alternative
ndards. The control efficiencies required to get to 2.0, 1.5, and 1.0
m3, respectively, are estimated to be 97.4%, 98.1%, and 98.7% as compared
h 95Z which is judged to be the maximum attainable.
16
-------
The absolute and relative magnitude (as a percent of annual revenue)
of the developed investment and annual ized costs are presented in Table 2-3.
For the lower fugitive emission rates, relative annualized control costs range
from 0.9% to meet a standard of 2.0 vg/m3 to 1.8% to meet a standard of
1.0 ug/m3- For the higher fugitive emission rates, relative annualized costs
are 7.7% to meet an ambient level of 3.9
2.1.3.2 Industry
The most critical factor in the model plant control cost assessment is
the fugitive emission rates. As indicated previously, these rates have been
derived from fugitive emission measurements at two of the six currently
existing primary lead smelters. However, site specific topographical,
meterological , and smelter size, configuration, and emission rates data are
required to assess the ambient and cost impact at these six smelters with
reasonable certainty. Assuming all primary lead smelters have the same site
specific characteristics as the model smelter, five of the six existing
smelters would probably have low fugitive emission rates and hence have
ambient impacts closer to 3.8 than to 78.2 ug/m3, monthly average. For
three (the Missouri smelters) of the five smelters this judgement is based
primarily on the presumption that since these smelters are newer they are
better controlled. For the other two of the five smelters (Montana and Texas),
"similarity to the hypothetical smelter in terms of size, lower measured
fugitive emissions, and/or lower measured ambient concentrations is the
basis for the Judgement. Higher measured fugitive emission and ambient
concentrations are the reasons for classifying the Idaho smelter (the
sixth smelter) as having ambient impacts closer to 78.2
17
-------
Table 2-3. CONTROL COSTS FOR THE MODEL PRIMARY LEAD SMELTER
Maximum Predicted Ambient Levels
(yg/m3 , monthly average)
Higher Fugitive
Emission Rates
3.9
2.0
1.5
1.0
Investment
OOO's of $
as a % of
annual revenue
10800
34.455
Annualized Cost
OOO's of $'s
as a % of
annual revenue
2400
7.7%
Lower Fugitive
Emission Rates
Investment
OOO's of S's
as a % of
annual revenue
dualized Cost
OOO's of $'s
as a % of
annual revenue
0
0
0
0
1300
4.0%
300
U.9X
1600
5.2%
400
1.2%
24UU
7.5%
600
1.8%
*The control efficiency of the best demonstrated control system is limited
according to engineering judgement to 95%. This is not sufficient to achieve
the alternative ambient standards for the model plant with higher fugitive
emission rates.
18
-------
To draw further inferences from the model plant assessment Is tenuous.
However, assuming production is proportional to the number of smelters
in each category and the model smelter costs can be extrapolated linearly to
higher production volumes, industry control cost estimates can be developed.
For 83.3% of the industry (current production share at the five plants
assumed to have low fugitive emission rates), the respective investment costs
o
in 1982 for meeting the 2.0, 1.5, and 1.0 ug/m standards are $11.9 million,
$15.3 million, and $22.4 million. The corresponding annualized costs are $2.8
million, $3.5 million and $5.3 million. For the remaining 16.7% of the industry
(current production share of the other smelter), control cost estimates are
those corresponding to 95% control efficiency. They are $20.5 million for
investment and $4.6 million for annualized costs.
2.1.4 Model Plant Closure Assessment
Using a discounted cash flow analysis technique, synthesized model plant
process economics, the aforementioned control costs, and assuming no other
lead emitters in the vicinity, the potential for closing the smelter on
financial grounds is assessed. Process economics (e.g., revenues, costs) were
developed using Bureau of Mines data as well as financial data from specific
companies.
Given the lower set of derived fugitive emission estimates, the model
smelter should not close regardless of the level of the standard. This finding
is true for several sets of circumstances. They include a range of marginal
tax rates and minimum acceptable return rates thought to be reasonable, full
19
-------
absorption of control costs, low profit margin and fully depreciated smelter,
and full absorption of operation and maintenance cost required for the proposed
Occupation Safety and Health Administration Lead Standard (100 ug/ro3, 40 hour
time weighted average and 60 ug/100 g whole blood).
Given the higher set of derived fugitive emission estimates, the ability
of the model smelter to achieve maximum (9558) control of fugitive emissions
(assuming this amount of control is satisfactory) and remain open is unclear.
The critical factors among previously described set of circumstances are the
marginal tax rates and the minimum acceptable return rates. Marginal tax rates
are the rates applicable to plant and not, for example, to the parent firm.
Minimum acceptable return rates are the profits available from the next best
investment opportunity. If the minimum acceptable return rates are not
realized, the plant will close; and, the next best investment opportunity will
be capitalized. If the tax and return rates are on the high end of the range
thought to be reasonable, it would be in the best financial interest of model
smelter to close. If the rates are lower, the model smelter with higher
fugitive emission rates should make the expenditures to achieve 95% control
and remain open.
20
-------
2.2 SECONDARY LEAD SMELTING
2.2.1 Industry Structure
The secondary lead smelting industry is a subset of the secondary lead
industry. The latter includes melters as well as smelters. Unfortunately,
industry structure statistics are only available for the secondary lead industry.
The secondary lead industry in the United States supplied 545,000 metric
tons or 39 percent of total lead consumption in 1974. Approximately 90
companies operating 130 plants produce lead and lead alloys for industrial
use from recycled materials, principally old batteries.8 Two companies, NL
Industries, Inc., and RSR Corporation, operating about 18 secondary plants,
account for over 50 percent of the total secondary lead production.9 Thirteen
other companies operating approximately 24 plants that manufacture storage
batteries and other metal products account for 45 percent of secondary lead
production.
Roughly two-thirds of all lead produced at secondary smelters is
antiroonial lead and goes into the manufacture of batteries. Therefore, demand
for secondary lead in the future is tied to growth in battery use. While new,
longer lasting batteries are becoming more and more popular, demand for
replacement batteries is still expected to be 35 percent of total demand each
year until the end of the decade.
Prices are a critical factor in determining the supply of secondary lead.
Primary lead prices affect secondary lead prices in a direct way. To the
extent that the secondary industry acts as a broker, any change in the price
of primary lead will be reflected in the price of scrap. Secondary producers,
21
-------
therefore, could expand or contract their collection effort depending on
changes in the price of lead. Although it may vary above and below the
primary lead price, the secondary soft lead price follows the average prices
quoted in Metals Week. Historical primary lead prices are provided in
Table 2-2.
2.2.2 Model Plant Specifications and Dispersion Model Results
To provide reasonably accurate predictions of ambient impact, the model
plant specifications should be consistent with the assumptions and data
requirements of the dispersion model. CRSTER, the dispersion model employed
in the assessment, is designed to predict ambient impacts of emissions from a
plant located in flat terrain with meteorological conditions representative
of the area. To avoid modeling an atypical situation, an actual smelter located
In flat terrain with local meteorological data available is the plant modeled.
The smelter produces about 30 metric tons of lead a day. However, this is
somewhat low when compared to the midpoint of the range of plant sizes specified
in the Control Techniques Document worksheets. The range specified there is
18 to 68 metric tons per day with the midpoint being 43 metric tons per day.
But the actual size distribution of the secondary lead smelters is unknown.
Horeover, an explicit attempt is made to avoid modeling atypical situations.
Hence, the size of the model plant is not adjusted upwards to reflect the
m'dpoint of a size range.
The model smelter has lead particulate stack emissions controlled to
iverage SIP process weight rates. These emissions have a negligible impact
m the maximum predicted ambient concentrations. In addition there are
22
-------
fugitive emissions assumed for the model smelter. These emissions are
uncontrolled and result in a maximum predicted concentration of 56.6 yg/m ,
monthly average for the midpoint (arithmetic mean) fugitive emission estimate.
For a low fugitive emission estimate the maximum predicted concentration is
33.2 ug/m3. For a high fugitive emission estimate the maximum predicted
concentration 1s 80.5 yg/m . The maximum predicted concentrations for all
fugitive emission estimates are extremely sensitive to the assumed release
height of the fugitive emissions. For example, increasing the release height
by 5 meters for the midpoint emission estimate reduces the maximum predicted
concentration from 56.6 ug/m3 to 20.0 vg/m3. However, the release height
assumed originally (10 meters) is thought to be typical of secondary lead
snelting release heights.
If the release height is typical and best demonstrated control cannot
achieve the standard, another consideration might be land acquisition. The
predicted maximum of 56.6 vg/m3 occurs 150 meters from the plant. If legally
feasible and cost effective, the plant may supplement the fugitive emission
control system by purchasing surrounding land which has ambient concentrations
greater than the standard. Under such a strategy, and with barriers to limit
access to these areas, the public would still be protected from the adverse
consequences of concentrations exceeding the standard. However, the feasibility
of land acquisition as a control strategy is beyond the scope of this assessment.
23
-------
2.2.3 Control Costs
As previously noted, the maximum predicted concentrations of 33.2 to
80.5 ug/m , monthly average exceed the considered alternative ambient standards
for all sets of fugitive emission estimates. Consequently, fugitive emissions
will have to be reduced to some degree at the model smelter regardless of the
level of the standard. Building evacuation to a fabric filter has been applied
to secondary lead smelters before. However, the achieved level of control
efficiency is not known. At present, 95% control efficiency seems to be
the limit applied by engineering judgement on this technically demonstrated
system.
Applying a 94% or 95% efficient control system to the model smelter is estimated
to require an investment of $1.8 million which is about 32% of annual revenue.
Corresponding annualized cost is $0.4 million which is about 7% of annual revenue.
A 94Z or 95% control efficiency will achieve the 2.0 yg/m standard for the
low fugitive emission estimates, but will not achieve any of the more
restrictive alternative ambient standards. Furthermore, 95% control efficiency
will not achieve any of the considered alternative ambient standards given the
dispersion modeling results for the midpoint and high fugitive emission rate
estimates. With the midpoint fugitive emission rate estimate, 95% control
efficiency results in a predicted maximum of 2.8 yg/m . With the high fugitive
emission rate estimate, 95% control results in a predicted maximum of
4.0 wg/m3.
2.2.4 Model Plant Closure Assessment
The model plant closure assessment for secondary lead smelting includes
24
-------
an analysis of high, average, and low profit margin smelters. The process
economics (e.g., revenues, costs) for these profit margins are developed using
publicly available financial data on NL Industries and RSR.
High profit margin model smelters with plants that are not fully depre-
ciated should be able to absorb the costs associated with the 94% or 95%
efficiency system and remain open. This finding holds within a range of
wrglnal tax rates and minimum acceptable return rates thought to be reasonable.
However, high profit margin model smelters that are fully depreciated will only
be able'to remain open under similar conditions if they have relatively low
marginal tax rates and minimum acceptable return rates.
Average profit margin model smelters not having fully depreciated plants
should be able to absorb the costs associated with the 94% or 95% efficiency
system and remain open under most conditions. However, with a 94% efficiency
system, this model smelter will close if the marginal tax rates and minimum
acceptable return rates are reasonably high. With a 95% efficiency system,
this model smelter will close if the minimum acceptable return rate is
reasonably high and the marginal tax rate is within a range thought to be
reasonable.
Low profit margin model smelters regardless of the depreciation circumstances
and average profit margin model smelters that are fully depreciated would
probably close rather than absorb the cost of a 94% or 95% efficiency system.
25
-------
2.3 PRIMARY COPPER SMELTING
2.3.1 Industry Structure
Eight companies with a combined total of 16 smelters make up the
U.S. primary copper smelting industry. Table 2-4 lists the companies,
plants, their locations, and capacities.
Smelting is an intermediate production process proceeding refining and
fabrication and following mining, ore benefielation, and/or scrap collection.
Consequently, the demand for smelted copper is derived from the demand for
refined and fabricated copper. Refined and fabricated copper is used primarily
by the construction, communications, and motor vehicle manufacturing industries.
Substitutes for refined and fabricated copper do exist. They include aluminum,
steel, and plastic.
Smelted copper can be supplied by secondary or primary producers either
located in the U.S. or elsewhere. Foreign producers supplied about 10% of
the 1974 U.S. demand for copper.11 Domestic producers on the other hand used
12
tout 4% of the 1974 production to satisfy foreign demand for copper.
Of the domestic production used to satisfy domestic demand in 1974,
about 45% was supplied by domestic secondary producers with the remaining 55%
teing supplied by primary producers.
Historical prices for domestic producers refined copper are presented
in Table 2-5.
The demand for domestic primary smelting output is projected to grow
it ft per year.13 However, limited domestic smelter capacity could constrain
this growth resulting in excess demand and upward price pressures for blister,
refined, and fabricated copper. Given future excess demand and upward price
26
-------
Table 2-4. PRIMARY COPPER SMELTERS
ro
Company
The Anaconda Company
ASARCO
Cities Service Corp.
Copper Range
Inspirational Consolidated
Copper
Kennecott Copper Corp.
Newmont Mining Corp.
Phelps Dodge
Smelter Location
Anaconda, Montana
El Paso, Texas
Hayden, Arizona
Tacoma, Washington
Copper Hill, Tenn.
White P1ne»M1ch1gan
Miami, Arizona
Hayden, Arizona
Hurley, New Mexico
McGIll, Nevado
Garfleld, Utah
San Manuel, Arizona
Ajo, Arizona
Douglas, Arizona
Morencl, Arizona
HI1dago, New Mexico
1974
Smelter Capacity Furnace Charge
(metric tons/yr)
680.000
523,000
871,000
544,000
68,000
82.0003
408,000
381,000
363,000
363,000
907,000
726,000
227,000
635,000
816,000
91,000
Measured as copper product.
Source: Background Information for New Source Performance Standards; Primary Copper, Z1nc, and Lead
Smelters - Volume I - Proposed Standards.U.S. EPA, Document Nol EPA-450/2-74-002a, October
1974, p. 6-3.Also ADL estimates for the EPA-MBO Study, forthcoming 1977.
-------
Table 2-5. COPPER PRICES
(average annual price, cents per kilogram)
F.O.B. Domestic Primary
Year Producer Refined Price*
1964 70.5
1965 77.2
1966 79.8
1967 84.2
1968 92.2
1969 104.7
1970 127.2
1971 113.3
1972 111.6
1973 129-9
1974 168.9
1975 140-0
1976 151.7
aSource: Metals Week
28
-------
pressures, domestic smelters will probably find it better to push forward
any increases in production cost. The alternative, pushing the costs back
to the ore and matte suppliers, could spell the loss of critical raw materials
from marginal ore and matte suppliers. However, if only one domestic smelter
Is faced with production cost increases, it may find it difficult to pass
them forward because of dependence on other domestic smelters following
its lead. Passing production costs forward becomes much more plausible,
however, if all or several domestic smelters incur similar production
cost increases.
2.3.2 Model Plant Specifications and Dispersion Modeling Results
As mentioned in the primary and secondary lead smelting assessments,
the requirements of the dispersion model affect the specifications of the
ndel plant. Terrain and meteorological conditions are the two major
Influences. In essence, the model plant should be located in flat terrain
irith nearby meteorological condition data available. Furthermore, to be
typical, the size and location of the model plant should correspond to an
actual plant.
The model smelter is located 1n flat terrain, has Tucson, Arizona
•rteorological conditions, and has furnace charge capacity of 635,000 metric
Urns per year. These specifications do correspond to an actual plant. More-
over, some of the other existing smelters do have similar characteristics.
For example, 5 of the 16 are located in flat terrain and 6 have meteorological
conditions similar to Tucson, Arizona.
29
-------
The charge rate of the model smelter is high when compared to the
industry average of 500,000 metric tons per year. However, no existing
Belters of that charge rate are located in flat terrain with nearby
leteorological condition data available. Consequently, the model smelter
size was not scaled up to the industry average.
Other model plant specifications Include the process equipment. These
specifications are for an actual plant with the aforementioned terrain,
rteorology, and size characteristics. The model plant has a reverberatory
Kiting furnace, Fierce-Smith converters, and multiple hearth roasters. Of
the 16 existing smelters, 12 have reverberatory smelting furnaces and 15
save Pierce-Smith converters. However, only four have multiple hearth roasters.
Ite others have either fluidized-bed roasters or no roasters at all. The
iffect of process equipment variations among smelters on ambient air quality
is presently unknown.
The model smelter is assumed to have both stack and fugitive emissions of
lead particulate. Stack emissions are assumed controlled to average SIP
articulate allowable process weight rates. The fugitive emissions are assumed
^controlled and to emanate from the roaster, reverberatory furnace, and
anverter buildings.
Stack emissions have a negligible predicted impact on ambient lead
toncentrations. Moreover, higher emission rates within a range thought to
te reasonable do not change this finding. Fugitive emissions do have a
wtlceable predicted ambient impact. However, fugitive emission estimates
ire dependent on the percent lead content of the materials handled. This
30
-------
percentage varies among smelters. The fugitive emission estimates felt to
be most typical of a middle estimate result in a predicted maximum ambient
concentration of 3.1 yg/m , monthly average. The fugitive emission estimates
felt to represent a reasonable lower limit on the percent lead content yield
a predicted maximum ambient concentration of 0.4 ug/m3, and, estimates for a
reasonable higher limit result in a maximum predicted concentration of
10.7 ug/m , monthly average.
2.3.3 Control Costs
2.3.3.1 Model Plant
As indicated above, the monthly maximum predicted ambient concentration
for the middle and reasonable higher limit fugitive emission estimates exceed the
alternative ambient standards (2.0, 1.5, and 1.0 yg/m3) at the model smelter.
Consequently, control costs corresponding to reduced fugitive emission rates
nhich will achieve the alternative ambient standards are developed. These
costs are specifically designed to approximate the least cost means of
achieving each standard. No costs are developed for the reasonable lower
limit fugitive emission estimates since the predicted maximum concentration
in that case is less than all considered alternative standards.
The investment, annualized control cost, and investment and annualized
cost as a percent of annual revenue for the model smelter are presented in
Table 2-6. The middle fugitive emission estimates investment costs range from
$5.2 million for the 2.0 yg/m3 standard to $9.8 million for the 1.0 ug/m3
standard. The corresponding annualized costs range from $1.1 million to
$2.1 million. With the reasonable higher limit fugitive emission estimates,
31
-------
Table 2-6. CONTROL COSTS FOR THE MODEL PRIMARY SMELTER
arable Higher Limit
Itlve Emission Rates
«tnent
I's of $'s
n % of annual revenue
ilized Cost
I's of $'s
l a % of annual revenue
Maximum Predicted Ambient Levels
(jfl/m3 monthly averaqe)
2.0
22,300
15.2
4,800
3.3
1.5
28,900
19.7
6,200
4.3
1.0
35,400
24.2
7,600
5.2
le Fugitive Emission Rates
ttnent
I's of $'s
11 X of annual revenue
«lized Cost
I's of $'s
li Z of annual revenue
5,200
3.5
1,100
0.8
7,600
5.2
1,600
1.1
9,300
6.7
2,100
1.4
ftable Lower Limit Fugitive
Won Rates
0*
0*
o*
tstnent and annualized costs are zero since the maximum predicted concentration
iless than all considered alternative ambient standards.
32
-------
corresponding investment costs range from $22.3 million to $35.4 million.
The annualized cost range from $4.8 million to $7.6 million.
2.3.3.2 Industry
A critical factor in drawing inferences from the model plant assessment
and applying them to the entire industry is the amount of control required. To
determine the amount of control required necessitates categorizing existing
swelters as indicative of the middle, reasonable lower, or reasonable higher
limit fugitive emission estimates. On the basis of the estimated amount of
lead in the furnace charge at full capacity, three existing smelters are believed
to be indicative of the middle fugitive emission estimate; seven are indicative
of reasonable lower limit fugitive emission estimate; and six are indicative
of the higher limit fugitive emission estimate. To draw further inferences
is tenuous. However, if it is assumed that .production is proportional to the
•
number of smelters in each category and the model smelter costs can be extrapolated
linearly to higher production volumes, industry control cost estimates can be
developed. The investment costs for the year 1982 are $184.4 million for a
standard of 2.0 yg/ra3, $241.8 million for a standard of 1.5 yg/m3, and $298.7
•illion for a standard of 1.0 vg/m . The corresponding annualized costs are
$40.0 million, $52.3 million, and $64.4 million.
2.3.4 Model Plant Closure Assessment
Using a discounted cash flow analysis technique and model plant process
economics synthesized from publicly available company financial reports, the
potential for closing the smelter on financial grounds is assessed. Given
the reasonable lower limit fugitive emission estimates, the model plant should
2
not close. With a maximum predicted concentration of 8.4 yg/m , no control
33
-------
expenditures are required. Given the middle fugitive emission estimate, the
model smelter should not close even though control costs must be expended to
achieve all alternative standards. This finding is true for a range of marginal
tax rates and minimum acceptable return rates thought to be reasonable, full
absorption of all control costs, and a fully depredated model plant.
Given the reasonable higher limit fugitive emission estimates, the ability
of the model smelter to remain open on financial grounds alone is unclear.
However, the potential for remaining open is greater with a standard of 2.0
•a 3
ug/m than with either a standard of 1.5 or 1.0 ug/m • With a standard of
2.0 ug/m certain marginal tax rate and minimum acceptable return rate combina-
tions within a range thought to be reasonable do permit a fully depreciated
plant absorbing all the control costs to remain economically viable. With the
.1.5 or the 1.0 ug/m3 standards, a fully depreciated plant absorbing all the
control cost is not economically viable under any marginal tax rate and minimum
acceptable return rate combinations judged to be reasonable.
34
-------
2.4 GREY IRON FOUNDRY CASTING
2.4.1 Industry Structure
There are about a 1000 establishments classified as grey iron foundries.14
firey iron is produced at these foundries in cupola, electric, and reverberatory
furnaces. The main output of the foundries is castings. These come in a
variety of sizes and shapes having different chemical and physical properties.
The grey iron foundry industry produces intermediate goods. Castings
become part of and/or are used in the production of automobiles and trucks,
construction machinery, railway equipment, electrical and farm machinery,
rolling mills, and machine tools. Consequently, the demand for grey iron
is a function in part of the demand for these products. Demand for grey iron
is projected to grow at 3..2% per year.
Historically, the average price of grey iron castings has risen slightly
faster than the Wholesale Price Index. However, what product that average
price represents is not clear. The selling price for grey iron varies
depending on the size, shape, and chemical and physical properties of the
product. In 1976 the average price of a grey iron casting was $340/metric
ton.16
2.4.2 Model Plant Specifications and Dispersion Modeling Results
The model foundry specifications are consistent with the assumptions
and data requirements of the dispersion model. For example, the model grey
Iron foundry is located in flat terrain with representative meteorological
condition data available. The model foundry specifications are also
consistent with the objective of modeling a realistic situation. For
example, like 60% of the grey iron foundry establishments, the model
swelter is located in an east north central state. Moreover, like 70%
35
-------
F the grey iron produced nationally, the model foundry produces grey
ron in a cupola furnace.18 In addition, the size of the model foundry,
.3 metric tons/hour melt rate, corresponds to an actual foundry
Dcated in flat terrain in an East North Central State.
The model smelter has lead particulate stack emissions controlled to
nverage SIP process weight rates for particulate. Stack emissions have a
legligible impact on the maximum predicted ambient concentrations. Fugitive
missions are assumed for the model grey iron foundry. These emissions are
issumed uncontrolled. For a low fugitive emission rate estimate, the maximum
predicted concentration is 0.3 ug/m3, monthly average. For a midpoint fugitive
emission rate estimate, the maximum predicted concentration is 1.8 ug/m .
tod, for a high fugitive emission rate estimate, the maximum predicted concen-
tration is 3.7 ug/m3, monthly average.
Z.4.3 Control Cost
No control costs need to be developed for the low fugitive emission rate
estimate since the predicted monthly average maximum concentration of 0.3
iig/m3 is less than any of the alternative standards (i.e., 2.0, 1.5, and
1.0 ug/m3). For the midpoint fugitive emission rate estimate control costs
ire developed for the 1.5 and 1.0 ug/m3 alternatives since the predicted
maximum (1.8 ug/m3) is greater than these levels. However, no control costs
ire developed for the 2.0 ug/m3 alternative since the predicted maximum is
less. For the high estimate, since the predicted monthly average maximum
3f 3.7 ug/m3 is greater than the three considered alternative standards,
:ontrol costs are developed for all three standards.
36
-------
The investment, annual 1 zed costs, and investment and annualized costs
as a percent of annual revenue are presented in Table 2-7. The control
system costed 1s side draft and canopy hoods which are ducted to a fabric
filter. For the midpoint fugitive emission rate estimate, investment cost
3
as a percent of annual revenue is 9.8% for both the 1.5 and 1.0 ug/m standards.
Annualized cost as a percent of annual revenue for both standards is 2.0%.
The reason the costs do not vary between the 1.5 and 1.0 ug/m standards
is that the control system costed is assumed incapable of distinguishing
between the required control efficiencies needed to attain both standards.
For the high fugitive emission rate estimate, the required control
efficiencies for the 1.5 and 1.0 ug/m3 standard are greater than for the
midpoint fugitive estimate. These greater efficiencies are assumed reflected
in different control system design and operating needs,"and hence in the cost.
Investment cost as a percent of annual revenue ranges from 9.8% for the
2.0 ug/m3 standard to 30.4% for the 1.0 ug/m3 standard. Corresponding
annualized cost as a percent of annual revenue ranges from 2.0% to 6.2%.
2.4.4 Model Plant Closure Assessment
The baseline process economics for the grey iron foundry model plant
are developed using financial data contained in Leo Troy's 1977 Almanac
of Business and Industrial Financial Ratios.
Assuming zero background concentrations of lead and no other lead emitters
in the area, the model grey iron foundry with low or midpoint fugitive
emission rates should remain open regardless of the level of the standard.
For the model with low fugitive emission rates, no control costs need
be expended since all alternative standards are predicted to be achieved.
37
-------
Table 2-7. CONTROL COST FOR THE MODEL GREY IRON FOUNDRY PLANT
Maximum Predicted Ambient Levels
(ug/m3 monthTy average)
2.0 1.5 KO
^Fugitive Emission Rates
-------
Hence, it is clear that the model foundry should remain open. For the
model with midpoint fugitive emission rates, no control cost expenditures
3
are required for the 2.0 wg/m standard. However, the model foundry
3 3
must expend control costs to achieve the 1.5 wg/m and 1.0 wg/m standards.
Given that these costs must be absorbed, the model foundry is fully
depreciated and the foundry faces a reasonable range of marginal tax and
minimum acceptable return rates; it should remain open on financial
grounds.
The model foundry with high fugitive emission rates requires control
expenditures for all alternative standards. For the 2.0 and 1.5 wg/m
standard, the model foundry should remain open on financial grounds
given the cost absorption, depreciation, and marginal tax and return
rate conditions mentioned previously. However, with the 1.0 wg/m
standard, a fully depreciated foundry absorbing the control costs should
close for most marginal tax rate and minimum acceptable return rates
within a range thought to be reasonable. Only with marginal tax and
minimum acceptable return rates at the low end of the reasonable range
will a fully depreciated model foundry absorb the control costs needed
to meet a 1.0 wg/m3 standard and remain open on financial grounds.
-------
GASOLINE LEAD ADDITIVES MANUFACTURING
1 Industry Structure
Four companies with a combined total of six plants comprise the
gasoline additives (lead alkyl) industry. Table 2-8 lists the
panics, their plants, capacities, and locations. Present annual
adty is 403 million kilograms of tetraethyl lead (TEL) equivalent.
wal production in 1974 was about 318 million kilograms of TEL equivalent
10
(about 80% of capacity.
Gasoline lead additives are mixed with gasoline to raise the Octane
*er and consequently, reduce engine knock. Low production cost and
$ effectiveness in raising the Octane Number resulted in widespread
i of lead additives with little competition from other compounds.
However, EPA's lead phasedown regulations will result in the
relopraent and acceptance of other compounds and hence, slow down
20
tore domestic production and consumption of gasoline lead additives.
though most analysts agree the future U.S. production of lead additives
<11 be less than it is today, the exact decline in future production is
known. Industry representatives are optimistic with regard to export
21
ssibilities even though domestic consumption will decline. However,
tters feel the export market growth may not materialize if foreign
untries also adopt lead phasedown regulations.
Even with a production decline, given tight energy supplies and
ejected increases in the demand for gasoline, future lead additives
fices should not decline. In 1976, pure TEL sold for about 223.3
-------
Table Z-a. u>s. GASOLINE ADDITIVE (Lead Alkyl) MANUFACTURING PLANTS
1974 Capacity
(millions of Kg
Company Plant(s) Locatlon(s) Tetraethyl Lead)
E.I. duPont de Nemours Antloch, California 154
& Company, Inc. Deepwater, New Jersey
Ethyl Corporation Baton Rouge, Louisiana 177
Pasadena, Texas
PPG Industries, Inc. Beaumont, Texas 54
Nalco Chemical Company Freeport, Texas 18
403
Source: Chemical Economics Handbook p. 671.5042 C, December, 1975.
-------
1964 124.6
1965 127.9
1966 126.3
1967 125.0
1968 124.1
1969 127.4
1970 128.5
1971 133.6
1972 135.4
1973 137.3
1974 157.6
1975 200.0
1976 223.3
Source: Chemical Economics Handbook, "Tetraethyl Lead and Tetramethyl Lead", Stanford Research
Institute, p. 671.5042 R, December 1975. and DuPont, 1976. Antiknock mix prices were
multiplied by 1.626 assuming 1.626 pounds of antiknock per pound of pure TEL.
-------
:.5.2 Model Plant Specifications and Dispersion Modeling Results
Of the six gasoline lead additives plants, all are located in flat
;errain, 74% of their combined capacity produces tetraethyl lead (TEL)
ising the sodium-lead alloy process, three are located in Southeast Texas,
aid the average capacity of each plant is 67,000 metric tons of TEL
njuivalent. The model plant used in the dispersion modeling is located
in flat terrain, produces TEL using the sodium-lead alloy process, has
touston, Texas meteorological conditions, and produces about 54 thousand
stric tons of TEL annually.
The model plant is assumed to-have lead stack emissions in both the
liarticulate and vapor phases. The lead recovery furnace stack emits
lead in the particulate phase and is controlled to the average SIP
allowable process weight rate. The process vents and sludge pit exhaust
stacks emit lead in the vapor phase and are uncontrolled. No fugitive
missions are assumed. The maximum predicted concentration for the lead
stack emissions is 15.7 ug/m , monthly average.
Z.5.3 Control Costs
2.5.3.1 Model Plant
Since 15.7 pg/m is greater than all the considered standards,
control costs indicative of reduced stack emission rates are developed
for all the considered standards. The costed control system includes
packed scrubbers and increased pressure drop on an existing venturi
scrubber. The model plant investment, annualized control cost, and
investment and annualized cost as a percent of annual revenue are presented
43
-------
'able 2-10. Investment control costs for the model plant range from $519
jsand for meeting the 2.0 yg/m standard to $521 thousand for meeting the
mg/m standard. Annualized control cost as a percent of product sales
ce is about 0.2%. for the three considered standards.
,J.2 Industry
A critical factor in the model plant impact assessment is the
Hired reduction in the baseline emission rates to achieve the alternative
lient standards. The required reduction could be greater for larger
.planter ^pr TEL plants clustered among other lead emitters.
ever, to tharextent that the required reductions at the six existing
aline lead additive plants are similar to those at the model plant,
aevement of all three alternative ambient standards would appear to
itechnically possible by all plants. Furthermore, if model plant
irol costs are related linearly to 1974 production, the estimated
-------
Table 2-10. CONTROL COSTS FOR THE MODEL GASOLINE
.LEAD ADDITIVES PLANT
Maximum Predicted Ambient Levels
(yg/ro3 , monthly average)
2.0 1.5 1.0
Investment
OOO's of $'s 519 519 521
as a % of annual 0.4% 0.4% 0.4%
revenue
tonualized Control Cost
OOO's of $'s 270 270 271
as a % of annual 0.2% 0.2% 0.2%
revenue
-------
o achieve all standards under a variety of circumstances. These include
ill marginal tax rate and minimum acceptable return rates thought to be
tasonable, fully depreciated plant, sustained production decline (25%),
n increase in the raw material lead price (as a result of the lead
rirfent air quality standard, i.e., 1.8%), and full absorption of all
mrtrol costs.
-------
2.6 LEAD-ACID BATTERY MANUFACTURING
2.6.1 Industry Structure
Currently there are about 200 lead acid battery manufacturing
plants in the United States ranging in size from about 50 to about
11,500 batteries per day.
Two major types of lead add storage batteries are manufactured in
the United States. Starting-Lighting-Ignition (SLI) batteries which are
used in auto, aircraft, and golf carts are one type. The units account
for more than 80 percent of the market.22 The second type includes
industrial storage batteries for such uses as low-voltage power systems
and industrial fork!ift trucks.
The market for lead-acid storage is composed of three segments.
The largest segment in terms of sales is the domestic replacement market.
This includes replacement batteries for automobiles, trucks, buses, farm
machinery, and heavy equipment. The original domestic equipment market
is the second largest segment. This includes batteries sold to producers
of new vehicles and equipment. The export sector is the smallest market.
This includes replacement batteries in existing equipment and batteries
for new equipment. The overall demand for batteries in these market
23
segments is expected to grow between 3.5* and 8.2% per year.
Prices for automobile SLI batteries are about $16 to $22 f.o.b.
plant and about $35 to $50 retail.24 Prices for industrial storage
batteries range from $200 to $11,500.25 Because batteries represent
such a small percentage of vehicle costs and appear to have few close
47
-------
substitutes, demand is thought to be price Inelastic. However, the
replacement demand market segment for batteries is thought to be less
rice inelastic than the new demand segment since useful battery life
'An be extended by improved maintenance, servicing, and repair.
1.6.2 Model Plant Specifications and Dispersion Modeling Results
Two model lead acid battery plants are used in the dispersion
ndeling. One is capable of producing 500 batteries per day; the other
is capable of producing 6500 batteries per day. These two plant sizes
ire thought to bound the reasonable range of actual industry plant.sizes.
'insistent with the dispersion model, both plants are located in flat terrain
rtth area meteorological condition data available. This situation is not
itypical since there is an actual battery plant with similar terrain and
cteorological features.
None of the model lead acid battery plants are assumed to have
\ig1tive emissions. All emissions emanate from point sources. These
oint sources are controlled to average SIP allowable process weight
ites. The maximum predicted concentration for the 500 battery per day
lant is 0.9 ug/m3, monthly average. The maximum predicted concentration
or a 6500 battery per day plant is 7.6 ug/m , monthly average.
.6.3 Control Cost
Since the maximum predicted concentration for the model 500 battery
er day plant does not exceed any of the considered standards, no control
wts are developed for that model plant. Control costs are developed for the
adel 6500 battery per day plant since the maximum predicted concentration
48
-------
(7.6 ug/m ) does exceed the considered standards of 2.0, 1.5, and
1.0 y
The only point source requiring control is the three process operation
stack. Required control efficiencies for this source are about 75
percent, 81 percent, and 88 percent for the respective standards of 2.0,
1.5, and 1.0 ug/m . A wet impingement scrubber is assumed to achieve
those control efficiencies at least cost. The investment cost 1s not
assumed to vary with the required control efficiency and is estimated at
$130 thousand. Operating costs are, however, assumed affected to some
degree by the required control efficiency. Respective annualized costs
for meeting 2.0, 1.5, and 1.0 standards are $57,000, $57,400, and $57,900.
Annualized cost as a percent of annual revenue (assumes price per battery
of $18.50) is about 0.1% for each of three standards.
2.6.4 Model Plant Closure Assessment
A low profit margin fully depreciated model 6500 battery per day
plant should be able to absorb the control costs associated with any of
the standards and still remain open. This finding holds for a range of
marginal tax rates and minimum acceptable return rates thought to be
reasonable. It also holds given the assumptions that raw material lead
prices increase as a result of the lead ambient air quality standard and
that this cost increase (1.8 percent) is absorbed.
Process economics were developed using published financial data for
Gould, Inc. and Northwest Industries, Inc.'s General Battery Division.
49
-------
3.0 OTHER AFFECTED SECTORS
3.1 MOBILE SOURCE ASSESSMENT
Several assumptions are used to project mobile source impact. These
include phasedown of lead in the gasoline pool, improved fuel economy,
and retirement of some older leaded gasoline using vehicles. For an ambient
standard of 2.0 yg/m3, one Air Quality Control Region (AQCR) 1s predicted
to require mobile source emission control in 1982. For an ambient standard
of 1.5, the number predicted is two AQCRs. And, for an ambient standard of
1.0, it is four AQCRs.
Vehicles using leaded gasoline are retired as they get older, and often
these vehicles are replaced with newer cars using unleaded gasoline. Hence,
the number, of AQCRs requiring mobile source emission control might be less
in for example 1985 than in 1982. In 1985, the number of AQCRs requiring
mobile source emission control is predicted to be one for the 2.0 and
3 3
1.5 ug/m standards, and two for the 1.0 ijg/m standard.
To achieve the alternative ambient standards in 1982 by retrofitting
existing leaded gasoline using light duty vehicles with 75% efficient lead
trap mufflers (if available) could affect the following number of vehicles.
For the 2.0 ug/m standard, the predicted number is 58,000. For the 1.5
ug/m standard, the predicted number is 106,500. And, for 1.0 vg/m3 standard
the predicted number of light duty vehicles affected is 1,300,000.
Of course retrofitting lead trap mufflers is not the only means of
controlling mobile source lead emissions. Reduction in vehicle miles
traveled (VMT) and further reduction in the lead content of gasoline are
alternatives. VMT reductions can be achieved many ways including
50
-------
carpooling, mass transit, and reduced trips. Further reduction in the lead
content of gasoline can also be achieved in many ways including using non-lead
gasoline additives and increasing the reforming capacity of the refineries.
Costs for these many ways to achieve alternatives to retrofitting lead trap
mufflers have not been developed. Consequently, relative cost effectiveness
of retrofitted lead traps 1s not known.
51
-------
3.2 STATE AND LOCAL AIR POLLUTION CONTROL AGENCY ASSESSMENT
Currently there are 55 state and 235 local air pollution control
agencies. Annually they spend approximately 7100 man-years of effort and
$157 million dollars implementing air pollution control regulations.
An ambient lead standard can impose additional requirements (costs)
on State and local air pollution control agencies to administer the
standard. The added requirements could take the form of more data gathering,
•enforcement, monitoring, laboratory, support, and management activities.
The previously described stationary and mobile source assessments are
the bases for estimating these added requirements. To the extent that
these bases understate the magnitude of the problem (for example the
number of source inspections) the added requirements for State and local
control agencies are also understated. But, even if the stationary and
mobile source assessments do not understate the magnitude of the problem,
control agency requirements could be understated for another reason.
The estimates of additional state and local control agency costs do not
include the requirements for developing completely new fugitive emissions
inventories. However, the estimates do include the requirements for
additional maintenance and update of existing inventories.
Some requirements (costs) are estimated to be incurred in the first year
only while others are on-going. First year requirements include the costs
for ambient monitors and laboratory equipment such as hi-vols and spectro-
photometers. In addition, there is non-recurring labor for activities
such as State implementation plan development and site preparation for monitors.
52
-------
No enforcement is assumed to take place during the first year while strategy
and regulations are being developed. Also, in the first year monitoring and
laboratory activities are limited while equipment is being installed. The
first year requirements in dollar terms range from $1.0 million with a
2
standard of 2.0 ug/m , monthly average to $1.7 million with a standard of
1.0 ug/ra . The relative magnitudes of these costs compared to current
expenditures are 0.6% and 1.0%. In terms of people, the corresponding
first year requirements are 30 man-years and 40 man-years. Compared to
current man-years expenditures, these correspond to 0.4% and 0.6%, respectively.
On-going or annual costs include operating costs but not depreciation
or interest. Examples of operating costs are maintenance, supplies, and
power for ambient monitors and laboratory equipment. Operating costs also
include a labor component. For example, there are the on-going labor-using
activities of mobile and stationary source inspection and enforcement.
On-going costs range from $1.4 million with a standard of 2.0 yg/m to
$2.8 million with a standard of 1.0 yg/m . The relative magnitude of
these costs are 0.9% and 1.8%. The corresponding man-year requirements
are 60 and 120. Compared to current man-year expenditures these correspond
to 0.9% and 1.7%, respectively.
First year and on-going costs for the 3 considered standards are
presented in Table 3-1.
53
-------
Table 3-1. STATE AND LOCAL AIR POLLUTION CONTROL
AGENCY STANDARDS ADMINISTRATION COST
Alternative Ambient Standards
(ug/m3 . monthly average)
270 1.5 l.Q
irst Year Cost
000's of $'s 1000 1400 1700
$ Expenditures as a
% of current
expenditures .6% .9% l.Q*
Man year requirements 30 40 40
Man-year requirements
as a % of current 0.4% 0,655 0.6%
expenditures
i-Going (Annual Costs)
000's of $'s 1*00 2300 2800
$ Expenditures as a %
of current expenditures 0.9% 1.5% 1.8%
Man-year requirements 60 100 120
Man-year requirements
as a % of current 0.9% 1.4% 1.7%
expenditures
54
-------
References
1. PEDCo-Enviroranental Specialists, Inc., Draft Document prepared for
the U.S. Environmental Protection Agency, Control Techniques For Lead
Air Emissions, April, 1977, p. xix.
2. Ibid., p. xix.
3. Mitre Corporation/Metrek Division and Office of Air Quality Planning
and Standards, U.S. Environmental Protection Agency, Draft Environmental
Impact Statement for the National Ambient Air Quality Standard for Lead,
November, 1977, Chapter 2.
4. Ibid., and background information in support of the aforementioned document.
5. Bureau of the Mines, U.S. Department of the Interior, Commodity Data
Summaries: 1977, p. 91. .
6. John Short and Associates, Inc., "Preliminary Technological Feasibility,
Cost of Compliance and Economic Impact Analysis of the Proposed OSHA
Standard for Lead", prepared for U.S. Department of Labor under Contract
No. J-9-F-6-004, 1/77, p. 34. See also Charles River Associates, Economic
Analysis of the Lead-Zinc Industry, prepared under contract for the U.S.
General Services Administration, revised April, 1969, p. 15.
7. Office of Air Quality Planning and Standards, U.S. Environmental Protection
Agency, Users Manual for the Single Source (CRSTER) Model, EPA-450/2-77-013,
July, 1977.
8. Bureau of the Mines, U.S. Department of Interior, Mineral Yearbook - 1974,
p. 731.
9. Federal Trade Conmission, Docket 18959, 4/20/76.
10. Ibid.
11. Stanford Research Institute, Chemical Economics Handbook: Copper.
February 1976, p. 736.1000C.
12. Ibid, p. 736.1000B.
13. Office of A1r Quality Planning and Standards, U.S. Environmental Protection
Agency, Background Information for New Source Performance Standards: Primary
Copper, Z1nc« and Lead Smelters, Volume I: Proposed Standards, EPA-450/2-74-002a,
' October, 1974, p. 6-14.
14. Bureau of the Census, U.S. Department of Commerce, 1972 Census of Manufacturers,
p. 33-B-6.
55
-------
U.S. Department of Commerce, 1977 U.S. Industrial Outlook, p. 82.
Office of Air Quality Planning and Standards, U.S. Environmental
Protection Agency, Standards Support and Environmental Impact Statement:
An Investigation of the Best Systems of Emission Reduction for Electric"
Arc Furnaces in the Grey Iron Foundry Industry, October, 1976, p. 8-32.
Bureau of the Census, U.S. Department of Commerce, op. cit., p. 33-B-6.
PEDCo-Environraental.Specialists, Inc., op. cit.. p. 4-166.
U.S. International Trade Commission, Synthetic Organic Chemicals, U.S.
Production and Sales. Also, Stanford Research Institute, Chemical
Economics Handbook: Lead Alkyls, December, 1975, p. 6715042E.
Federal Register. December 6, 1973, 38 (234) 40CFR80: 33734-33741.
Stanford Research Institute, op. cit.. pp. 6715042R-S.
Barkand, R. A. A Report by the Battery Council International Statistical
Comnittee. Replacement Battery; Industry Forecast 1975-1979. Globe Union,
Inc. Milwaukee, Wisconsin, p. 1.
Office of A1r Quality Planning and Standards, U.S. Environmental Protection
Agency, Draft Standards Support and Environmental Impact Statement: Control
of Emissions from the Manufacture of Lead-Add Storage Batteries. September,
1977, p. 3-9 and p. 8-16.
Compiled from Company responses to U.S. Environmental Protection Agency
Inquiry regarding product prices. The inquiry was issued under Section 114
of the Clean Air Act as Amended, 1970.
Conversation between the staff of JACA Corporation and two battery companies:
Illinois Battery on September 8, 1977, and Moore Battery Company on
September 14, 1977.
56
------- |