TO:	EPA Docket No. EPA-HQ-OAR-2003-0146

		

FROM:	Bob Lucas, EPA/SPPD 

DATE:	July 27, 2007

SUBJECT:	Marine Vessel Loading:  Control Options and Impact Estimates 

I.	Purpose

This memorandum documents the methodology used to identify the control
options and estimate the costs and environmental impacts for marine
vessel loading control options. 

II.	Background

Section 112(f) of the Clean Air Act Amendments (CAA) directs the U.S.
Environmental Protection Agency (EPA) to assess source categories
regulated under Section 112(d) of the CAA and determine whether any
human health or environmental risks remain from the continued emissions
of hazardous air pollutants (HAPs) following implementation of maximum
achievable control technology (MACT) standards.  The CAA further states
that if the MACT standards do not reduce lifetime excess cancer risk to
the most exposed individual to less than one in one million, EPA must
set additional standards to protect human health and the environment, in
accordance with the interpretation set forth in the Benzene NESHAP. 
Additionally, the EPA is required to review these technology-based
standards and to revise them “as necessary (taking into account
developments in practices, processes and control technologies)” no
less frequently than every 8 years, under CAA section 112(d)(6).  The
Refinery MACT 1 (40 CFR Part 63 Subpart CC) was promulgated over 8 years
ago and is now being reviewed.   

III.	Summary of Existing Refinery MACT 1 Marine Vessel Loading Control
Requirements

Refinery MACT 1, subpart CC requires control of marine vessel loading
terminals with HAP emissions greater than 10 tons per year of a single
HAP or greater than 25 tons per year total HAP when loading commodities
with vapor pressure greater than or equal to 10.3 kilopascals [1.5
psia].  Refinery MACT 1 marine vessel loading must meet the control
requirements specified in 40 CFR part 63, subpart Y (National Emission
Standards for Marine Tank Vessel Loading Operations).  Subpart Y defines
Marine tank vessel loading operation as: 

any operation under which a commodity is bulk loaded onto a marine tank
vessel from a terminal, which may include the loading of multiple marine
tank vessels during one loading operation.  Marine tank vessel loading
operations do not include refueling of marine tank vessels.

The applicability section of subpart Y indicates that Sources with
emissions of 10 or 25 tons per year are subject to emissions standards
requiring 97 percent emissions reduction for existing sources and 98
percent emissions reduction for new sources.  Sources with emissions of
10 or 25 tons per year means major sources(s) having aggregate actual
HAP emissions from marine tank vessels loading operations at all loading
berths of 9.1 Mg (10 tons) or more of each individual HAP or 22.7 Mg (25
tons) or more of all HAP combined.

Commodities with vapor pressures greater than or equal to 10.3
kilopascals (1.5 psia) are subject to the requirements.  Typical
commodities loaded at petroleum refineries marine terminals with vapor
pressures greater than 10.3 kPa would include gasoline and other
products such as naphtha.  Other products such as toluene and kerosene
have vapor pressures less than 10.3 kPa (1.5 psia).  Distillate fuel oil
No. 2 (or diesel gasoline) has vapor pressures less than 0.5 psia.

IV.	Identification of Control Options

Control options were reviewed for marine vessel loading with lower HAP
emissions cutoff values than the current NESHAP levels (i.e., 10 or 25
tons).  One marine vessel loading control options was identified.

  

Option:  Require emissions reduction for marine tank vessel loading
operations with HAP emissions greater than 7.5 tons per year for single
HAP or greater than 18.75 tons per year for multiple HAP and for
commodities with vapor pressure greater than or equal to 10.3
kilopascals.  (Emissions reduction requirements are 97 percent for
existing sources and 98 percent for new sources.)

V.	Impact Estimates for Model Marine vessel loadings

Model Plant Marine Vessel Loading Operations 

The marine terminal emissions cutoffs for the control option are HAP
emissions greater than 7.5 tons per year for single HAP or greater than
18.75 tons per year for multiple HAP.  The 18.75 ton/yr cutoff is
represented by Model Plant 1 at 20 ton/yr multiple HAP, and the 7.5
ton/yr cutoff is represented by Model Plant 2 at 8.0 ton/yr single HAP. 
An analysis was conducted for two commodities typically loaded at marine
terminal operations at Refinery MACT 1 sources:  a gasoline analysis and
a naphtha analysis.

Gasoline commodity:  A corresponding annual gasoline throughput was
estimated from the HAP emissions cutoff levels:  first, a back
calculation of VOC emissions from an estimated percentage of HAP in VOC
emissions was conducted, and then a back calculation of gasoline
throughput from VOC emissions based on an AP-42 VOC emissions factor was
conducted.  A gasoline throughput of approximately 2.58 million barrels
per year is estimated to emit HAP emissions at 20 ton/yr for multiple
HAP (Model Plant 1), and a throughput of approximately 2.53 million
barrels per year is estimated to emit at 8.0 ton/yr for single HAP Model
Plant 2).  (For example, based on average refinery gasoline vapor phase
compositions from Refinery MACT 1, it is assumed that VOC emissions are
approximately 10.8 percent by weight HAP.  The VOC emissions were back
calculated from the total HAP emissions of 20 tons per year to be 184.0
tons per year VOC.  Using an emissions factor of 3.4 lb VOC per 1,000
gallons of gasoline loaded from Table 5.2-2 of AP-42 section 5.2
[assumed typical overall loading situation emissions factor, for barge
loading and dedicated normal service], the gasoline throughput was back
calculated from 184.0 tons per year of VOC emissions and was estimated
to be 2.57 million barrels per year.)

Naphtha commodity:  A corresponding annual naphtha throughput was
estimated from the HAP emissions cutoff levels:  first, a back
calculation of VOC emissions from an estimated percentage of HAP in VOC
emissions was conducted, and then a back calculation of naphtha
throughput from VOC emissions based on an AP-42 VOC emissions factor was
conducted.  A naphtha throughput of approximately 4.01 million barrels
per year is estimated to emit HAP emissions at 20 ton/yr for multiple
HAP (Model Plant 1), and a throughput of 2.11 million barrels per year
is estimated to emit 8.0 ton/yr for single HAP (Model Plant 2).  (For
example, based on average refinery naphtha vapor phase compositions from
Refinery MACT 1, it is assumed that VOC emissions are approximately 13.5
percent by weight HAP.  The VOC emissions were back calculated from the
total HAP emissions of 20 tons per year to be 147.8 tons per year VOC. 
Using an emissions factor of 1.75 lb VOC per 1,000 gallons of naphtha
loaded, the naphtha throughput was back calculated from 147.8 tons per
year of VOC emissions and was estimated to be 4.01 million barrels per
year.  The emissions factor was estimated using equation 1 from section
5.2 of AP-42, based on a saturation factor of 0.5 assumed submerged
loading for barges [this assumption provides a higher emissions factor
than the 0.2 saturation factor for marine vessel loading to ships]; a
vapor pressure based on naphtha of 1.9 psia, molecular weight of 80
lb/lb mole; and temperature of 80ºF.)

The model plants for marine vessel loading controls from the proposal
TSD for Marine Vessel Loading Operations were used to estimate control
costs.  The model plants in the proposal TSD for Marine Vessel Loading
were based on model terminals developed by the Marine Board of the
National Council on Engineering and Technical Systems.  The models
included both crude loading and product loading (the term “product”
is used to mean any commodity other than crude oil), however, only the
product loading models were considered for this Refinery MACT1 analysis,
as loading of crude from a petroleum refinery was considered unlikely. 
The model plants for product loading represented commodity loading on
inland river barges and ships and on barges and included various
commodity throughput ranges.  Two model plants for product loading from
the TSD for Marine Vessel Loading Operations were selected (Model Plant
5C and Model Plant 7A, from Table 3-1, Description of Model Terminals)
to simulate marine vessel loading operations at Refinery MACT 1 sources
that would be subject to the control options, as these two models
matched the range of throughputs of 2.11 million barrels per year to
4.01 million barrels per year estimated for the HAP emissions cutoffs
for Model Plant 1 and Model Plant 2.

Model Plant Emissions Estimates

For this analysis, the baseline emissions correspond to the model plant
HAP emissions levels of 8.0 tons per year single HAP or 20 tons per year
multiple HAP.  Based on average refinery gasoline vapor compositions
(Refinery MACT 1), benzene emissions were estimated to be 0.626 percent
of the total VOC emissions, and hexane emissions were estimated to be
7.6 percent of VOC.  Based on naphtha vapor compositions, benzene
emissions are estimated to be 1.06 percent and hexane emissions are
estimated to be 14.2 percent of total VOC emissions.  Baseline emissions
for HAP for the control option are shown in Table 1.

Table 1.  Model Plant Emissions of Refinery Marine Vessel Loading

Control Option:	HAP Emissions

	Gasoline commodity	Naphtha commodity

Model Plant 1 (Multiple HAP emissions 20 ton/yr)

Throughput, bbl/yr	2.58 million	4.01 million

Total HAP, ton/yr	20	20

Hexane, ton/yr	8.15	15.2

Benzene, ton/yr	1.15	1.57

Model Plant 2 (Individual HAP emissions 8.0 ton/yr

Throughput, bbl/yr	2.53 million	2.11 million

Total HAP, ton/yr	19.6	10.5

Hexane, ton/yr	8.0	8.0

Benzene, ton/yr	1.13	0.826



Emissions reductions for control of emissions from marine vessel
loadings were estimated based on control with a thermal oxidizer
achieving an efficiency of 97 percent.  The emissions reductions for HAP
are shown in Table 2.  The range of emissions reductions are shown by
Model Plant 1 with gasoline as the commodity at 19.4 ton/yr emissions
reduction and by Model Plant 2 with naphtha as the commodity at 10.2
ton/yr emissions reductions.

Table 2.  Model Plant Emissions Impacts of Refinery Marine Vessel
Loading Control Options

Commodity	Model Plant 1:  Multiple HAP cutoff, 20 ton/yr	Model Plant 2: 
Single HAP cutoff, 8.0 ton/yr

	HAP Baseline emissions, ton/yr	HAP Emission Reduction from Baseline,
ton/yr	HAP Baseline emissions, ton/yr	HAP Emission Reduction from
Baseline, ton/yr

Gasoline	20	19.4	19.6	19.1

Naphtha	20	19.4	10.5	10.2



Model Plant Control Costs 

Control costs were taken from the proposal TSD for Marine Vessel Loading
Operations (Model Plant 5C, from Table 3-6 for Capital and Annual Costs:
 Model 5C; and Model Plant 7A, from Table 3-10 for Capital and Annual
Costs:  Model 7A); the costs for these two models were averaged. 
Capital costs for vessel retrofits from the original TSD analysis were
not included in the current analysis as it was assumed that vessel
retrofits likely have already occurred because many of the vessels would
also service marine vessel terminals that are subject to the control
requirements of subpart Y.  Rough estimates were made to scale the
capital and annual costs for the average model plant to 2006 costs.  As
the cost estimates were originally developed in 1987 dollars, the
capital costs were escalated to 2006 dollars using the Chemical
Engineering Plant Cost Index (CEPCI).  The average CEPCI in 1987 was
323.8; the average CEPCI in 2006 was 499.6.  The scaled 2006 capital
costs for model plants 5C and 7A are shown in Table 2.

Annual operating costs were also escalated to 2006 dollars.  Current
loaded labor rates and current energy unit costs were used to update
costs to 2006.  The model plant control costs for the control option
were considered to be average costs per marine vessel terminal.

Table 2.  Model Plant Capital and Annual Costs for Thermal Oxidizer
Control of Refinery Marine Vessel Loading

Costs	Model 5C	Model 7A	Average model

	Thermal Oxidizer, 1987 $	Thermal Oxidizer, 2006 $	Thermal Oxidizer,
1987 $	Thermal Oxidizer, 2006 $	Thermal Oxidizer, 2006 $

Capital Costs a

Incinerator	$290,000	$447,000	$371,000	$572,000	$510,000

Inert gas generator	$63,400	$97,800	$312,000	$481,000	$290,000

Water system	$15,800	$23,400	$29,200	$45,100	$34,700

Other major equipment	$118,000	$182,000	$337,000	$520,000	$351,000

Piping	$139,000	$214,000	$1,020,000	$1,570,000	$894,000

Instrumentation	$29,500	$45,500	$69,600	$107,000	$76,500

Engineering, startup, contingencies	$164,000	$253,000	$535,000	$825,000
$539,000

TCI	$819,000	$1,260,000	$2,680,000	$4,130,000	$2,700,000

Annual costs 

Labor b	$1,760	$7,220	$1,760	$7,220	$7,220

Maintenance c	$36,200	$49,800	$68,200	$81,800	$65,800

Natural gas d	$19,900	$45,700	$96,600	$222,000	$134,000

Electricity e	$2,310	$2,970	$10,400	$13,400	$8,160

Overhead f 	$22,800	$34,200	$42,000	$53,400	$43,800

Property taxes, insurance, and administration g	$32,800	$50,600	$107,000
$165,000	$108,000

Capital recovery h	$127,000	$196,000	$389,000	$600,000	$398,000

TAC	$243,000	$386,000	$715,000	$1,140,000	$765,000

a  Rough estimates of capital costs were made to scale the 1987 costs to
2006 costs based on CEPCI of 323.6 for 1987 and 499.6 for 2006.

b  Assume 0.5 hr/shift for the incinerator and 0.5 hr/shift for the IG
generator.  Operator labor for 2006 is $28.90/hr.  	

c  Assume 0.5 hr per shift for the incinerator and 0.5 hr/shift for the
IG generator; multiplied by 2 for parts; addition of $32,800 for model
plant 5C and $64,800 for model plant 7A.  Maintenance labor for 2006 is
$33.99/hr.2

d  Annual natural gas usage is 5,802,000 ft3 for model plant 5C; annual
natural gas usage is 28,163,000 ft3 for model plant 7A.  Natural gas
costs in 2006 were $7.88/1000 ft3.  

e  Electricity usage was not reported in proposal TSD for Marine Vessel
Loading Operations and neither was the unit cost in 1987.  As an
alternative, the costs were scaled up based on a ratio of unit costs for
1990 to unit costs for 2006.  Electricity costs were $0.0474/kWhr in
1990 and were $0.0609/kWhr in 2006.  (Electricity unit costs were not
found for 1987; the closest available year was 1990.),  

f  Overhead is 60 percent of labor and maintenance costs.

g  Taxes, insurance, and administration costs are 4 percent of capital
costs.

h  Capital recovery costs were estimated as 0.1675 of capital costs
without piping costs plus 0.1175 of piping capital costs.  CRF estimated
based on interest rate of 10 percent, equipment life of 10 years, and
piping life of 20 years.



Model Plant Emissions Reductions and Control Cost Impacts

The model plant impacts are summarized in Table 3, including the total
capital costs, total annualized costs, and the cost effectiveness.  The
impacts for Model Plant 1 with gasoline and the impacts for Model Plant
2 with naphtha commodity are shown as these represent the range of the
cost effectiveness values that would be expected at marine vessel
loading operations at the emissions cutoff for the control option.

Table 3.  Model Plant Impacts of Refinery Marine Vessel Loading Control
Options a

Model Plant	Total Capital Investment, $	Total Annualized Cost, $	HAP
Emission Reduction from Baseline, tons	Total Cost Effectiveness for 
HAP, $/ton HAP

Model Plant 1:  Emissions 20 tpy multiple HAP (Gasoline commodity)	$2.70
million	$0.765 million	19.4 b	$39,400/ton HAP b

Model Plant 2:  Emissions 8.0 tpy single HAP (Naphtha commodity)	$2.70
million	$0.765 million	10.2 c	$74,800/ton HAP c

a  Average model costs from Table 2 were applied to both Model Plant 1
and Model Plant 2.

b  For the gasoline commodity, the emissions reductions based on the
single HAP emissions cutoff of 7.5 ton/yr would be 17.9 ton/yr and the
cost effectiveness would be $42,800/ton HAP.

c  For the naphtha commodity, the emissions reductions based on the
single HAP emissions cutoff of 7.5 ton/yr would be 9.58 ton/yr and the
cost effectiveness would be $79,800/ton HAP.



Technical Memorandum – Marine Vessel Loading:  Control Options and
Impact Estimates

July 27, 2007

Page   PAGE  6 

Technical Memorandum

 VOC/HAP Emissions from Marine Vessel Loading Operations – Technical
Support Document for Proposed Standards.  EPA Publication No.
EPA-450/3-92-001a.  U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards.  Research Triangle Park, NC.  May 1992.

   Bureau of Labor Statistics, May 2006 labor rates (accessed June
2007).

   Natural Gas Monthly, June 2007.  Table 3, Selected National Average
Natural Gas Prices, 2002-2007.  Energy Information Administration.

   Electric Power Monthly with data for April 2007.  Table 5.3, Average
Retail Price of Electricity to Ultimate Customers:  Total by End-Use
Sector, 1993 through April 2007.  Energy Information Administration.

   Electric Power Monthly with data for September 2003.  Table 5.3,
Average Retail Price of Electricity to Ultimate Customers:  Total by
End-Use Sector, 1990 through September 2003.  Energy Information
Administration.

