c/EPA
United States       Industrial Environmental Research  EPA-600/2-79-069
Environmental Protection  Laboratory            March 1979
Agency         Research Triangle Park NC 27711


Research and Development	



Control Technology


Evaluation for Gasoline


Loading of Barges

-------
                 RESEARCH REPORTING SERIES


Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination  of  traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

    1. Environmental Health Effects Research

    2. Environmental Protection Technology

    3. Ecological Research

    4. Environmental Monitoring

    5. Socioeconomic Environmental Studies

    6. Scientific and Technical Assessment Reports (STAR)

    7. Interagency Energy-Environment Research and Development

    8. "Special" Reports

    9. Miscellaneous Reports

This report has been assigned to the  ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment,  and methodology  to repair or prevent en-
vironmental degradation from  point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
                       EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policy of the Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.

This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                     EPA-600/2-79-069

                                             March 1979
Control  Technology  Evaluation
for  Gasoline  Loading of Barges
                       by

                D. Gee and W. M. Talbert

                   Pullman Kellogg
             16200 Park Row, Industrial Park Ten
                 Houston, Texas 77084
                Contract No. 68-02-2619
                     Task No. 9
               Program Element No. 1AB604B
             EPA Project Officer:  Irvin A. Jefcoat

          Industrial Environmental Research Laboratory
            Office of Energy, Minerals, and Industry
             Research Triangle Park, NC 27711
                    Prepared for

          U.S. ENVIRONMENTAL PROTECTION AGENCY
             Office of Research and Development
                 Washington, DC 20460

-------
                            CONTENTS

Table of Contents   	   ii
Index of Figures	    v
Index of Tables   	   vi

1.0    INTRODUCTION	    1

       1.1  PURPOSE	    1
       1.2  SOURCES AND TECHNIQUE 	    2

2.0    SUMMARY	    3

3.0    CONCLUSIONS AND RECOMMENDATIONS  .	    6

       3.1  CONCLUSIONS 	    6
       3.2  RECOMMENDATIONS . . .	    8

4.0    DISCUSSION	    9

       4.1  DESCRIPTION AND OPERATION OF A GASOLINE
            BARGE LOADING TERMINAL AND A BARGE   	    9
       4.2  DISCUSSION OF CONTROL TECHNOLOGIES IN
            RELATION TO GASOLINE LOADING  	   12

            4.2.1  Carbon Adsorption  .	   12
            4.2.2  Refrigeration	   16
            4.2.3  Thermal Incineration 	   19
                               11

-------
                 CONTENTS  (Cont)

4.3  GENERAL DISCUSSION OF SAFETY TECHNOLOGY ....   22

     4.3-1  Fire and Explosion   	   23

     4.3.2  Barge Overfilling  	   29

4.4  APPLICATIONS OF SAFETY AND CONTROL
     TECHNOLOGIES   .	29

     4.4.1  Thermal Incineration by Dilution  ....  36

     4.4.2  Thermal Incineration-with Saturation  .  .  36

     4.4.3  Thermal Incineration with Fuel Gas
            Blanketing	39

     4.4.4  Thermal Incineration with N2
            Blanketing  . .'	41

     4.4.5  Thermal Incineration with Inert
            Gas Generation	43

     4.4.6  Carbon Adsorption with Saturation ....  43

     4.4.7  Carbon Adsorption with N2 or
            Inert Gas Generator	44

     4.4.8  Refrigeration with Saturation 	  44

     4.4.9  Refrigeration with N2 or Inert
            Gas Generator   	45

     4.4.10 Other Design Considerations 	  45
                         111

-------
                        CONTENTS  (Cont)

       4.5  EMISSIONS AND CONTROL EFFICIENCIES  	   46

            4.5.1  Primary Emissions  	   46

            4.5.2  Secondary Emissions  	   55

       4.6  ECONOMICS	   58

            4.6.1  Capital Costs	   59

            4.6.2  Annualized Costs	  .   65

            4.6.3  Economic Analysis  	   71

References    	   80
Appendices
       A.   Legend	A-l
       B.:.  Capital Cost Data for Safety
            and Control Modules	  B-l
     - C.   Companies Supplying Pricing and
            Technical Information 	  C-l
                              IV

-------
                        INDEX OF  FIGURES
Number                        .                                Paqe

4.2.1  Carbon Adsorption Module    	    13
4.2.2  Refrigeration Vapor Recovery Module   	    17
4.2.3  Thermal Incineration Module  	    20
4.3.1  Hydraulic Flash Arrestor  	    28
4.3.2  Barge Modifications  .	    30 .
4.3.3  Barge Vapor Collection Manifold   	    31
4.4.1  Loading and Collection System Layout  	    33
4.4.2  Dilution Module	    37
4.4.3  Saturation Module	-	    38
4.4.4  Fuel Gas Blanketing Module   	    40
4.4.5  N  Blanketing Module	    42
4.6.1  Cost Effectiveness Projections
       Carbon Adsorption  	    76
4.6.2  Cost Effectiveness Projections
       Refrigeration    	'	    77
4.6.3  Cos.-t. Effectiveness Projections
       Thermal Incineration	    78
4.6.4  Cost Effectiveness Projections
       Comparison   	    79

-------
                        INDEX OF TABLES
Number                                                       Page

4.U.I  Predicted Vapor Compositions	    34
4.5.1  Gasoline Barge Vapor Composition  	    48 •
4.5.2  Barge Vapor Compositions After Safety Module  ...    49
4.5.3  Manufacturers Predicted Vapor Control Unit
         Efficiencies  	    51
4.5.4  Control efficiencies Reported In" Literature ....    52
4.5.5  Preducted Annual HC Emission Rates For Control
         Cases Lb/Yr HC Emitted	    56
4.6.1  Capital Cost Of Safety Modules	    60
4.6.2  Capital Cost For Control Modules	    62
4.6.3  Capital Costs Of Barge Loading Emission
         Control Systems 	    64
4.6.4  Anriaalized Costs For Safety Modules 	    67
4.6.5. Annualized Costs For Control Modules	    68
4.6.6  Annualized Costs For One Gasoline Barge
         Loading Emission Control System 	    70
4.6.7  Cost Effectiveness For Control Cases	    72
4.6.8  Summary Of Results Of Cost Effectiveness Curves .  .    75
                              VI

-------
                          SECTION 1

                        INTRODUCTION
1.1      PURPOSE
         This study undertakes both an assessment of feasibility
         for developing  new vapor control systems for  loading
         barges  with  gasoline and development  of  background
         information  for  use  in designing a  demonstration
         facility.  Specifically, the study was designed to:

         A.  Determine the  feasibility of  controlling  gasoline
             vapors during barge-loading operations  by  using
             carbon  adsorption, refrigeration,  or thermal
         -    incineration technologies.

         B.  Identify and solve  the problems of  safety hazards
             related to  vapor  control.

         C.  Determine the achievable emission  level  and  ex-
             pected efficiency for each control technology.

         D.  Determine the secondary emissions  resulting from
             implementation of  each  gasoline  vapor control
             technology.

         E.  Project the capital  and annualized  costs  for each
             vapor control alternative.
                            -1-

-------
1.2       SOURCES AND TECHNIQUE

          Equipment manufacturers were  contacted to determine the
          effectiveness, cost,  and  operating history  of  their
          vapor control units.   Several  alternatives were  pro-
          posed  for rendering the  potentially flammable  vapor
          mixtures  from barges non-flammable.  These alternatives
          included  dilution with  air,  saturation with gasoline,
          and blanketing with fuel gas  or inert gases.  Combina-
          tions  of  safety and control  modules were studied  to.
          evaluate  costs, safety, emissions, and reliability.
                              -2-

-------
                           SECTION  2

                            SUMMARY
Feasibility,  safety,  vapor emissions  and  costs constitute  the
basis of evaluation in this study for  controlling gasoline-barge
loading emissions.  The evaluation involves application of carbon
adsorption,  refrigeration, and  thermal  incineration  control
technologies  to control gasoline-barge  vapors.  Gasoline  loading
emissions  consist  of two portions,  an  arrival portion and  a
generated one.  Arrival emissions consist of  evaporated  residual
gasoline from a previous load.   Generated emissions are  produced
as the  barge is  loaded with gasoline.   A hydrocarbon  loading
emission  factor  of 4 #/1000 gallons  of  gasoline  loaded,  for
uncleaned gasoline  barges, was  used  for  economic analysis.   In
practice, remitted vapors have a hydrocarbon concentration ranging
from 456 to over  50% by volume.   A  significant amount  of  these
vapors'are within the 1.45& to 8.H% nominal  limits of flammability
and they represent  a  serious safety hazard  since they introduce
flammable mixtures  into the vapor  control systems.  Should an
ignition occur within the vapor control system,  not only would it
be hazardous  to  the vapor control system, but a  flame front could
travel by connecting  pipe to the interior of  the barge and ignite
the vapors within the barge.   This  could  result in a  fire  and
explosion that could  demolish the barge and dock facilities,  and
result in loss of life.

Dilution,  saturation, and inertion  have  been investigated as
primary  safety methods  for  rendering  the  barge vapors
                             -3-

-------
non-flammable  before collection and disposal  or  recovery.  Safety
modules were developed using these methods for  combination  with
the three control  technologies.  A secondary  protection device (a
hydraulic  flash  arrestor) is used  to  prevent transmission of
flames between the barge and control system in  case  an  ignition
occurs  that is caused by a primary system malfunction.   Each
combination of safety module and  control module  was studied first
for compatibility in  safety  and  control.  The more promising
schemes were  subjected  to  a  detailed  analysis  of safety,
reliability, operability, control efficiency, and cost  (capital
and annualized) .   Eleven cases passed the initial screening and
became the basis  for the balance  of the  study.

After studying the design and  operation-of gasoline-barge loading
terminals operated by  different oil companies, a hypothetical
gasoline-barge loading terminal  has  been formulated  for  use in
this report.   Since design and operation of barge terminals and
shipping procedures vary substantially for the  various operators,
the hypothetical  barge terminal incorporates features  common to
actual terminals  although it can  not be  construed to  be  typical
of all  terminals.  The hypothetical terminal uses an annual
gasoline throughput of six million barrels per  year.  Two loading
berths" for gasoline barges are assumed.   Maximum liquid  gasoline
loading  rate  was 4200  gpm (6000 barrels  per hour)  for  the
terminal.  The gasoline barges are assumed to  be 20,000  barrel
liquid capacity each, and they are in dedicated  gasoline service.
The gasoline  barge  loading emissions  for the terminal  in the
uncontrolled base  case are calculated to be 1,008,000 #HC/yr.

Control equipment/manufacturers  were contacted for  information
concerning capital cost estimates, control efficiencies,  utility
requirements,  theory of operation, safety, operating  history, and
other technical features.  The vendors'  costs for vapor  control
                              -4-

-------
units for  similar applications when  using  carbon adsorption  or
refrigeration,  were roughly equal.   The  costs  of  thermal
incineration  in the same applications were substantially  less.
Control efficiencies estimated by  the  manufacturers  for the
various cases were 99.9/6 for thermal  incineration, 98 to 9956 for
carbon adsorption, and 91 to 9.4$  for  refrigeration.

The economics of applying vapor control to  gasoline barge loading
are studied here, and estimates of  total  installed capital  costs,
annualized costs, and cost effectiveness  are calculated.   Capital
costs for  barge  modifications are also  estimated.  The safety
modules are found to contribute significantly  to the annualized
and capital  costs  of the control cases.  The proportion  of
annualized cost attributed to the  safety  module ranges  from 64$
to 98$ of  the total.  The safety  module portion of total capital
cost ranges from 24$ to 70$.  Cost  effectiveness is calculated  by
dividing the  pounds of HC controlled  (compared to the base  case)
by the annualized cost.  The most  cost effective case is carbon
adsorption with inert gas generation  at 8.50 #HC/$.  The  next  two
most cost  effective systems also use inert gas generation, and
are thermal-incineration (8.01 #HC/$) and refrigeration  (6.08
#HC/$).   At  the  annual throughput  of six million barrels per
year,  none of the schemes are profitable, and all represent  a  net
operating  loss.— Projections of cost  effectiveness are  made for
other throughputs and show that higher throughputs increase cost
effectiveness.   Some of the recovery systems are  capable  of
recovering gasoline  whose value  is greater  than  the direct
operating  cost.
                              -5-

-------
                           SECTION 3

                CONCLUSIONS AND RECOMMENDATIONS
3.1 •      CONCLUSIONS
          A.   Installing vapor emission controls systems on
             gasoline barges  introduces the potential  for large
             scale  catastrophic  fires  and  explosions of
             significantly greater magnitude  than exists in
             present day gasoline-barge loading facilities.

          B.  Excluding safety considerations,  application of any
             of  several technologies  to destroy or  capture
             gasoline emissions is  straightforward and  presents
             no  unusual problems.

          C.   The control efficiencies of the vapor  control
             systems  in  descending  order are;   thermal
             incineration,  carbon adsorption,  and refrigeration.
             The overall control system efficiencies range  from
             86% to 97$ reduction in HC.

          D.  Capital costs for the control systems range  from
             $34.7,000  to $973,000.  Capital  costs for  cases
             using thermal incineration are significantly lower
             than  those using  carbon  adsorption or
             refrigeration.
                             -6-

-------
E.  Annualized costs for  the control systems  range from
    $114,000 to $215,000.  Those  systems  using inert
    gas  generators  for their  safety  modules have the
    lowest annualized costs, and the  least  costly
    combination of safety and  control  modules is carbon
    adsorption with inert gas  generation.

F.  Safety modules are the largest contributors to the
    capital and annualized costs of the systems.  From
    64£  to 98% of the annualized  costs and 2H% to 7056
    of capital costs are  contributed by  the  safety
    modules.

G.  Cost  effectiveness, expressed in  pounds  of
    hydrocarbons controlled per  dollar  of annualized
    cost,  range from 8.50 #HC/$  to  4.20  #HC/$.   The
    most  cost  effective systems  use  inert  gas
    generation for the  safety module.   The  best  of
    these uses  carbon  adsorption  for the  control
    module.

H.  Cost effectiveness increases  with increased gaso-
    line  throughput.   At higher throughputs  some
    recovery schemes have a positive cash flow.  At the
    6  million barrels per year throughput  used in the
    study, all control cases result in a net operating
    loss.

I.  The cost and emissions evaluations developed in
    this report are based on a  hypothetical  barge
    terminal.  A specific terminal has its own unique
    design, throughput, operation,  and location which
    differ from the hypothetical  one; therefore, the
                    -7-

-------
              cost  and emissions data generated in this  report
              cannot  be applied  directly to a specific terminal.
3.2       RECOMMENDATIONS
          A.   The  design of any control system for emissions from
              gasoline  barges should provide means for rendering
              the  vapors non-explosive as  they exit the  barge
              manifold.

          B.   A secondary, passive device should be installed  at
              the  exit  of the barge vapor manifold, and at  other
              critical  locations in the system to prevent passage
              of a deflageration or detonation should the primary
              (active)  system fail.

          C.   Hydraulic flash  arresters should be used  as  the
              secondary protection devices, and  the type used  in
              acetylene service should be considered for  use  in
              gasoline  vapor service.

          D.   Any  flash arrester chosen for use in an emission
              control  system should be tested and certified by an
              independent laboratory prior  to installation in a
              commercial facility..
                              -8-

-------
                           SECTION 4

                          DISCUSSION
4.1       DESCRIPTION AND OPERATION OF A  GASOLINE-BARGE LOADING.
          TERMINAL AND A BARGE

          Initially, this control  technology  study for gasoline
          barge  loading considers some  of  the  operational and
          physical features of the loading  terminal.   Although
          there is no standard  gasoline-barge loading  facility,
          and  individual installations  vary widely,  there are
          some common functions and details.   The terminals are
          normally multi-purpose  facilities that are adjacent to,
          or part of, a refinery,  and usually they are owned by
         ;the  oil  company that owns the  refinery.  The one or
          more docks from which  barges  are  loaded may also be
          used for ships, and  in  addition  to  gasoline, they are
          often_ used to load  other  liquid  petroleum  products
          including  fuel oils,  ke.rosene, naphtha, etc.   It  is not
          uncommon  for the docks to  be used  for unloading
          petroleum materials - most often, crude oil from  ships.
          The docks usually use dedicated  or  segregated lines for
          different  services,  and  since individual lines serve  a
          particular product solely, the docks tend to  be filled
          with piping, valves,  loading arms, and hoses.  Since
          space  is at a premium, the docks  generally  have no
          fixed machinery, but  this is also a function  of safety
          considerations.  The crowding  is  especially true at
          older  facilities where the available dock  space has
                              -9-

-------
been used  for  facility expansions.  The  circumstances
just described  are normal; however,  some  facilities do
have docks used exclusively for gasoline  or  barges.

The gasoline loading pumps are usually located in the
tank farm  area.   Generally,  dedicated pumps,  storage
tanks, and suction  lines are used.  Totalizing flow
meters may be  used on the loading lines  as  part of an
inline blending station or custody transfer  system.  A
significant potential for contamination between leaded.
and unleaded gasoline  stocks exists, and,  therefore,
the  storage  and transfer  facilities  are usually
dedicated  for  each grade.  The loading rates for  barges
vary between individual  operators but are  usually in
the range  of 3000 bph (2100 gpm)  to  8000  bph (5600 gpra)
at design  flow.   The loading  lines can be throttled to
a lower flow rate.  Individual arrangements will vary
widely if  the  same  pumps or dock manifolds are also
used for  ship  loading.

-The barges used in gasoline service  are somewhat  easier
to typify  than  the loading terminals.  Barges  used in
gasoline  service  are  usually 20,000 bbl capacity and
have.^gight separate  cargo compartments.   The  overall
dimensions of  the barge are approximately 50 feet wide,
250 feet  long  and 13 feet deep.

The barges are most  commonly in dedicated gasoline
service,  and  they are  not  ballasted,  cleaned,  or
degassed  between -loadings.  A common  submerged  manifold
is used for loading  and unloading, and a series of
valves permits  isolation of tanks along  the manifold.
                    -10-

-------
The barge is equipped  with two pumps for  unloading
cargo and  stripping  the  compartments.  The pumps  are
located on the  rear deck and  are  driven  by diesel
engines.   A diesel fuel oil tank and a slops  tank  are
standard equipment.  Pressure/vacuum relief  valves  are
mounted on a header.and  configured  to relieve  either
single tanks or  pairs of tanks.

The loading operation  for barges is similar  fo'r most
terminals.  After the barge is moved into position at'
the dock,  it is  made fast by mooring lines.  The  cargo
loading hose is  lifted  into position (manually  or by
hoist) and bolted  both  to the barge loading  manifold
flange and the dock manifold flange.  The barge  opera-
tor prepares for loading by opening  the proper  valves
in the loading manifold and by opening the ullage caps
on the compartments  to be filled.  Opening  the  ullage
caps serves two  purposes;  it permits displacement of
vapor  and thereby  relieves pressure, and  it allows
manual gauging of the cargo level in the tank.   Vapor
release through  the  ullage caps, rather than through
the pressure/vacuum  relief valve prevents  a defective
valve  from overpressurizing  a barge compartment.
Although  no flow rating for  the  valves has been
obtained, it appears that the  valves are intended
primarily  for breathing applications and are undersized
for  loading applications.   Manual  gauging   is
accomplished by inserting a calibrated rod  into  the
tank to determine the liquid level.  The barge operator
inspects the liquid  level visually throughout loading
operations and advises the onshore operator to  reduce
the flow rate as the tank approaches the desired level.
                    -11-

-------
         Initial  filling is performed  by  gravity feed.   All
         valves along  the flow path are opened allowing  the
         gasoline  to flow into the barge.  After ensuring that
         gasoline  is flowing  to the correct  tanks, the loading
         pump is started.  When the tank level approaches full,
         the  barge  operator signals the onshore operator to
         reduce the pumping rate and begin  closing the valves.
         Shutting  down operations roust be handled slowly to
         avoid  hydraulic transient shock  or "water  hammer."
         After the shore side valves are closed, the gasoline_
         remaining in the hose  is  drained into the barge  and  the
         hose is disconnected.

4.2      DISCUSSION OF CONTROL  TECHNOLOGIES IN RELATION TO GASO-
         LINE LOADING

         As the introduction to this paper  indicates,  the three
         different  types of control technology being^studied
         with respect to gasoline-barge  loading are refrigera-
         tion, carbon adsorption,•and thermal incineration.   The
         '..following paragraphs discuss the three methods indivi-
         dually.

4.2.1    Carbon Adsorption
         Carbon adsorption utilizes the affinity of  activated
         carbon  for hydrocarbon  compounds to remove gasoline
         vapors from air.  A typical carbon adsorption system
         consists of two  or more carbon adsorber  beds and  a
         regeneration system for the carbon beds.  (See Figure
         U.2.1.)  Two or more beds are  necessary  to maintain the
         unit onstream  because one bed is kept  in adsorption
         service  while  the other bed is  being regenerated.   The
         number and  size of beds are determined  by the  loading
                             -12-

-------
                          TREATED  VAPOR
                          TO ATMOSPHERE
                                    FL'AME
                                  ARRE5TOR-
I
M
OJ
                                                         RECYCLE
                              CARBON
                           "ADSORPTION
(—,M
  MOV
LIQUID RING
VACUUM PUMP
            SEPARATOR
                          , i MOV     MOV

                          -§o-
         INLET
         GASOLINE VAPOR
&~
                                                              GASOLINE
                                                              WAT
              )LIN
              ER
                            GASOLINE
                                                     COOLER
                                                                                         FROM
                                                                                        STORAGE
                                                                                     GASOLINE
                                                                                      STORAGE
                                                                                       PUMP
                                                                                    >  TO
                                                                                    !STOPAGE
                                                                   GASOLINE RETURN
                                                                        PDWP
                          Figure 4.2.1  Carbon  Adsorption  Module

-------
period, the time required  for  regeneration,  and the
working capacity of  the  activated  carbon under the
operating  conditions.  Regeneration  (in  situ)  can be
performed  by vacuum or steam  or  both.   Each of these
methods relies on elevating  the  vapor  pressure  of the
adsorbed hydrocarbons relative to the absolute pressure
in the  void space  at  the -bed.   The  gasoline  vapors
desorb from the carbon particles into  the void space
where they are removed.

In the vacuum regeneration  system  (Figure U.2.1) the
carbon bed is placed under a high vacuum,  approximately
25mm Hg,  by a liquid-ring-seal  vacuum pump.   Vacuum
regeneration is especially useful  where temperature
limitations exist  due to polymerization  or  safety
problems.  The gasoline vapor  desorbed and evacuated is
reclaimed  by condensation or adsorption into gasoline
liquid. After desorption (which regenerates the bed)
the bed is returned to service.   Advantages of vacuum
regeneration include .an  uncontaminated recovery
^product, low operating temperature,  and fewer utility
requirements.

Steam  regeneration uses direct  steam contact with
carbon to  heat the  carbon and  desorb gasoline vapors.
The flow of  steam through the  bed carries the gasoline
vapor  out for  reclamation.   The  purged  steam and
gasoline  vapors are  condensed  by  cooling  and the
gasoline liquid  is  decanted  and  recovered.  After steam
desorption  and  purging, the  beds are cooled  before
being returned  to service.   Advantages of  steam
regeneration  include higher  working capacity and
increased  desorption of heavier  compounds.
                    -14-

-------
Hybrid  regeneration systems utilizing both vacuum  and
indirect  steam heating  have been designed for  carbon
adsorption  systems.  These techniques  permit additional
flexibility in  the  types of hydrocarbons treated by
combining the advantages of steam and  vacuum.

This section of  the  barge loading study presumes  the
use of  a  carbon  adsorption system incorporating  vacuum
regeneration and  adsorption at atmospheric  pressure.
The system was  selected  because it  has been  used by
several  operators for  gasoline-truck  loading,  and
because records  of the operating results  are available.
Approximately 10  units in the United States are  now in
service  for gasoline-truck, loading, and  they have
design  flowrates  of up to 6000 gpm.

                        NOTE
     For  convenience, flow  rates of  vapors  dis-
     placed by liquid tank will be given  in gpra
     rather than  the more familiar ft3/min.

The efficiency  of recovery for  carbon  adsorption is
high;  as much  as 99%  recovery in  some  commercial
installations.

The estimated costs obtained from the  manufacturer  for
the two units considered in the study  are $335,000  and
$229,000  depending on the safety technique  used with
the system. The  vendor package is complete except  that
utility connections  must  be made prior  to operation.
The carbon  adsorber beds, vacuum pump, gasoline  trans-
fer pumps,  adsorber-separator,  and the  associated
piping  and  instruments are included,  and all  are skid
mounted.  The package does not include any supplemental
safety devices  although explosion proof  electrical
                  -15-

-------
          construction is used.   The  danger  of spark generation
          in  the vacuum pump is  minimal  and  an over-rich condi-
          tion is maintained.   Hot  spots in  the adsorber beds  are
          unlikely because of  the working capacity used, the  type
          of  components  adsorbed,  and the  frequency of  bed
          cycling.

4.2.2     Refrigeration The refrigeration system studied utilizes
          two cooling  stages operating  at  approximate
          temperatures of 35°  F  and  -100° F.   (Refer  to Figure.
          4.2.2.)   The  principle  of  operation  is  simple
          condensation  at atmospheric  pressure  caused  by
          refrigerating the vapor mixture  to temperatures below
          the boiling points of the h-ydrocarbon components.
          Since gasoline  is a  mixture of compounds,  each with a
          unique vapor pressure- temperature  curve,  cooling to
          any given  temperature will condense components that
          exist as liquids at  that temperature and  atmospheric
          pressure  and  only   to the extent  permitted  by
          vapor-liquid equilibrium'.

          Of  the three vapor control  technologies  considered in
          this study, refrigeration is most likely  to  be impacted
          by  the various  mixtures of  gasoline.   For instance,  one
          recent, detailed analysis of a full-range,  motor gaso-
          line lists  over  200  different compounds with a range of
          boiling points  from  -43° F,  to 421° F, all  of which, of
          course, would condense in  the -100°  F environment of
          the refrigeration system.   On the other hand,  gasoline-
          loading vapor analyses  by two oil  refineries showed a
          significant difference  in vapor composition for gaso-
          line.  One  refiner gave the volume  percentage  of G   and
          C2  compounds in  the  vapor  at ambient temperatures as
          0.02$  while another  gave 3.38$ as the  C   and  C2
          composition at  ambient  temperatures.
                             -16-

-------
 CONDENSER
    AIR
 PRECOOLER


   J/\l
  CONDENSER
     AIR
     n
    J/NL  .
  PRECOOLER
REFRIGERATION
    UNIT  n
ELECTRICAL
UTILITY
HOOK-UP
                 /
    LOW
 TEMPERATURE
REFRIGERATION
  SYSTEM
                                           RECOVERED
                                             VAPOR
                                           FROM BARGE
                                                         TREATED
                                                         VAPOR TO
                                                        ATMOSPHERE
                                                                       COOLING
                                                                       RECOVERY
                                                                       SECTION

                                                                       VAPOR
                                                                     CONDENSER
                                                                     CONDENSED
                                                                    ! GASOLINE
                                                                     TO  RECYCLE

                                                                     CONDENSED
                                                                    'WATER  TO
                                                                     DISPOSAL
                   Figure 4.2.2  Refrigeration Vapor Recovery Module

-------
                   NOTE
     EPA  does not consider methane  and ethane
     emissions a problem  since they do not  enter
     into the  reaction  that produces  smog.

While the refiners' analyses may not be significant
from an environmental protection  point-of-view, they
are important with respect to  this basic study of  vapor
control by refrigeration  because  C   and C  compounds
(primarily methane  and  ethane)  are unlikely to be
removed to any significant degree  by  cooling to -100°F.
The normal boiling  points  of methane and ethane are
-259°F and -127°F  respectively.   However,  a small
amount will be absorbed  into heavier  compounds and thus
removed.   In  fact, overall recovery efficiencies of 94/6
have been observed.  In  the first  or precooler stage,
the inlet vapors are cooled to 3M°F  which allows high
removal of water vapor as well  as the higher  boiling
hydrocarbon components.   In the  next step, vapors are
introduced to  the  low-temperature  vapor  condenser,
which operates  between -80°F  and -115°F,  where
additional condensible  HC vapors are  removed.  The
condensed liquids  are separated into hydrocarbon and
water components by decanting.   The  recovered  gasoline
from the  condenser is returned to  storage, the water  is
sent to the wastewater system, and the non-recoverable
HC  and  air  are  vented  to  the  atmosphere.   The
refrigeration system uses a cooling recovery section
for  increased  efficiency.  Incoming  vapors to the
condenser pass  through the cooling recovery section and
are cooled  by  heat  exchange with  the  exiting
non-recoverable  vapors.   The exiting  vapors warm  to
approximately 75°F during this process step.
                   -18-

-------
         A warm brine system  is  included  in  the  equipment for
         periodic  defrosting of the  low  temperature vapor con-
         denser section.  The defrosting operation is performed
         during non-loading  periods  and  requires one  to two
         hours.  Applications which  require dual condensers are
         those in  continuous service or  those  which operate in
         high humidity and build up  ice  rapidly.  The  condensers
         are cycled  between defrost and operating modes.  The
         models used for evaluation and costing  in this study
         include dual condensers.

         Gasoline vapor recovery by  refrigeration  for  truck
         loading  operations has been accepted by some oil
         companies and approximately 70  units- are  now in  use.

         The price estimates, obtained  from  the manufacturer's
         local representative, for the  two refrigeration vapor
         recovery units described  in  the study  are $300,000
         each.  The  vendor package is complete except that
         utility connections  must be  made prior  to operation.
         The  skid mounted  package  contains   the cascade
         refrigeration  units,  defrost  system, condensers,
         separator,  recovered gasoline  pump, and the piping and
         instruments.  The unit  is  constructed to  meet
         applicable explosion  proof  codes,  but it does not
         normally  include any additional safety devices  such as
         flame  arreators.   However,  the  precooler  and  vapor
         condense-r  coils  (Operating  at approximately  -100°F)
         could be  regarded as wetted flame arresters.

4.2.3     Thermal Incineration

         Thermal  incineration (Figure  4.2.3)  disposes of
         gasoline vapors  by  burning  rather  than by  product

                             -19-

-------
to
o
i
                                TREATED VAPOR
                                TO ATMOSPHERE
 PILOT
BURNER
                        FUEL
                    V-

                     VAPOR
                    BURNER
                              I	

                           AIR DAMPER
                                                  STACK
MAIN BURNER
                                         , INLET GASOLINE
                                          VAPOR
                                       WATER SEAL-
                           Figure 4.2.3 Thermal Incinerator Module

-------
recovery  methods.   It  provides the simplest  and  most
direct control  technology for  gasoline  and other
hydrocarbon  vapors.  A typical  unit  as  provided  by one
manufacturer is described in this  study.  The collected
gasoline vapors  are  injected  into  the  combustion
chamber of  the incinerator through a burner manifold.
A temperature of approximately  1500°F is maintained in
the chamber.

The residence time  in  the chamber is  on the  order of
one second.   Pilot  burners provide ignition  and  main
burners (supplementally fueled) are  used, as necessary,
to maintain the  proper combustion  temperature.
Automatic air dampers admit combustion  air and  are set
on temperature control.  The efficiency  of thermal
incineration  is  higher  than  that  of competing
technologies.  Over 99.9% of the hydrocarbon vapors are
converted to non-hydrocarbon combustion products  such
as CO  or CO.  With little or no incinerator modifica-
tions, similar results are obtained  for a wide range of
hydrocarbon  compounds, and, of course,  this  is  one of
thermal incineration's chief advantages.  Thermal in-
cinerators  are available from several  manufacturers,
and they have been installed by  a number of  gasoline-
truck loading terminal operators.  The  estimated  cost
given  by one manufacturer of an incinerator  with a
design flow~of 5000 gpm,  is $64,000.  Their estimate of
a duplex'unit capable of  10,000 gpm is  $128,000.   The
price includes the incinerator, monitoring devices, and
protective  water seals.   The vendor supplied  package
requires only utility connections for operation.
                   -21-

-------
         The incinerator operates at atmospheric pressure.  The
         total pressure drop across the  unit  is approximately
         12" H 0,  and it is due almost  entirely to the water
         seals on  the vapor inlet.  The  thermal incinerators
         investigated for the report  include several  safety
         features:

         1)  As previously mentioned, for  greater reliability,
             two water seals are installed in the vapor  inlet.

         2)  An interlock for low-water  seal  level is used  to
             prevent startup.

         3)  Ultraviolet radiation detectors  are incorporated
             into the pilot and main burner interlock controls.

         4)  Emergency shutdown is used  to  prevent damage  by
             excessive temperature.

         5)  The incinerator shutdown and  interlock systems  have
             a safety interface with the  vapor collection and
             gasoline loading systems.

         6)  The incinerator burners  themselves are designed
             with quenching slots  to prevent  flashback  in the
             combustible vapor stream.

4.3      GENERAL DISCUSSION OF SAFETY TECHNOLOGY

         The two major safety problems  associated with gasoline
         barges  and vapor control  systems  are fire  (and/or
         explosion) and overfilling.
                               -22-

-------
          The  following study considers  safety problems  as  they
          exist and as they would be  if  vapor  control and  safety
          technology were used.

4.3.1      Fire and Explosion
          With the current gasoline loading  system, it is routine
          for  fuel and air, two of the  three elements required to
          initiate a fire or explosion,  to be  present around the
          barges.  They are there  because  the heavier-than-air
          gasoline vapors displace, the  air  near the deck  of the
          barge.   The third  element,  ignition,  is  a  constant
          danger  since the potential for static  electrical and
          friction sparks., improper or  malfunctioning electrical
          equipment, unauthorized smoking,  operating tug  boats,
          etc., can never be controlled  absolutely.  Until  now,
          fire prevention philosophy has  been  to  accept  the
          presence of combustible  mixtures both  inside  and
          outside of the barge, and to  concentrate on eliminating
          ignition sources  while hoping the  air currents  are
          sufficient to disperse  the  vapors  and  keep  them too
          lean to ignite.

          Addition of a vapor control system would eliminate one
          problem since vapors would no longer be  released  near
          deck level.  However, a new problem  caused  by  connect-
          ing  an  on-shore control facility  to  the  barge -  with
          both containing explosive mixtures - would  be  created.
          Since the  facilities would be connected,  new   ignition
          sources would be created.  Effectively,  there  would be
          a closed system, and  a chain ignition  could  occur in
          either  direction.  Ignition from  the shore  to  the  barge
          would  be of most  concern  since  blowing up  the barge
          would have the more  disasterous  consequence.  Safety
                             -23-

-------
problems attendant  to this closed system are  more
severe  than  the safety problems with the  present system
since ignition using the current  loading technique
should,  at worst, cause a flash fire on the  deck of the
barge.

Potential ignition  sources in the  onshore equipment
include  rotating parts, static  electricity,  tramp metal
in lines, and  external fires.

The philosophy adopted in this  study for  protection of
barges  and vapor control systems from fire/explosion
utilizes two systems  employed  together.   These  systems
are:

1)  A primary  (active) system  that  renders the vapors
    emitted from the barge non-flammable by  one of
    three means;         -                    ~

    a)   Dilution with  air,

    b)   Replacement  or reduction of air by fuel gas,
        N  ,  or combustion gases (also called blanketing
        or inerting) ,

    c)   Saturation of  the vapors with gasoline.

2)  A secondary (passive) system uses a  single device
    that would preclude allowing a flame  front  (defla-
    gration)  or a detonation  to pass either  to the
    barge or to the  dockside facility.  Both components
    must be  used in  the overall system.  The  primary
    system,  no matter how well designed, can  have
                   -24-

-------
             equipment fail and  allow  exploding vapors to  enter
             the vapor control system.  Should an ignition occur
             under these conditions, the  secondary system  must
             be installed to prevent the  flame from reaching the
             barge and creating  a major catastrophe.

U.3.1.1    Primary (Active) System.- The primary system protects
          the barge loading area  from fire or explosion by alter-
          ing the barge vapors to make  them non-combustible.  The
          principles of the  three  active  fire prevention  tech-
          niques are discussed in the following paragraphs.

          A.   Dilution. -  Dilution  of  the  gasoline  vapors
             recovered from the  barge  constitutes  the simplest
             technique for rendering the  mixture non-flammable.
             Dilution is accomplished  by adding and  mixing
             enough  air with  the vapors to reduce  the  HC
             concentration below the lower explosion limit (LEL)
             (approximately  1.4? gasoline by volume).   The
             initial  HC  concentration  determines the  amount of
             dilution required.

             To determine  the volume  of air required  for dilut-
             ing a given HC concentration, the maximum  concen-
             tration  by  volume for a  given set of conditions is
             established,  and  the volume of air  required to
             reduce this to 1.4? by volume is calculated.  The
             theoretical maximum HC concentration  is  the ratio
             of the gasoline  true vapor  pressure  (TVP) to the
             absolute (atmospheric) pressure of  the system.  For
             a TVP of 7.4  psia, the theoretical maximum  is 50.4?
             HC by volume.  To  reduce the HC concentration to
              the LEL,  (ratio  of HC  to LEL  or 50.4:1.4)
                             -25-

-------
    the volume of air required  is  35  additional volumes
    of air per volume of initial mixture.   Allowing a
    safety factor of  four  (25% LEL) the required  air
    volume is 143 times the initial  volume.  Therefore,
    the maximum efficient dilution air  flow  rate is 143
    times the gasoline vapor flow  rate.

B.  Inertion. - Another means of rendering the mixtures
    non-flammable is by inertion.  This technique  uses
    gases such as N2, C02>  treated flue gases,  or  fuel
    gas, to replace  air in  the  barge  vapors  and  reduce
    the percentage of oxygen present.  Depending on the
    type of  inert gas used  and the HC mixture,  the
    maximum  oxygen  concentration'allowable in  the
    mixture is approximately 12-15?  by volume.   Inert
    gas can  be  introduced  to  the gasoline vapors  by
    several methods.   The preferred method is  to  admit
    inert gas into the barge as  the liquid gasoline  is
    unloaded  by the  receiver.   Theoretically, this pre-
    cludes  an explosive mixture  being encountered.   As
    the gasoline is  unloaded, the  air that would  norm-
    ally enter the barge  and  create  an explosive mix-
    ture is replaced  by the  inert  gas.  During unload-
    ing and the  return transit,  gasoline would evapo-
    rate as it normally does, but  it would do  so  in  an
    essentially  inert atmosphere.  When the  barge  is
    reloaded  with gasoline  the  non-explosive mixture  is
    displaced into  the  vapor  collection-treatment
    system.   The amount  of  vapor that is treated  is  at
    a  minimum because  no  other gases  are added.

C.  Saturation. - Saturation  of  hydrocarbon vapors
    renders  the  mixtures  non-flammable by enriching  the
                    -26-

-------
             HC concentration  of the stream significantly above
             the upper  explosion  limit.  The upper  explosion
             limit is about  8.4$, and the saturation  concentra-
             tion (which varies with temperature)  ranges  from 30
             to 50$  for normal  temperatures.  The  vapors  are
             saturated by passing them through a vessel,  speci-
             fically designed  for this purpose, where gasoline
             liquid is sprayed into the stream.  As the gasoline
             spray  evaporates,  the HC  concentration  of  the
             stream increases.

M.3.1.2   Secondary  (Passive) Systems.-  Several  secondary
          devices are available  to prevent flame  or  flashback
          propagation in  the  vapor control  system.  The  devices
          are based on one of two principles  although some  may
          utilize both.  One  device uses  metal tubes  or  plates
          which offer a  large  surface area to the vapor flow.
          The principle is to provide a large metal  surface which
          acts as a heat  sink and quenchs the fire  preventing its
          propagation. The other type uses  a liquid seal  leg to
          disrupt the continuous vapor flow  and  prevent flame
          propagation.  A variation of the  liquid  seal type  has
          been used to contain  and stop detonations  in  acetylene
          plants.  Their effectiveness was demonstrated  in  a plant
          accident where  a large  detonation,  between  acetylene
          plants, in  a pipeline  was contained (Refer to Trade
          Journal #1)-.  A demonstration  of the usefulness in
          gasoline "vapor  service  is warranted.  This  device is
          shown in Figure 4.3.1.   Several  secondary  protection
          devices are used to  isolate various portions  of  the
          system (e.g., the barge is separated from  the  active
          portion of the  safety module and both will be isolated
          from the vapor  control  module) .  A  flame arrester is
          also  required for  the  vent stack of the  carbon
          adsorption and refrigeratio.n  vapor recovery  units
                            -27-

-------
LIQUID
MAKEUP
                      CHECK
                      VALVE
                              »— INLET
                                         OUTLET
                                       SCHEDULE 80
                                      THREE-WAY
                                      VALVE
                               DRAIN
      Figure 4.3.1  Hydraulic Flash Arrestor
                       -28-

-------
          because these systems can  emit flammable vapors.

M.3.2     Barge Overfilling
          The  present method used  to load barges requires one  or
          more open hatches per compartment  for the operator  to
          gauge liquid levels by direct sight.  While this  method
          is extremely simple and  reliable, it poses emission and
          safety  problems.  Vapor  control  for gasoline barges
          would require discontinuation of this system and  use  of
          a closed  gauging system.   Since  the consequences  of
          overflowing a compartment  are severe, a combination  of
          a gauging system and  warning devices  to  indicate
          excessive liquid levels  is warranted.  A direct reading
          level indicator utilizing  a stainless steel  tape and
          float (housed in a stilling well) are used.   (Refer  to
          Figures 4.3.2 and 4.3.3.)   High  level  floats with
          externally  mounted  switches provide the warning and
          they could  be used  to  shutdown  the systems^  As  an
          alternate to the level indicator, a  transparent window
          for  an  observation port  could be used  for direct
          sighting,  provided  lighting and  a method to prevent
          obscuring the level are  furnished.

4.4       APPLICATIONS OF SAFETY AND CONTROL TECHNOLOGIES

          There are fifteen combinations  of  safety and control
          modules available when studying  the three control
          technologies and the five  safety  technologies.  Refer
          to Figures  4.2.1, 4.2.2,  and  4.2.3 for  the carbon
         'adsorption,  refrigeration,  and  thermal incineration
          controls, and to Figures 4.4.2, 4.4.3, 4.4.4,  and  4.4.5
          for  the safety technologies using dilution, saturation,
          and  inerting with three  different gases.  Also refer  to
          Figure 4.4.1 for the overview or system block  diagram.
                              -29-

-------
                                                       MODIFIED ULLAGE
                                                             CAP
o
I
    PRESSURE VACUUM
    RELIEF VALVE

     ULLAGE
      CAP
   (OPENED DURING")
   v  LOADING  ^
RELIEF HEADER
CONNECTS TWO
CARGO COMPARTMENTS
          WARNING
          DEVICES
                        .	VAPOR
                       tOMPARTMENT
FLAME ARRESTOR
                                         XVVAPOR COLLECTION
                                         /     MANIFOLD
                                    EXTEND TO
                                    WITHIN  6"
                                    OF BOTTOM
                                           MODIFIED ULLAGE
                                         /CAP  INTERNALLY
                                         '  LIGHTED  TRANS-
                                           PARENT COVERED
                                           INSPECTION PORT

                                        FIXED CALIBRATED
                                        GUAGEING STICK


                                   f   DUAL (REDUNDANT) LEVEL
                                   L— SWltCHES AT DESIGN
                                      MAXIMUM LIQUID  LEVEL
       FXIST. BARGE VAPOR
        SYSTEM  (TYPJ
                  MODIFIED  BARGE VAPOR
                       SYSTEM
                            Figure  4.3.2  Barge Modifications

-------
                    PV VALVE SIZED TO PROTECT
                    ALL TANKS DURING LOADING
i
U)
M
I ,
      EACH PV VALVE PROTECTS PAIR OF
      TANKS SIZED FOR BREATHING ONLY
         FUEL OIL
        .   AND
        SLOP TANKS
»CARGO PUMPS
 PUMP "ENGINES
                     Figure  4.3.3  Barge Vapor Collection Manifold

-------
A number  of  these combinations can be eliminated  from
analysis  because of incompatability between the control
and safety modules for technical or economic  reasons.
The  dilution  scheme is  incompatible with carbon
adsorption or refrigeration due to  the high cost of the
control  equipment  and  low removal  efficiencies.
Enrichment of the barge  vapors with fuel  gas is  also
incompatible with carbon adsorption or  refrigeration
because virtually all of the light  HC components  would
be emitted to the atmosphere at an  economic loss.

Eleven cases have been investigated.  Thermal  incinera-
tion  is  used with  all  five safety modules.   Carbon
adsorption and  refrigeration are used with  saturation,
with  inerting  using  N2> and with  inerting   using  an
inert gas generator.  The inert gas generator used  in
the study burns  natural gas under carefully controlled
conditions to  produce  an oxygen-free  pro-duct  of
scrubbed  and cooled combusion gases.

The predicted exit vapor compositions from  each of the
safety modules  is given in TABLE 4.4.1.  Compositions
are given for the three different initial HC concentra-
tions  used  (5%,  18$,  and 50%).   These  compositions
represent the vapors that are received by the  treatment
units.

The control  efficiencies and economics that  are  deve-
loped in  this report  are based on an initial HC  con-
centration of 1856.  The design gasoline loading rate is
6000 barrels per hour (bbl/hr) or 4200 gpm.  The annual
gasoline throughput  for the  barge terminal is six
million  barrels per  year.   It is assumed   that 300
barges, each 20,000 _b.bl capacity, are loaded each  year.
                     -32-

-------
U)
I
                                                   TREATED
                                                    VAPOR
 VAPOR
"CONTROL
.MODULE
                                                                                GASOLINE
                                                                                .STORAGE
                                                                                 TANKS
                                                                      GASOLINE
                                                                     PUMP HOUSE
                   Figure 4.4.1   Loading and Collection System Layout

-------
7.7 ppm
50.4 ppm
168 ppm
123.9 ppm
99.96$
26.6 p pm
182 ppm
605.5 ppm
- 445.9 ppm
99.87$
.01$
.05$
.17$
.12$
99.65$
                           TABLE 4.4.1
                  PREDICTED VAPOR COMPOSITIONS

Safety Technique:   Dilution with air to maintain 25$ L.E.L.
                   (= .35% total HC by volume) @ 50$ initial
                   concentration

Design Flow Rate = 353,000 GPM (47,000 ACFM)
Vapor Analyses for Design
Initial Total HC            5%            18$             50$
Components by Volume
c, +ca

C3
s
C  + Heavier
Air
Additional Volumes of
  Air Added               141.86         141.86       141.86

Safety Technique:   Fuel Gas is used to blanket the  system and
                   maintain 50$ fuel gas in the vapors
Design Flow Rate = 9900 GPM (1325 ACFM)
Initial Total HC            5$            18$             50$
Components ($ by Volume)
C  + C        _              .06            .19           .53
C                             .36-           1.30         3.61
C3                          1.20            4.33        12.10
C  and Heavier                .89            3-19         8.90
Fuel Gas*                  50             50           50
0        .                   9.98            8.6          5.25
 2
Additional Volumes of
  F.G. Added                1.0             1.0          1.0

*Fuel Gas can be assumed  to be 100$
                            -34-

-------
.66
4.32
14.4
10.62
30
.66
4.32 ,
14.4
10.62.
30
1.06
7.22
24.03
17.70
50*
                        TABLE 4.4.1  (Cont)
                   PREDICTED VAPOR COMPOSITIONS
Safety Technique:  Saturate vapors to 30% HC
Design Flow Rate = 5900 GPM (8'00 ACFM)
Vapor Analysis for Design
initial Total HC           5%             18%            50%
Components  (%)

C3
C4
Cj. and Heavier
Final Total HC
Volume Increase  due to
  Gasoline  Saturation   .   36%             17%

*_Fully Saturated
 Safety Technique:   Inert  Gas  Blanket
 Design Flow  Rate =  4950 GPM (660  ACFM)
 Vapor Analysis  for  Design
 Components  (%)""
 Cl + C2
 C3
 C4
 C  and Heavier
 Inerts
 O.
"5%
.11
.72
2.40
1.77
95
0
18%
.38
2.60
.8.65
6.37
82
0
50%
1.06
7.22
24.03
17.70
50
0
                                -35-

-------
4.4.1      Thermal Incineration  by  Dilution
        .  (Figures 4.2.3 and 4.4.2)

          A dilution system reduces  the HC concentration  from  the
          initial concentration (as high  as  50%)  to 25% L.E.L.
          (.35% HC) .  Dilution  is  accomplished using a 47,000  CFM
          blower  which would- be  10056  spared.   The  vapors are
          diluted upstream of the  blower,  a  potential  spark
          source.  The barge vapors  are introduced into the suc-
          tion side of the blower  at 50% of  the maximum loading
          rate, which is 2100 gpm.   This step increases the  load-"
          ing time and was taken to  reduce  the  size and cost of
          the system.  This safety module  is simple to operate
          and extremely reliable.  However, it is compatible only
          with  thermal incineration.  The  diluted  vapors are
          transferred to the thermal incinerator and raised in
          temperature using supplementally  fueled burners.  In
          the event of incinerator failure, the vapors are  safely
          released to atmosphere  since they  are not flammable.
          Interlocks are provided  on the incinerator to prevent
          operation if the dilution  system  fails.  An emergency
         "shutdown for loading  pumps, in case of dilution system
          failure,  is desirable.   Overall  system  simplicity
          insures very high reliability.   Disadvantages of this
          system  are the high fuel  usage  necessary  to burn  the
          lean vapors and the large  flowrate  which necessitates
          large and expensive equipment.

4.4.2      Thermal Incineration  with  Saturation
          (Figures 4.2.3 and 4.4.3)

          Vapors  displaced  from  the barge go directly  to  the
          saturator  tower.   There they are  contacted  with
                             -36-

-------
                   6" VAPOR RETURN LINE
  6  MARINE   .
    LOADING 1^
     HOSE
I
to
                                                         WEATHER CAP



*«t
*tn


i
\

n^ 	 ,


UJ
cr
CO
^bin .. ff
J, VV rO^1 — —
1 V \
[ I
\ /
LJ
o:
CO
" ' \ ^^^-36" DILUTION AIRLINE
^ '» n AMF
^RE^r?r°N rARRESTOR-\ DILUTED VAPORS
ORIFICE / \ ^ TO VAPOR
J.^yv ° ^ rnr\|TRni
A' . UNIT
5 \ BLOWERS
g V. 47000 ACFM
(2-!00^ UNITS)
i
i
i
!£ " FROM
DX] 	 » i 	 W*vl* / \ bw^ULIlNL
GASOLINE ( *7 	 "5blORAGE
i— CONTROL 2±1 . ..AniMr TANKS
. , . . . ,r |T i OAl J N(i
\/Al VF \_ i_w'-*L-"iNvj
v L PUMP
*:
0
Q
                             Figure 4.4.2  Dilution Module

-------
         r
Ul
CD
1
C
I
\
•-•
l
\ 6" MARINE
v LOAD ING
HOSE
1
I (PC)
1 1 MP
/ LJ
/ . r,
o:
CD
/
"1
k Y J
hi 	
A SATURATED VAPOR
6" r~\ 	 V3 — o — (sj— HTO VAPOR
RI n\ /F& yrr CONTROL CON i ROL UNI i
C-)l_xy WC.r\.-.J S i "Lt \/Al \/P
OvJv-' '^\^i 1 1 JTT-
(M^T°C%) ! p- / ! LOU PRESSURE
\UNITS / ! _, <: Ji ' irTEAM
^ CONTROL VA^VE j ," ^ CONDENSATE
c "\ Y -\ ' : ' KLIUHN
/" "S. YiWAVxnWihS? 1 , 'J'^Sw
(FIC)
6«L" i ii
/-*-
SATURATIONy Q_
TOWER .
_t_
f^* i ' f^Mji
PQ* 	 1 )^ U-*vT*
CONTRC
VALVE
*:
o
Q
« ' >; < \
_ , ,A> y 4 .. -.(Fc INO Fl D\J
w X^^^ — T \r-yiMU TLUW (^LQC.,^
p ^^HFJvrER'^CQN"^01-
Q76MMBTU/HR. VALVE !
"7^ CLOSE
. vL%bPENl
rfe) r
VLLSTOP,"
(^r\ r RE^ipri i|_ATinM
V^X H — PUMP
0 l5 GPM |" GASOLINE
r-l-l - ' 	 < TDr>M
MAKE-UP STORAGE
VALVE
INCLINE
\\ f _T\ ^ FROM
XX STORAGE
LOADING TANKS
PUMP
Figure 4.4,3  Saturation Module

-------
         gasoline  which raises  the concentration  of  HC in the
       *  vapors well  above the upper explosion limit (UEL) .  The
         gasoline is  recirculated by a  small  pump through a
         steam heated  heat exchanger to maintain the gasoline at
         the temperature  required  for adequate vaporization.  A
         demister on  the outlet  removes  entrained droplets.
         Level switches control  the gasoline  make-up  valve and
         prevent the  pump from operating dry.   The  steam control
         valve is  set on  temperature control  with  a flowswitch
         override.  Liquid gasoline flow rate can be varied  with
         the vapor flow  rate  by  means of a fixed ratio  control-1"
         ler.  The vapors are transported by blower to the ther-
         mal incinerator. Combustion air is added  in  the incin-
         erator because  the  vapors are  too_ rich for  efficient
         direct combustion.   However, no  supplemental  fuel  is
         necessary  and  only the  pilot burners are  used  to
         maintain combustion.   Upon failure  of the saturator,
         the thermal incinerator is automatically prevented from
         operating.   A  disadvantage of a  saturator is  its
         relative complexity   compared  to  the other safety
        r.modules, and this reduces its overall  reliability.

4.4.3-    Thermal  Incineration with Fuel  Gas Blanketing
         (Figures U.2.3 and  4.4.U)

         Fuel  gas blanketing of the barge  vapors is accomplished
         by mixing equal volumes  of  fuel gas  and vapors which
         cause blanketed vapors to be  5056  fuel  gas and  too rich
         to burn.  Assuming  the fuel  gas is mostly methane, the
         UEL of. which is 13.5$,  a safety factor of  4  is used.
         The fuel gas blanketing system  resembles  the  dilution
                            -39-

-------
                                  4' FUEL GAS LINE
                                                                        r FROM REFINERY
O
1
i
^
6 MARINE LOADING HOSE
y ^
f
\
i
V '
r
k
\
CHECK
VALVE
_^
NO FLOW !
STOP i
I
8% .A-
BLOWERS
(2-lOO?o
rZilT"" ^J_
r—tXll 	 ^^ 	 1
LU
1 n
/ a
/ <
/ m
/NJW \ \ . I/ M
CONTROL
g VALVE
0
Q
FUEL GAS
REGULATOR
~i
RATIO
, W/MIN.
r" P
— Q— 6y— — o
CONTROL
VALVE
1325 ACFM
UNITS)
i
/" -^N
LOADING
PUMP
3 FUEL GAS
t
CONTROL . '
FLOW BLANKETE
— [s}-*$ TO VAPOR
UNIT
, GASOLINE
' J STORAbh
SYSTLM
D VAPORS
CONTROL
FROM
TANKS
                        Figure 4.4.4  Fuel Gas Blanketing Module

-------
 module closely except that  fuel  gas is used instead  of
 dilution air.

 A  fuel gas regulator maintains the flow of fuel  gas  at
 the  proper pressure and a restriction orifice  prevents
 the  barge  vapor  flow  rate from  exceeding  a
 predetermined maximum.   The  over-rich gas is  sent  to
 the  incinerator by  a  blower where  combustion  air  is
 added.  No supplemental fuel  is  necessary and  only  the
 pilot burners are necessary.   The reliability is almost-
 the  same as the dilution system  and the same interlocks
 and  safeguards apply.

 Thermal Incineration with N g-Blanke'ting
 (Figures 4.2.3 and 4.4.5)

 The  proposed N2 blanketing  scheme  varies  from  the
 earlier  safety schemes  in that the  barge vapors  are
 made safe  at the barge  unloading  terminal.   This  is
 done by introducing the N2 gas as the barge is emptied,
"and  the  gas  replaces  the  air  that  would normally  be
 allowed to enter the barge.  The N2 gas is stored  as a
 liquid  and vaporized  as necessary  in an ambient  air
 vaporizer.  When the  barge  is  loaded, the  expelled
 vapors contain a mixture of  gasoline and N2  which  is
 non-flammable.  These vapors  are transported by  blower
 to the incinerator.   Since  the average HC concentration
 (1856) is above the UEL, only  combustion air and  pilot
 burners are necessary for incineration.

 The  pressure vacuum relief  valves  must be operating
 correctly  to prevent  excess air infiltration from
 forming an explosive mixture in the  barge.   Sampling
                    -41-

-------
6" MARINE LOADING HQSE
                                  HtXH-
 INERT GAS LINE

 '4' /
                                 UJ
                                       N2 REGULATOR VALVE
                                   o
                                   o
                                   Q
AMBIENT TEMPERATURE L|QU|D N  DEWAR
          VAPORIZER STORAGE TANK.
                                       6' MARINE LOADING HOSE
                              GASOLINE UNLOADING SCHEME
                    TO GASOLINE
                    STORAGE
6 MARINE LOADING HOSE
                                BLOWERS 700 ACFM
                                (2-IOO;0 UNITS)tf&
                                                   ^l^•
                    INERTED VAPOR
               )	•$ TO VAPOR
                    CONTROL UNIT
                                                GASOLINE   U-4*-
                                                CONTROL  2r^
                                                 VALVE   LOADING
                                                          PUMP
                  _e GASOLINE FROM
                  "^ STORAGE TANKS
                               GASOLINE LOADING SCHEME
                      Figure 4.4.5  N2 Blanketing Module

-------
         the  barge  vapors might be required  before  gasoline
         loading is  started.  N2blanketing is simple,  and  its
         reliability is  high.

4.4.5     Thermal Incineration with Inert Gas Generation
         (Figures 4.2.3  and 4.4.5)

         Except  for using an  inert gas generator  rather  than
         liquid  N2  as the blanketing medium,  this scheme  is
         similar to the N2  blanketing scheme.   The  inert  gas_
         generator supplies scrubbed and cooled  combustion  gas
         to the barge.  An inert  gas generator burns a stoichio-
         metric mixture  of fuel  (gas or oil)  and  air  carefully
         to obtain  a gas with virtua-lly  no  free oxygen.   The
         inert gas generator is  more complex than the N2 storage
         and  vaporizer system  and, therefore,  is   considered to
         be  somewhat less reliable  -  but  not  enough so  to
         warrant its rejection.   The system offers  many of the
         same advantages as  N 2  gas and is  equally compatible
         with the control equipment.

4.4.6    Carbon Adsorption with  Saturation
         (Figures 4.2.1  and 4.4.3)

         The  saturation module  operation is discussed  in detail
         in   Section 4.4.2.   The carbon adsorption system
         receives  the  saturated mixture downstream  of  the
         blowers.   The effect  of saturation is to increase the
         removal efficiency of  the  carbon  adsorption unit;
         however, due to -the  higher HC loading, the beds must be
         sized larger  than  they would  if  receiving  vapors
         directly from the barge. Switching from the  adsorbing
         mode to the  regenerating mode  is initiated  automati-
         cally by a  timer.  All  valves  are actuated  automati-
         cally.  The desorbed  vapors are  passed through a
                             -43-

-------
 adsorbing/condensation separator  column of  cool
• gasoline from storage.  Condensed water is decanted in
 a separator.   The  water is sent to sewer for subsequent
 treatment and disposal.  The recovered gasoline  product
 is returned  to storage.   Because  no segregation of
 vapors from leaded  and unleaded stocks is proposed, the
 coolant gasoline is leaded to prevent contamination of
 unleaded gasoline.  Reliability for this case is lower
 than  average  for the cases  studied due  to  the
 complexity.   It is  possible for the  exit vapors to be
 flammable,  even  though  the influent vapors  are
 over-rich.  To  prevent  a  fire that  originates  at the
 vent from  propagating,  flame arresters and careful
 positioning of  the vent is necessary-.

 Carbon  Adsorption with N 2  or  Inert  Gas  Generator
 (Figures  4.2.1 and 4.4.5)

 These  cases are sufficiently  similar that they  can be
 discussed together.   The•operations of  the  N2  module
 and inert gas generator module  are discussed in 4.4.4
 and 4.4.5 respectively.  The  carbon adsorption  system
 operates  in the same way as that used with saturation.
 However,  due to the  lesser  HC loading, the  carbon
 requirements are lower.  The  possibility  for  flammable
 exit gases is reduced  by the presence of N 2 or inert
 gas.  With respect  to  reliability relative to the other
 systems, this system ranks about average.

 Refrigeration  with  Saturation
 (Figures 4.2.2 and  4.4.3)

Operation of the saturation module  has  been  described
in Section 4.4.2.   The refrigeration  unit is designed
                      -44-

-------
        . to operate automatically without  operator attention.
         Condensed  water and HC are  collected and separated.
         Recovered gasoline is sent  to  leaded gasoline  storage
         tanks  to prevent -contamination.   The refrigeration
         system also handles  defrosting  automatically.   It  is
         possible that the ex-it vapors  will be  in the  flammable
         range.   Therefore, flame  arresters and proper placement
         of the vent is mandatory.   The  overall complexity  of
         this case  is among  the highest studied and  for this
         reason reliability is affected.

4.4.9    Refrigeration with N? or  Inert Gas Generator
         (Figures 4.2.2 and 4.4.5)

         The operation of the N 2 blanketing and inert gas gener-
         ators  is discussed in Sections 4.4.4 and 4.4.5.  Due to
         their  similarity they can be discussed together.  The
         type of  refrigeration system for use with  these two
         safety modules need not be  changed from that  used with
         .saturation because  they  are designed for saturated
         v'apors.  The possibility  of flammable exit  vapors  is
         low due  to the presence of N2 or inert gas.   Based  on
         the degree of complexity for the overall systems, the
         reliability is about average for the cases  studied.

4.4.10   Other  Design Considerations
         The features of individual gasoline  barge terminals are
         significantly different than those of  the hypothetical
         terminal.  These differences would,  of course, change
         the specific design  conditions.  For example, if the
         cargo  services for the barge were switched from gaso-
         line to  diesel oil, gasoline vapors  (the arrival
                              -45-

-------
          component) would be  displaced when  diesel was loaded.
          Although  the HC concentration  of  the evolved vapors
          would be lower than  for gasoline  loading,  they would
          still  present significant  safety  and  emissions
          problems.  It would,  therefore, be desirable to control
          this mixture of gasoline and diesel  vapors.  However,
          if gasoline and diesel  were loaded  at  separate docks,
          the gasoline barge vapor control  system would have to
          serve  the diesel oil  loading  dock.   This  is  not a
          design consideration  used on the model and would  affect.
          the conclusions as well as the design and operation of
          the control facility.

          Some means for handling condensed and entrained liquid
          hydrocarbons within the vapor  collection  system is
          necessary.  Condensed HC is a problem particularly  with
          respect to the saturation and inertion safety modules,
          because the lines are full of vapors that condense due
          to nocturnal cooling  when the  system is  shut  down.
          This problem can be  solved by sloping the vapor collec-
         T..tion lines to a liquid  knock-out  drum.   The condensed
          liquids  will drain to the drum and  can  be  removed
          periodically.  Another  solution  for the condensation
          problem is to purge  the system with  air by  continuing
          to  operate  the cont.rol  system after loading is
          completed.

U.5       EMISSIONS AND CONTROL EFFICIENCIES

4.5.1      Primary Emissions

          A hypothetical, non-explosive composition  for vapors
          emitted  from the barges is used  in this study.  The

                              -46-

-------
 composition of the  stream is altered by various sa.fety
 modules  to  insure  that explosive  mixtures  are never
 sent to the vapor recovery system.  Composition of the
 unaltered  vapor as it leaves  the  barge is shown  in
 TABLE 4.5.1.  The hydrocarbon vapor has a concentration
 of  185& by volume,  which is equivalent to an emission
 factor of  4#H.C/1000  gallons  of gasoline filled,
 assuming an average molecular  weight of 66 #/#mole.
 These figures are based on test data developed  by  Exxon
 and as reported in  EPA-450/3-76-038a,  page  127.   For
 full  loading "of a  20,000 bbl  capacity barge, the HC
 loss,  therefore,  is 3360 pounds  when the vapors are
 allowed to vent directly  to atmosphere.  Since  this is
 the method by which barge vap.ors are being handled, the
 3360 pounds HC loss per barge loading  is the emission
 quantity used for  the  base case.  For an annual  rate of
 300 barges, the predicted loss  is  1,008,000  pounds of
 hydrocarbons per year  for the hypothetical terminal.

 The stated efficiencies -for the  vapor  control  modules
:were  obtained from the  equipment  manufacturers,  or
 their representatives,  for optimum operation  under the
 vapor conditions given  to them.  These vapor  conditions
 are listed in TABLE 4.5.2.

 TABLE  4.5.3 describes the manufacturers'  predicted
 efficiencies.  For  comparison,  a  short survey of EPA
 test reports of tank-truck loading controls was per-
 formed  to  obtain  some actual, measured results of
 control efficiencies.   (EPA Reports PB-243-3&3,
 PB-275-060, EMB-77-GAS-19, EMB-76-16,  and EMB  77-18.)
 The results of this survey are  shown  in TABLE 4.5.4,
 and  they indicate that  for  carbon  adsorption, the
                    -47-

-------
                            TABLE 4.5.1
                 GASOLINE BARGE VAPOR COMPOSITION
COMPONENT                       VOLUME CONCENTRATION  (%)

C-L + C2                                .38

C3                                    2.60


C4                                    8-65

C  and Heavier                        6.37

Air                                     82


Total HC                                18
                                -48-

-------
7.
. 50.
168
123.
99.
7 ppm
4 ppm
ppm
9 ppm
96%
26.
182
605.
445.
99.
6 ppm
ppm
5 ppm
9 ppm
87%
.01%
.05%
.17%
.12%
99.65%
                            TABLE 4.5.2
           BARGE VAPOR COMPOSITIONS AFTER SAFETY MODULE
Safety Technique:  Dilution with air to maintain 25% L.E.L.(= .35%
                   total HC by volume) @ 50% initial concentration
Design Flow Rate = 353,000 GPM (47,000 ACFM)
Vapor Analyses for Design
Initial Total HC           5%             18%           50%
Components by Vol.
Cl + C2
C3
C4
C5 + Heavier
Air
Additional Volumes of
  Air Added                141.86        141.86       141.86

Safety Technique:  Fuel Gas is used to blanket the system and
                   maintain 50% fuel gas in the vapors
Design Flow Rate = 9900 GPM (1325 ACFM)
Initial Total HC           5%             18%           50%
Components (% by Vol.)
C-L + C2                    .06            .19           .53
C3                         .36           1.30          3.61
C4                        1.20           4.33         12.10
C5 and Heavier             .89           3.19          8.90
Fuel Gas*                  50             50            50
02                        9.98           8.6           5.25
Additional Volumes of
  F.G. Added              1.0            1.0           1.0
*Fuel Gas can be assumed to be 100% CH,
                                      4
                               -49-

-------
.66
4.32
14.4
10.62
30
.66
4.32
14.4
10.62
30
1.06
7.22
24.03
17.70
.. 50*
                       TABLE 4.5.2  (CONT.)
           BARGE VAPOR COMPOSITIONS AFTER SAFETY MODULE
Safety Technique:  Saturate vapors to 30% HC
Design Flow Rate = 5900 GPM (800 ACFM)
Vapor Analysis for Design
Initial Total HC           5%             18%           50%
Components (%)
c1+c2
C3
C4
C- and Heavier
Final Total HC
Volume Increase due to
  Gasoline Saturation      36%            17%

* Fully Saturated
 Safety Technique:  Inert Gas Blanket
 Design Flow Rate = 4950 GPM  (660 ACFM)
 Vapor Analysis  for Design
 Components  (%)
 Cl  + C2
 C3
 C4
 Cp  and Heavier
 Inerts
'5%
.11
.72
2.40
1.77
95
0
18%
.38
2.60
.8.65
6.37
82
0
50%
1.06
7.22
24.03
17.70
50
0
                                -50-

-------
                            TABLE 4.5.3
      Manufacturers Predicted Vapor Control Unit Efficiencies
Control Technology

Carbon Adsorption




Refrigeration
Thermal Incineration
Removal Efficiency

      98.3%

      99  %


      91.4%

      93.9%



      99.9%
      Comments
For use with N_ or
Inert Gas Generator
For use with
Saturator

For use with N_ or
Inert Gas Generator
For use with
Saturator


For use 'with all
Safety Modules
                                -51-

-------
                             TABLE 4.5.4
           CONTROL EFFICIENCIES REPORTED IN EPA LITERATURE
                         Range of    Test Period      EPA Report
Control Technology      Efficiency      Length       Reference No.

Carbon Adsorption       91.0-99.5/6     3 days        EMB-77-GAS-19

Refrigeration           80.4-93.1$     3-4 days      PB-275-060
                                                     EMB-76-GAS-16
                                                     EMB-77-GAS-18

Thermal Incineration    99.8$          1 year        PB-243-363
                              -52-

-------
control unit had  a  three  day average recovery  effi-
ciency of  95.9$, with a range (for  a  single  day)  of 91
to 99.5/6.   A  recovery efficiency of 91$ occurred when
the carbon adsorption system failed to cycle  properly
between the  twin  beds,  and the system operated pri-
marily on  one bed.  Outlet HC concentrations  from the
control unit for  individual runs ranged  from  0% to
12.77$ by  volume  expressed  as  propane.   The  daily
averages  ranged  from 3-5$ to 0.2$.

Although  interpreting these  reported  test results
provides  useful  data, there is no direct applicability
to the present barge loading case because  differences
in vapor  flow rates and HC concentra'tions between truck
and barge  loadings prevent comparison  of numbers.  The
data, however,  could  be typical of  variations  that
occur with actual  operations.  The  important difference
is in HC  concentrations; .the average  concentration for
truck vapors  entering the recovery  unit ranges from 30
to 55$ for trucks with vapor  balance,  while  barge
vapors for the conditions of this study vary from 5$ to
over 50$,  with an  average  value of 18$.   In  general,
for a given carbon adsorption unit,  when inlet concen-
trations  are  reduced, the recovery  efficiency  is   also
reduced.   This occurs because the exit concentration is
relatively fixed as long as breakthrough or  overloading
the carbon bed has not occurred.

The  reported test results  for  three refrigeration
systems gave  recovery efficiencies  of 80.4$, 93.1$, and
84.4$.  (Detailed  descriptions were obtained  for only
one  of the units.)   The unit  that achieved  84.4$
recovery  was  reported to have experienced  refrigerant
                    -53-

-------
leakage  which resulted in  the  condenser operating at a
higher-than-design  temperature.   Efficiencies for
individual  runs ranged from 49  to 91$ on one day and 71
to 95%  on the other day.   The reported inlet  HC
concentrations in the vapor were in  the range  of 10.8
to  30.5$  with  an average  of 15.4$.   These  HC
concentrations compare  favorably with expected  barge
vapor HC concentrations.  Exit vapor concentrations of
1.7 to 4.8$ (as propane)  were observed in the  tests;
the average was 3.2$.

A  reduction in control efficiency  occurs  in
refrigeration systems as the inlet HC concentration is
reduced.  This occurs because  outlet concentration is
set by the temperature and pressure  of operation, and
it does not vary  with inlet concentration.

Reported test results for thermal incineration  systems
have  shown that  very high control efficiencies are
obtainable.  Reductions  of 99.8$ occurred in normal
operation  at  a  truck  loading  terminal,  and HC
concentrations  between 1 and  45  ppm (expressed as
methane)  were  reported  at  the outlet   of the
incinerators.

The claimed efficiencies  of the control  technologies
correspond to the efficiencies demonstrated  in  testing
under normal circumstances (i.e.,  in the  absence  of
malfunctioning  equipment).  A greater effort  is made
under test circumstances  to ensure proper  operation  of
the equipment than  under  normal  day-to-day operation.
Therefore  it  is  reasonable  to  expect that the
efficiency of the control units in normal  use  will  not
be the  same as  obsefved in test situations, or  as given
by the  manufacturers.       -
                    -54-

-------
         However, due to the limited  data base, it is impossible
         to predict reasonably the normal  operating efficien-
         cies.  It can be said, however,  that the results  are
         likely to be lower than  results achieved in testing.

         The  data that have been  obtained  demonstrate that
         thermal  incineration,  at  99.856, has  the highest control
         efficiency of the technologies  evaluated.  The  second
         most efficient technology is carbon  adsorption at 98.3/6
         and 99%  reduction, depending on the  safety system used.

         Emission rates for each  of the  control cases have  been
         calculated using the control unit  reduction efficien-
         cies from TABLE 4.5.3  and an estimated collection  sys-
         tem efficiency of 9856.  These rates  are given  in  TABLE
         4.5.5.   The collection  efficiency is theoretically
         10056.  Most of the vapor  collection  systems in  service
         for  truck loading  depend on vapor displac-eraent by
         differential pressure  to  transport vapors to treatment.
         They have experienced  collection  efficiencies from 3056
         to greater than  90/6.   The single  most  prevalent
         reasons  for the leakage  have been  poor sealing  at  the
         truck to vapor return  line connection and malfunction
         of pressure relief valves.  It  is reasonable to  expect
         a  higher efficiency  in  the barge  vapor  collection
         system  because bolted-flange  piping connections  are
         used.

4.5.2    Secondary Emissions
         The secondary emissions  produced by  the adoption of any
         of the control or  safety  technologies are minor,  and
         they will be discussed by type  and source rather than a
         quantitatively.
                              -55-

-------
                             TABLE 4.5.5
        PREDICTED ANNUAL HC EMISSION RATES FOR CONTROL CASES
                          LB/YR HC EMITTED
           (Percent reduction from Uncontrolled Base Case)


                           Safety Modules

                                   Fuel Gas    N?     Inert Gas
            Dilution  Saturation   Blanket   Blanket  Generator
Control
Modules

Carbon
Adsorption

Refrigeration


Thermal
N/A
n N/A
21,150
(97.9)
50,061
. (95)
13^,028
(86.7)
35,246
(96.5)
N/A
N/A
49,887
(95)
36,954
(96.3)
105,111
(98.6)
21,150
(97.9)
36,954
(96.3)
105,111
(89.6)
21,150
- (97.9)
N/A - Not Applicable
                              -56-

-------
Gaseous secondary  emissions are produced  by the com-
bustion processes  of the inert gas  generator and the
thermal incinerator.  The combustion gases  evolved will
contain C02,  CO, unburned HC, NOX,  and SOX  (if the fuel
contains sulfur).  The inert gas product from the  inert
gas generator is a water-cooled and scrubbed  combustion
gas; therefore,  a  portion of the contaminants are re-
moved.  Tests on other thermal incinerators in gaso-
line-tank-truck  loading  service have reported measured
concentrations of  CO as  high as 35  ppm and NOxas high
as 10 ppm.   Refrigeration and carbon adsorption control
modules are  not  producers  of  gaseous  secondary
emissions.

Liquid secondary emissions  from the safety  and control
modules include  a  number of contaminated water streams.
The inert gas generator  produces a cooling  water stream
which  is used for contact cooling  of the combustion
gases.   The  absorbed C02 ,  NOxand SOX (if  present)
cause the cooling  water  to  become acidic and  corrosive,
and they must be  sent to waste water disposal.  The
refrigeration system condenses water vapor  and gasoline
vapors together  and  separates  them  by decanting.  The
water  is  contaminated  by  hydrocarbons and must  be
properly disposed.   A  final  source of secondary liquid
wastes is the overflow stream  from the water  seals used
to  isol.ate  portions of the vapor  collection  system.
This  overflow is  contaminated  by  the hydrocarbon
vapors.
                    -57-

-------
          Solid  emissions are not a  problem  with the  safety
          modules  or  with the refrigeration  technology.  Also
          particulate emissions  from the thermal incinerators are
          predicted to be minor.  The  carbon adsorption produces
          small carbon fines  as  the carbon particles abrade.  The
          most significant solids disposal  problem occurs when
          the  carbon  in  the  beds must  be replaced,  but  the
          replacement interval is on  the  order of 10  or more
          years and then the  problem can be resolved by returning
          the  carbon t.o the manufacturer  for  disposal  or
          regeneration.
4.6       ECONOMICS
          Capital (total erected) and annualized  (operating and
          depreciation)  costs were calculated  for each of the
          control cases studied.  All costs are given in  terms of
          U.S.  dollars (second quarter  1978).   The basis and
          discussions  of the capital and  annualized costs are
          presented in  the following paragraphs.

          The costs of  the modules were calculated separately and
          then assembled into specific control cases.  By adding
          together the  capital and annualized costs of the safety
          module and control module, the capital or annualized
          cost for the  specific  control case  was  obtained.   The
          cost  of .using a particular control module will  vary
          with the type of safety module  used, and is dependent
          on  the  degree of compatibility between  the  safety
          module  and the control module.   Therefore,  the  cost
          effect of the safety module on the  control module is
          apparent by direct comparison.
                              -58-

-------
4.6.1     Capital Costs
         The capital cost  estimate for any particular  safety or
         control module  represents the  total  cash outlay
         required for the  design, procurement,  and  installation
         of the equipment  and  material within  the  module.   The
         costs  for  major  equipment  items  were  obtained  from
         vendors as engineering  estimates or budget prices.  The
         major  equipment  items include the  specialty  vapor
         control units,  the inert  gas generator, and  the
         blowers.  Bulk commodity items  such as  piping,  steel-
         work,  foundations,  electrical  supply,  and paint were
         estimated  and  priced by  Pullman  Kellogg's  cost
         estimating department.  An  allocation for minor spare
         parts  was made as a  percentage of the  total material
         cost;  however, no spare or  backup control units were
         included.  It was assumed  that  no  additions in utility
         plant  capacities for cooling water, steam, electricity,
         sewers, or fuel  gas were  required  (utility cost rates
         include depreciation).  Only the costs  for short  (200
          feet)  local utility distribution connections were in-
          cluded.  Sales and use  taxes were  estimated as a  per-
          centage of the total material cost.   Home office costs,
          insurance, commissioning,  contingency,  and contractor
          overhead and  profit  were  estimated as  percentages  of
          the  material  and labor costs.'

          The  safety module includes all  items  from the  loading
          dock  to the  control module connection.   The total
          capital  costs for each of the safety  modules are  shown
          in TABLE 4.6.1.   The  itemized cost  listings  for the
          safety modules  are in  Appendix B.  The  costs  for
          fuel-gas  and  nitrogen-blanketing modules  are grouped at
          $200,000,   a  n d _ t h e~  remaining  modules  are
                              -59-

-------
              TABLE 4.6.1
    CAPITAL COST OF SAFETY. MODULES
MODULE                           COST

Dilution                       $330,800

Saturation                      307,200

Fuel Gas Blanketing             209,100

N  Blanketing                   198,300

Inert Gas Generator             364,100
               -60-

-------
grouped between $300,000 and $365,000.   It should  be
noted that the safety module utilizing N? for blanket-
ing does not include  the  capital cost for the liquid  N2
storage tank and vaporizer.  This  equipment is leased
and thus  becomes an operating  expense  rather than a
capital expenditure.  If these  items  were purchased
rather  than leased, N2  blanketing would  not  be the
least expensive module, and it would  most likely cost
about $320,000.

The control module  cost includes the vapor control unit
and the associated  utility connections.   Total capital
costs  for each of the control  modules  are  shown  in
TABLE 4.6.2.                       '        ,     .

The costs  for thermal  incineration as a  group are much
less expensive than the competing  technologies.  The
capital cost of thermal -incineration control schemes
range  from $148,800 to $297,600, depending  on the
safety module applied.  Control module  costs for use
with saturation and N2 or inert gas  generator modules
are lowest since flow  rates are not increased substan-
tially, as compared to dilution or fuel gas blanketing
which raise flow rates substantially.  Capital cost for
a thermal  incineration module  to be used with the
dilution safety module was not obtained;  however, the
excessive  operating cost  makes it  prohibitive  due  to
high fuel  requirements.

Carbon adsorption is the highest capital  cost techno-
logy.   Carbon adsorption with saturation  costs $661,800
and carbon adsorption with inertion cost  $458,300.  The
difference in cost  is caused because larger  carbon beds

                       -61- -

-------
                            TABLE 4.6.2
                 CAPITAL COST FOR CONTROL MODULES
Carbon Adsorption
         - For Use W/ Saturation               $661,800
         - For Use W/N2 or Inert Gas Generator $458,300

Refrigeration
         - For Use W/Saturation    -            $614,800
         - For Use W/N_ or Inert Gas Generator $614,800

Thermal Incineration
         - For Use W/Dilution                     N/A
         - For use W/Saturation                $171,100
         - For use W/Fuel Gas Blanketing       $297,600
         - For Use W/N2                        $148,800
         - For Use W/Inert Gas Generator       $148,800
                                -62-

-------
are required  for saturation  than for inertion due  to
the gasoline  added.

Refrigeration  modules are identically priced  at
$614,800  for  use with saturation  or inertion since the
refrigeration duty is based on  a  saturated gasoline-air
mixture.

The cost  of each case studied is  shown in TABLE 4.6.3.
As the  table  indicates,  thermal  incineration systems
establish the lowest capital  expenditure, from a  low  of
$347,100  to a high of $506,900  and an average cost  of
$459,800.  The  carbon  adsorption systems range  from
$656,600 to $969,000 with an  average  figure  of
$814,000.   The  refrigeration  systems range  from
$813,000  to $972,900 and average  $902,600.  For any  of
the three control technologies, the lowest capital  cost
is  incurred when nitrogen  inert blanketing  is
employed.

The capital cost for  barge modifications reported  by
others  ranged from $50,000 to $150,000 (1976 dollars)
per barge.  (Refer to EPA-450/3-76-038a, page 159).  A
detailed  description  of  the  modifications covered  by
these reported costs could not  be determined however.
The total erected costs for modifying the barges  to
conform .with  the modifications  shown  in Figures 4.3.2
and 4.3.3 (with  the exception  of the direct sighting
port) are estimated to  be $152,000 per barge.  These
costs  do not include the  costs "due  to cleaning,
degassing, or lost revenue during the period the barge
is removed from  service. These costs are all incurred
by the  barge  owner.
                   -63-

-------
                                          TABLE 14.6.3
                   CAPITAL COSTS OF BARGE LOADING EMISSION CONTROL SYSTEMS ($)
Emission
Control Module Type

Carbon Adsorption

Refrigeration

Thermal Incineration
                                             Safety Module Type
Dilution
N/A
N/A
-
Saturation
969,000
922,000
478,300
Fuel Gas
Blanketing
N/A
N/A
506,700
N
Blanketing
656,600
'813,100
347,100
Inert Gas
Generator
816,400
972,900
506,900

-------
M.6.2     Annualized Costs
          Components included in annualized costs are utilities,
          maintenance, labor,  capital recovery charges, and
          gasoline recovery credits  (if any).  Utility costs
          include  electricity,  fuel  gas,  N2  gas, steam,  cooling
          water, and gasoline.

          The costs for the utilities  are:
             Electricity               $.OU/KWH
             Natural Gas               $2.73/1000 ft3
             N2 Gas                    $2.90/1000 ft3
                                        (delivered as liquid)
             N2 Storage and
              Vaporizer                $1350'/month
             Steam                     $3-50/1000 Ib
             Cooling Water             $.15/1000 gal
             Gasoline                  $.40/gal

          The unit costs for these utilities  are representative
          of  prices  obtainable  on  the  Texas Gulf  Coast.
          Maintenance costs for  equipment have been estimated as
          a percentage of the capital  cost.

          The percentages used  for maintenance costs  of the
          safety modules are:
             Dilution                  2%
             Fuel Gas' Blanketing       1%
             Inert' Gas Generator       3%
             N2 Blanketing             1%
             Saturation                5$

          The percentages used  for maintenance costs  of the
          control modules  are:
                              -65-

-------
    Carbon Adsorption         3%
    Refrigeration             2%
    Thermal Incineration      1%

Additional labor (over the  uncontrolled case)  has been
assumed to be one extra operator  during the loading
periods for both safety control modules.  The labor
rates are  approximately union  scale  for the  area and
include  fr i.n ge  benefits.   A  labor  rate  of
$12/operator-hour is used.   Capital charges are based
on a 10/6 annual interest  rate  and 15 year equipment
life  with a capital recovery factor of 0.13147.
Administrative expenses are  calculated as ,4$ of the
capital cost.  A credit for gasoline recovery  is taken
at $.40/gal  for  those cases where a reduction in
gasoline losses is achieved.

TABLE 4.6.4 contains the annualized costs for each of
the safety modules.   The annualized cost for the inert
gas generator is lowest at  $95,600, with dilution only
slightly  higher at  $99,200.   The  remaining  safety
modules rank  upward from  fuel  gas blanketing to
gasoline saturation,  to N 2 blanketing.  These three
modules range from $142,500 to $162,200.

Annualized costs for the control modules are listed in
TABLE 4.6.5.  The cost for  each  control module depends
on the  safety module  used with it.   For the control
modules that  recover  gasoline  (carbon adsorption and
refrigeration),  the  annualized  costs  are  reduced
dramatically  when the saturation safety module is used.
This is due to  the  credit  for  the  increased gasoline
                     -66-

-------
                            TABLE 4.6.4
                ANNUALIZED  COSTS  FOR SAFETY MODULES ($)

                                                       Inert Gas
             Dilution  Saturate  Fuel Gas      N-      Generator
Gasoline        -      76,800
Electricity  11,936       328       597           298        578
Fuel Gas        -        -       91,973        -           8354
N2              -                            113,903
Steam           -        2919
Cooling Water   -        -          -          -           2520
Total
  Utilities  11,936    80,047    92,570      11-4,199    11,452

Capital and Ad-
  ministrative
  Costs      56,670    52,675    38,854      34,002     61,403
Maintenance   6,610    15,360     2,.091       1,983     IX),743
Labor        24,000    12,000    12,000      12,000     12,000
Total
  Annualized
  Cost       99,216   160,082   142,515     162,184     95,598
                                -67-

-------
                            TABLE 4.6.5
                ANNUALIZED COSTS FOR CONTROL MODULES
                                 ($)
Control Module
Safety Module
      Carbon Adsorption
 Saturation      N_ or
                 Inert Gas
                 Generator
                     Refrigeration
                Saturation
               N2 or
              Inert Gas
              Generator
Utilities           3,640
Capital and
  Administrative
  Costs           113,479
Maintenance and
  Labor            19,854
(Gasoline Credit) (130,447)
Total Costs
   6,526
                 2,432
   78,584

   13,749
  (76,186)
   18,579
                  5,960
  105,420

  ,12,296
 (121,252)
    2,424
                5,960
105,420

 12,296
(70,814)
 52,862
Control Module
Safety Module
Utilities
Capital and
  Administrative
  Costs
Maintenance and
  Labor            N/A
 (Gasoline Credit)   (0)'
Total Costs        N/A
                                Thermal Incineration
Dilution
453,464
  N/A
Saturation
      473
    29,342

     1,711
     (0)
Fuel Gas
Blanketing
      946
   51,030

    2,976
      (0)
N_ or Inert
Generator
     473
  25,515

   1,488
     (0)
                31,526
                 54,952
                27,476
N/A = Not Applicable
                                -68-

-------
recovered.  Thus  the reduction in control module
annualized cost when using  saturation  rather than
blanketing with carbon adsorption  is  65%,  and with
refrigeration it  is 95.4$.   Ranking  the  average
annualized costs for each control technology  starting
with the  lowest indicates carbon adsorption  is  least
costly, followed by  refrigeration, and finally  by
thermal incineration.  The annualized cost  calculation
for  thermal incineration  used  with dilution was
abandoned due to a utility  cost of $453,000, which
makes it  non-competitive.  The utility cost is  for  the
fuel gas  that must  be  used to burn the dilute  gasoline
vapor mixture.

The annualized costs for the complete systems  which
include the safety and control modules  are  shown  in
TABLE  4.6.6.  The  costs range  from  $115,000  to
$215,000.  The lowest  cost for each  control technology
is achieved when an  inert gas generator  is  used  as  the
safety module.  As a group,  the carbon adsorption
systems have the lowest average annualized  cost  at
$153,800.  The average costs for  refrigeration  and
thermal incineration systems are  virtually identical at
$175,300 and $175,500 respectively.  In each case
(except  for the dilution-thermal incineration)
annualized cost was greater  due to  the safety  module
than due-to the control module.

If the barges are owned by the terminal operator,  the
capital charges are  amortized along  with  the  dockside
equipment.  However,  if the barges are  leased  from  an
independent barge operator  (which  appears to  be  the
more common procedure) tjb_e costs are  reflected  in  the
                   -69-

-------
                                         TABLE 4.6.6

          ANNUALIZED COSTS FOR ONE GASOLINE BARGE LOADING EMISSION CONTROL SYSTEM
    Emission Control
      Module Type
    Carbon Adsorption

    Refrigeration

    Thermal Incinera-
     tion
           Total Annualized Cost for Each Combination
             of Safety and Control Modules ($1,000)
Dilution
I

  N/A

  N/A

 >550
Saturation

    165

    160

    190
 Fuel Gas
Blanketing

    N/A

    N/A

    195
 Nitrogen
Blanketing

   180

   215

   190
Inert Gas
Generator

   115

   150

   125
o
I

-------
         lease price,  since  the  barge operator must recover the
         cost of modification.   How these costs are reflected  in
         leasing prices depends  on the terms negotiated between
         the barge operator  and  the oil company.

         From 64.4$ to 98.5$ of  the total  annualized cost is a
         function of safety which has the  effect of obscuring
         the relative costs  of  the control modules themselves.

4.6.3     Economic Analysis
         This analysis will  center on the determination of cost
         effectiveness for each of  the  control cases.   Cost
         effectiveness must relate  some  parameter of cost  to
         some  parameter of performance.   The annualized cost
         will be used as the cost parameter.   The performance
         parameter will be  the  net  reduction in HC emissions
         over the uncontrolled  or base case.  It is assumed  that
         the annualized cost of the uncontrolled case is zero.
         The cost  effectiveness will be  measured in terms  of
         pounds of HC controlled annually per dollars  spent
         annually.  Thus the  cost  effectiveness index  is
         expressed as # HC/ $  and the  higher the  index,  the
         greater the cost effectiveness.  The cost effectiveness
         of each control case is shown in TABLE 4.6.7.  The  cost
         effectiveness ranges from 4.2  #  HC/$ to 8.50 0 HC/$.
         The most  cost effective system  is carbon adsorption
         with  ine^t gas generation at 8.50  # HC/$.  The next two
         cases  are thermal  incineration  and  refrigeration,  both
         using  inert gas generation,  at  8.01 # HC/$ and 6.08 #
         HC/$  respectively.

         As an  alternate approach to  evaluation of  the control
  ..„     systems,  consider  the  additional cost of  the product
                             -71-

-------
                             TABLE 4.6.7
                COST EFFECTIVENESS FOR CONTROL CASES
            LB. HC CONTROLLED/$ OF ANNUALIZED COST ($/LB)
                         Safety Module Type

                                   Fuel Gas          Inert Gas
            Dilution  Saturation  Blanketing   N	Generator
Emissions
Control Module
     Type     	   #HC($/LB)     	  #HC($/LB)  #HC($/LB)

Carbon         N/A    5.75(0.17)     N/A   5.37(0.19)  8.50(0.12)
Adsorption

Refrigeration  N/A    5.37(0.19)     N/A   4.20(0.24)  6.08(0.16)

Thermal        N/A    5.08(0.20)     4.85  5.20(0.19)  8.01(0.12)
Incineration
N/A - Not Applicable
                              -72-

-------
shipped by barge when using a vapor  control  system.
For the proposed 6 x 10 ° bbl annual  gasoline through-
put,  the effect of the  least cost  effective  control
scheme  (refrigeration  with N2 blanketing at  4.2 # HC/$)
on the  cost of gasoline  is approximately one dollar per
1230 gallons or  0.08 
-------
cases.   The  annual  throughputs used varied  from three
to 24 million  barrels per year.   The  projected  results
are shown in TABLE  4.6.8.  Figures 4.6.1, 4.6.2, 4.6,3,
and 4.6.4 are curves of cost  effectiveness  versus
annual  throughput for the various control cases.   The
information  in TABLE 4.6.8 is plotted  in -these figures.
When the cost  effectiveness index for  recovery  schemes
reaches  12.8  # HC/$,  a break even point is  reached,
i.e.  the value of  the gasoline  recovered is  equal to
the annualized-cost required to  recover it.  Above this
point the vapor  recovery system begins  to pay  back
direct  operating  costs  and  capital charges.   If  only
direct  operating  expenses are considered and  capital
charges  neglected,  the break  even point'would  be
lower.
                     -74-

-------
                                     TABLE 4.6.8

             SUMMARY OF RESULTS OF COST EFFECTIVENESS  CURVES  (HC/$)
Throughput
(10b BBL/YR)

Carbon Adsorption'
 w/Saturation
 w/N2
 w/Inert Gas Gen.

Refrigeration
 ^/Saturation
 w/N2
 w/Inert Gas Gen.
  1
Thermal Incin.
 w/Saturation
 w/Fuel Gas Blanket
 w/N2
 w/Inert Gas Gen.
12
18
24
2.89
3.31
3.82
5.75
5.37
8.50
8.61
6.78
14.38
11.47
7.80
21.98
17.16
9.19
46.56
22.81
10.08
105.70
2.73
2.55
2.86
5.37
4.20
6.08
7.96
5.36
9.72
10.47
6.21
13.88
15.30
7.40
24.24
19.89
8.17
38.68
3.56
3.37
3.96
4.70
5.08
4.85
5.20
8.01
5.92
5.69
5.81
10.48
6.46
6.22
6.17
12.40
7.10
6.87
6.58
15.15
7.47
7.24
6.80
17.04

-------
u
ac
=»=
 O
 U
   70
   60
   50
   40
H  30
en
en
0)
c
Q)

•rH
-P

g  20
U-l

w
   10
CARBON ADSORPTION with
  A - Saturation Module
  B - N_ Blanketing
  C - Inert Gas Generator
                                    12
                         15
18
21
24
                    Annual Throughput 10  BBL/YR
      Figure 4.6.1  Cost  Effectiveness Projections Carbon Adsorption


                                 -76-

-------
  70
  60

  50
X
w
Q

M

en


1 4°
w
>
M
EH
U
W

h 30
O
U
  20
  10
REFRIGERATION W/

A - SATURATION

B - N2 BLANKETING

C - INERT GAS GENERATOR
                                 12
                      15
18
21
24
    ANNUAL THROUGHPUT 10  BB&/YR
             Figure 4.6.2
       Cost Effectiveness Projections
       Refrigeration
                               -77-

-------
  70
  60
u 50
X
=«=


X
w
Q
2
cn 40
w
w
S
W
EH
U
W
O
U
  30
  20
  10
THERMAL INCINERATION WITH

A - SATURATION


B - FUEL GAS BLANKETING


C - N2 BLANKETING


D - INERT GAS GENERATOR
                                        VIRTUALLY

                                        INDISTINGUISHABLE
                                         15
                                    18
    ANNUAL THROUGHPUT  106 BBL/YR
                                                       21
                                                   24
          Figure 4.6.3  - Cost  Effectiveness Projections

                         Thermal  Incineration

                      '.        -78-

-------
u
ffi
\
=*:

X
w
Q
2
M
CO
W
U
W
Cfl
o
u
   70
            A - CARBON ADSORPTION W/INERT GAS GENERATOR

            B - REFRIGERATION  W/INERT GAS GENERATOR

            C - THERMAL  INCINERATION W/INERT GAS GENERATOR,
   60 -
    50
40
    30
20
    10
                                    12
                                        15
18
21
24
       ANNUAL THROUGHPUT 10  BBL/yr*
            Figure 4.6.4  Cost Effectiveness Projections
                          Comparison

                                 -79-

-------
                          REFERENCES
BOOKS
1.   Hughes,  John R., "Storage and Handling of Petroleum Liquids:
     Practice and  Law," Charles Griffin  and Company Limited,
     London,  Great  Britian, W.C.-2,  196?

2.   Oil Companies  International Marine Forum,- "International Oil
     Tanker and  Terminal Safety Guide," Second  Edition,  Halsted
     Press, John Wiley and Sons Inc., New York,  1974

3.   Wooler,  R.  G., "Tankerman1s Handbook,"  Second  Edition,
     Edward W.   Sweetman, Publisher, New York 1950

REPORTS

1.   "API Bulletin  2514:  Evaporation Loss  from Tank Cars, Tank
     Trucks,  and Marine  Vessels," American  Petroleum Institute,
     Washington, D.C., 20037, 1959

2.   "API Bulletin 25-14-A:   Hydrocarbon  Emissions  from Marine
     Vessels Loading  of  Gasolines," American Petroleum Institute,
     Washington, D.C., 20037, 1976
                              -80-

-------
3.   Amoco  Oil  Company; "Demonstration of Reduced  Hydrocarbon
     Emissions  from Gasoline Loading Terminals," EPA Report No.
     EPA-650/2-75-OH2,  EPA, Research Triangle  Park,  N.C.   27711,
     1975  '

M.   Betz  Environmental Engineers, Inc., "Gasoline Vapor Recovery
     Efficiency Testing at Bulk Transfer Terminals  Performed at
     Pasco-Denver Products Terminal," EPA Report No.  76-GAS-17,
     EPA,  Research Triangle Park, N.C. 27711,  1975

5.   Betz  Environmental Engineers, Inc., "Air  Pollution  Emission
     Test  - Diamond  Shamrock  Gasoline Terminal; Edwards Vapor
     Control System; Denver, Colorado," EPA Report No. 76-GAS-16,
     EPA,  Research Triangle Park, N.C. 27711

6.   Betz  Environmental Engineers, Inc., "Texaco  Westville Sales
     Terminal;  Westville, New Jersey," EPA Report No.  77-GAS-18,
     EPA,  Research Triangle Park, N-.C. 27711

7.   Environmental Protection  Agency, "Control  of  Hydrocarbons
     from  Tank  Truck Gasoline Loading Terminals," EPA Report No.
     EPA-450/2-77-026,  EPA, Research Triangle  Park,  N.C.   27711

8.   EPA,  "Compilation  of Air Pollutant Emission Factors, (1977
     Supplement 7)," EPA, Research Triangle Park, N.C. 27711

9.   Pullman Kellogg, a Division of Pullman Inc., "Evaluation of
     Control Technology  for  Benzene  Transfer Operations,"
     prepared under EPA Contract No. 68-02-2619,  Work Assignment
     No.  2; EPA,  Research Triangle Park, N.C.  27711

10.  Radian Corp., "A Study of Vapor Control Methods for Gasoline
     Marketing  Operations Volume II - Appendix," EPA Report No.
     EPA-M50/3-75-OU6-b, EPA, -Research Triangle Park,  N.C.  27711
                              -81-

-------
11.  Radian Corp., "Background Information  on Hydrocarbon Emis-
     sions from  Marine Loading Terminal  Operations, Volumes I &
     II,"  EPA Report  No.  EPA-450/3-76-038  a,b;  EPA,  Research
     Triangle Park, N.C.  27711

12.  Scott Environmental Technology,  Inc.,  "Control Characteris-
     tics  of  Carbon  Beds  for Gasoline Vapor Emissions,"  EPA
     Report No.  EPA-600/2-77-057, EPA  Research Triangle Park,
     N.C.  27711

13.  Scott Environmental Technology,  Inc.,  "Air Pollution Emis-
     sion Test - Phillips Fuel Company;  Hackensack, New Jersey,"
     EPA Report  No. 77-GAS-19; EPA, Research  Triangle Park, N.C.
     27711
Trade Journals

1.    Southerland,  M.E.,  and H.W.  Wegert.  "Flash  Arresters
     Successfully Forestall An Acetylene Catastrophe". The Oil
     and Gas Journal, March 13, 1972,  pp.  73-75.
                              -82-

-------
                       APPENDIX A

                         LEGEND
EQUIPMENT AND PIPING SYMBOLS
         • INSTRUMENT CONTROL
        "* LINE
        -$ PROCESS FLOW LINE
 5—da—J
CONTROL VALVE
 5	tX	$  BLOCK VALVE

          MOTOR OPERATED
          VALVE


 5—f\j	$  CHECK VALVE
        -5 FLAME ARRESTOR
                          1	5 ORIFICE  PLATE
                                         FLEXIBLE HOSE
                                   SJ	5  LIQUID SEAL POT
                       /• -\ "
                               CENTRIFUGAL PUMP
                                         BLOWER
          PRESSURE^VACUUM
          VENT VALVE
Instrument Abbreviations

FIC - Flow Indicating Controller
FS  - Flow Switch
I   - Interlock  "
LI  - Level Indicator
LS  - Level Switch
PI  - Pressure Indicator
PIC - Pressure Indicating  Controller
TIC - Temperature Indicating Controller
XA  - Warning Alarm  (Audible)
XL  - Warning Light
                          -83-

-------
          APPENDIX B
       CAPITAL COST DATA
FOR SAFETY AND CONTROL MODULES
             -84-

-------
                      CAPITAL COST DATA
                      MODULE -  DILUTION
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight  (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion
Contractors OH & .P
Contingency

Total Erected Costs
COST(MATERIAL AND LABOR)
       (.$ xlOOO)
        71.0
         8.0
         4.0
        99.1
         8.0
        10.6
         3.0
         1.1
        20.9
        20.9
         4.3
         3.1
         1.1
         2.6
        52.2
        20.9

       330,8
                            -85-

-------
                      CAPITAL COST DATA
                      MODULE - SATURATION
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight  (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion
Contractors OH &. P
Contingency
COST(MATERIAL AND LABOR)
      ($ xlOOO)
         5.2
        12.4
        26.7

         4.0
        22.5
        10.0
        54.6
         7.0
        45.5
         3.0
         2.0
        19.2
        19.2
         3.5
         2.9
         0.9
         2.4
        48.0
        19.2
Total Erected Costs
       307.2
                             -86-

-------
                      CAPITAL COST DATA
               MODULE  -  FUEL  GAS  BLANKETING
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion
Contractors OH & P
Contingency
COST(MATERIAL AND LABOR)
      (.$ xlOOO)
        19.7

         U.O
         1.6
         6.0
        48.7
         4.0
        41.4
         3.0
         1.0
        13.1
        13.1
         2.2
         2.0
         0.5
        ' 1.6
        32.7
        13.1
Total Erected Costs
       209.1
                            -87-

-------
                      CAPITAL COST DATA
                    MODULE - N  BLANKETING
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight  (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion
Contractors OH &,P
Contingency

Total Erected Costs
COST(MATERIAL AND LABOR)
      (.$ xlOOO)
        19.7

         4.0
         4.0
         6.0
        51.4
         4.0
        30.8
         3.0
         1.0
        12.4
        12.4
         2.3
         1.8
         0.6
         1.5
        31 .0
        12.4

       198.3
                            -88-

-------
                      CAPITAL COST DATA
                MODULE - INERT GAS GENERATOR
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight  (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion
Contractors OH & P
Contingency

Total Erected Costs
COST(MATERIAL AND LABOR)
       (.$ xl.000)
        19.7
        91.7
         4.0
         8.0
         6.0
        57.4
         U.O
        30.8
         3.0
         1.5
        22.6
        22.6
         6.0
         3.4
         1.5
         2.8
        56.5
        22.6

       364.1
                            -89-

-------
                      CAPITAL COST DATA
            MODULE - CARBON ADSORPTION W/INERTION
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight  (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion
Contractors OH &._P
Contingency
COST(MATERIAL AND LABOR)
      ($ xlOOO)
        231.3

         12.0
          4.0

         30 .0

          3 .0
          2.5
         28.3
         28.3
          9.8
          4.2
          2.4
          3.5
         70.7
         28.3
Total Erected Costs
        458.3
                             -90-

-------
                      CAPITAL COST DATA
          MODULE - CARBON ADSORPTION W/SATURATION
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion
Contractors OH &_P
Contingency

Total Erected Costs
COST(MATERIAL AND LABOR)
       (.$ oclOOO)
        338.4

         12.0
          4 .0
         11.5
         36-0

          3.0
          3.5
         40.8
         40.8
         14.2
          6.1
          3.5
          5.1
        102.1
         40.8

        661.8
                            -91-

-------
                      CAPITAL COST DATA
       MODULE - REFRIGERATION W/INERTION OR SATURATION
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight  (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion
Contractors OH & P
Contingency
COST(MATERIAL AND LABOR)
      (.$ xlOOO)
         303.0

          12.0
           4.0
          12.3
          42.0

           3.0
           3.2
          38.0
          38.0
          12.8
           5.7
           3.2
           4.7
          94.9
          38.0
Total Erected Costs
         614.8
                            -92-

-------
                      CAPITAL COST DATA
         MODULE - THERMAL INCINERATION W/SATURATION
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight  (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion
Contractors OH &.P
Contingency
COST(MATERIAL AND LABOR)
      (.$ xlOOO)
        75.4

         9.2
         2.3
         7.1
         6.9

         3.5
         1.1
        10.6
        10.7
         3.5
         1.3
         1.1
         1.2
        26.6
        10.6
Total Erected Costs
       171.1
                            -93-

-------
                      CAPITAL COST DATA
    MODULE - THERMAL  INCINERATION W/FUEL GAS BLANKETING
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion .
Contractors OH &- P
Contingency
COST(MATERIAL AND LABOR)
      .-(.$ xlOOO)
         131.2

          16.0
           4.0
          12.4
          12.0
           6.0
           2.0
           2.0
          18.4
          18.4
           5.8
           2.8
           2.0
           2.2
          46.0
          18.4
Total Erected Costs
         297.6
                            -94-

-------
                      CAPITAL COST DATA
         MODULE  -  THERMAL INCINERATION W/INERTION
DESCRIPTION
Exchangers
Towers
Pumps & Compressors
Special Equipment
Fire & Safety Equipment
Concrete Work
Steel Work
Piping
Electrical
Instruments
Insulation & Paint
Freight (Unallocated)
Field Construction Costs
Home Office Costs
Sales & Use Taxes
Insurance
Spare Parts
Project Completion
Contractors OH fr-P
Contingency
COST(MATERIAL AND LABOR)
      (.$ xlOOO)
         65.5

          8.0
          2.0
          6.2
          6.0

          3.0
          1.0
          9.2
          9.2
          2.9
          1.4
          1.0
          1.1
         23.0
          9.2
Total Erected Costs
        148.8
                            -95- '

-------
                          APPENDIX C
                                                  Item
Air Products & Chemicals, Inc.
260 North Belt East, Suite 200
Houston, Texas 77060

Buffalo Forge Co.
Fred P. Heinzmann Co., Inc.
1425 Blalock  Rd., Suite  220
Houston,  Texas 77055

Edwards Engineering Corp.
Hendricksen Co.,  Inc.
 p. 0.  Box 55565
 Houston,  Texas 77055

 Hydrotech Engineering, Inc.
 p. 0.  Box 45042
 Tulsa, Oklahoma 74145

 Liquid Air Inc.  -
 p.  0.  Box  15313
 Houston,  Texas  77020

  National Air Oil Burner Co., Inc.
  3717  Montrose Blvd.,  Suite 431
  Houston, Texas 77006
Nitrogen Blanketing
  Systems
Blowers
 Refrigeration
 Vapor Recovery Units
 Carbon Adsorption
 Vapor Recovery Units
  Nitrogen  Blanketing
    Systems
  Thermal Incineration
                               -96-

-------
                                TECHNICAL REPORT DATA
                         (Please read Inunctions on the reverse before completing)
1. REPORT NO.
 EPA-600/2-79-069
                                3. RECIPIENT'S ACCESSION/ NO.
4. TITLE AND SUBTITLE
 Control Technology Evaluation for Gasoline Loading
 of Barges
                                5. REPORT DATE
                                 March 1979
                                6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

 D. Gee and W. M. Talbert
                                8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Pullman Kellogg
 16200 Park Row, Industrial  Park Ten
 Houston, Texas  77084
                                10. PROGRAM ELEMENT NO.
                                 1AB604B
                                11. CONTRACT/GRANT NO.
                                 68-02-2619, Task 9
12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC  27711
                                13. TYPE OF REPORT AND PERIOD COVERED
                                Task Final: Thru 11/78	
                                14. SPONSORING AGENCY CODE
                                  EPA/600/13
is. SUPPLEMENTARY NOTES
                              project officer is Irvui A.  Jef coat , MD-62, 919/541-
 2547. For details contact B.A. Tichenor at same phone; Jefcoat is no longer with
16. ABSTRACT
          The report gives results of a study to determine the feasibility, safety,
 and cost of methods to control the emission- of hydrocarbon vapor during the loading
 of gasoline barges. Approximately 4 Ib of hydrocarbons are emitted per 1000 gal.  of
 gasoline loaded; annually about 1 million Ib of hydrocarbons  may be emitted at a ter-
 minal pumping 6 million barrels of gasoline.  Vapor control  techniques evaluated
 included carbon adsorption, refrigeration, and thermal incineration. Hydraulic flash
 arresters prevent flame transmission between the barge and vapor control system.
 Safety methods  are also required to render the barge vapors non-flammable prior
 to collection and transport. Safety methods evaluated were dilution, saturation, and
 blanketing with  fuel gas, liquid nitrogen, and inert gas. All  combinations of vapor
 control/safety systems were evaluated. In terms of cost, dilution was determined
 non-applicable for all three vapor control methods; fuel gas blanketing was  appli-
 cable only with  thermal incineration. Costs for the applicable combinations range
 from #0. 12/Ib of hydrocarbon vapor (for inert gas blanketing combined with carbon
 adsorption) to S'O. 24/lb (for liquid nitrogen blanketing/refrigeration). These costs
 correspond to gasoline cost increases of 0.048j^ to 0.096^/gal. ,respectively. More
 accurate  estimates await demonstration of one or more vapor control/safety systems
17.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                                          b.IDENTIFIERS/OPEN ENDED TERMS
                                            c.  COSATl Field/Group
 Pollution
 Gasoline
 Fueling Systems
 Barges
 Hydrocarbons
 Vapors
Carbon
Refrigeration
Incinerators
Nitrogen
Blanketing  -
Pollution Control
Stationary Sources
Carbon Adsorption
13 B
21D

13A
07C
07D
07B
                                 07A,13H
18. DISTRIBUTION STATEMENT
 Unlimited
                                          19. SECURITY CLASS (This Report)
                                          Unclassified
                                             21. NO. OF PAGES
                                                   103
                    20. SECURITY CLASS (Thispage)
                    Unclassified
                                                                   22. PRICE
EPA Form 2220-1 (9-73)
                                        -97-

-------