903R83004
                             U.S. EPA Region III
                             Regional Center for Environmental
                              Information
                             1650 Arch Street (3PM52)
                             Philadelphia, PA 19103
         VAPOR  CONTROLS  FOR
      BARGE  LOADING  OF GASOLINE
                 by

      PEDCo Environmental,  Inc.
         1006 N.  Bowen Road
       Arlington,  Texas  76012
       Contract No.  68-02-3512
          Task Order No.  43
           Project Officer
             Eileen Glen
U.S.  ENVIRONMENTAL PROTECTION AGENCY
             REGION III
  PHILADELPHIA,  PENNSYLVANIA  19106
            December 1983                 '    > Fn-ircnmental

-------
                                   CONTENTS
Figures                                                                  iii
Tables                                                                    iv
Summary                                                                    v

1.    Introduction                                                          1

     1.1  Sources and Amounts of Emissions                                 1

2.    Emission Control Techniques                                           4

     2.1  Refrigeration                                                    4
     2.2  Absorption                                                       6
     2.3  Incineration                                                     8
     2.4  Carbon Adsorption                                               10
     2.5  Control System Reliability                                      10

3.    Cost Analysis                                                        13

     3.1  Model Loading Rates                                             13
     3.2  Assumptions                                                     13
     3.3  Capital Investment                                              14
     3.4  Annual Operating Costs                                          14
     3.5  Cost-Effectiveness                                              18
     3.6  Economic Impact of Controls                                     18

4.    Regulatory Analysis                                                  20

     4.1  Draft Regulation                                                20
     4.2  Regulation in Other Areas                                       22
     4.3  Fire and Explosion Hazards                                      23
     4.4  Compliance and Monitoring Techniques                            24

References                                                                25

Appendices
     A    State of California Air Resources Board Resolution 78-59       A-l
     B    Santa Barbara County Air Pollution Control  District Rule 327   B-l
     C    Compliance Test Method and Monitoring Techniques - Tank
           Truck Loading Terminals EPA Report No.  EPA-450/2-77-026       C-l

-------
                                    FIGURES






Number                                                                   Page



   1      Refrigeration Vapor Recovery Unit                                  5



   2      Absorption Vapor Recovery Units                                    7



   3      Incineration Vapor Control  Unit                                    9



   4      Carbon Adsorption Unit                                           11



   5      Barge-Mounted Refrigeration Vapor  Recovery Unit                  12

-------
                                    TABLES


Number                                                                   Page

   1      VOC Emission Sources and Factors for Gasoline Barges              2

   2      Capital  Investment Costs for Gasoline Barge Loading Vapor
           Control  Systems                                                 15

   3      Annual  Costs and Cost-Effectiveness for Vapor Control
           Systems                                                          16

   4      Economic  Effect of Gasoline Barge Vapor Control  Systems
           on Gasoline Prices                                              19
                                       i v

-------
                                    SUMMARY

     Barges are commonly used for the bulk shipment of gasoline from coastal
refineries to metropolitan storage terminals.   The loading of barges at refin-
eries is a source of uncontrolled volatile organic compound (VOC)  emissions.
Vapor-control systems such as refrigeration,  absorption,  incineration,  or
adsorption can be employed to reduce such emissions.   Control efficiencies are
80 to 95 percent for refrigeration,  70 to 92  percent for  absorption, 98 to 99
percent for incineration,  and 99 percent for  adsorption.   The refrigeration,
absorption, and adsorption systems also recover gasoline,  which makes these
control systems economically attractive.
     Currently, one carbon adsorption unit is  used for the loading of benzene
into barges and several  are used for the unloading of gasoline from barges.
The barge loading operation is in Corpus Christi,  Texas.   The economic  analy-
sis of costs indicates that carbon adsorption  is a cost-effective  VOC control
method.  For carbon adsorption, the  estimated  credit is from $0.45 per  1000
gallons of gasoline loaded.
     Currently, the only on-barge control system used is  refrigeration.   A
barge-mounted, refrigerated vapor control unit is  being installed  for crude
oil loading operations in Santa Barbara County, California, and will be in
operation in the near future.  Santa Barbara  County is the only area that
requires barge loading vapor controls.  Economic analysis  of projected  costs
of gasoline truck loading at bulk terminals indicates that on-barge refrig-
eration is also cost-effective.  For on-barge  refrigeration, the estimated
credit is $0.10 cents per 1000 gallons of gasoline loaded.  Costs  of other
control methods vary up to $0.72 per 1000 gallons  of gasoline loaded.

-------
                               1.   INTRODUCTION

     During the past several years, the Office of Air Quality Planning and
Standards (OAQPS) of the U.S.  Environmental  Protection Agency (EPA) has devel-
oped a series of Control Techniques Guidelines (CTGs) for volatile organic
compounds (VOC) to assist state and local agencies in developing regulations
for VOC control.  Although the CTGs cover major VOC source categories from an
overall nationwide perspective, several VOC source categories that are not
covered by these CTG documents are major contributors to the ozone problem
within given areas.
     Air pollution control agencies in the Philadelphia Air Quality Control
Region (AQCR) have asked for guidance in determining whether VOC controls are
available for non-CTG sources and for information to assist them in developing
appropriate regulations.  One such VOC source is the loading of gasoline
barges.

1.1  SOURCES AND AMOUNTS OF EMISSIONS
     The Philadelphia area is the site of several petroleum refineries and
also a transportation hub for shipping.  Thus, the handling of petroleum
products such as gasoline is a potentially large source of VOC emissions in
the area.  Barges are frequently used for intracoastal transport of gasoline.
Table 1 shows emission sources and factors for loading gasoline barges.  The
range of hydrocarbon emission factors far exceeds the differences in emission
factors for inland waterway barges, ocean barges, and ships.  The average
emission factors cited in AP-42 for ships and ocean barges are lower than for
inland waterway barges, but the average emission factor for barges falls into
all three ranges.  Because most of the gasoline is for domestic use, it is
appropriate to use the inland waterways emission factors.

-------
                       TABLE  1.   VOC  EMISSION  SOURCES  AND
                           FACTORS  FOR GASOLINE  BARGES
      Emission  source
     Barge condition
      Emission factor
      Barge  loading
     Barge transit
Cleaned and vapor-free
Uncleaned - dedicated
 service
Average condition
Average
1.2 lb/103 gal transferred

4.0 lb/103 gal transferred
4.0 lb/103 gal transferred
3.0 Ib/wk per 103 gal
 transferred
      Emissions depend on the condition of the barge at the time of loading.
Barge breathing emissions during transit are relatively small when compared
with  loading losses.  In this document, an emission factor of 4.0 Ib per 1000
gallons of gasoline transferred was used because gasoline barges are normally
dedicated to one service and thus do not have to be cleaned after each load.
     Gasoline barges range in capacity from 10,000 to 50,000 barrels.  Accord-
ing to the literature, the average barge capacity is 20,000 barrels  (840,000
gallons).2'3'4  Loading rates may vary from 1000 to 4200 gallons per minute
per barge.2'3'4  Thus, for an average 20,000-barrel barge, loading can take
from 3 to 14 hours.
     Although facilities for gasoline barge loading are not standardized, they
do have features.   Terminals are normally multipurpose facilities that are
adjacent to or part of a refinery,  and they are owned by the refinery.   The
docks where the barges are loaded may also be used by ships and are often used
to load other liquid petroleum products.   It is assumed that the terminals use
submerged fill  pipes for loading to avoid unnecessary splashing and to reduce
emissions.   Because the docks use dedicated lines for various products, they
tend to be crowded with piping,  valves,  loading arms,  and connecting  hoses.
Crowding is particularly evident at older facilities where expansion  has
encroached on available space.   Unless the demand for a specific product is
very great,  a dock usually is equipped to unload one barge at a time.  It
might be possible  to unload a kerosene barge and an unleaded gasoline barge
simultaneously,  but not two unleaded gasoline barges.

-------
     The gasoline loading pumps are normally in the storage tank farm area of
a refinery.   To prevent mixing of leaded and unleaded gasoline stocks,  separ-
ate storage  and transfer facilities are used for each grade.   After the barge
is moored at the dock for loading,  the cargo loading hose is  lifted into place
and bolted to the loading manifold  and dock piping.   The barge operator opens
the loading  manifold valves and the barge compartment caps.   Opening the
compartment  caps permits release of displaced vapors and manual  gauging of the
gasoline level.   Loading is started by gravity feed to ensure that the  valves
along the flow path are open.   The  loading pump is then started.   As the com-
partment fills, manual  gauging continues.   When the compartment is almost
full, the barge operaion alerts the onshore operator to reduce the flow rate
and to begin shutting the valves in the supply line.   When the shoreside
valves are closed,  residual gasoline in the connecting line is drained  to the
barge before it is  disconnected.   Gasoline and air are displaced through the
open hatch during filling.   At the  average gasoline loading rates, uncon-
trolled VOC  emissions range from 480 to 960 Ib/h.   Therefore, for a barge
loading terminal with an annual capacity of six million barrels per year, VOC
emissions may total 504 tons per year.   Although some refineries considered
this information proprietary,  a telephone survey of the refineries in the AQCR
found that barge usage  ranged from  zero to four million barrels per year.

-------
                        2.   EMISSION CONTROL TECHNIQUES

     Because a barge loading facility is similar in many respects to a bulk
terminal  where gasoline trucks are loaded,  the control  methods will  be simi-
lar, except that the systems must be much larger to accommodate the higher
gasoline transfer rates at barge loading facilities.   Several  systems could be
used to control  VOC emissions from gasoline barge loading operations.

2.1  REFRIGERATION2
     A refrigeration vapor-recovery system recovers displaced  gasoline vapors
by condensation at cryogenic temperatures and atmospheric pressure.   Figure 1
is a flow diagram of a refrigeration vapor-recovery system.
     The vapors collected from the various  gasoline-loading operations are
routed into one of two identical vapor-processing trains.   A dehydrator first
cools the gasoline vapors to a temperature  of 25° to 35°F by direct contact
with finned tube cooling coils.   A major portion of the moisture and a small
portion of the heavier hydrocarbons in the  vapors are condensed in the dehy-
drator.  Vapors then flow from the dehydrator to the condenser, where they are
cooled to a temperature of -80° to -100°F and most of the remaining gasoline
vapors are condensed.   Methane,  which cannot be condensed at -100°F, is vented
from the condenser along with the treated air.
     Water and oil condensates collected in the dehydrator are separated in a
gravity oil-water separator.  The water is  routed to wastewater treatment, and
the recovered organic  material goes to storage.   The product recovered by the
refrigeration vapor recovery unit is a mixture consisting mainly of propane,
butanes,  and pentanes.   Because of their high vapor pressures, these recovered
products must be stored and handled in closed pressurized systems.   Recovered
product is pumped regularly from storage back to the refinery.

-------
1
GASOLINE
i DADINC M.
VAPORS

OIL-WATER
jLrAKAlOR
1
OIL/WATER''
CONDENSATE c^
DEHYDRATOR CONDENSER
25°F TO 35UF -80UF TO -10
I
*i |
i j,,
xi «»•««
t !
i
ti !i
DEHYDRATOR CONDENSER
^ 25UF TO 35UF -80UF TO -10
CONDE
PRO[
i '
. , f
mi -wiipp ^,
SEPARATION
OIL/WATER
rnunruc«Tc 	 ^UOTFB Tn nKPfKfil
^ T^rFrrTrfPl „ PRODUCT RETURNED
^ CONDENSED *" Tn RFFTNFRY
PRODUCT \ 1U RtHNtRY
DENSED
ODUCT
•> VENTED ATP


GE
AT I ON

,, M. VFNTFD AIR

NSED
)UCT^^
^NfTFl^Fri ^. PRODUCT RETURNED
PRODUCT TO REFINERY
v^TORAGE^
Figure 1.   Refrigeration vapor recovery unit.

-------
     Refrigeration for the dehydrator and condenser is supplied by a two-stage
refrigeration unit.   A large volume of cold coolant is stored in the system to
supply instantaneous cooling capacity on demand.
     Two vapor-processing trains are used so that one train can be defrosted
without interruption of the vapor processing.   The cold air vented from the
condenser creates a moisture plume in cold or humid weather.   To avoid the
formation of plumes, some designs incorporate air reheat by the use of waste
heat from the refrigeration unit.
     The control efficiency for this system is estimated to be 80 to 93 per-
cent.4  The efficiency varies according to the methane content of the vapor.
An average control efficiency for truck loading at bulk terminals is reported
to be 87 percent5; thus, we have used that figure to estimate the efficiency
of refrigeration control for barge loading.

2.2  ABSORPTION2
     The absorption vapor-recovery unit absorbs hydrocarbons from gasoline
vapors into a lean oil stream from the refinery.   Figure 2 presents two lean
oil absorption systems.
     One proposed lean oil absorption system operates at low pressure and
ambient temperature.  Gasoline-loading vapors are compressed to 20 psig by a
blower or liquid ring compressor.  The compressed vapors then contact a lean-
oil stream in a packed-bed absorber, where hydrocarbons are absorbed by the
lean oil.  Projected flow rates are 500 barrels/h of lean oil per 1000 acfm of
vapor.  Purified air is vented from the absorber unless methane is present in
the vapors.   Methane cannot be absorbed; it passes through the column and is
vented with the treated air.
     The lean oil source is cat-cracker feed from the refinery.  A one-month
supply of lean oil is stored at the vapor recovery site.  Rich oil laden with
absorbed light hydrocarbons is recycled to the lean oil storage tanks until
the vapor pressure of the lean oil is too high for effective absorption.  When
this occurs, the enriched oil is returned to the refinery, and the lean oil
storage tanks are replenished with fresh cat-cracker feed.
     As a means of avoiding some of the potential safety problems associated
with blowers and compressors, a second type of lean oil absorption system has
been proposed, which uses air eductors for the primary motive force.  Because

-------
uj    a:
o:
o    §
                                                         c
                                                         13
                                                          o
                                                          L)
                                                          QJ
                                                          i-
                                                          O
                                                          Q.
                                                          rO
                                                          Q.
                                                          5-
                                                          O
                                                          I/)
                                                         _Q
                                                         
-------
air eductors produce a slight vacuum in the lean oil  absorber,  effective
absorption requires low temperatures.   The lean oil  is cooled to 40°F by
refrigeration.   Lean oil  flow rates for the refrigerated absorption vapor
recovery unit have been estimated at 200 barrels/n per 1000 acfm vapor.
     Some absorber designs call  for constant lean oil  flow rates that exceed
the flow rate required for effective hydrocarbon recovery at maximum loading
rates.   Other absorber designs include instrumentation and control  systems for
regulating lean oil flow rates with demand to conserve energy.
     The efficiency of this control system (71 to 92  percent5)  is a function
of the amount of methane in the gasoline vapors.   A system efficiency of 87
percent, based on tank truck operations,5 was used for the evaluations.

2.3  INCINERATION2
     The incineration vapor-control unit eliminates gasoline loading vapors by
burning them.  The flow diagram of a typical incineration vapor-control  unit
is shown in Figure 3.
     Gasoline vapors from marine loading operations are drawn through a  flame
arrester and a saturator.  In the saturator, gasoline vapors are enriched with
hydrocarbons through contact with recirculating gasoline product so that the
vapor concentration exceeds the upper explosive limit.  Some systems use
propane instead of gasoline as the enrichment source.
     The enriched or saturated vapor is conveyed by blower to a knockout drum,
where condensates are allowed to settle.   From the knockout drum, the gasoline
vapors enter an incinerator or flare after passing through a water seal.  In
the incinerator or flare, combustion air is mixed with the gasoline vapors to
bring them back into the combustible range before they are burned.
     Incinerators require sophisticated instrumentation and control systems
for maintenance of proper fuel-to-air ratios for all  gasoline vapor flow rates
and concentrations.  Most incineration units are equipped with  oxygen analy-
zers, hydrocarbon analyzers, and temperature sensors  for the control of  satur-
ator operation and combustion air flow rates.
     The efficiency of incinerators for the control  of gasoline loading  vapors
is well over 99 percent.   Hydrocarbon concentrations  in the flue gas vented
from incineration units are well below I percent by volume.  We have used a 99
percent efficiency for our incinerator evaluation.

                                       8

-------
o
co
                                      o
                                      oo
                                      CJ3
                                                          i
                                                                                                c
                                                                                                Z3
                                                                                                c:
                                                                                                o
                                                                                                u
                                                                                                 o
                                                                                                 Q.
                                                                                                 ro
                                                                                                 C
                                                                                                 o
                                                                                                 03
                                                                                                 S_
                                                                                                 O>
                                                                                                 C

                                                                                                 u
01
S-
                                                                                                 cn

-------
2.4  CARBON ADSORPTION6
     Gasoline loading vapors are adsorbed on activated carbon and are removed
from the carbon by vacuum stripping.   The gasoline vapors are recovered by
absorption in gasoline from the bulk storage tank.   The adsorber has two
carbon beds.   During operation, one bed is loaded while the other bed is being
stripped.   The carbon adsorption unit currently in use for benzene loading has
a capacity in excess of 7100 barrels/h.   The units used in barge unloading of
gasoline have capacities ranging between 2840 to 7100 barrels/h.  The units
are shore-mounted and can be used for the unloading of any type of vessel.
The system manufacturer chose vacuum stripping in preference to steam strip-
ping because it eliminates the need for stainless steel construction in the
recovery system.   Figure 4 is a flow diagram of the unit.
     The manufacturer warrants 97.5 percent recovery of gasoline vapor, but
claims that better than 99 percent recovery is obtained.   The systems current-
ly in use are working well.   Maximum VOC emissions have been 30 milligrams per
liter loaded, whereas the EPA standard is a maximum of 35 milligrams per liter
(personal  communication from D. Buxton,  McGill, Inc., November 30, 1983).

2.5  CONTROL SYSTEM RELIABILITY
     All of the control systems discussed have been successfully used and
tested in tank-truck gasoline-loading terminals.5  Except for the carbon
adsorption unit,  none has been used to control gasoline vapors from barges.  A
skid-mounted refrigeration unit has recently been installed on a barge in
Santa Barbara to condense VOC from crude oil loading.  Although it has not yet
been operated, the refrigeration unit is guaranteed to operate at a minimum
efficiency of 90 'percent (personal communication from J.  English, Santa Bar-
bara County Air Pollution Control Board, December 1, 1982).   The manufacturer
states that 95 to 99 percent VOC recovery will be realized at vapor concentra-
tions ranging from 5 to 20 volume percent (personal communication from
B. Collett, Chapman Engineering Company, December 6, 1982).   Figure 5 is a
drawing of this control system.
                                       10

-------
   AIR
   VENT
     CARBON
    ADSORPTION
       BEDS
INLET
VAPOR
                             AIR RECYCLE
                        VACUUM PUMP
                                                           GASOLINE
                                                            SUDPLY
                                                       PU'.'P
                                     ABSORBER
                                     SEPARATOR
                                               FUf.'P
GASOLINE
 RETURN
          Figure 4.   Carbon  adsorption unit.
                              11

-------
                                                                a>
                                                                >
                                                                o
                                                                u
                                                                QJ
                                                                o
                                                                a.
                                                                c
                                                                o
                                                                ro
                                                                t.
                                                                OJ
                                                                CD
                                                                 OJ
                                                                 S_
                                                                -a
                                                                 OJ
                                                                 I
                                                                 ai
                                                                 O)
                                                                 s-
                                                                 fO
                                                                CQ
                                                                ai
                                                                s-
12

-------
                               3.   COST ANALYSIS


     This section presents an economic analysis of vapor control  systems for

gasoline barge loading.


3.1  MODEL LOADING RATES

     Barge loading rates vary from 1000 to 4200 gallons per minute.2'4  Prac-

tice in the industry, however, is to state loading rates in terms of barrels

per hour.  Since a petroleum barrel contains 42 gallons, the model  plant

loading rates in barrels per hour are as follows:

          Small  terminal           1400 barrels/h for 1 barge berth
          Medium terminal          3000 barrels/h for 1 barge berth
          Large terminal           6000 barrels/h for 2 barge berths

     Ship terminal loading rates,  which are higher than those for barge termi-

nals, typically range from 5,000 to 10,000 barrels per hour.2


3.2  ASSUMPTIONS

     The capital and annual operating costs for the vapor control systems are

based on several assumptions:

     1.   Model  terminal sizes are representative of those in the Philadelphia
          AQCR.

     2.   Barge terminal loading facilities are used 2000 hours per year.
          Daily usage rates for barge terminals range from zero to 24 hours.
          Terminals unload barges as quickly as possible to free the barges
          for service.   Two thousand hours per year is a conservative esti-
          mate;  however, if the utilization rate proved to be higher, the
          already favorable economics for vapor recovery would be improved.

     3.   Cost data for vapor control systems at gasoline tank-truck loading
          terminals can be used to project barge-loading vapor control costs;

     4.   Capital investment costs and operating costs can be updated to
          September 1982 by use of the Plant Cost Index of Chemical Engineer-
          ing magazine.


                                       13

-------
     5.    Utility costs for the operation of each vapor control  system can be
          updated to September 1982 by use of the appropriate ratio of unit
          costs from U.S.  Department of Energy magazine, Monthly Energy Re-
          view.
     6.    Annual  credits for recovered gasoline are based on the September
          1982 wholesale gasoline price.

3.3  CAPITAL INVESTMENT
     The required investment for a vapor control  system varies with the barge
loading rate, the physical layout of the barge terminal, and the type of vapor
recovery system.   The costs in Table 2 for onshore installations include
equipment, piping, instrumentation, and all  installation charges.   They have
been developed from cost data for gasoline tank truck loading terminals and
vendor quotations8 (personal communication from D.  Buxton, McGill,  Inc.,
March 9, 1983).  Site-specific requirements  may cause actual costs  to vary
from the cited figures by as much as 30 percent.

3.4  ANNUAL OPERATING COSTS
     Table 3 presents the estimated annual costs  for the various vapor control
systems.  These costs include direct operating costs, capital recovery
charges, and gasoline recovery credits.
     The direct operating costs include costs for utilities and system main-
tenance.  Costs for utilities were based on  the September 1982 commercial
rates of $0.0484/kWh7 for electricity and $3.28 per 1000 cubic feet2 for
natural  gas.  The power consumption was adjusted for refrigeration  and absorp-
tion vapor-control systems and the fuel consumption of incineration systems at
gasoline tank-truck loading terminals to barge loading terminals by the ratio
of their loading rates.4  The power consumption for adsorption units was esti-
mated from the motor horsepower, and it was  assumed that the absorption strip-
per would use the same power as an absorption recovery unit.
     System maintenance costs were estimated as a percentage of the capital
investment costs:  3 percent for refrigeration, adsorption, and absorption
systems and 2 percent for incinerator systems5 (personal communication from
D. Buxton, McGill, Inc., March 9, 1983).
                                       14

-------
                 TABLE 2.   CAPITAL INVESTMENT COSTS FOR VAPOR
                  CONTROL SYSTEMS FOR GASOLINE BARGE LOADING
                           (September 1982 dollars)
    Barge terminal    Number of
    loading rate,      barge
      barrels/h       berths

A.   Onshore installation

        1400            1

        3000            2

        6000            2

B.   On-barge installation

        1400            1
                                   Type of vapor control  systemc
                                  Refri-
                                 geration
                                  517,000
Absorp-   Incinera-  Adsorp-
  tion       tion      tion
 587,000  422,000
                                  853,000    1,017,000   690,000
320,000

380,000
                                1,170,000    1,486,000   898,000     485,000
                                  282,000
NOTES:
b
 Investment  costs  include  charges  of  $110,000  per  barge  for  required  vapor
 collection  equipment  upon the  barge  (Reference  2).

 Values  agree  within 20  percent with  updated values  developed  by  Pullman-
 Kellogg,  Inc.  (Reference  4).

'Skid-mounted  unit at  a  unit  cost  of  $250,000  f.o.b.  factory (personal  com-
 munication  from B. Collett,  Chapman  Engineering Company,  December  6,  1982).
                                       15

-------
                          TABLE 3.   ANNUAL COSTS AND
                  COST-EFFECTIVENESS OF VAPOR CONTROL SYSTEMS
                           (September 1982 dollars)


                                         Barge loading rate,   barrels/h
Refrigeration (Onshore)
                                          1,400
                                                     3,000
         6,000
                 b
No.  barge berths
Control efficiency, percent
Controlled emissions, tons/yr
Gasoline recovered, 103 gal/yr
Annual costs
  Utilities3
  Maintenance
  Capital charges"
  Gasoline recovery credit1"
  Net cost (credit)
Cost-effectiveness, $/ton
 VOC controlled

Refrigeration (On-barge)

No.  barge berths
Control efficiency, percent
Controlled emissions, tons/yr
Gasoline recovered, 103 gal/yr
Annual costs
 Utilities6
 Maintenance
 Capital charges"
Gasoline recovery credit1"
Net cost (credit)
Cost-effectiveness, $/ton VOC
 control led

Absorption

No.  barge berths
Control efficiency, percent
Controlled emissions, tons/yr
Gasoline recovered, 103 gal/yr
Annual costs
  Utilities3
  Maintenance    ,
  Capital charges
  Gasoline recovery credit
  Net cost (credit)
Cost-effectiveness, $/ton
 VOC controlled
                b
1
87
204.6
73.2
11,900
15,500
88,600 ,
(76,100)
39,900
195
1
87
204.6
73.2
7,500
8,500
48,400 .
(76,100)
(11,700)
(57)
1
87
204.6
73.2
7,700
17,600
100,700 .
(76,100)°
49,900
2
87
438.5
156.9
25,500
25,600
146,300
(163,200)
34,200
78



2
87
438.5
156.9
16,600
30,500
174,400
(163,200)
58,300
2
87
877.0
313.8
51,200
35,100
200,600
(326,400)
(39,500)
(45)



2
87
877.0
313.8
33,100
44,600
254,800
(326,400)
6,100
                                            244
133
                                       16

-------
TABLE 3.   (continued)
Inci neration
                                         Barge loading rate,  barrels/h
                                          1,400
                           3,000
          6,000
No. barge berths
Control efficiency, percent
Controlled emissions, tons/yr
Gasoline recovered, 103 gal/yr
Annual costs
  Utilities3
  Maintenance    ,
  Capital charges
  Gasoline recovery credit
  Net cost (credit)
Cost-effectiveness, $/ton
 VOC controlled

Carbon Adsorption

No. barge berths
Control efficiency, percent
Controlled emissions, tons/yr
Gasoline recovered, 103 gal/yr
Annual costs
  Utilities
  Maintenance    ,
  Capital charges
  Gasoline recovery credit
  Net cost (credit)
Cost-effectiveness, $/ton
 VOC controlled
c
1
99
232.8
Nil
4,400
8,400
72,400
Nil
85,200
366
1
99
232.8
83.3
11,700
9,600
54,900
(86,600)
(10,400)
2
99
499.0
Nil
9,500
13,800
118,300
Nil
141,600
284
2
99
499.0
178.5
22,600
11,400
65,200
(185,600)
(86,400)
2
99
997.9
Nil
19,000
18,000
154,000
Nil
191,000
191
2
99
997.9
357.1
45,000
14,600
83,200
(371,400)
(228,600)
                  (45)
(173)
(229)
 Updated and adjusted costs based upon Reference 8.
k
 Estimated as 17.147 percent of capital investment.

cGas-oline recovered is priced as $1.04 per gallon with a specific gravity of
 0.67.

 Parentheses indicate a financial credit.
£1
 Personal communication from B. Collett, Chapman Engineering Company, Decem-
 ber 6, 1982.
                                       17

-------
     Capital  charges,  which are the indirect costs associated with the opera-
tion of the various systems, include a capital  recovery charge and an allow-
ance for property taxes, insurance, and administrative costs.   The annual
capital recovery charge, based on a 15-year life2 and a 10 percent interest
rate, is 13.1 percent of the capital investment.8  The allowance for taxes,
insurance, and administrative costs is 4 percent of the capital  investment.2
     Gasoline recovery credits apply to the refrigeration, absorption, and
adsorption control -methods because these systems recover condensed gasoline
vapors.  The gasoline credit was based on the September 1982 wholesale gaso-
line price of $1.04 per gallon (personal communication from-T. Elsass, Tresler
Products, Division of Ashland Oil Company, December 20, 1982).  The gasoline
recovered is equal to the amount of gasoline controlled by the systems.   No
gasoline recovery credits are involved with the use of incineration systems.

3.5  COST-EFFECTIVENESS
     The cost-effectiveness of the control system is determined by dividing
the annual costs for each system by the annual  amount of VOC removed.  Table 3
shows that on-barge refrigeration and carbon adsorption have negative control
costs because credits for recovered gasoline are greater than the capital  and
operating costs of the control system.  Costs of the other control options
range from $7 to $366/ton of VOC controlled.

3.6  ECONOMIC IMPACT OF CONTROLS
     The costs of installing and operating a vapor-control system for gasoline
barge loading would be passed on to consumers.   The gasoline price increase
for each vapor control system was determined by dividing the annual operating
cost by the annual gasoline loading throughput.  Table 4 shows a possible
nega'tive cost increase; however, the cost data developed and used in this cost
analysis are not based on actual operation of a barge-loading vapor-control
system.  The main purpose of this financial analysis is to demonstrate the
technical and economic feasibility of vapor-control systems for such loading
operations.  The operations of carbon adsorbers on gasoline barge loading and
the Santa Barbara on-barge refrigeration unit should be monitored to compare
actual performance with vendor claims.

-------
                TABLE  4.   ECONOMIC  EFFECT ON  USING VAPOR CONTROL
                  SYSTEMS  DURING THE  LOADING  OF GASOLINE BARGES
Control system
Refrigeration
On-barge
Onshore
Absorption - onshore
Incineration - onshore
Adsorption - onshore
Annual volume3
loaded, 106 gal/yr
118
118
252
504
118
252
504
118
252
504
118
252
504
Annual cost, Sept.
1982 dollars
(ll,700)b
39,900
34,200
(39,800)
49,900
58,300
6,100
85,200
141,600
191,000
(10,400)
(86,400)
(228,600)
Added cost
(credit)
cents/103 gal
(10)
34
14
(8)
42
23
1
72
56
38
(9)
(34)
(45)
 Annual volume based on 2000 hours operation.
 Parentheses indicate a profit for system operation.

     Based on average hydrocarbon emission factors for gasoline loading opera-
tions cited in AP-42,1 emissions per 1000 gallons of gasoline transferred are
lower for ships and ocean-going barges.  Although potential hydrocarbon emis-
sions would be lower, the scale of operations would be greater.   This would
allow the use of larger-capacity control devices.  As shown in Table 3, this
would have a very favorable impact on the economics of control.   For a barge
loading rate of 6000 barrrels/h, a negative annual control cost of $80,000
would result, even if the gasoline recovery credit were reduced by 49 percent.
(The average emission factor for ships is 60 percent of the average emission
factor for barges.)  It might also be feasible to use an adsorber with a
larger loading capacity,  which would improve the economics.
                                       19

-------
                            4.   REGULATORY ANALYSIS

     This section contains a proposed draft regulation for VOC controls at
barge gasoline-loading terminals, a discussion of compliance monitoring tech-
niques, and a description of potential  problem areas.

4.1  DRAFT REGULATION
     The draft regulation on VOC control  of barge gasoline-loading terminals
in the Philadelphia AQCR is as follows:
       CONTROL OF EVAPORATIVE LOSSES FROM THE LOADING OF GASOLINE BARGES
A.   Definitions
     1.   "Barge" means any tank vessel  not equipped with means of self-
          propulsion.
     2.   "Marine terminal" means all permanent facilities used in whole or in
          part to load gasoline into cargo vessels for bulk transport.
B.   Applicability
     The provisions of this Regulation  shall  apply to the loading of gasoline
into barges from any marine terminal.
C.   Coast Guard Rules
     Nothing in this Regulation shall be construed as superseding or conflict-
ing with applicable United States Coast Guard regulations.  Any person who
believes that any provision of this Federal rule supersedes or conflicts with
United States Coast Guard Regulations shall submit documentation to the Con-
trol Officer a minimum of 60 days prior to the scheduled compliance date for
that provision.
D.   Emissions From Loading Gasoline
     1.   No person shall cause or allow the loading of gasoline into gasoline
          barges from any marine terminal unless the gasoline vapors displaced
                                       20

-------
          from the barge during loading are directed to a system where at
          least 80 percent by weight of the organic compounds  in the vapor are
          recovered or destroyed.

     2.    The owner or operator of any marine terminal  used to load gasoline
          into gasoline barges, or the owner or operator of such vessel,  shall
          demonstrate to the satisfaction of the Control Officer,  by emission
          tests,  engineering evaluations, or other means of reasonable preci-
          sion and accuracy that:

          a.    The control practices or equipment selected to  achieve compli-
               ance will reduce the organic compounds in the vapor to the
               extent required by section D.I;  and

          b.    There is a reliable method for determining the  effectiveness of
               such control practices or equipment on a routine basis.

E.    Recordkeeping

1.    The owner or operator of any marine terminal shall keep operating records
     regarding the gasoline barge loading activities for that  terminal.   These
     records  shall be maintained at the respective marine terminals and shall
     be made  available to the Control Officer upon request.  They shall  in-
     clude,  but are not limited to:

     a.    The dates for each loading event, the identification of the barge
          being loaded, the company immediately responsible for the operation
          of  the barge, and the legal owner of  the barge;

     b.    The amount of gasoline loaded into each barge; and

     c.    A written document, signed by both the person responsible for the
          operation of the barge and the person responsible for the operation
          of  the marine terminal,  attesting to  the fact that their respective
          portions of the organic vapor control system, as required by section
          D.I of this regulation,  were operated as designed during each load-
          ing event.

F.    Compliance Schedule

     1.    Any owner or operator subject to the  provisions of this Regulation
          shall comply with the following increments of progress following
          adoption of the Regulation and promulgation of safety criteria for
          organic vapor control systems by the  United States Coast Guard:

          a.    Within 12 months, submit final control system plans to the
               United States Coast Guard for approval.

          b.    Within 24 months, submit to the  Control  Officer a final  system
               control plan that describes the  steps that will be taken to
               achieve compliance with the provisions of this  regulation.
               Such plans shall include specific statements that:
                                       21

-------
               1.    All  barges loaded from the terminal  shall  have a designed
                    system that is compatible with that of the terminal;

               2.    The  barge operators are aware of the design and operation-
                    al  requirements of the marine terminal;  and

               3.    Terminal  operators will not service a barge that is in-
                    compatible with the recovery system.

          c.    Within 30 months,  negotiate and sign all  necessary contracts or
               issue purchase orders for securing an emission  control  system.
               The proposed equipment installation schedule  shall be provided
               to the Control Officer at this time.

          d.    Within 54 months,  complete installation of emission control
               equipment.

          e.    By July 1,  1987, assure final  compliance with the provisions of
               this Regulation.

     2.    Nonavailability of specific emission control equipment or of a
          specific emission control system or a specific method for achieving
          compliance with any provision of this regulation shall not consti-
          tute relief from such provision if other types of  control equipment,
          systems, or methods acceptable to the control  officer are available.

     3.    Any gasoline barge or marine terminal operating after July 1, 1987,
          shall be operated in full compliance with the provisions of this
          Regulation.


4.2  REGULATION IN OTHER AREAS

     Recognizing the need for reducing VOC emissions from marine vessels,  the

California Air Resources Board passed resolution 78-59 on November 16, 1978.

This resolution is attached as Appendix A.
     In 1978, Santa Barbara County, California, adopted Rule 327 to regulate

VOC emissions from marine vessels.  The current rule, presented in Appendix B,

establishes a schedule requiring final compliance with emission regulations by

July 1,  1985.  As of now,  with only 2h years remaining to achieve compliance,

no control systems other than the barge-mounted refrigeration  unit have been

planned.   In light of this deadline, the control agencies in California expect

that the affected companies to be submitting control system  applications  in

the near future (personal  communication from J. English, Santa Barbara County

Air Pollution Control Board,  December 1, 1982).
                                       22

-------
     The U.S.  Coast Guard requires an inert gas blanketing system for a self-
propelled vessel with a dead weight displacement, of 20,000 tons (or greater)
and built after 1979.  The vessel also must have a high-capacity washing
machine for cleaning products off the sides of the vessel.  The Coast Guard
has no requirement, however, for an inert gas system for either a carbon
adsorber or a loading/unloading system.   According to a personal communication
from Petty Officer Dillon, U.S.  Coast Guard, on September 27, 1983, the inert
gas blanketing requirement does not apply to barges in any location.
     Coast Guard regulations state that all barges must be inspected every two
years.   The applicable regulations are published in the Code of Federal Regu-
lations.   The citations are:
     1.   For manned barges, 33 CFR §§155-156; 46 CFR §§30-40; and
     2.   For unmanned barges, 46 CFR §151.

4.3  FIRE AND EXPLOSION HAZARDS
     As for any installation handling gasoline, safeguards are required to
prevent fires and explosions from occurring either on the barge or in the on-
shore vapor control units.  Because vapor concentrations may be in the explo-
sion range for gasoline, special system requirements must be met to prevent
static electricity discharges and sparking.  Those special requirements in-
clude:
     1.   Explosion-proof electric equipment and wiring;
     2.   Spark-proof blowers or conductors; and
     3.   Equipment grounding devices.
     The spread of fire from barge to terminal or terminal to barge must be
guarded against by flame arresters and water seal drums for isolation of
system components.
     Provisions of the safety criteria established by the U.S. Coast Guard
will govern the design of any vapor control system that is used.
                                       23

-------
4.4  COMPLIANCE AND MONITORING TECHNIQUES
     Manufacturer certification for barge vapor control  systems is the pre-
ferred compliance demonstration technique.   This certification could be based
on the performance of similar systems at gasoline tank-truck loading termi-
nals.   According to the carbon adsorber manufacturer (personal communication
from D.  Buxton, McGill, Inc., March 23, 1983),  cited recovery efficiencies for
this document are based on emission test results on 10 operating units.
Because only two systems (out of 330) are used  for barge vapor recovery,  the
manufacturer considers operating characteristics to be nearly identical  for
tank truck and barge loading terminals.  Hence, system efficiencies are inter-
changeable for the two applications.   These compliance techniques are pre-
sented in Appendix C and can be suitably modified for use in controlling
emissions from gasoline barge loading.
                                       24

-------
                                 REFERENCES
1.   U.S.  Environmental  Protection Agency.   Compilation of Air Pollutant Emis-
    sion Factors.   AP-42, July 1979.   pp.  4.4-7 and 4.4-10.

2.   Burklin, C.  E.,  et al.   Background Information on Hydrocarbon Emissions
    From Marine Terminal Operations,  Volume 1.   EPA-450/3-76-038a, 1976.

3.   Engineering Science, Inc.   Summary of Technical Information for Selected
    Volatile Organic Compound Source  Categories.   Arcadia, California.
    EPA-400/3-81-007, May 1981.   p.  4-3.

4.   Gee, D., and W.  M.  Talbert.   Control  Technology Evaluation for Gasoline
    Loading of Barges.   EPA-600/2-79-069,  March 1979.

5.   U.S.  Environmental  Protection Agency.   Control of Hydrocarbons from Tank
    Truck Gasoline Loading Terminals.   Emission Standards and Engineering
    Division, Chemical  and Petroleum  Branch, Research Triangle Park, North
    Carolina.  EPA-450/2-77-026, October 1977.

6.   McGill,  Inc.   The McGill  Adsorption/Absorption Gasoline Vapor Recovery
    System.   Sales brochure.   Tulsa,  Oklahoma.   1980.

7.   U.S.  Department of Energy.  Monthly Energy Review, July 1982, p. 90.

8.   Neveril, R.  B.   Capital and Operating Costs of Selected Air Pollution
    Control  Systems.  EPA-450/5-80-002, December 1978.
                                      25

-------
           APPENDIX A

     STATE OF CALIFORNIA AIR
RESOURCES BOARD RESOLUTION 78-59
               A-l

-------
                     Resolution 78-59

                     November  16,  1978
WHEREAS, marine vessel operations in California Coastal Waters
account for substantial quantities of air pollutants;

WHEREAS, the California Health and Safety Code and  the federal
Clean Air Act Amendments of 1977 require extraordinary efforts
to achieve the state and federal ambient air quality standards
throughout California;

WHEREAS, the California Air Resources Board recognizes the need
to reduce air pollution in a cost-effective fashion that will
minimize economic hardships;

NOW, THEREFORE, BE IT RESOLVED, that the Air Resources Board
adopts in principle the Model Rule for the Control  of Sulfur
Oxides and Organic Ga-s Emissions from Marine Vessel Operations;

BE IT FURTHER RESOLVED, that the Board instructs its staff to
schedule a formal Workshop no later than March 1979 to examine
remaining technical and economic questions concerning the Model
Rule;

BE IT FURTHER RESOLVED, that the Board instructs the staff to
modify the rule to specifically:

     (1)  Provide   that U.S.-flag vessels will not  suffer undue
         discrimination;

     (2)  insure that normal bunker operations will not be
         subject to the vapor recovery requirements of the Rule;

     (3)  eliminate any portions of the Rule which would put
         California ports at a significant er .iomic disadvantage
         when compared to other west coast ports;

     (4)  assure that any systems used for emissions control
         will not jeopardize vessel safety;

BE IT FURTHER RESOLVED, that the Board directs the staff to
determine the availability and cost of low-sulfur bunker fuel;

BE IT FURTHER RESOLVED, that the Board instructs the Executive
Officer,  following the formal Workshop,   to  reschedule considera-
tion of the Model Rule before the Board.

-------
BE IT FURTHER RESOLVED,  that the Board instructs  the  Executive
Officer to confer  with  the U.S.  Environmental Protection Agency,
U.S. Coast Guard,  International Marine Consultive  Organization
and other appropriate organizations to determine whether an
acceptable international control program can be developed in
lieu of the proposed Model Rule.
                                   I certify that the above is a true
                                   and correct  copy of Resolution 78-59
                                   as passed by the Air Resources Board.

-------
            APPENDIX B

     SANTA BARBARA COUNTY AIR
POLLUTION CONTROL DISTRICT RULE 327
                3-1

-------
RULE  327.   ORGANIC LIQUID CARGO VESSEL LOADING.




     A.   Definitions.




         1.   "Barge" means any tank vessel not equipped




with  means  of  self-propulsion.




         2.   "Marine Terminal" means all permanent facilities




used  in whole  or  in part to load organic liquid cargo into




organic liquid cargo vessels,  excepting existing barge




loading facilities as  long as  their annual throughput, as




cf the adoption of this Ruler  does not increase.




         3.   "Organic  Liquid Cargo" means organic liquid




loaded into a  vessel to be transported from one location




:o another  location as a commodity.



         4.   "Organic  Liquid Cargo Vessel" means any tanker,




 freighter,  barge,  vessel, boat or ship used for the bulk




 •.ransport of more  than 250 barrels of organic liquid cargo.




     B.   Applicability,




         The provisions of this Rule shall apply to the loading




 *i organic  liquid  cargo into organic liquid cargo vessels




 •res any  marine terminal.




     C.   Coast Guard Rules.




         Nothing  in this Rule  shall be construed as superseding




  '" "-nflicting with United States Coast Guard Regulations.


  Irt-r -
   '•  j-erson  who believes that any provision of this Rule super-


  •.., •

   ~*3 °r  conflicts with United States Coast Guard Regulations




   "*•- submit documentation to  the Control Officer a minimum

   ,» ,

    '• cays  prior   to  the scheduled compliance date for that
                             -338-

-------
     D.  Emissions from Loading Organic  Liquid.

         1.  No person shall  load or allow  the  loading of

organic liquid cargo into any organic  liquid  cargo  vessel

from any marine terminal unless the weight  of non-methane

organic vapors emitted during loading  is reduced  by the


application of the best control technology  available at the

time specified in the final control plan submitted  to the


Control Officer.                                             ';
                                                             i*
         2.  The owner or operator of  any marine  terminal

used to load any organic liquid cargo  into  any  organic liquid

cargo vessel, or the owner or operator of such  vessel, shall
                                                            - ,<-
                                                           " v "-
demonstrate, to the satisfaction of the Control  Officer, by

en-ission tests, engineering evaluation,  or  other  means of

reasonable precision and accuracy that:                    "*,


             a.  The control practices or equipment selected

to achieve compliance will reduce the  organic vapor emission*


to the extent required by section B.I; and

             b.  There is a reliable methodology  for deters-ia"

ing the effectiveness of such control  practices or  equipment
                                                         .: ;*£;.' :r-
on a routine basis.                                      "~t""•-'
                                                        ^f-.?'-.''•
     E.  Record Keeping.

         1.  The owner or operator of  any marine  terminal

shall keep operating records regarding the  loading  ac

for that terminal.  These records shall  be  maintained at

respective marine terminals and shall  be made available to


the Control Officer upon request.  These records  shall


but are not limited to: ,
                             -339-

-------
              a.   The date(s)  of  each  loading event,  the name



 of the vessel being  loaded,  the  company  immediately  responsible



 for the operation of the vessel  and the  legal owner  of the




 vessel;



              b.   The type and amount  of  organic  liquid cargo



 loaded into  each  vessel;  and



~"~             c.   A written document signed  by both the person



 responsible  for the  operation of the  vessel and  the  person



 responsible  for the  operation of the  marine terminal which



 attests  to the fact  that their respective portions of the




 organic  vapor control system,  required by section B.I of this



 Pule,  was operating  as designed  during each loading  event.



     F.  Compliance  Schedule.



         1.   Any  owner or operator subject  to the provisions



 of  this  Rule  shall comply with the following increments of



 progress following the promulgation of safety criteria for



 organic  vapor emission control systems by the United States



 Coast  Guard,



              a.   Within  12 months - submit  final control



 system plans  to the  United States Coast  Guard for approval.



              b.   Within  24 months - submit  a final control



 plan which describes the steps that will be taken to achieve



 compliance with the  provisions of this Rule to the Control



 Officer.  Such control plans  shall include  specific  informa-




 tional statements that:



                  1)   All vessels loading through this terminal




 shall  have a  designed system which is compatible with that of




 the terminal;  and





                              -340-

-------
                 2)  The vessel operators are aware of the


design and operational requirements of the marine terminal;


and


                 3)  Terminal operators will not service a


vessel which is not compatible with their recovery system.


             c.  Within 30 months - negotiate and sign all


necessary contracts for emission control systems or issue


orders for the purchase of component parts to accomplish


emission control.  The proposed schedule of equipment installa-


tion by vessel name and terminal location shall be provided

                                                          . w
to the Control Officer at this time.


             d.  Within 54 months - complete construction or


installation of emission control equipment.


             e.  By July 1, 1985 assure final compliance


with the provisions of this Rule.


         2.  The non-availability of specific emission control


equipment or of a specific emission control system or method


to be used for the purpose of achieving compliance with any


provision of this Rule shall not constitute relief from such

                                                          1 *,
provision if other types of emission control equipment,


systems, or methods acceptable to the Control Officer arc-


available.


         3.  Any organic liquid cargo vessel or marine term-


inal, operating after July 1, 1985, shall be operated in *a**


compliance with the provisions of this Rule.
                             -341-

-------
               APPENDIX C

  COMPLIANCE TEST METHOD AND MONITORING
TECHNIQUES - TANK TRUCK LOADING TERMINALS
     EPA REPORT NO.  EPA-450/2-77-026
                   C-l

-------
              6.0   COMPLIANCE TEST METHOD AND MONITORING  TECHNIQUES





6.1   COMPLIANCE TEST METHOD



      The recommended compliance test method  as detailed in  Appendix  A



can be used to determine emissions  from bulk  terminal  gasoline  vapor  control



equipment under conditions of loading leak-free tank  trucks  and trailers,



and leak-free operation of the vapor collection arid  processing  systems.



Direct measurements of volume and concentration of vapor processor  emissions



are made to calculate the total  mass of vented hydrocarbons.  This  total



mass emitted is divided by the total volume of liquid gasoline  loaded



during the test period to determine the mass  emission factor.



      To insure that the vapor collection and processor are operating under




  leak-free conditions,  qualitative monitoring  should be conducted using a



  combustible gas indicator  to indicate any leakage from the tank truck or



  trailer carqo  compartments and all  equipment  associated with  the control



  system.  Any  incidence of  direct  hydrocarbon  leakage would indicate  that



  corrective actions are required prior to  further compliance testing.







      The test period specification  is intended to allow inclusion of the



 typical daily variation in loading  frequency  in each repetition and three



 repetitions are specified  in order  to include the normal  day-to-day variations



 in loading frequency.



      For terminals employing intermittent vapor processing systems,  each



 test repetition must include at least one fully automatic operating  cycle



 of the vapor processing unit.
                                  6-1

-------
      This procedure is applicable to determining hydrocarbon emission
rates from systems serving tank truck or trailer loading only.   For
those facilities employing a single control  system to process vapors
generated from both tank truck and trailer loading and fixed roof
storage tank filling, no storage tank filling may occur during  the
duration of test repetition.
      Source testing may not be required after initial compliance
testing or if preconstruction review indicates the equipment will
achieve compliance.  In such cases, the performance parameters  of the
vapor control system would be checked and compared with compliance
tests of other installations using the same system design.
6.2   MONITORING TECHNIQUES
      The vapor collection system and associated vapor control
equipment must be designed so that under maximum instantaneous  loading
rates, the tank truck pressure relief valves will not vent.
      An intermittent monitoring approach is recommended.  In this
type of program, a portable hydrocarbon analyzer would be used  to
determine the processing unit exhaust hydrocarbon concentration and a
combustible gas indicator would be used to detect any incidence of leaks
from the cargo tanks and vapor collection lines at specified intervals.
      Such a procedure would require the establishment of a  control
equipment exhaust concentration level at which the compliance with a
mass emission factor regulation is assured.
                                 6-2

-------
      There are currently available  instruments that have a dual range of
0-100 percent LEL and 0-100 percent  by volume of hydrocarbons as propane.
The cost of this type instrument is  approximately $500.  A disadvantage
of this type instrument is that the  accuracy of the measurements at 4 to 5
percent hydrocarbon level is about +_ 20 percent.  This may not provide
the precision necessary to differentiate between complying and non-
complying operation.  It would, however, detect gross deviations from
design operation.  An additional disadvantage is that comparative
calibrations would be necessary to relate the monitoring results to the
reference test procedure concentration measurements.
      Portable hydrocarbon analyzers based on FID or NDIR principles are
also available at costs ranging from $1500-$4000.  These instruments
have the advantage of being the most precise measurement techniques
available.  Also, since these techniques are used for hydrocarbon
measurements in the reference procedure, no comparative testing is
necessary to establish relative accuracy of the monitoring technique.
      For leak monitoring alone, many versions of combustible gas
indicators with 0-100 percent LEL spans are available.  The cost of this
type of unit would range from $200 to $500 depending on the particular
vendor and instrument features.
      In addition  to  the  use  of instruments  monitoring  control  equipment
process  variables  (  principally temperature  and  pressure)  can  give  a  good
indication of performance.   The primary variables  of interest  and  the
approximate  values  that would  indicate  acceptable  performance  are  listed
on page  3-1.
                                  6-3

-------
6.3   AFFECTED FACILITY
      In developing terminal  regulations,  it is suggested  that the
affected facility be defined  as the tank truck gasoline loading
stations and appurtenant equipment necessary to load  the tank  truck
compartments.
6.4   STANDARD FORMAT
      It is recommended that  the following provisions be written
into the tank truck gasoline  terminal  loading regulations.
      1.  Gasoline is not to  be discarded  in sewers or stored  in
open containers or handled in any other manner that would result
in evaporation.
      2.  The allowable mass  emissions of hydrocarbons from control
equipment are to be 80 milligrams per liter or less of gasoline
loaded.
      3.  Pressure in the vapor collection lines should not exceed
tank truck pressure relief valve settings.
      Test procedures for determining allowable hydrocarbon
emissions are detailed in Appendix A.
                               6-4

-------
                             APPENDIX A

A.I   EMISSION TEST PROCEDURE FOR TANK TRUCK GASOLINE LOADING TERMINALS
      Hydrocarbon mass emissions are determined directly using flow meters
and hydrocarbon analysers.   The volume of liquid gasoline dispensed is
determined by calculation based on the metered quantity of gasoline at  the
loading rack.  Test results are expressed in milligrams of hydrocarbons
emitted per liter of gasoline transferred.
A.2  APPLICABILITY
     This method is applicable to determining hydrocarbon emission rates
at tank truck gasoline loading terminals employing vapor balance collection
systems and either continuous or intermittent vapor processing devices.
This method is applicable to motor tank truck and trailer loading only.

A.3  DEFINITIONS
  3.1  Tank Truck Gasoline Terminal
     A primary distribution point for delivering gasoline to bulk plants,
service stations, and other distribution points, where the total gasoline
throughput is greater than 76,000 liters/day.
  3.2  Loading Rack
     An aggregation or combination of gasoline loading equipment arranged
so that all loading outlets in the combination can be connected to a tank
truck or trailer parked in a specified loading space.
  3.3  Vapor Balance Collection System
                                     A-'

-------
     A vapor transport system which uses direct displacement by the liquid



loaded to force vapors from the tank truck or trailer into the recovery



system.



  3.4  Continuous Vapor Processing Device



     A hydrocarbon vapor control system that treats vapors from tank trucks



or trailers on a demand basis without intermediate accumulation.



  3.5  Intermittent Vapor Processing_Deyice



     A hydrocarbon vapor control system that employs an intermediate vapor



holder to accumulate recovered vapors from tank trucks or trailers.  The



processing unit treats the accumulated vapors only during automatically



controlled cycles.





A.4  SUMMARY OF THE METHOD



     This method describes the test conditions and test procedures to be



followed in determining the emissions from systems installed to control



hydrocarbon vapors resulting from tank truck and trailer loading operations



at bulk terminals.  Under this procedure, direct measurements are made to



calculate the hydrocarbon mass exhausted from the vapor processing equipment.



All possible sources of leaks are qualitatively checked to insure that no



unprocessed vapors are emitted to the atmosphere.  The results are expressed



in terms of mass hydrocarbons emitted per unit volume of gasoline transferred.



Emissions are determined on a total hydrocarbon basis.  If methane is present



in the vapors returned from the tank trucks or trailers, provisions are



included for conversion to a total non-methane hydrocarbon basis.





A.5  TEST SCOPE AND CONDITIONS APPLICABLE TO TEST



  5.1  Test Period



     The elapsed time during which the test is performed shall not be  less






                                 A-2

-------
than three 8-hour test repetitions.



  5.2  Terminal Status During Test Period



     The test procedure is designed to measure control system performance



under conditions of normal operation.   Normal operation will  vary from



terminal-to-terminal and from day-to-day.  Therefore, no specific criteria



can be set forth to define normal operation.   The following guidelines are



provided to assist in determining normal  operation.



  5.2.1   Closing of Loading Racks



     During the test period, all  loading  racks shall  be open  for each product



line which is controlled by the system under test.  Simultaneous use of more



than one loading rack shall occur to the  extent that such use would normally



occur.



  5.2.2   Simultaneous use of more than one dispenser on each  loading rack



shall  occur to the extent that such use would normally occur.



  5.2.3   Dispensing rates shall be set at the maximum rate at which the



equipment is designed to be operated.   Automatic product dispensers are



to be used according to normal operating  practices.



  5.3  Vapor Control System Status During Tests



     Applicable operating parameters shall be monitored to demonstrate that



the processing unit is operating at design levels.  For intermittent vapor



processing units employing a vapor holder, each test repetition shall include



at least one fully automatic operation cycle of the  vapor holder and processing



device.   Tank trucks shall be essentially leak free  as determined by EPA Mobile



Source Enforcement Division.





A.6  BASIC MEASUREMENTS AND EQUIPMENT REQUIRED



  6.1   Basic measurements required for evaluation of emissions' from gasoline



bulk loading terminals are described below.  The various sampling points





                                      A-3

-------
are numbered in Figure 1.


        Sample Point                         Measurements  Necessary

1.  Gasoline dispensers            - Amount dispensed

2.  Vapor Return Line              - Leak check  all  fittings

3.  Processing unit exhaust        - Temperature of  vapors exhausted

                                   - Press.of vapors exhausted

                                   - Volume of vapors exhausted

                                   - HC concentration of vapors
                                                                     *
                                   - Gas chromatograph analysis  of HC

                                   - Leak check  all  fittings  and vents

  6.2  The equipment required for the basic measurements are  listed below:

  Sample_Point                            Equipment  and Specifications

        2                          1 portable combustible  gas detector,
                                      (0-100% LEL)

        3                          1 flexible thermocouple with  recorder

                                   1 gas volume  meter, appropriately sized
                                      for exhaust flow rate and  range

                                   1 total  hydrocarbon analyzer  with recorder;
                                      (FID or NDIR type, equipped to read out
                                      0-10% by volume hydrocarbons as propane
                                      for vapor  recovery processing device;  or,
                                      0-10,000 ppmv  HC as  propane for incin-
                                      eration processing devices)

                                   1 portable combustible gas detector  (0-100%
                                      LEL)

Miscellaneous                      1 barometer

                                   1 GC/FID w^olumn to separate C, - C-,
                                      alkanes
 *
  Required if methane is present in recovered vapors
 •*
  Required if methane is present in recovered vapors or if incineration is
  the vapor processing technique.
                                      A-4

-------
A.7  TEST PROCEDURES
  7.1  Preparation for testing includes:
  7.1.1  Install an appropriately sized gas meter on the exhaust vent of
the vapor processing device.  A gas volume meter can be used at the exhaust
of most vapor recovery processing devices.  For those where size restrictions
preclude the use of a volume meter; or when incineration is used for vapor
processing, a gas flow rate meter (orifice, pitot tube annubar, etc.) is
necessary.   At the meter inlet, install a thermocouple with recorder.  Install
a tap at the volume meter outlet.  Attach a sample line for a total  hydro-
carbon analyzer (0-10% as propane) to this tap.  If the meter pressure is
different than barometric pressure, install a second tap at the meter outlet
and attach  an appropriate manometer for pressure measurement.  If methane
analysis is required, install  a third tap for connection to a constant volume
                                   *
sample pump/evacuated bag assembly.
  7.1.2  Calibrate and span all instruments as outlined in Section 9.
  7.2  Measurements and data required for evaluating the system emissions
include:
  7.2.1  At the beginning and end of each test repetition, record the volume
readings on each product dispenser on each loading rack served by the system
under test.
  7.2.2  At the beginning of each test repetition and each two hours thereafter,
record the  ambient temperature and the barometric pressure.
  7.2.3  For intermittent processing units employing a vapor holder, the unit
shall be manually started and allowed to process vapors in the holder until
the lower automatic cut-off is reached.  This cycle should be performed
immediately prior to the beginning of the test repetition before reading in
7.2.1 are taken.  No loading shall be in progress during this manual cycle.
 Described in Method 3, Federal  Register, V36, n247, December 23,  1971.
                                     A-5

-------
  7 2.4  For each cycle of the processing unit during each test repetition,



record the processor start and stop time, the initial and final gas meter



readings, and the average vapor temperature, pressure and hydrocarbon



concentration.  If a flow rate meter is used, record flow meter readouts



continuously during the cycle.  If required, extract a sample continuously



during each cycle for chromatographic analysis for specific hydrocarbons.



  7.2.5  For each tank truck or trailer loading during the test period, check



all fittings and seals on the tanker compartments with the combustible gas



detector.  Record the maximum combustible gas reading for any incidents of



leakage of hydrocarbon vapors.  Explore the entire periphery of the potential



ieak source with the sample hose inlet 1 cm away from the interface.



  7.2.6  During each test period, monitor all possible sources of leaks in



the vapor collection and processing system with the combustible gas indicator,



Record the location and combustible gas reading for any incidents of leakage.



  7.2.7  For intermittent systems, the processing unit shall be manually



started and allowed to process vapors in the holder until the lower automatic



shut-off is reached at the end of each test repetition.  Record the data in



7.2.4 for this manual cycle.   No loading shall be in progress during this



manual cycle.





A.8  CALCULATIONS



  8.1   Terminology



       T         = Ambient temperature (°C)
        a


       P,         = Barometric pressure (mm Hg)



                 = Total volume of liquid dispensed from all controlled

                   racks during the test period (liters)
V
                 = Volume of air-hydrocarbon mixture exhausted from the

                   processing unit (M )
                                  A-6

-------
       V         = Normalized volume of air-hydrocarbon mixture exhausted,
        es         NM3 @ 20°C, 760 mmHg

       C         = Volume fraction of hydrocarbons in exhausted mixture
                   (volume % as C3H,Q/100, corrected for methane content
                   if required

       T         - Temperature at processing unit exhaust (°C)

       Pe        = Pressure at processing unit exhaust (mm Hg abs)

       (M/L)     = Mass of hydrocarbons exhausted from the processing unit
                   per volume of liquid loaded, (mg/1 )

  8.2  Processing Unit Emissions

     Calculate the following results for each period of processing unit

operation:

  8.2.1  Volume of air-hydrocarbon mixture exhausted from the processing

unit:

      Ve  =  Vef  -  Ve.,or           (m3)

      V   =  totalized volume from flow rate and time  records.


  8.2.2  Normalized volume of exhausted mixture:

      V    = (0.3858 °K/mmHg) VePe              NM3 @  20°C,  760 mmHg
       es         Te* 273.2
8.2.3  Mass of hydrocarbons exhausted from the processing unit:



                              '
                         mgC3H8             (   }
  8.3  Average Processing Unit Emissions

  8.3.1   Average mass of hydrocarbons emitted per volume of gasoline  loaded:

      (M/L)_  = ^        (mg/liter)

                ~Lt

A. 9  CALIBRATIONS

  9.1  Flow Meters

     Use standard methods and equipment which have been  approved  by the
                                       A-7

-------
Administrator to calibrate the gas meters.
  9.2  Temperature^ Recording Instruments
     Calibrate prior to the test period and following the test period using
an ice bath (0°C) and a known reference temperature source of about 35°C.
Baily during the test period, use an accurate reference to measure the
ambient temperature and compare the ambient temperature reading of all
other instruments to this value.
  9.3  Total hydrocarbon analyzer
     Follow the manufacturer's instructions concerning warm-up and adjust-
ments.   Prior to and immediately after the  emission test, perform a
comprehensive laboratory calibration on each analyzer used.   Calibration
gases should be propane in nitrogen prepared gravimetrically with mass
quantities of approximately 100 percent propane.   A calibration curve
shall be provided using a minimum of five prepared standards in the range
of concentrations expected during testing.
     For each repetition, zero with zero gas (3 ppm C) and span with 70%
propane for instruments used in the vapor return  lines and with 10%
propane for instruments used at the control device exhaust.
     The zero and span procedure shall be performed at least once prior to
the first test measurement, once during the middle of the run, and once
following the final test measurement for each run.
     Conditions in calibration gas cylinders must be kept such that con-
densation of propane does not occur.  A safety factor of 2 for pressure and
temperature is recommended.
                                     ,1-8

-------