EPA-600/2-77-057
February 1977
Environmental Protection Technology Series
                IL  CHARACTERISTICS  OF CARBON  BEDS
                       FOR  GASOLINE  VAPOR  EMISSIONS
                                    Industrial Environmental Research Laboratory
                                          Office of Research and Development
                                         U.S. Environmental Protection Agency
                                   Research Triangle Park, North Carolina  27711

-------
               RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency,  have been  grouped  into five series. These five broaa
categories were established to facilitate further development and application or
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:

     1.   Environmental Health Effects Research
     2.   Environmental Protection Technology
     3.   Ecological Research
     4.   Environmental Monitoring
     5.   Socioeconomic Environmental  Studies
This report has been  assigned to the  ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop ana
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new  or improved technology required for the control ana
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-600/2-77-057

                                   February 1977
    CONTROL CHARACTERISTICS

       OF  CARBON BEDS  FOR

    GASOLINE  VAPOR  EMISSIONS
                    by

    Michael J. Manos and Warren C. Kelly

     Scott Environmental Technology, Inc.
           2600 Cajon Boulevard
       San Bernardino, California  92411
           Contract No.  68-02-2140
              ROAP No. 21AXM
        Program Element No. 1AB604
     EPA Project Officer: Max Samfield

 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

-------
                              ACKNOWLEDGMENT

             Scott Environmental Technology, Inc., would like to thank
the following representatives and their companies for their generous
cooperation in providing activated carbon samples for evaluation in this

investigation into control characteristics of activated carbon beds for
hydrocarbon vapor emissions.
             The samples tested are identified by code letter only in the
test matrix in the Appendix and are not identified by manufacturer.

                             Mr. Edward G. Polito
                             Environmental Services Engineering
                             Carbon Department
                             Chemical Division of Westvaco
                             Covington, Virginia 24426
                             Mr. Blaine R. Joyce
                             Technical  Service Director
                             Activated Carbon Products
                             Union Carbide Corporation
                             11709 Madison Avenue
                             Cleveland, Ohio 44107
                             Mr. Bernard L. GrandJacques
                             Activated Carbon Division
                             Calgon Corporation
                             Post Office Box 1346
                             Pittsburgh, Pennsylvania 15230
                             Mr. John R. Conlisk
                             Product Development Department
                             ICI United States, Inc.
                             Wilmington, Delaware 19897
                                      11

-------
                           DEFINITION OF TERMS
Charge


Bleedthrough

Breakthrough


Hold


Strip
(Purge Strip)


Strip
(Vacuum Strip)

Cycle

Weathering


RVP

Total Charged
Weight

Heel
Working
Capacity

Acclimated
NDIR

HC

C C14

RH
Flow of gasoline vapor and air mixture upward
through carbon bed

Low concentration of unadsorbed vapor detected at
outlet of carbon bed during charge

Event when outlet vapor from carbon bed exceeds
preset limit terminating charge

Period of inactivity after a charging or stripping
mode

Flow or air down through the carbon bed to desorb
hydrocarbons


Desorption of hydrocarbons by reducing the pressure
in the carbon bed

Repetitive sequence of charge, hold, strip, hold

Gradual reduction in volatility of gasoline due
to loss of higher partial pressure components

Reid Vapor Pressure (a measure of fuel volatility)

Total weight of adsorbed vapor at breakthrough
Total weight of residual vapor not desorbed from
carbon at end of strip mode

Difference between total charged weight and heel
Condition of carbon achieved when the weight of the
carbon has stabilized while being purged with
constant moisture air

Non-dispersive infrared (analyzer for hydrocarbons)

Hydrocarbons

Carbon Tetrachloride

Relative humidity
                                      iii

-------
                       CONVERSIONS AND EQUIVALENTS
1 gallon
1 inch
1 pound/square inch
1 inch H20 (column)

1 inch ^0 (column)

1 pound
1 foot
1 cubic foot
1 grain H^O/pound
Temperature °F
     -100 °F
     - 10 °F
        0°F
       36 °F
       49 °F
       50 °F
       69 °F
       75 °F
       86 °F
       90 °F
      100 °F
      150 °F
      200 °F
3.785 liters
2.54 centimeters
0.07  kilograms/square centimeter
0.00254 kilograms/square
centimeter
0.073 millimeter of mercury
(column)
454 grams
0.305 meters
0.028 cubic meters
0.0293 grams H20/kilogram
°C
     -73°C
     -23°C
     -18°C
       2°C
       9°C
      10°C
      21°C
      24°C
      30°C
      32°C
      38°C
      66°C
      93°C
                                    IV

-------
                             Table  of Contents
                                                                   Page  No.

ACKNOWLEDGMENT 	    ii
DEFINITION OF TERMS	   ill
CONVERSIONS AND EQUIVALENTS	    iv
LIST OF FIGURES	    Vi
LIST OF TABLES	vii
1.0   INTRODUCTION	1-1
2.0   PROGRAM SUMMARY	2-1
      2.1  SUMMARY OF TEST RESULTS	2-1
      2.2  PROJECTION OF RESULTS TO A SERVICE STATION VAPOR
           CONTROL SYSTEM  	  2-3
3.0   EXPERIMENTAL PROCEDURES AND RESULTS  	  3-1
      3.1  REPEATABILITY-BASELINE TESTS  	  3-1
      3.2  EFFECT OF GASOLINE COMPOSITION  	  3-3
      3.3  CARBON MANUFACTURERS  	  3-7
      3.4  CANISTER DESIGN	3-18
      3.5  STRIP METHOD	3-21
      3.6  AMBIENT TEMPERATURE 	  3-32
      3.7  ONE THOUSAND CYCLE ENDURANCE TEST	3-35
      3.8  PRESATURATED CARBON 	  3-37
      3.9  SELECTIVE HYDROCARBON RETENTION 	  3-42
4.0   TEST EQUIPMENT	4-1
      4.1  TEST CHAMBER	4-1
      4.2  AUTOMATIC TEST FIXTURE	4-1

APPENDIX A   	A-l
APPENDIX B   	B-l
APPENDIX C   	C-l
APPENDIX D   	D-l

-------
                              List of  Figures
Number                                                        Page No.
   1      FUEL COMPOSITION EFFECTS ON WORKING CAPACITY. .  .       3-6
   2      WORKING CAPACITY OF DIFFERENT CARBONS 	       3-10
   3      ACTIVATED CARBON PARTICLE SIZE COMPARATOR ....       3-11
   4      CARBON MESH SIZE AND FORM VS. WORKING CAPACITY.  .       3-12
   5      CARBON SURFACE AREA VS. WORKING CAPACITY	       3-14
   6      CARBON DENSITY VS. WORKING CAPACITY 	       3-16
   7      CARBON C C14 ACTIVITY VS.  WORKING CAPACITY.  . .  .       3-17
   8      EFFECT OF CANISTER L/D RATIO ON WORKING
          CAPACITY	       3-19
   9      MASS TRANSFER ZONE CONCEPT	       3-20
  10      PURGE RATE EFFECT ON WORKING CAPACITY 	       3-23
  11      PURGE TEMPERATURE EFFECT ON WORKING CAPACITY. .  .       3-25
  12      RELATIVE HUMIDITY EFFECT ON WORKING CAPACITY. .  .       3-28
  13      STRIP VACUUM EFFECT ON WORKING CAPACITY 	       3-30
  14      VACUUM STRIP TEMPERATURE EFFECT ON WORKING
          CAPACITY	       3-34
  15      AMBIENT TEMPERATURE EFFECT ON WORKING CAPACITY.  .       3-36
  16      WORKING CAPACITY AFTER EXTENDED CYCLES	       3-39
  17      PRESATURATED CARBON EFFECT ON WORKING CAPACITY.  .       3-41
  18      FLOW SCHEMATIC FOR TEST APPARATUS	       4-2
                                    VI

-------
                              List of Tables

Number                                                         Page  No.
   1      REDUCTION IN ACTIVATED CARBON WORKING CAPACITY
          AT ADVERSE CONDITIONS 	       2-7
   2      INCREASE IN ACTIVATED CARBON WORKING CAPACITY
          WITH SYSTEM DESIGN AND CARBON SELECTION 	       2-8
   3      TYPICAL HYDROCARBON CONCENTRATION PROFILE  OF
          STRIPPED GAS DURING AIR STRIPPING 	       2-10
   4      COST EFFECTIVENESS OF VAPOR CONTROL SYSTEMS WITH
          ACTIVATED CARBON	       2-12
   5      CARBON WORKING CAPACITY—BASELINE TESTS	       3-2
   6      EVIDENCE OF ADSORBED LEAD	       3-4
   7      FUEL COMPOSITION EFFECTS ON WORKING CAPACITY. .  .       3-5
   8      CARBON PROPERTIES VS. WORKING CAPACITY	       3-8
   9      CANISTER CONFIGURATION EFFECTS ON WORKING
          CAPACITY	       3-18
  10      PURGE RATE EFFECT ON WORKING CAPACITY 	       3-22
  11      PURGE TEMPERATURE EFFECT ON WORKING CAPACITY. .  .       3-24
  12      RELATIVE HUMIDITY EFFECT ON WORKING CAPACITY. .  .       3-27
  13      VACUUM STRIP EFFECT ON WORKING CAPACITY 	       3-29
  14      COMPARISON OF BLEEDTHROUGH LEVELS  	       3-31
  15      HEATED VACUUM STRIPPING VS. WORKING CAPACITY. .  .       3-33
  16      AMBIENT TEMPERATURE EFFECT ON WORKING CAPACITY.  .       3-35
  17      WORKING CAPACITY  AFTER EXTENDED CYCLES 	       3-38
  18      SATURATED CARBON EFFECT ON WORKING CAPACITY  . .  .       3-40
  19      GAS CHROMATOGRAPH ANALYSES OF VAPORS	       3-43
  20      ANALYSIS OF HYDROCARBON HEEL FROM  1000-CYCLE
          TEST	       3-45
                                   vi i

-------
                                     1-1
                           1.0 INTRODUCTION
          Recently proposed regulations have specified emission controls
which would reduce hydrocarbons entering the atmosphere from the marketing
of gasoline.  These emissions occur primarily from two modes of operation:
bulk delivery of fuel to service station underground tanks and refueling
of motor vehicles.  Early  control  system designs operated on a pressure
 balance  principle which  simply meant  that  vapors were transferred back to the
container being emptied, from the container being filled, through a separate
vapor hose.  The driving force for this vapor transfer was the small
pressure difference between the containers.  Early tests showed this
approach worked well for the bulk delivery mode, however, during vehicle
refueling vapors were not contained unless an adequate seal  was maintained
between the vapor recovery dispensing nozzles and vehicle fill necks.  A
wide variety of vehicle  fill neck designs made the design of a sealing-
type vapor recovery nozzle difficult.
          Alternately, systems were designed with vapor blowers which
generated a reduced pressure in the vapor recovery dispensing nozzle to
suck vapors in when poor seals occurred.   Excess air  ingested  into  the
vapor recovery system  stimulated additional vaporization of the liquid
fuel resulting in vapor  growth situations.  Systems were designed using
activated carbon beds to capture and temporarily store these gasoline
vapors with ultimate disposition of the vapors effected by incineration,
condensation, and absorption techniques.  Periodic regeneration of the
carbon was deemed necessary to maintain a level of "working capacity"
available for continuous use.
          This program considered the practical working capacity of
activated carbon to cyclically adsorb gasoline vapor which would otherwise
be lost to the atmosphere such as during gasoline transfer operations at
a service station.  Quantitative measurements were made in the laboratory
which were extrapolated  to represent typical operation of a carbon control
system at a service station pumping 50,000 gallons of gasoline per month.
*English-to-metric conversions,  2nd equivalents  presented  on  page iv.

-------
                                    1-2
          Tests were performed on carbon from four manufacturers.
Selection of the carbon materials used in this program was based on
published performance specifications, commercial availability, bulk cost
and suppliers'  experience in using their products with gasoline vapor
emissions.  Previously vapor saturated carbon and different lots of the
same carbon were tested.  Three different granular mesh sizes were tested.
The carbon was subjected to vapor from gasoline with and without lead and
at two volatility levels.  Tests were conducted at three ambient temperatures
and three humidity levels.  Two carbon bed lengths and two carbon bed cross-
sectional areas were tested.
          Vapor was desorbed at three air flow rates, three temperatures,
two vacuum pressures and three vacuum temperatures.  The ability of carbon
to effectively adsorb vapor after 1000 cycles was determined.  A complete
specification of the test variables is presented in the Test Matrix in
Appendix B.

-------
                                    2-1
                           2.0  PROGRAM SUMMARY

2.1  SUMMARY OF TEST RESULTS
          Eight types of activated carbon from four manufacturers were
evaluated to determine working capacity, basically defined as the amount
of gasoline vapor which could be cyclically adsorbed per 100 grams of virgin
activated carbon.  Tests were conducted at various levels of fuel volatility,
lead content, carbon bed shape, ambient temperature/humidity, purge air flow
rate/temperature and vacuum stripping pressure/temperature.
          A "baseline" set of conditions was established against which the
effects of variables were compared.  The "baseline" tests used a coal derived
12 x 28 mesh activated carbon material designated W-l in a container which was
approximately 13.3 cm (5.2 inches) tall by 6.5 cm (2.6 inches) in diameter.
Gasoline vapors were generated by bubbling air at 75° F. and 12% relative
humidity through a leaded fuel having a volatility rating of 8.8 psi by the
Reid Vapor Pressure (RVP) method.  The vapors flowed upward through the carbon
bed at 2000 cc/min.  Charging of the carbon continued until a breakthrough
level of 0.33% hydrocarbons (as propane) was observed in the effluent at
which time charging ceased and the canister remained inactive for one minute
prior to purging.
          Purging was accomplished via countercurrent (downward) flow of
75° F. 12% relative humidity air flowing at 4000 cc/min.  When the effluent
dropped to a concentration of 2.50% hydrocarbons (as propane), purging ceased
and the carbon bed stood inactive for one minute prior to the next charge
mode.
          The test results indicated the .following:
          a) For eight  "baseline" tests the working capacity averaged
             6.03 grams of gasoline vapor per 100 grams of virgin
             activated carbon.
          b) The working capacity was slightly lower on non-leaded fuel.
          c) A high volatility fuel resulted in noticeably greater
             working capacity.

-------
                          2-2
d) A carbon bed container twice as tall (same cross-
   sectional area) as the "baseline" case resulted in a
   slightly higher working capacity, and a container
   of twice the cross-sectional area (same height) had
   slightly lower capacity.

e) Air purge flow rates of five and ten times the "base-
   line" rate yielded slightly lower working capacities,
   however, purge times were approximately l/5th and l/10th
   the "baseline" time, respectively.

f) Heated purge air tests indicated no significant change in
   working capacity.

g) Higher relative humidities resulted in decreased working
   capacity.
h) Activated carbon exposed to extended periods of vapor
   saturation did  not appear to be rendered inactive
   although the working capacity was decreased.

i) Working capacity decreased at lower ambient temperatures
   and increased at higher ambient temperatures for the same
   volatility fuel.
j) Vacuum stripping at 25 mm Hg absolute pressure resulted in
   lower working capacities than "baseline", however, preheating
   the carbon bed to 200° F. before stripping yielded higher
   capacities.   Vacuum stripping at 100 mm Hg absolute pressure
   was ineffective.

k) Two samples  of pelletized carbon having a larger particle
   size (lower mesh numbers) had slightly lower working
   capacity.
1) Working capacity appeared to vary directly with activated
   carbon specifications such as surface area (Nitrogen BET
   Method) and C Cl^ activity and inversely with apparent
   density.

m) Different production batches of the same product were found
   to vary noticeably in performance.

n) The hydrocarbon residual  or heel remaining on the activated
   carbon material after normal stripping techniques is comprised
   mainly of Cg - Cg compounds.

o) After 1000 operating cycles on one test carbon, working
   capacity was approximately 70 - 75% the capacity at the 20th
   cycle.

-------
                                   2-3
2,2    PROJECTION OF RESULTS TO A SERVICE STATION VAPOR CONTROL SYSTEM
            The data obtained during this program can be very useful  for
determining the applicability of activated carbon adsorption systems  for
control of gasoline vapor emissions.  Existing service station control
systems using activated carbon all  have capability for on-site regeneration
although one of the earlier proponents of activated carbon envisioned periodic
replacement of carbon beds and a centralized regeneration facility.   This
section of the report will briefly discuss the nature of the service  station
gasoline vapor emission problem and types of existing control systems and
presents sample calculations of how to determine activated carbon require-
ments.  For the following discussions, a typical service station is
considered to be one pumping 50,000 gallons/month.
     2.2.1 Service Station Vapor Emission Levels
            Gasoline vapor emissions at service stations generally result
from three modes of operation:
            (a) Displacement of HC vapors from the underground storage
                tanks during bulk delivery
            {b) Displacement of HC vapors from vehicle fuel tanks
                during refueling operations
            (c) Accidental spillage of gasoline during refueling
                of vehicles
Additionally, gasoline vapors can be emitted in the form of  "breathing
losses" from storage tanks due to temperature and barometric pressure
changes, regardless of whether or not there is any service station
activity.
            Early estimates and measurements of the quantity of vapors
attributable to the various modes of activity vary considerably.  The
majority of the emission problem results from (a) and (b) both of which
have been found to vary with gasoline temperatures, volatilities, flow
rates, etc.  However, estimates of between three and six grams vapor per
gallon of fuel transferred cover most conditions.  Estimates of spillage

-------
                                   2-4
losses and breathing losses are both on the order of 0.5 grams per
gallon of throughput and obviously would be subject to  large potential
variations depending on conditions which influence their magnitude.
            For the purpose of this discussion, potential emissions for
categories (a) and (b) will be assumed to average 4.5 grams/gallon and
potential breathing losses 0.5 grams/gallon.  Spill losses may be
influenced by attendant handling practices and hence may be beyond the
control capabilities of any control system; however, proposed regulations
in this area have been written to limit the number of spills per 100
refuel ings.
            A 50,000 gallon/month service station being supplied by an
8,500 gallon tank truck would receive deliveries at least every four to
five days.  During bulk delivery, approximately 8,500 gallons of vapor
are displaced from the underground tanks.  With most existing vapor
control systems at least 90 percent of these vapors are displaced back
to the emptied tank truck compartments leaving 850 gallons of vapor (at
4.5 grams/gallon) or 3825 grams of vapor to be controlled.  If the
compartments are unloaded at 400 gallons per minute and two products are
unloaded simultaneously, then the entire unloading operation would take
place in 10.6 minutes.   The rate of loading on the control system would be
360 grams of vapor/minute.
            To estimate typical vapor loading of the adsorbers during vehicle
refueling many assumptions must be made.  If one assumes that all of the
vapor displaced from the vehicles refueled is collected by the system and
that excess air is ingested at a rate equal to 20% (conservative), then
for every ten gallon fill ten gallons of vapor and two gallons of air will
be drawn into the system and ten gallons of liquid fuel leave the system.
The two excess gallons of air can create additional vaporization of liquid
in the underground tank so that about three gallons of excess vapor-air
mixture may result from each ten gallon refueling.  Again, it is emphasized
that numerous factors such as temperature, fuel volatility, collection blower

-------
                                   2-5
setting, piping configuration and dispensing nozzle design  affect  this  ratio,
however, three gallons of excess vapor-air mixture per ten  gallons dispensed
is believed to be a reasonable estimate.   Therefore, a potential  loading  of
the adsorber equal  to 13.5 grams HC per ten gallon fill  is  assumed and  2125
gallons/day (8500 gallons/four days) is dispensed at the typical  service
station.  Loading of the adsorber would be 2869 grams/day (13.5/10 X 2125).
            Based on the above assumptions, it would appear that  the adsorber
needs to be sized for the bulk delivery load of 3825 grams  and that periodic
regeneration modes after each refueling operation would insure that maximum
adsorber capacity is available any time a bulk delivery arrives.   In actual
practice, most operating systems do regenerate after each refueling operation.
     2.2.2 Sizing of Activated Carbon Adsorbers
            Among the numerous factors which affect the design of  a service
station vapor control system are:
            •Emission control capability
            •Costs:  capital, operating and maintenance
            •Safety
            •Reliability
Assuming that there are minimum acceptable standards for safety and reliability
of a control system, there still exists a large potential latitude of design
associated with controlling "X" amount of hydrocarbon vapors for  "Y" dollars.
The following discussion is not intended to cover the subject of cost effective-
ness or cost tradeoffs for various system designs.
            The previous subsection of this report presented typical values
for tank truck capacity, gasoline vapor density, displacement efficiency
and gasoline delivery rate to estimate the typical loading conditions that
an activated carbon adsorption system might encounter.  Variations in
actual  loading due to variations in operating condition is inevitable,
however, data is not available to predict the emission rate for various
combinations of variables.  Similarly, the subject program attempted to
identify the working capacity of various activated carbon materials in
conjunction with fuel volatility, lead content, strip technique/temperature,
canisters shapes,  etc.  Because of the limited number of actual tests

-------
                                   2-6
performed, performance data often must be projected from one  set of
conditions to another.  Readers of this report are hence cautioned that
projected results should be substantiated with additional  testing prior
to making serious commitments.
            To determine the amount of activated carbon required to
adsorb a known weight of gasoline vapors, one would apply  the general
formula:
            WTCARB = 100 WTVAPR/WORKCAP
            WTCARB = Weight of activated carbon required,  grams
            WTVAPR = Weight of gasoline vapor to be adsorbed, grams
           WORKCAP = Working capacity of activated carbon, grams vapor/
                     100 grams activated carbon
The working capacity of the activated carbon can vary due  to  conditions
beyond the control of the system designer such as ambient  conditions,
fuel composition, manufacturing variations.  Table 1 shows the working
capacities for a "baseline" set of conditions and provides an indication
of the relative change in working capacity with changes in the uncontrolled
variables.   Since worst case conditions are bound to occur, the  designer
must size his adsorber accordingly.   As indicated,  adverse conditions
could theoretically reduce the working capacity from 6.03 to 3.31 grams
vapor per 100 grams activated carbon.   Mathematically combining  the  percent
reductions  in working capacity for the three adverse cases may provide a
distorted picture of the worst combination  due to synergistic effects or
cancellation of effects.  For example, since the  relative humidity is a
function of ambient temperature,  80% relative humidity at 75° F.  would
contain about 2% times  as  much water as 80% relative humidity at 50° F.
In this regard, it would be advisable  to investigate more completely the
activated carbon characteristics  in  low temperature, high humidity environ-
ments rather than to evaluate the variables separately.

-------
                                    2-7
                                  Table 1
                       REDUCTION  IN  ACTIVATED  CARBON
                  WORKING CAPACITY AT  ADVERSE  CONDITIONS
                                                       Standardized
                                                       Working  Capacity  in
                                                       Grams  Vapor  per  100
 	Condition	                      Grams  Activated  Carbon
 Baseline conditions  on W-l  carbon;  i. e.,
 75° F.  ambient temperature, 12%  relative
 humidity, 9  psi  Reid Vapor  Pressure
 leaded  fuel                                                  6.03
 (a) At  50° F.  (% reduction  from  baseline)                    5.67  (  6.050
 (b) At  80% relative  humidity (%  reduction  from              4.68  (22.4%)
     baseline)
 (c) With unleaded fuel (% reduction from baseline)           5.15  (14.6%)
 Net effect of  all reductions from baseline                  3,76  (37.7%)
             A number of factors investigated during this program such as
stripping method, stripping rate, stripping temperature and container
shape give a designer certain flexibility to optimize the working capacity
of the activated carbon.  Table  2  shows how these factors could
theoretically be exploited.  Again, we end up at a combination of
conditions that were not directly evaluated during this program and
which should be examined in more depth to confirm the assumption.
Referring back to general equation WTCARB = 100 WTVAPR/WORKCAP, a 3825
gram load would require 42,833 grams of activated carbon Y-l (94 Ibs.)
for control.  At 0.274 grams/cc density this would equal 156,325 cc (5.52
ft.3) of material required to handle the bulk delivery.
             The aforementioned projections do not take into account
either degradation in working capacity with time or designing in a safety
factor (or overcapacity) to accommodate atypical conditions.

-------
                                    2-8
                                 Table 2
                   INCREASE IN ACTIVATED CARBON WORKING
                      CAPACITY WITH SYSTEM DESIGN AND
                             CARBON SELECTION
                                                    Standardized
                                                    Working Capacity in
                                                    Grams Vapor per 100
 	Condition	                              Grams Activated Carbon
 Baseline condition on W-l activated carbon;
 i. e., 75° air strip, strip rate 10 bed
 volumes per minute, canister shape has height/
 diameter of 2:1                                            6.03
 (a) With 150° air strip (% increase)                       6.40 (6.1%)
 (b) With 200° vacuum strip (% increase)                    6.78 (12.4%)
 (c) With canister shape h/d of 4:1 (% increase)            6.77 (12.3%)
 (d) With Y-l activated carbon (% increase)                11.35 (88.2%)
 Net effect of increases from (b), (c) and (d)             14.32 (137.6%)
 Net effect of factors (a), (b) and (c) from Table A
 with (b), (c) and (d) from this Table                      8.93 (48.0)

             Both  the  air strip and vacuum  strip techniques appear  to have
 the capability  for  further  increasing  the  working capacity.  The relatively
 unimpressive performance on the  hot air strips resulted  from not allowing
 the carbon to cool  off prior  to  successive charges.   The choice of which
 particular strip  technique  is  to  be used essentially  determines the means
 of recovering or  disposing  of the  hydrocarbons.
            One of the most important considerations in system design is
obviously cost.   Initial cost estimates from the various suppliers indicated
activated carbon costs of $0.54 - $0.88 per pound.  Being relatively
inexpensive, it is apparent that increasing the size of the adsorber 50%,
for example, from 94 pounds carbon to 141  pounds carbon may be significantly
less costly than incorporating certain hardware items which improve working
capacity.

-------
                                   2-9
    2.2.3 Disposition of Hydrocarbons
            Current in-use systems which utilize air stripping either
incinerate the hydrocarbon/air mixture  in a direct flame burner or in a
catalytic oxidizer.  Table   3  shows the percent hydrocarbons in the
stripped gas as a function of time during a typical 16% minute strip mode
(from test 3, cell #1).  The theoretical proper mixture for combustion of
these vapors is estimated to be 2.7% with a flammable range estimated to be
from 7.8% to 1.6%.  As indicated, a large amount of dilution air is needed
for the first few minutes.  By the tenth minute no more dilution air is
theoretically required (although it is  probably desirable).  By the 16th
minute, the concentration is near the lower flammability limit with no
dilution air.  A catalytic oxidizer would have the capability to continue
oxidizing the stripped gas beyond the 16% minute period.
            The average concentration for the strip mode on Table  3  was
about 7.5% gasoline vapors.  With the concentrations characteristic of
this stripped gas, condensation by refrigeration would require very low
temperature possibly in the area of -100° F.
            Stripping via a vacuum pump has the advantage of not diluting
the stripped gas with dilution air.   Although not measured during the course
of this program, the HC concentration would be expected to be essentially
100 percent.   The absence of air would make it possible to recover a sizable
portion of these hydrocarbons by absorption in the underground tank (sparging)
or condensation (either compression-refrigeration or straight refrigeration).

-------
                    2-10
                   Table 3

  TYPICAL HYDROCARBON CONCENTRATION PROFILE
    OF STRIPPED GAS DURING AIR STRIPPING

                             Average Volume
  Time                      % Gasoline Vapor
 0-1 min.                      35.3 est.
 1 - 2 min.                      19.2
 2-3 min.                      12.7
 3 - 4 min.                       9.3
 4 - 5 min.                       7.1
 5 - 6 min.                       5.7
 6-7 min.                       4.7
 7-8 min.                       4.0
 8-9 min.                       3.4
 9-10 min.                      2.7
10 - 11 min.                      2.4
11 - 12 min.                      2.3
12 - 13 min.                      2.1
13 - 14 min.                      2.0
14 - 15 min.                      1.9
15 - 16 min.                      1.8
16 - 16*2 min.                      1.7

-------
                                   2-11
     2.2.4 Service Station Vapor Control System Costs
             As of the writing of this report, three types of control
systems using activated carbon were in use at service stations.  One system
employs activated carbon to temporarily store captured vapors which are
subsequently air stripped and burned by direct incineration.  A second type
differs primarily in that a catalytic oxidizer is used to dispose of the
vapors.  The third type incorporates a refrigerated condenser to liquefy part
of the excess vapors, the noncondensable portion being captured by the
activated carbon.  Vacuum stripping periodically removes the adsorbed vapors
at which time they are injected in the chilled condensate for return to the
underground tank.
             Table 4 presents data on relative cost factors for the various
types of systems.  Although there is disparity between annualized capital
costs, operating costs and estimated control efficiency, cost effectiveness
figures of approximately $0.20 per pound of HC controlled are indicated.

-------
                                                    Table 4

                      COST EFFECTIVENESS OF VAPOR CONTROL SYSTEMS WITH ACTIVATED CARBON
System Type
Carbon Adsorption
w/ Direct Incineration
Carbon Adsorption
w/ Catalytic Oxidation
Carbon Adsorption
w/ Refrig. Condensation
Annual i zed
Cost, $/Yr.
$2,795
$2,991
$3,304
Refueling
HC Controlled,
Ibs/yr.
7,569
7,569
7,569
                                                                   Bulk Delivery       Control
                                                                   HC  Controlled    Effectiveness
                                                                                .        $/1b.
Ibs/yr.



7,137***



7,137***



7,137***
                                                                                       0.190



                                                                                       0.203



                                                                                       0.224
ro

i—»
ro
Refrig.
Condensation
$1,252
3,761
2,554
0.198
   information extracted  from  "Cost Analysis—Service Station  Emission Controls  (Vehicle Refueling)",
     £nh«;U 19^6' Conoco' Process Engineering Department,  Ponca City, Oklahoma.  Data is applicable to
   a 60,000 gal/mo, station and assumes 90% control of vehicle refueling emissions  (@ 5 qms/qal) and
   100% control of diurnal breathing losses.


 **Information extracted  from APCA Paper #75-54.1 "Cost Effectiveness of Gasoline Vapor Recovery Systems "
   Presented at 68th Annual Meeting June 15 - 20, 1975/  Data is applicable to a 40,000 gal/mo, station
   and assumes 96.3% control of all service station emissions.  Paper does not state whether or not
   activated carbon is used.


***Assumed to  be 90% control based on 5 gms vapor/gallon of fuel  delivered.

-------
                                    3-1
                   3.0   EXPERIMENTAL  PROCEDURES  AND RESULTS
          In this section the individual  test results and the  procedures
followed will be discussed in detail.   Full  and complete evaluation  of  the
parameters of interest for the four basic types of carbon would  have required
many thousands of individual  tests.  The  data obtained during  this  program
represent an initial  step toward evaluating   activated carbon  performance
characters!tics, however, additional effort  is indicated to quantify carbon
working for certain combinations of adverse  or optimum conditions.
          A tabulation of the test results is shown in Appendix  A,  and  the
matrix of test conditions for each individual test is contained  in  Appendix  B.

3.1  REPEATABILITY-BASELINE TESTS
          The working capacity of one carbon was  measured eight  times under
identical conditions.  One hundred sixty grams of type W-l carbon was charged
with leaded nine pound RVP (nominal) gasoline vapor at 2000 cc/min. and stripped
at 4000 cc/min. with 75°F air.  The 12 x 28 mesh  carbon was tested in a 65 mm
diameter, 133 mm tall canister.  The pressure in  the canister was held at
atmospheric during both charge and strip.  These same conditions were
established on all four canister weighing test stations.  Only the actual
gasoline volatility was subject to minor but unavoidable differences due to
weathering.  The concentration of  gasoline vapor was measured for each test.
Moisture content in the charge-strip air was held constant by saturation at
36°F in a moisture trap.  The activated carbon was  acclimated to this same
moisture level prior to initiating each test.
          Test results are tabulated in Table 5.   The average value for working
capacity was 6.03 grams of gasoline vapor per 100 grams of activated carbon
with an average  gasoline  vapor concentration of  53.6  volume percent  (measured
as propane).

-------
                   3-2
                 Table 5
CARBON WORKING CAPACITY - BASELINE TESTS

                        Observed
Carbon
Type
W-l
W-l
W-l
W-l
W-l
W-l
W-l
W-l

Test
No.
1-1
1-2
2-1
2-6
3-1
4-1
4-2
4-14
Mean Baseline
Date
1976
8/20
9/17
8/20
9/2
8/20
8/20
8/23
9/13

Standard Deviation
Working Capacity
gm/100 gm
5.88
5.78
5.89
6.06
5.91
6.17
6.17
6.44
6.03
0.22
Vapor Concentration,
Volume % as Co
52
52
55
50
56
53
54
59
53.6


-------
                                    3-3
3.2  EFFECT OF GASOLINE COMPOSITION
     2.3.1 Lead Content
             Prior to performing tests of activated carbon working capacity*
an experiment was performed to quantify the effect of lead compounds (from
gasoline anti-knock additives) in the gasoline vapor.   Two canisters containing
approximately 160 grams of activated carbon were subjected to  80  cycles  of
operation (approximately 3.84 m  of gas).   Vapors flowing into one canister
originated from a commercial  premium fuel  having a lead content of 2.79
grams/gallon, and vapor to another canister was generated from a  commercial
unleaded fuel with a lead content of 0.017 grams/gallon.  A third canister was
supplied an equivalent amount of vapor free air.
          The amount of lead on the activated carbon, determined  by spectro-
graphic analyses, indicated that minute quantities of lead compounds present
in the vapors were depositing on the carbon as shown in Table  6.   Accumulation
of lead at the maximum rate of 16.5 y grams/gram for 80 cycles of operation
indicates that it would take about 1500 cycles to accumulate enough lead to
have a measurable influence on the working capacity weight measurements.
     3.2.2  Leaded vs. Unleaded Fuel
          As mentioned previously, eight replications of the "baseline"  test
conditions were conducted using leaded fuel and carbon designated W-l.
Similarly, the same conditions were used twice in evaluation of carbon X-l.
Tests performed on each of these two carbons using unleaded fuel  resulted
in decreases in working capacity from 9 to 14% as shown in Table 7.  One
possible explanation for this effect is that high octane compounds such as
aromatics (benzene, toluene, xylene) are present in higher percentages
in the unleaded fuel.  The ring-like structure of these aromatics may require
higher energy levels to remove from the carbon micropores, hence less are
desorbed under normal stripping techniques.

-------
                                    3-4
                                   Table  6
                          EVIDENCE  OF  ADSORBED  LEAD
Carbon
Type
Z-l
Y-l
Y-l
Total
Adsorbed Lead
y gm/gm
3.6 ± 0.1
16.5 ± 0.1
0.88 ± 0.05
Gasoline
Lead
GM/Gal
0.017
2.79
(Vapor
.Specifications
Nominal
RVP Ib.
9
9
Free Air Only)
     3.2.3  Effect of Volatility
          A high volatility unleaded fuel  was obtained from a specialty fuel
refiner for comparison with the 9 psi RVP fuels.  These data, shown in
Table 7, indicate working capacities of 27% and 29% greater with the 14 psi
unleaded fuel  over the 9 psi RVP unleaded fuel for the W-l carbon and the
X-l carbons, respectively.  Figure 1 presents the results graphically.
Butanes are usually added to gasoline to Increase volatility and are a major
constituent of the vapor-air mixture.  The activated carbon appears to easily
adsorb and desorb these relatively small molecules resulting in higher working
capacities for the higher volatility  fuel.
          With  changes  in  working  capacity due to  fuel  volatility  being
more pronounced than had been anticipated, it was desirable to "correct"
test results to a common volatility base.   To accomplish this correction, gas-
oline vapor concentration was used as the indicator of fuel volatility since
actual RVP tests were performed only on the fresh, as-received fuel.

-------
                                    3-5
                                  Table 7
               FUEL COMPOSITION EFFECTS ON WORKING CAPACITY
Carbon Test Observed Working
Type No. Capacity gm/100 gm
W-l
W-l
W-l
Baseline*
4-10
4-5
6.03
5.19
6.69
Vapor Concen-
tration, Vol %
53.6
55
98
Reid Vapor
Pressure, psi
8.8
8.7
13.8
Lead Content
gm/gal
2.79
0.017
0.003
         % Change in
         Working Capacity =  (6.03  - 5.19)76.03 = 13.9% decrease for unleaded
         % Change in
         Working Capacity =  (6.69  - 5.19)/5.19 = 28.9% increase for high RVP

  X-l   1-5, 4-11         5.52  avg.          52.5 avg.        8.8            2.79
  X-l     3-10           5.04               56               8.7            0.017
  X-l     4-7            6.38               94              13.8            0.003
         % Change in
         Working Capacity =  (5.52  - 5.04)/5.52 = 8.7% decrease  for unleaded
         % Change in
         Working Capacity =  (6.38  - 5.04)75.04 = 26.6%  increase for high RVP
*Data  shown  is  average  for  eight  baseline  tests

-------
                               3-6
s.
«
u
°  C
o  b
L.
o
Q.
(J
10
a.
ID
5  4
o
                 Leaded

                 Gasolin
                Unleaded
                Gasoline
                    9 Ib. RVP
                                                 Ib.  RVP
                               A W-l Carbon


                               Q X-l Carbon
     0
20          40          60          80


 Gasoline Vapor Concentration,

 (expressed as % propane)
                                                                 100
                            Figure 1

          FUEL  COMPOSITION  EFFECTS ON WORKING CAPACITY

-------
                                    3-7
          The slope of the volatility trend line for W-l  carbon  in  Figure  1
was used to correct the data to 53.651 when 9 RVP fuel  was used and  to  98%
vapor concentration when 14 RVP fuel  was used.   The slope of this line is:
For nominal 9 pound RVP tests, the data are corrected to 53.6%, therefore;

                Where:
WCC = WCQ + 0.0338 (53.6 - % concQ)
                                WC  = corrected working capacity
                                  \*
                                WC  = observed working capacity
                            % cone  = observed vapor concentration
For nominal 14 pound RVP tests, the data are corrected to 98.0%, therefore:
                                WCc = WCQ + 0.0338 (98.0 - % concQ)
          Since the above correction equations are based on just two levels of
volatility, the linear approach was the only feasible method, however, a
curvilinear relationship probably exists.  The equations were used during the
data analysis to minimize variability in the program test results and should
not be applied to any situations or conditions other than the subject test data.
All further references to working capacity test results in this report should
be interpreted as corrected working capacity unless otherwise stated.

3.3  CARBON MANUFACTURERS
          Working capacity was directly compared between activated carbons from
four different manufacturers.  The type, form and mesh size were nearly the
same for the four samples.  Moisture content in the samples was not the same.
Before testing each sample was acclimated to be in equilibrium with the moisture
in the charge and strip air.   Important properties of each carbon are summarized
in Table 8.

-------
                                     3-8
                                   Table 8
                   CARBON PROPERTIES VS. WORKING CAPACITY
                                Carbon
Test
No.
(8 Basel
1-11
1-11
1-12
1-10
1-9
1-5
4-11
1-3
1-3
1-4
Date
1976
ine tests)
9/7
9/15
8/26
9/16
9/13
8/19
8/19
9/2
9/17
8/27
Fype*
W-l
W-2
W-2
W-3
W-5
W-7
X-l
X-l
Y-l
Y-l
Z-l
Mesh
12x28
8x10
8x10
4x6
12x28
12x28
8x30
8x30
14x35
14x35
12x30
Surface Density Corrected
Area as Tested Activity Working Capacity
rr/gm gm/cc % C CU gm/100 am
1000
1000
1000
1000

1100
960
960
1200
1200
625
0.380
0.412
0.412
0.388
0.338
0.447
0.452
0.452
0.274
0.274
0.411
58.7
61.3
61.3
68.9

66.4
63.8
63.8
80**
80**
42.1
6.03
4.64
4.99
5.26
7.12
4.52
5.49
5.62
11.37
11.36
4.22
i^ 	
 *See Appendix D for  additional  information on material
**Reported 60-100 % C C14

-------
                                    3-9
          The same mass of 160 gm of each acclimated carbon  was  placed  in  canisters
and tamped to approximately the same compaction.   The working  capacity  in  grams
of hydrocarbon adsorbed per 100 grams of acclimated virgin  activated  carbon  are
presented in Figure 2.  Sample Y had nearly twice the capacity of Sample W and
nearly three times the capacity of Sample Z.  Sample X had  slightly less capacity
then Sample W.  Complete specifications for the carbon samples tested are
presented in Appendix D.  Only Sample Visa type made from wood. The  other
three are coal based.  All four samples were granular in form.  The estimated
costs varied from $54 to $88 per 100 pounds depending on the quantity purchased.
Sample W was received in a composition board drum with no apparent moisture
barrier.  Sample X was received in a steel pail sealed with a metal lid with
no 0-ring.  Sample Y was received in a sealed metal can.  Sample Z was  received
in a sealed collapsible plastic container.
     3.3.1  Carbon Production Lots
          Carbons from the same manufacturer with the same published  specifica-
tions but from different lots were tested (designated W-l and W-5).   The second
lot, W-5, had 18% greater capacity than the first.   When actually measured,  the
acclimated densities of the first and second lots were found to  be 0.38 and
0.34 respectively.
     3.3.2  Carbon Particle Size
          Activated carbons with differing particle size from one manufacturer
were tested.  Other properties were nearly the same.  The two larger  particle
(lower mesh) carbons were in pellet form while the smallest had a granular
form.  All are coal based.  The carbon with the smallest particle (highest
mesh size) had the highest working capacity.  The least capacity was  exhibited
by  the middle mesh size while the largest particle  (lowest mesh) carbon had
a capacity between the two.  Activated carbon is screened to establish a
desired  particle  size distribution.  A comparator scale shown in Figure 3  indicates
relative  particle sizes.  Working capacity measurements for carbons with
different particle size are presented  in Table 8 and shown in Figure 4.

-------
                            3-10
 >> 10
•M O
•^
 O E
 ID O)
 O.
 * O
O O
  t—t
 cn-^.
 c s.
•i- O
-* Q.
 S- (0
O >
     12
     10
      8
      0
2nd  lot  (W-5)

baseline
               W-l

              Coal
        X           Y

      Coal         Wood

      Activated Carbons
  Z   Mfg.

Coal  Base
                         Figure 2

          WORKING CAPACITY OF DIFFERENT CARBONS

-------
                    3-11
         STANDARD MESH   OPENING   PARTICLE
          Tytor   U.S.   mm   inches
            4    4    4.70  0.185
            6    6    3.33   .131
            8    8    2.36   .094
           10   12    1.65   .065
12
14
16
20
24
28
32
35
42
48
60
80
100
ISO
200
250
325
400
14
16
18
20
25
30
35
40
45
50
60
80
100
140
200
230
325
400
1.40
1.17
0.991
.833
.701
.589
.495
.417
.351
.295
.246
.175
.147
.104
.074
.061
.043
.038
.056 •
.047 •
.039 •
.033 •
.028
.023
.020
.016
.014
.012
.0097
.0069
.0058
.0041
.0029
.0024
.0017
.0015
                   Figure  3
ACTIVATED CARBON  PARTICLE  SIZE  COMPARATOR

-------
                         3-12
re
o
01
o
o
s-
o
CL
to
en
en
c
    0
                    2nd  lot  (W-5)

                    baseline
             4x6
             Coal
             Pellet
              W-3
8 x 10
Coal
Pellet
 W-2
12 x 28
 Coal
Granular
  W-l
12 x 28
Coconut
Granular
  W-7
Mesh No,
Base
Form
                       Figure 4
    CARBON MESH  SIZE AND FORM VS. WORKING CAPACITY

-------
                                   3-13
     3.3.3 Surface Area
             Activated carbon granules contain  an  internal  micropore  structure
which exhibits an enormous surface area.   The published  surface  areas  vary
widely between the different carbons.  Working  capacity  as  a function of surface
area is listed in Table 8 and plotted in Figure 5.  Samples W,  X and  Z are
coal based carbons.  Sample Y is a wood based  carbon.   The  highest working
capacity was obtained with the highest surface  area, Sample Y.
     3.3.4  Apparent Density
          Activated carbon has an affinity for both hydrocarbon vapor and water
vapor.  Hydrocarbon vapor can be driven off during original activation and
later during reactivation.  Water vapor, however,  may be adsorbed during {hot
steam) activation.  Moisture may also be adsorbed by activated carbon whenever
it  is exposed to the atmosphere and until it reaches equilibrium with the
moisture level in the surrounding air.
          The apparent bulk density of the carbons, as they were received for
this program, varied from 0.31 to 0.45 gm/cc; the moisture level varied from
0.2 to 14.8% by weight (see Appendix D).  Preliminary testing showed that
widely varying moisture levels in the carbon could affect  the gravimetric
canister weight measurements.  Dry carbon adsorbed both gasoline vapor and
moisture during charge but desorbed only hydrocarbons during strip since the
strip air had higher moisture than the carbon.  Conversely wet carbon desorbed
moisture during both charge and strip because both the charge carrier air and
the strip air were drier  than the carbon.
          To minimize these effects,  it was decided that the moisture level in
both charge and strip air should  be  held constant.  It was further decided
that before test,  the moisture in each carbon sample should be brought  to
equilibrium with  the moisture in  the  charge and strip air  so that  the net
weight change in  the canister would  be due only to the gain and loss of
gasoline vapor.

-------
                            3-14
     12
 ro

 U
 01


O

O
 o
 Q.

 5
£
•r-
o
to
o.

J_
o
6
     4  —
                  600          800        1000        1200



              Indicated Carbon Surface Area. m2/gm
                        Figure  5


        CARBON SURFACE AREA  VS. WORKING CAPACITY

-------
                                    3-15
          Moisture was controlled by the apparatus  described  in  Section 4.2.5.
The air was chilled in a moisture condenser to 36°F.   Condensation  in  the
bottom of the condenser trap maintained the relative  humidity at 100%.  When
this air was delivered to the test apparatus, the temperature was raised  to
75°F.  The new relative humidity at these conditions  was determined to  be
12 +2% from a psychrometric chart.
           Each carbon was acclimated to equilibrium with this moisture level
before testing.   Finally a sample of each carbon was baked in a 150 C oven
until no  further  weight loss could be detected.  The density and percent
moisture  for each carbon, as published, as received, as acclimated before
test and  dry are  presented in Appendix D.
           The working capacity of each carbon as a function of density is
presented in Figure 6.  The units for working capacity are grams of hydrocarbon
adsorbed  per 100  grams of activated carbon acclimated to 12% relative humidity
air.  The units for density are  grams per cubic centimeter of acclimated
carbon.   Working  capacity is seen to increase strongly with  reduced acclimated
carbon density.   One  sample, X,  fails to fit this characteristic.  It is the
heaviest  of  the samples and held  the least moisture  during test.
     3.3.5  Carbon Activity
           Published data for each carbon  include the ability to  adsorb carbon
tetrachloride, C  Cl^.  This ability is called C Cl^  activity and is defined
as  the weight percent of C Cl^  that can  be adsorbed  on  the carbon.  These
data are  listed in Table  8.
           Working capacity as a  function  of  C Cl,  activity is plotted  in
Figure 7. Activity among the coal  base  carbon seems to  have little affect on
working capacity. The  greater  capacity  of the one wood  base carbon is
distinguished by  the  highest C  Cl^  activity  of the carbons tested.

-------
                          3-16
   12
I
S 10
 en
o
o
S-
o
Q.
10
8
O
re
Q.
(O
O

CD
C
                                             Q  X
                                         Z   ©
                 0.2
                         0.3
0.4
0.5
             Measured Bulk Density  .  gm/cc
            (Acclimated to 12% RH)
                       Figure 6

          CARBON DENSITY VS. WORKING CAPACITY

-------
                         3-17
   12
8  10
o
o
o
CL
01
$
•r-
U
CT>
O
    8
                                        X-l
                         O Z-l
           20          40          60          80         100


              Indicated C C14 Activity,    % by weight
                       Figure 7



      CARBON C C14 ACTIVITY VS. WORKING CAPACITY

-------
                                   3-18
3.4  CANISTER DESIGN
         Tests were performed on carbon beds having two configurations
different from the baseline canister.   One can had a cross-sectional  area
and diameter equal to the baseline can but was twice the length (height)
and twice the volume.  The other can had a length equal to the baseline can
but was twice the cross-sectional area ("Y7* x diameter) and twice the volume.
The L/D ratios were 4.0, 3.0 (baseline) and 1.4 respectively.  Charge and
strip flow rates were held the same as in the baseline tests.
          Working capacity of the carbon in three different canister  configura-
tions is listed in Table 9, and results are shown graphically in Figure 8.
In Figure 8 working capacity is seen to improve with increased L/D ratio within
the range tested.  The increased working capacity for the taller canister is
explainable using Mass Transfer Zone (MTZ) theory.  Basically, this means that
the carbon bed, upon being charged, contains a zone which is fully saturated
with hydrocarbons and a zone where the saturation level varies from full to empty
as indicated at the top of Figure 9 .   For all practical purposes, this
zone contains half of its maximum capacity.  Any additional hydrocarbons
added to the bed will cause these zones to  shift  slightly  and a breakthrough
is detected in the effluent.  Hence, the capacity at breakthrough of the
example carbon bed is 75% of maximum capacity if the bed length is twice the
length of the MTZ.  Additional carbon bed length increases the capacity at
breakthrough as shown at the bottom of Figure 9.

                                   Table 9
             CANISTER CONFIGURATION EFFECTS  ON WORKING  CAPACITY

      Test
      No.
       2-3
    (Baseline)
       2-2
Can
Confiq.
Hx2A
HxA
2HxA
Area>
cm2
66
33
33
L/D
1.4
2.0
4.0
Charge
Velocity
cmVsec
0.5
1.0
1.0
Corrected
Working Capacity,
qm/100 gm
5.77
6.03
6.77

-------
                         3-19
o
o
o
Q.
CO
u
re
CL
re
o

en  _
c  5
s.
o
                  1.0          2.0         3.0


           Carbon  Bed   Length/Diameter Ratio
4.0
                        Figure  8


   EFFECT  OF CANISTER  L/D  RATIO  ON WORKING CAPACITY

-------
                                      3-2U
Vapor
Flow
                         Length of Bed
                        	L
   (example)
MTZ =  1/2  L
                                                     In  zone  "A" the carbon is fully
                                                     In  tK*? Wlrh Mrocarbons  ^
                                                     *n  the Mass Transfer Zone (MTZ)
                                                     ? Sclent condition exists a*
                                                     indicated by the curved 1 ne
                                                     25m ?9r?C °f saturation var es
                                                     from fully saturated at «f« Jf
                                                     f     available at V  and
                                                              b°Ut half "tlllied over

  % of Full
  Capacity
  Utilized
                          2345
                             Ratio  L/MTZ
     10
                                 Figure  9
                        MASS  TRANSFER  ZONE  CONCEPT

-------
                                     3-21
          If one assumes a MTZ of about 6.6 cm in length for bed lengths  of
13.3 and 26.6 cm, the longer bed would theoretically have about 12.5% more
capacity at breakthrough than the shorter "baseline" canister.   Actual  test
results indicated a working capacity 12.3% greater than "baseline"  tests.
Quadrupling the bed length over the "baseline" case of 13.3 cm would
theoretically increase the working capacity 18.8% over the baseline case.
          The above predictions are based on an absolute MTZ of 6.6 cm in
length which should not change appreciably for a full size adsorber bed.
Therefore, while the effect of the MTZ is important for short bed depths,
a full size service station type system would  be expected to be several feet
deep and would be using greater than 95% of maximum capacity.
          The fact that the H x 2A can had a slightly lower capacity than the
"baseline" case is attributed to a potentially poorer distribution  of vapors
at the inlet which may have resulted in "channeling" through bed.

3.5   STftIP METHOD
          Several practical methods to strip  (remove) adsorbed hydrocarbon
vapor from charged activated carbon were  proposed.  Working  capacities of
carbon stripped by the methods  are discussed  in  the following  sections.
      3.5.1  Air Purge Flow Rates
          Carbon was vapor-charged upward after which air was passed in the
opposite direction downward through the carbon bed.  Three different air
purge flow rates were tested.  Canister outlet gage pressure was regulated
to zero  (+1.5 in H?0) during most of  the  strip when both air and high concen-
tration  vapor was issuing  from  the carbon bed.  Only during  the last moments
of the strip, when air and low  concentration  vapors issue, did the pressure
fall  further to approximately -5.0  in. H20.
          Working capacity of carbon  stripped at three different flow rates
is listed in Table lOand  shown  graphically  in Figure 10.   The  working capacity
tended to diminish as air purge flow  rate increased, however,  the  time to
strip varied  inversely with  the flow  rate indicating that approximately  the
same  total  volume of  strip air  was  required for  all  three  tests.

-------
                                   3-22
                                  Table 10
                    PURGE RATE EFFECT ON WORKING CAPACITY
TP«;t. Purge Rate
No. cc/min BVPM*
(Baseline) 4000
2-4 20000
2-5 40000
9
45
90
cm/ sec
2
10
20
Purge Corrected
Time Working Cap.
min gm/100 gm
12 6.03
3 4.99
1.5 5.21
volumes per minute
Stripping Temperature
          Two carbons  were  charged  with  vapor  from  9  psi and 14 psi  RVP
gasoline at 75°F in the  normal  manner after which heated purge  air was  used
to strip the vapor from  the carbon.  Air temperature  at  the  canister  purge
air inlet was recorded during  each  strip.   Several  minutes were required  to
achieve the required temperature during  each strip.  This temperature delay
would be expected in a typical  full size system to  come  up to temperature
after an air heater was  turned on.
          Working capacity  of  two carbons  for  two gasoline vapors  is  listed
in Table 11.  The results are presented in  Figure 11.   As indicated,  there was
no increase in working capacity by stripping with hot air.   Although  it was
presumed that hot air stripping would significantly increase working  capacity,
two factors diminished the  effect.   First, the total  BTU content of the heated
air was low due to its low  specific heat value compared  to the specific heat
and mass of carbon.  This means that the 200°F air  only  contained enough heat
to raise the carbon bed  about  20-30°F during  a 12 minute purge mode.   The
second factor was that the  carbon bed did not  cool  down  to ambient temperatures
during the short one minute hold period prior to the subsequent charge mode
and the slightly elevated temperatures during charge would tend to offset
the slightly improved stripping ability of the hot  air.
           Increasing  working capacity would require  either much hotter air
temperatures  (low  kindling  point of  some  carbons would  preclude very high air
temperatures),  heating  of  carbon directly such  as  with  steam coils - resistance
heaters  -  etc.,  and adopting a  cooldown mode  prior to successive charge  modes.
The  feasibility  of any  of  the  above  methods would  involve comparing  the  gains
in working capacity vs.  the equipment cost, operating cost and energy  con-
sumption  factors.

-------
                        3-23
 S-
 (0
 o

 E 6
 01

o
o
 s-
 o
 o.
 (O
0
10
o

en
                10          20
         Purge Strip Flow Rate,
       30
  1000 cc/min
40
                 5          10

          Purge Air Velocity,
cm/sec
       15
20
                      Figure 10

        PURGE  RATE EFFECT ON WORKING CAPACITY

-------
                    3-24
                   Table 11
PURGE TEMPERATURE EFFECT ON WORKING CAPACITY
Test
No.
(Baseline)
4-3
4-15
4.4
4-16
1-5
4-11
4-12
4-13
4-5
4-6
4-7
4-8
Carbon
Type
W-l
W-l
W-l
W-l
W-l
X-l
X-l
X-l
X-l
W-l
W-l
X-l
X-l
Nominal
RVP
Ib.
9
9
9
9
9
9
9
9
9
14
14
14
14
Purge Air
Temp.
OF.
75
150
150
200
200
75
75
150
200
75
150
75
150
Corrected
Working Cap.
gm/100 gm
6.03
6.40
6.49
5.82
6.10
5.49
5.62
4.87
5.65
6.69
6.44
6.52
6.45

-------
o

"£  7
(O
O
O
a.
re
>
o
a.
IO
en
t-

O
                        3-25
                               unleaded 14 Ib. RVP
         W-l - -
          X	
leaded 9 Ib
                                               CD —
          50
 100
150
200
         Purge  Strip Temperature,  °F.

         (No cool down after  hot strip)
                      Figure  11


    PURGE TEMPERATURE EFFECT  ON WORKING CAPACITY

-------
                                   3-26
          3.5.3  Relative Humidity
          Relative humidity in the charge and strip air was controlled at
three different levels to observe the effect on working capacity.   Air was
bubbled through water at regulated temperatures and pressures to establish
12, 40 and 80% relative humidity.  See Table 12 for a summary of these conditions.
Before testing, the carbon was acclimated to the moisture level in the air  to
be used for charge and strip.  The data reported are grams of hydrocarbon
adsorbed per 100 grams of carbon as acclimated to the moisture level  tested.
Working capacity as a function of relative humidity of the charge and strip
air is presented in Figure 12 .  Working capacity is seen to diminish  with increases
in relative humidity.
          Carbon adsorbs more moisture at higher humidity levels.   The density
of the carbon used in each test is also plotted on Figure 12.  The relationship
of working capacity to carbon density as discussed in Section 3.3.4 is upheld.
Although it could not be determined accurately from three data points, the
carbon working capacity is believed to vary with relative humidity as indicated
by the dashed S-shaped curve.
     3.5.4  Vacuum Stripping
          Tests were performed to determine activated carbon working  capacity
using vacuum stripping techniques.  Vacuum was regulated at two different
absolute pressures--!00 mm Hg and 25 mm Hg.  No purge air flow was permitted
during the vacuum strip operations.  Therefore, no sample was available for
analyzing hydrocarbon concentration to sense completion of strip.  Instead
the duration of the vacuum strip operation was timed to last approximately
81/2 minutes.
          A one-minute delay was provided by the automatic control system
after completion of vacuum strip before the start of the subsequent charge.
Vacuum from the preceding strip was not totally lost in only one minute.

-------
                                   3-27
                                  Table  12
               RELATIVE HUMIDITY  EFFECT  ON WORKING  CAPACITY
Saturation
Test
No.
(Baseline)
1-15
1-16
Press.
PSIA
32
14
14
Temp.
°F.
36
49
69
Relative
Humidity
% @ 75° F.
12
40
80
Corrected
Working
Capacity
gm/100 gm
6.03
4.95
4.68
W-l Carbon
Density
As Tested
gm/cc
0.380
0.411
0.452
% H20
by Wt.
2.5
10.0
18.1
         Vacuum stripping did remove vapor from the carbon bed but with no air
purge flow.   A small quantity of vapor in the top of canister (isolated during
vacuum strip) remained.  The concentration of this vapor was reduced during
vacuum strip but was still higher than the "charge breakthrough" setting.
Consequently, each charge started into a partial vacuum and immediately drove
out a slug of sufficiently high concentration vapor to signal the automatic
control logic to terminate charge prematurely and skip on to the next strip
mode.  To overcome this problem, the "charge breakthrough" setting was raised
to a higher concentration value.  This allowed the charge mode to continue
about 15 seconds longer than normal.
         Working capacity of carbon stripped by vacuum is listed in Table 13.
The results of testing for the W-l carbon are presented in Figure 13.  Working
capacity improves as absolute pressure during vacuum strip is reduced.  Vacuum
strip working capacities, at the pressures tested, are lower than air purge
strip capacities.  Extrapolation of these results indicates that vacuum
stripping at an absolute pressure of 5 mm Hg may increase working capacity
to be equal to comparable "air purge strip" working capacity.

-------
                                3-28
c
o
.O
V_
(O
O
o
o
o
o.
<0
o
•O

&   4
O
                  20
40
60
80
        Relative Humidity,    %
        (in  charge and strip air)


                              Figure 12

            RELATIVE  HUMIDITY EFFECT ON WORKING CAPACITY
                                                                      0.6
                                                                      0.5  8
                                                                     0.4
                                            gjS
                                            at I/I
                                            o ^

                                            li
                                                                     0.3
                                            
-------
                  3-29
                 Table 13
VACUUM STRIP EFFECTS ON WORKING CAPACITY
Test
No.
3-2
3-3
3-4
3-4
3-4
3-4
3-16
3-16
3-16
Carbon
Type
W-l
W-l
W-l
W-l
W-l
W-l
X-l
X-l
X-l
Nominal
RVP
Ib.
9
9
9
9
9
9
9
9
9
Strip
Press.
mm Hg
100
100
25
25
25
25
25
25
25
Corrected
Working
Capacity
gm/100 gm
0.33
1.40
3.90
4.05
3.90
4.42
3.46
3.80
4.18

-------
                   3-30
to
o


o>

o
o
o
a.
re
o>
o
re
a.
fO
o
s.
o
   0
                                 75° F.
        Premature charge

         breakthrough    —»—A
                                          1
0          50         100          150


  Vacuum Strip Absolute Pressure, mm Hg
                                                      200
                  Figure  13

  STRIP VACUUM  EFFECT  ON WORKING CAPACITY

-------
                                   3-31
         Of particular  concern for these vacuum stripping tests was the
presence of a low concentration of hydrocarbons in the canister effluent
during charge modes.  This phenomenon is referred to as "bleedthrough" and
was not evident with air purging.  Bleedthrough hydrocarbon levels appeared
to continuously increase with the number of operating cycles for both carbon
types W-l and X-l.  Table 14 shows a comparison of bleedthrough levels.

                                 Table 14
                     COMPARISON OF BLEEDTHROUGH LEVELS
                     Bleedthrough:  hydrocarbon concentration expressed
                     as volume % propane equivalent
  Test No.           @ 4th Cycle         @ 10th Cycle         @ 20th Cycle
  Baseline               0.1%                0.135                 0.1%
  (air purge)
    3-2*                 3%                  5%          Too high to complete test
    3-3*                 3%                  4%                   5%
    3-4*                 1%                  2%                   3%
*See Table 13for description of test conditions
         Referring  to Test No. 3-4, one  observes  that  the  bleedthrough level
is approximately 6% of the charge gas concentration of 50% C3 and is  increasing
steadily.  To achieve an overall control system efficiency of 90% only low
levels of bleedthrough could be tolerated.
     3.5.5 Hot Vacuum Stripping
         Each of two carbons were charged with vapor  from  two gasolines after
which the carbon beds were subjected to  vacuum stripping at two eleveated
temperatures in an  attempt to  improve working capacity.  All of the strips
were performed at the better of  the two  absolute  pressures (25 mm Hg)
investigated in the previous section.
         Hot vacuum strip cycles were not performed 21 times as  in  other  tests.
Instead, the carbon bed was  first exercised  through nine automatic  charge  and
vacuum strips at 75° F.  The test canister was then weighed, charged  at 75° F.

-------
                                   3-32
and reweighed to determine the 75° F. working capacity.  After placing the
canister in a 150° F. oven for 15 minutes, the vacuum stripping commenced
and continued for  eight and one-half minutes (during which time the canister
remained inside the oven).  Immediately following the completion of the strip
mode, the canister was weighed hot, allowed to cool at room temperature for
45 minutes, charged with 75° F. vapors and then reweighed.  The difference in
these previous two weight measurements was used to compute the working capacity
for the 150° F. vacuum strip.  The aforementioned steps were repeated this time
using a 200° F. oven to determine working capacity for a 200° F. vacuum strip
temperature.
         Working capacities with vacuum stripping at three test temperatures
are presented in Table 15 and are plotted in Figure 14.- Strong improvement in
working capacity is seen with rising vacuum strip temperature.  The working
capacity relation to gasoline volatility and carbon manufacturer is identifiable
at all temperatures.  Additionally, bleedthrough hydrocarbon levels decreased
appreciably at the elevated temperatures within one or two operating cycles.

3.6  AMBIENT TEMPERATURE
         The working capacity of carbon as affected by ambient temperature was
investigated.  One carbon was charged with the vapor from nine RVP gasoline
after both the carbon and gasoline were stabilized at three different temperatures,
Consequently, the results reflect not only the response of carbon to temperature
but also to increases in the vapor pressure of the gasoline.    (See Section 3.2)
         Working capacity of carbon at different carbon/gasoline temperatures is
listed in Table 16 and shown graphically in Figure 15.  Working capacity increased
with ambient temperature rise of both carbon and gasoline.  The increase in vapor
concentration of the same RVP gasoline with rising temperature is also shown.

-------
                   3-33
                 Table 15
HEATED VACUUM STRIPPING VS WORKING CAPACITY

Test
No.
3-4



3-11


3-12


3-6

3-7
3-5
3-16


3-13


3-14


3-9
3-8
3-15

Carbon
Type
W-l
W-l
W-l
W-l
W-l
W-l
W-l
W-l
W-l
W-l
W-l
W-l
W-l
W-l
X-l
X-l
X-l
X-l
X-l
X-l
X-l
X-l
X-l
X-l
X-l
X-l

Nominal
RVP
Ib.
9
9
9
9
9
9
9
9
9
9
14
14
14
14
9
9
9
9
9
9
9
9
9
14
14
14

Strip
Press.
mm Hg.
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25

Strip
Temp, °F
75
75
75
75
150
150
150
200
200
200
75
75
150
200
75
75
75
150
150
150
200
200
200
75
150
200
Corrected
Working
Capacity
gm/100 qm
3.90
4.05
3.90
4.42
5.33
5.40
6.09
6.58
6.71
7.05
5.42
4.97
5.41
7.16
3.46
3.80
4.18
4.65
6.58
5.24
6.21
7.08
6.18
5.48
5.05
5.95

-------
                          3-34
o
JO
i_
(O
u
O
O
(O

&   4
en
O
        Unleaded
        14  Ib.  RVP
                                                   X-l
                                          X-l Carbon

                                          W-l Carbon
                       9  Ib.  RVP  leaded
             I
             50          100          150         200


               Vacuum Strip Temperature,  °F.
                         Figure  14

   VACUUM  STRIP  TEMPERATURE  EFFECT ON WORKING CAPACITY

-------
                                   3-35
                                 Table 16
              AMBIENT TEMPERATURE EFFECT ON WORKING CAPACITY
Ambient
Test Temp .
No. op.
1-13
(Baseline)
1-14
50
75
86
Vapor
Cone.
Vol. %
32
53.6
62
Working Capacity
gm/100 gm
Observed Corrected*
4.97
6.03
6.50
5.70
6.03
6.21
         *Corrected to the Baseline vapor concentration of 53.6%

         Working capacity at different carbon temperatures (gasoline temperature
and volatility constant) was not investigated directly.  But this information
may be inferred by correcting these data to the baseline vapor concentration
of 53.6% at 75° F.  When these data are plotted on Figure 15, the lesser
slope of the resulting line indicates the reduced effect of carbon temperature
only on working capacity at constant volatility.  The difference in slope
between the "carbon temperature only" line and the "combined carbon gasoline
temperature" line indicates the effect of "gasoline temperature only" on
working capacity.  This difference increases with vapor concentration and
compares well with the results in Figure 1 in Section 3.2.

3.7  ONE THOUSAND CYCLE ENDURANCE TEST
         The capacity of carbon to adsorb gasoline vapor during an extended
period was investigated.  After the standard 21-cycle test was completed on
a canister, an additional 1000 cycles were performed.  The apparatus was run
continuously for 271 hours.  Canister weight measurements were recorded
approximately every 100 cycles.  Vapor concentration in the generator was
recorded periodically to account for weathering of the gasoline.  Fresh
gasoline was used twice to  replace weathered gasoline.

-------
                                    3-36
     6.5
     6.0
  s.
  (0

*i
o
     5'5
10
o \

en O
c a.
•i- 
i-
O E
     5.0
Corrected for
Volatility
                                 Observed
                                                      70
                                                      60
                                                           r. CVJ
                                                           C Z
                                                           o
                                       (O
                                       J- 
-------
                                   3-37
         The combined canister, carbon and adsorbed vapor weight was recorded
periodically after strip and after the succeeding charge.  The tare weight of
the canister and of the carbon bed were subtracted from these measurements to
obtain the net adsorbed vapor weight.  The remainders, the stripped heel and
the total charged vapor weights are given in Table 17.  The difference is
working capacity which is also given.
         These data are presented graphically in Figure 16.  The top line is
total charged vapor weight throughout 1023 cycles.  The middle line is the
heel weight.  The bottom line is the difference between the upper two or
working capacity.  A decrease in working capacity of 29% occurred between the
21st and the 1023rd cycle.  The heel is seen to increase throughout the test.
Apparently, the carbon bed was accumulating more hydrocarbon molecules during
each charge than were being removed during the subsequent strip with an
attendant increase in weight.  Total charged vapor weight is also seen to
increase with cycles.

3.8  PRESATURATED CARBON
         The working capacity of two carbons, W-l and X-l, was determined
after they had been presaturated with gasoline vapor.  Each  sample  was
continuously charged until there was no further weight gain.  Then  the
samples were subjected to the normal test sequence except  that the  first
charge was deleted.  Working capacity for presaturated carbons is reported
after 21 cycles  the same as  for normal tests.  The data  are  assembled in
Table 18.  The total vapor weight,  heel and working capacity are tabulated
for both carbons.
         Working capacity of presaturated carbon  is presented in Figure 17.
Neither presaturated carbon  sample  has as great a capacity as it has unsaturated.
After 20 cycles  the capacities were still rising  and  might have recovered  their
full capacities  given more cycles.   An additional eight  cycles  run  on one  sample
showed working capacity was  still being restored.

-------
                  3-38
                Table 17
WORKING CAPACITY AFTER EXTENDED CYCLES
Cycle
1
2
3
10
21
100
200
291
388
462
633
725
823
948
1022
1023
Date
1976
9/2
9/2
9/2
9/2
9/2
9/4
9/5
9/6
9/7
9/8
9/10
9/11
9/12
9/13
9/14
9/14
Uncorrected gm/100 gm
Total Working
Vapor Heel Capacity
12.94
15.81
17.41
18.22
19.44
21.22
22.34
23.47
24.09
24.38
24.91
24.94
25.03
25.03
24.91
25.09
0
8.59
10.97
12.22
13.38
15.44
16.50
18.16
18.91
19.19
19.84
20.06
20.28
20.38
20.59
20.72
12.94
7.22
6.44
6.00
6.06
5.78
5.84
5.31
5.18
5.19
5.07
4.88
4.75
4.65
4.32
4.37
Vapor
Cone.
Vol. %
50
50
50
50
50
53
52
52
51
50
57
56
54
53
53
53
Corrected
Working Cap.
qm/100 gm
13.06
7.34
6.56
6.12
6.18
5.80
5.89
5.36
5.27
5.31
4.95
4.80
4.74
4.67
4.34
4.39

-------
                3-39
            Total  Charged
            Vapor
            (Observed)
                           Heel
                           (Observed)
             Working  Capacity
              (Corrected)
                    i   i  i   11 i     i   i
1              10             100

    Adsorb-Desorb Working Cycles

               Figure 16

WORKING CAPACITY AFTER EXTENDED CYCLES
1000

-------
                  3-40
                 Table 18
SATURATED CARBON EFFECT ON WORKING CAPACITY
Uncorrected gm/100 gm
Cycle
1
2
3
12
21
29
1
2
3
10
21
Carbon
Type
W-l
W-l
W-l
W-l
W-l
W-l
X-l
X-l
X-l
X-l
X-l
Total
Vapor
30.84
29.78
29.16
27.66
27.19
27.03
34.60
32.98
32.60
31.30
30.70
Heel
30.84
26.16
25.34
23.38
22.63
22.25
34.60
30.23
29.54
28.04
27.16
Working
Capacity
0
3.63
3.81
4.28
4.56
4.78
0
2.75
3.06
3.26
3.54
Vapor
Cone.
Vol. %
53
53
53
53
53
53
50
50
50
50
50
Corrected
Working Cap.
gm/100 gm
0
3.65
3.83
4.30
4.58
4.80
0
2.87
3.18
3.38
3.66

-------
                               3-41
TJ
O)
10
>
                 Total Charged Vapor
                 ~  —


                 Heel                Presaturated Carbon
               Iotal Charged
               apor
Activated Carbon
                 working Capacity
        Presaturated Carbon
                 5               10

             Adsorb-Desorb Working Cycles
                             Figure 17

         PRESATURATED CARBON EFFECT ON WORKING CAPACITY
                          20

-------
                                   3-42
3.9  SELECTIVE HYDROCARBON RETENTION
         One of the program objectives was to determine if there was a
tendency for the activated carbon to selectively retain certain hydrocarbons.
Two approaches were used to (a) first determine the nature of the vapors which
are retained on the carbon via comparison of the vapors entering the adsorber
to the vapors being stripped, and (b) secondly determine the composition of
the heel which accumulates in the carbon over a long time period.
     3.9.1 Comparative Vapor Analyses
         Syringe samples were taken of the gasoline vapors that were generated
in the 55-gallon drums for analysis using gas chromatography (GC).  Concurrently,
bag samples were taken of the stripped vapors from various tests for analysis by
the same method.  The GC procedure separated the hydrocarbons esentially into
five groups:
                 •C2 (ethylene + ethane)
                 •C3 (propylene + propane)
                 •C4 (normal and iso-butane plus 04 isomers)
                 •C5 (normal and iso-pentane plus C5 isomers)
                 •Heavy hydrocarbons (components with a boiling point
                  above 40° C.)
Since G£ and C3 components are usually very minor, the area under the C4 peak
and the area under the heavy HC peak was used to provide a ratio of light to
heavy hydrocarbons.  If this ratio is higher in the stripped vapors than in
the generator vapors, one might conclude that heavy HCs are being retained
by the activated carbon and vice-versa.
         Table 19 shows a summary of the analytical results where the ratio  of
C4/heavy hydrocarbons entering the adsorber is denoted by "A", and the  ratio
of C4/heavy hydrocarbons being stripped is designated  "B".  Values of B/A
greater than 1.00 indicate a tendency to retain heavier hydrocarbons.   All of
the results pertain to 21-cycle tests except for the 1000-cycle test, 2-7/16.
As shown, there are 12 out of 17 tests where the B/A is greater than unity
indicating a tendency to retain heavy hydrocarbons and three out of 17  cases
where a slight tendency to retain light hydrocarbons is indicated.

-------
                                                       Table  19

                                       GAS CHROMATOGRAPH ANALYSES OF VAPORS
Ratio of Heavy Hydrocarbons


Test
No.
1-1
1-3
1-8
1-9
1-10
1-13
1-14
2-5
2-6
2-7/16

4-3
4-4
4-5
4-6
4-8
4-16



Description of Test
Baseline*
Y-l Carbon
Saturated W-l Carbon
W-7 Carbon
W-5 Carbon
50 op. Ambient
90 op. Ambient
40,000 cc/min strip
"Baseline"
1000-cycle test @ 435 cycles
1000-cycle test @ 927 cycles
150° F. air strip W-l carbon
200° F. air strip W-l carbon
14 RVP fuel W-l carbon
14 RVP fuel & 150° air strip, W-l carbon
14 RVP fuel & 150° air strip, X-l carbon
2000 air strip, W-l carbon
A
Vapors
Entering
Adsorber
.94
1.08
.94
.94
.81
1.30
.95
.61
.69
.71
.99
.79
.72
2.58
2.60
2.26
1.06

B
Desorbed
Vapors
1.00
1.04
.94
1.02
.92
1.47
1.04
.95
.65
.71
1.05
.92
.74
3.02
2.47
3.36
1.64


B
A
1.06
.96
1.00
1.08
1.14
1.13
1.09
1.56
.94
1.00
1.06
1.16
1.03
1.17
.95
1.49
1.55

Working
Capacity
gms/100 gins
5.78
11.42
4.56
4.47
7.10
4.97
6.50
5.16
6.06
5.19
4.38
6.31
5.63
6.69
6.34
6.31
6.28
                                                                                                                    GO
                                                                                                                    I
                                                                                                                    -pa
                                                                                                                    OJ
* = Baseline conditions are W-l carbon, 12 x 28 mesh, 75° F.  air  strip at  4000  cc/min  ambient
    temperature 75° F.  9  psi RVP fuel

-------
                                   3-44
Due to the inherent variability associated with the measurements and the
extremely small amounts of heavy hydrocarbons retained each cycle, the
extraction results discussed below are more enlightening.
     3.9.2 Analysis of Extracted HC
         A sample of activated carbon from the 1000-cycle test was desorbed
using carbon disulfide.  This mixture was injected into a chromatograph and
analyzed by mass spectrometry.  The results are shown on Table 20.  Compared
to a calculated heel before desorption of 173.4 milligrams/gram, the total
desorbed weight of 172.7 milligrams per gram indicates that nearly all of the
retained hydrocarbons were extracted and that they were primarily 65 - Cg
components as shown on the table.  The selective retention of molecules larger
than GS is believed to be caused by the fact that these large molecules are
firmly lodged in the micropore structure of the carbon and require higher
energy levels to effect removal than do the smaller molecules.  This would
be especially true if the larger hydrocarbon molecules and the micropores are
of comparable diameters.

-------
                     3-45
                    Table 20
          ANALYSIS OF HYDROCARBON HEEL
              FROM 1000-CYCLE TEST
                               Milligrams Per Gram Charcoal
2-Methylpentane                           1.12
3-Methylpentane                           0.67
n-Hexane                                  1.19
2.4-Dimethylpentane                       2.67
2-Methylhexane                            8.47
n-Heptane                                 8.21
2,2,3-Trimethylbutane                    45.88
2,5-Dimethylheptane                      14.52
2,3,3-Trimethylpentane                    8.64
Toluene & Methyl ethyl heptane             33.11
2,2,5-Trimethylhexane                     6.01
Methyloctane                              5.57
2-Ethylheptane                            2.30
3-Ethylheptane                            3.01
m,p-Xylene                                18.83
o-Xylene                                  4.06
Cumene                                    4.55
Methyl ethyl benzene                       0.54
Ethyl toluene                               2.55
1,2-Diethylbenzene                        0.78
      Total                              172.68

-------
                                    4-1
                            4.0  TEST EQUIPMENT

4.1  TEST CHAMBER
         All testing was conducted in Scott's environmental  chamber in
San Bernardino, California.  This chamber is approximately 15 feet wide,
9 feet high and 35 feet long.  The entire chamber is fully insulated and
has temperature and humidity control.  Temperature limits are -10° F. to
100° F. controllable to ±2° F.  Humidity is controllable from ambient to
100 percent relative.  Testing during the program was conducted at 50°, 75°
and 90° F.  The reason the environmental chamber was selected was due to the
fact temperature affects the working capacity of carbon.  By closely controlling
the ambient temperature, variance was minimized.

4.2  AUTOMATIC TEST FIXTURE
         Four individual carbon canister weighing stations were placed  in the
environmental test chamber.  Vapor was delivered to each station by  a gasoline
vapor generator.  Vapor breakthrough from each canister was  sensed by non-
dispersive  infrared analyzers  (NDIR).  Automatic charge and  strip modes were
controlled  by an electronic  logic console.  Vapor concentrations in  the four
generators  were measured using one analyzer and a manually controlled four-
way valve.  The subsystems are described in the following sections.   Figure  18
shows a flow diagram for the apparatus.
     4.2.1  Hydrocarbon  Instrumentation
         Five NDIR  (non-dispersive infrared)  analyzers were  assembled  in a flow
bench.  One analyzer was connected to each  of the four canister weighing stations
and one was valved  to analyze  the headspace of any  one of the  four gasoline  vapor
generators.
         The first  four analyzers were  adjusted  to  respond at  full scale to
approximately 40%  propane  (CsHs).  The  precise response  curves were  established
with calibration gases  of  2.05,  18.77 and  35.48%  propane  diluted with nitrogen
 (N2).   Propane  calibration gases  may not be mixed  in  air because of  the hazard
of explosion at the high  pressures  in gas  cylinders.   The  instruments were
zeroed  using ambient  air  from the same  compressor used to  acclimate  the charge
and strip  air.

-------
                                                             generator vapor  analyze
compressor
and tank
             trap  '  humid
                   I  ifier
vapor generators
    flow during
    charging

    flow during
    stripping
                                                                                                         ro
test
canisters
                        HC analyzers
                                          Figure 18
                             FLOW  SCHEMATIC  FOR TEST APPARATUS

-------
                                    4-3
         Three solenoid valves were used to individually select which
source of vapor would be analyzed.   One valve carried charge breakthrough
vapor from the top of the canister headspace to the analyzer.  The second
valve carried vapor-free air directly from the compressor to the analyzer
during the one-minute duration hold period (between each charge and strip
operation) so that the instrument was properly zeroed before each reading.
A third valve allowed measuring vapor concentration at the bottom of the
canister during the purge mode.
         Oilless teflon coated diaphragm pumps were used to deliver the
vapor to each HC analyzer.  Flow was regulated by a needle valve and
indicated by a rotameter.  Analyzed vapor was directed to a large manifold
vented to the roof.  Pump capacity in excess of the flow required by the
analyzer was bypassed through a backpressure regulator to the vent manifold.
Total delivery delay time of the vapor from the canister to the analyzer was
less than five seconds.
         The fifth hydrocarbon analyzer was used to measure the concentration
of vapor in the hydrocarbon generators.  This analyzer was adjusted  for  full-
scale response on 100% propane.  The precise response curve was established
with calibration gases of 18.77, 35.84, 66.5,83 and 99.5% propane  diluted
with nitrogen.  Zero response was  set with air from  the  test chamber.  Two
manual and two electric  valves were used to select which generator would be
used to deliver vapor  from  the generators  to the analyzer.   The flow was
regulated by a needle  valve and  indicated  by a rotameter.  After  analysis
the  vapor was directed  to the  vent manifold.
     4.2.2 Continuous  Weighing Instrumentation
          Four nearly  identical canister weighing station were  constructed.
Each canister was  supported by a load  cell  transducer.   Weight increase  during
charge and weight  decrease  during  strip were  recorded  on chart paper.   The
load cell was calibrated against known weights with  the  canister  suspended  in
place  so  as  to  include the  inherent  spring rate  of the  interconnecting tubing.
Under  actual cyclic  test conditions, the recorded  weight traces were
found  to  be  extremely  sensitive  to pressure  and  temperature  fluctua-
tions.   Manual weighing  of  test  canisters  was  employed to monitor
working  capacity.

-------
                                    4-4
         A tube introduced charge vapor/air mixture to the bottom of the
canister.  A circular perforated manifold distributed the vapor uniformly
through a fiberglass wool pad which supported the carbon bed.  The vapor
flowed up through the Tied and was adsorbed until the capacity of the carbon
was exceeded at breakthrough.  Charge effluent passed up through the bed and
out through a tube in a rubber cork at the top of the canister.
         The vapor was delivered by the regulated pressure in the vapor
generator described in Section 4.2.4.  Flow was governed by a solenoid valve,
regulated by a neelde valve and indicated by a rotameter.  A backpressure
regulator maintained atmospheric pressure inside the canister during charge.
A pressure gage indicated canister head space pressure, and a thermocouple
located approximately one-half inch beneath the top of the carbon bed was
used to monitor bed temperature.
             For purging strip air was introduced through the top tube  to the
head space of the canister and passed down through the carbon bed, desorbing
vapor carrying it out through the glass wool, manifold and tube  in the  bottom
of the canister.  The strip air flow was pressure regulated, throttled  by
needle valve, indicated by rotameter, governed by solenoid valve and carried
by the top tube to the head space of the canister.  Another  backpressure
regulator maintained atmospheric pressure inside the canister during strip.
The gage and thermocouple indicated as during charge.
             Station #1 was modified to incorporate a second stage moisture
control described in Section 4.2.5.  Higher relative humidity was required
for two experiments.  Station #2 was modified to incorporate a  high  strip flow
rate rotameter and large capacity pump to accommodate the  high  strip air flows
required.  Station #3 was modified to incorporate a vacuum pressure  regulator
and vacuum pump to perform the vacuum strip tests.  An oven  was  set  up  next
to this station in which hot vacuum strips were performed.   Station  #4  was
modified to incorporate a strip air heater with which to  perform the hot
purge air strip tests.
     4.2.3 Control Logic
             The activated carbon test system was controlled by an  electronic
logic console.  Internally divided into four independent  channels,  this console
acquired the gasoline vapor concentration signal sensed  by an  NDIR  analyzer
from the outlet of each carbon canister during  charge  (top)  or  during  strip

-------
                                    4-5
(bottom).  Time delay relays were arranged to deliver power to the electric
valves in each canister weighing test station.  The electric valves were
energized in a sequence required to deliver (charge) gasoline vapor to the
carbon canister and remove (strip) vapor from the canister.  The canister
was isolated (hold) for one minute both before and after each charge during
which the canister was available for weighing.
             Vapor concentration limits were independently set to terminate
charge and strip in each weighing station.  Charge was terminated when vapor
concentration emerging from the outlet (top) of the carbon bed exceeded
(break through) the limit.  Strip was terminated when the vapor concentration
emerging from the outlet  (bottom) of the carbon bed fell below the limit.
             During vacuum strip tests no strip sample was available with
which to sense HC concentration.  A time delay relay was provided to control
the period during which the canister was vacuum stripped.
             An array of  color coded light emitting diodes was provided  for
each canister weighing test station.  The performing mode  in  the  test cycle
was indicated for each canister.  Overriding  toggle switches  were provided
for manual operation of the system.  A cycle  counter was provided for each
canister.  A rotary switch was provided with  which  to  select one  of  four
gasoline vapor generators to analyze.  The vapor  generator analyzer  and
recorder were controlled  from the logic console.   The  individual  canister
weight transducers were calibrated with controls  in  the  logic console.   Zero,
coarse and fine trim potentiometers were  provided.
      4.2.4 Hydrocarbon Vapor Generator
              Four  hydrocarbon vapor  generators were constructed—one for each
carbon canister weighing  station.  The  55-gallon  barrel  in which  the gasoline
was received was modified to generate  vapor.   A  sintered metal  bubbler  was
installed  through  the  V  bung and was  submerged  near the bottom.   Pressurized,
moisture controlled  air was delivered  from the compressor  to the  bubbler.  An
individual  pressure  regulator was  provided for each generator and a  solenoid
valve governed  air flow during  charge  only.   Check valves  protect the solenoid
valve and  regulator from  reverse flow  of  gasoline in  the event of lost  pressure.

-------
                                    4-6
             A multiport standpipe was screwed into the two-inch bung
opening.   Five pipe taps were provided in the standpipe.  One led to a
pressure  gage near the pressure regulator which supplied the bubbler.
Pressure  was maintained constant and as low as possible (4 - 6" ^0).  The
second tap carried charge vapor to the canister weighing station described
in Section 4.2.2.  A third tap carried generated vapor, when selected, to an
NDIR hydrocarbon analyzer described in Section 4.2.1.  Selection is made only
during charge so that the analysis is representative of the vapor being
adsorbed  by the carbon.  A fourth tap led to a pressure relief vent valve
to protect the barrel from rupture in case of overpressure.  The fifth tap
led to a  normally open solenoid valve which opened to vent the barrel when-
ever the  air compressor electrical power was lost or turned off.  This feature
provided  further safety (as during a weekend shutdown) by preventing thermal
pressure  buildup and possible rupture or overflow of gasoline from the barrel.
             The gasoline barrels were placed in the termperature controlled
test chamber several days before testing so that the liquid temperature and
vapor concentration could be stabilized.
     4.2.5 Moisture Controlled Air Compressor
             The necessity to control the percent of moisture in the carbon
for experimental purposes was made evident by the inconsistency between the
adsorption characteristics of wet and dry carbon.  With each succeeding charge
cycle, dry carbon exhibited an even higher total weight.  Succeeding heels
increased but at a lesser rate which resulted in an  increasing working capacity
between total and heel.  Wet carbon displayed little or no increase  in total
adsorbed vapor, heel or working capacity after the first charge.  All three
weights were less than for dry carbon after several  cycles.
             For consistency in experimental results,  it was decided that all
measurements should reflect only  the increase or decrease of hydrocarbon
vapor.  Therefore, the moisture content of the carbon  had to be acclimated  to
the moisture content of the charge and strip air.  For consistency throughout
the program, the moisture content of the air had to  be constant.

-------
                                    4-7
             A moisture controlled air compressor system was assembled and
located in a room adjacent to the test chamber both to remove the compressor
noise and to prevent ingestion of possible background vapor.
             A large capacity oil!ess diaphragm pump was mounted on a pressure
tank.  A moisture condensing trap in a refrigerated water bath was mounted
next to the pump.
             The thermostat on the water bath was set at 36° F.--as cold as
possible without freezing the water.  A pressure switch controlled tank
pressure between 12 and 24 psig.  Moisture in the air leaving the tank
condensed in the trap.  The air exited the trap saturated with moisture at
an average of 18 psig.  Saturation at this pressure and temperature determines
that the specific humidity is 14.7 grains t^O/lb. dry air.  This moisture level
results in a relative humidity of 12% at the test conditions of 75° F. and 0
psig.  All carbon types were acclimated with this air to either add or remove
moisture until equilibrium was observed in stabilized total carbon weight.
             For those tests where higher relative humidity was required, an
additional temperature controlled water bath was installed.  Two water filled
bubblers were placed  in the bath.  One provided moisture to the charge air just
before it entered the vapor generator.  The other bubbler  provided moisture  to
the  strip air just before it entered the canister.  The  bath temperatures
surrounding the  bubblers was set  at  49° F. for 40% relative  humidity  air  and
69°  F.  for 80%  relative humidity.
     4.2.6 Carbon Bed Canister
             The can  commonly used  for aerosol sprays was  selected  for carbon
bed  evaluation.  This can was able  to withstand  the  vacuum pressures  involved
in  this  program.  The spray  valve disc was removed  from the  top to  expose  a  hole
large  enough  to  pour  in  the  carbon.   A 3/8" diameter  tube  manifold  was  soldered
in  the side of the  can with  one  leg sticking  out.   The  manifold was  formed  in
a circle with distribution  holes  around  the  inside bottom  of the  can.  Struts
were added  to  strengthen  the  leg  of the  tube  sticking  out.  A one hole  rubber
cork with a 3/8" diameter  tube  was  placed in  the top of the can.

-------
                                    4-8
             A fiberglass wool pad, 1/4" thick, was placed on top of the
circular manifold to support the carbon bed.  A ball of glass fiber was
stuffed inside both of the 3/8" tubes at the top and bottom of the canister
to prevent escape of the carbon particles.
             The effective volume of this canister was 440 cc.  Two sets of
canisters were made so that one set could be tested while the other set was
being loaded.  Two additional cans were fabricated for investigation of
different canister configurations.  One had twice the length of the standard
can but equal diameter.  The other had twice the cross sectional area
("Y2~  x diameter) as the standard can but equal length.

-------
                                  A-l


                               APPENDIX  A

                             Correspondence
                               Form  Letter
Scott Environmental Technology has been awarded a contract by the
Environmental Protection Agency to evaluate the properties of
activated carbon in conjunction with gasoline vapor emission control
at service stations.  A literature search revealed that your company
was a supplier of activated carbon.

The purpose of this letter is to solicit available information
about activated carbon.  It would be greatly appreciated if you
would fill out and return the attached questionnaire as well as
forward any information you have regarding your activated carbon
product.  All data submitted will become the property of the EPA
and will eventually be made public.

Thank you for your time and effort spent responding.

                                    Sincerely,
kd
Attachment
                                    Michael J. Manos
                                    Project Manager
                                    Environmental Services Department

-------
                                  A-2
                              Attachment
                    ACTIVATED CARBON QUESTIONNAIRE
                    Company           ,	 _

Do you presently manufacture activated carbon?          Yes	    No_
Are you presently a supplier of activated carbon?       Yes	    No_
If you are only a supplier, who manufactures the
 activated carbon?           Company	
                             Address
                             Phone
Do you supply an activated carbon suitable for
 hydrocarbon adsorption and desorption?                 Yes	    No_
                                             Identification No.	
                                             Type Carbon	
Is this activated carbon used to control service
 station hydrocarbon emissions?                         Yes	    No_
If yes, where?	

Do you regenerate activated carbon?                     Yes	    No_
If yes, method used 	

Do you manufacture or supply a service station
 hydrocarbon emission control system?                   Yes	    No_
If yes, briefly describe:	
Name and phone number of individual in your company to contact if
additional questions arise.
                                   Name
                             Phone  No.

-------
835 North Cassady Avenue
P.O. Box 2526
Columbus, Ohio  43216
                                                                     JMN  i 9  i.j

                              Attachment                         !^/tf& Eiwironmejiiai
                    ACTIVATED CARBON QUESTIONNAIRE

                    Company    Barnebev-Chenev


Do you presently manufacture activated carbon?          Y.es_x_    No	
Are you presently a supplier of activated carbon?       Yes  y    No	

If you are only a supplier, who manufactures.the
 activated carbon?           Company		
                             Address         	
                             Phone
 Do you  supply  an  activated  carbon  suitable  for
 hydrocarbon adsorption  and desorption?                  Yes ,x     No	
                                              Identification No. Type AC & Type VG
                                              Type Carbon  Coconut shell

 Is  this activated carbon used to control service
  station hydrocarbon emissions?                         Yes	    NO_XJ_

 If  yess where?	.	


 Do you regenerate activated carbon?                     Yes	x_    No	
 If yes, method used 	Combination thermal and steam regeneration


 Do you manufacture or supply a service station
  hydrocarbon emission control system?                   Yes	    No x
 If yes, briefly describe;  However, we manufacture several different

   panel and canister adsorbers which could be used effectively	



 Name and phone number of individual in your  company to  contact if
 additional  questions arise.,
                                    Name      Gary W. Hartman	
                              Phone  No.     614 258-9501
                                        Attachments:   Charcoal  literature  package;
                                                       T-74,  264,  142

                                        Under separate cover:  1-lb each Types AC and
                                                                    VG Charcoals

-------
      (CALGON
A-4
RECEIVED

  FEB  13 1976
     SUBS.D.ARY OF MERCK I CO.. ,NC.


ACTIVATED CARBON DIVISION  CALGON CORPORATION CALCON CENTER BOX 1346  PITTSBURGH, PA. 15230 (412)923-2345


                                    February 10,  1976
   Mr. Michael Manos
   Scott Environmental  Technology
   2600 Cajon Boulevard,
   San Bernardino,  California  92411

   Dear Mr. Manos:

   Confirming our  recent  telephone conversation, we  are  sending
   you a 5 gallon  sample  of  BPX 8 x  30 which is  suitable for the
   recovery of gasoline vapors at service stations.

   Enclosed you will  find  with the questionnaire which you recently
   sent a report which  summarizes information gathered during our
   investigation programs  in regard  to:

        a)  Carbon  systems efficiency.

        b)  Activated carbon working capacity.

        c)  Working capacity and life with air sweep
            regeneration.

        d)  Working capacity and life with vacuum
            regeneration.

        e)  Heel characteristics.

   We are confident that  this  report will be of  interest to you.

   We hope that our data  and commentary are useful.   Please
   advise if there  are  any questions or additional ways  in which
   we may assist.
                                    Very truly yours
                                    Development Engineer
   BLG:rmk

   Enclosure

   cc:   Mr. R.  L. Cooley
                             PITTSBURGH ACTIVATED CARBONS

-------
                                 .  A-5
                             Attachment
                   ACTIVATED CARBON QUESTIONNAIRE
                   Company Calgon  Corporati on

Do you presently manufacture activated  carbon?          Yes X    No
Are you presently a supplier of activated carbon?       Yes %    No_
If you are only a supplier, who manufactures the
 activated carbon?           Company	
                            Address
                            Phone
Do you supply an activated carbon suitable  for
 hydrocarbon adsorption and desorption?                Yes X     No
                                            Identification No .   R p
                                            Type Carbon   8x30
 Is this activated carbon used to control service
 station hydrocarbon emissions?                        Yes x     No _
 If yes, where?1" San  Diego County.  Our carbon  is  used  by three
               equipment  manufacturers.

 Do you regenerate activated  carbon?                 .  Yes X     NO _
 If yes, method used   Thermally __
 Do you manufacture or supply a  service station
  hydrocarbon emission control system?                  Yes	    No
 If yes, briefly describe:	
 Name and phone number of individual in your company to contact if
 additional  questions arise.
                                  Name Bernard Grandjacoues
                             Phone  No.  412-923-2345

-------
  1C I United States Inc.
                                               WILMINGTON, DELAWARE 19897
                                                          (302) 658-931 1
                            ElVED                    (301) 
-------
                                    A-7
                              Attachment
                    ACTIVATED CARBON QUESTIONNAIRE

                    Company
Do you presently manufacture activated carbon?          Yes \f    No_
Are you presently a supplier of activated carbon?       Yes_ /    No_
If you are only a supplier, who manufactures. the
 activated carbon?           Company _ . . .
                             Address
                             Phone
Do you supply  an  activated  carbon  suitable  for
 hydrocarbon adsorption  and desorption?                  Yes"\f_    No_
                                              Identification No.
                                              Type Carbon  / _ j_ V I
 Is  this  activated  carbon used to control service              -vz-v-*-    <*-/
  station hydrocarbon emissions?                         Yes _    No__\_
 If  yes,  where? _    .. __ . _____ •. -
 Do you regenerate activated carbon?                     Yes -    No_JL_
 If yesj method used ____ ___ _ _ _ _. ._., . . -
 Do you manufacture or supply a service station
  hydrocarbon emission control system?                   Yes _    No__\

 If yes, briefly describe: _ ___ ___
 Name and phone number of individual in your company to contact  if
 additional questions arise.
                                    Name_	yj c ^ K ft  ( f ^
                              Phone  No.   3 d 3^ -  /T 7

-------
                                             A-8
                      UNION CARBIDE CORPORATION

                      CARRON PRODUCTS DIVISION
                          MADISON AVENUE, CLEVELAND, OHIO 44107 • TELEPHONE: 216-433-8600
   ADDRESS REPLY TO:
        P. O. BOX 6087
CLEVELAND, OHIO 44101
                                                      February 10,  1976
       Mr. Michael J. Manos
       Scott Environmental Technology Inc.
       2600 Cajon Blvd.
       San Bernadino, California 92411                         DECE

       Dear Mr. Manos:                                           rt d 1. 8
            In your activated carbon questionnaire attached  £6°'y6u£f(H6TeetefrT
-------
                                    A-9
                              Attachment

                    ACTIVATED CARBON QUESTIONNAIRE
                    Company   Union  Carbide Corporation



Do you presently manufacture activated carbon?          Yes X     No_
                                                            Y
Are you presently a supplier of activated carbon?       Yes       No_

If you are only a supplier, who manufactures, the
 activated carbon?           Company
                             Address
                             Phone
Do you supply an activated carbon suitable for
 hydrocarbon adsorption and desorption?                 Yes X     No
                                             Identification No.
                                             Type Carbon   See Attached

 Is  this activated  carbon used to control service
  station hydrocarbon emissions?  See attached           Yes	   No	
 If  yes, where?	See Attached	
 Do you regenerate activated  carbon?                     Yes X      No	

 If yes, method used    Thermal	


 Do you manufacture or supply a service station
  hydrocarbon emission control system?                   Yes	    NoL_x
 If yes, briefly describe:	
 Name and phone number of individual in your company to contact if
 additional questions arise.
                                    Name jAllan  Koeppel	
                              Phone  No. (312)  454-2000  -  Ext.  2102

-------
                                      A-10

Westvaco
                                      January 27,  1976
Mr. Michael  J.  Manos
Scott Environmental  Technology
Environmental  Services Department
2600 Cajon Boulevard
San Bernadino,  California  92411
                                             x
Dear Mr.  Manos:

In reference to your letter dated January 20, I  have  enclosed the  completed
questionnaire and would like to take a minute now to  review Westvaco's
achievements in the  vapor emission control  area.

Please check back in your records for my letter  to Mr.  Dyrle Quick dated
September 26, 1975.   This letter contained several enclosures outlining
the various patent work performed by Westvaco in the  early 1970's.

Most of the design concepts in the vapor control  field  were developed by Don
Tolles at our Charleston Research Laboratory.  Don's  work included the
following:

1.  "Methods of Controlling Hydrocarbon Vapor Emissions from Fuel  Storage
    Tanks"

2.  "Adsorber Designs Especially Suited for Applications Involving Regen-
    eration by Evacuation or Low Temperature Purge"

3.  "Regeneration Process for Adsorbents and Disposal of Recovered Vapors"

4.  "Adsorption and Regeneration Method Using Annular Cartridge Adsorption
    Elements"

5.  "Control of Refueling Losses Using Activated Carbon Adsorption" -
    Canister design and laboratory performance testing of a refueling
    vapor emission control suitcase-type adsorber, fully portable
   Chemical Division
   Carbon Department
   Covington, Virginia 24426
   Telephone: 703-962-2111

-------
Mr. Michael J. Manos                     A-ll
Page 2
January 27, 1976
6.  Technical assistance in the development of the RBW/Tokheim System.

Most of the reports and prototype information are currently not available
for out-of-company distribution and are listed as "confidential."  Release
of portions of this data could be arranged under a nondisclosure agreement.
However, please note that some sections of these reports were copied and
sent to Mr. Dyrle Quick on September 26, 1975.

Westvaco can provide suitable prototype design, including the fabrication
of an  experimental test unit along with the normal complement of engineering
consulting.  Anything  larger than the  initial pilot units would probably
have to be subcontracted out, but we would like a chance to review that
design before declining.

'Westvaco has been very active in refueling emission control applications;
we have the talent and the capability  and would certainly appreciate the
chance to get involved once more.  Samples of our products can be sent
upon request.  Product data information on our wood-based WV-A can be sent
upon request.

Read over the enclosed literature and  product data bulletins and let us
know what else we can  do to help out.

                                        Sincerely yours,
                                         Edward G.  Polito
                                         Environmental  Services  Engineer
 EGP/ckj

 Enclosures

-------
                                   A-12
                              Attachment
                    ACTIVATED CARBON QUESTIONNAIRE
                    Company    Westvaco	

Do you presently manufacture activated carbon?          Yes X     No_
Are you presently a supplier of activated carbon?       Yes X     No_
If you are only a supplier, who manufactures.the
 activated carbon?           Company	
                             Address
                             Phone
Do you supply an activated carbon suitable for
 hydrocarbon adsorption and desorption?                 Yes X     No_
                                             Identification No.WV-H and/or WV-A
                                             Type Carbon bituminous coal/wood based
Is this activated carbon used to control service
 station hydrocarbon emissions?                         Yesj(      No	
If yes, where?  Westvaco wood-based WV-A,  14x35,  and  coal-based,  8x30,  used in
                automotive industry in evaporative  fuel  canisters

Do you regenerate activated carbon?                     Yes	    No  X
If yes, method  used	
Do you manufacture  or  supply  a  service  station
 hydrocarbon  emission  control system?                   Yes	   NoX
If yes, briefly  describe:  Nqte:  However, we do produce activated carbon
 <:mtaMp for SUCh  Um'tS	
 Name and  phone number of  individual  in your  company  to  contact  if
 additional questions  arise.
                                   Name Edward G. Polito, Environmental Services Eng
                              Phone   Ho. JJQ3) 962-6081	

-------
                                            A-13
                                                                       Scott

                                           January  27,  1976
           Mr. Michael  J. Manos
           Project Manager
           Environmental  Services  Dept.
           Scott Environmental Technology  Inc.
           260 Cajon  Blvd.
           San Bernardino,  California  92M1

           Dear Mr. Manos:

           Enclosed,  please find the  completed  questionnaire sent to Witco
           in your  letter of January  20,  1976.

           Please feel  free to telephone  or  write me for any additional data
           your program may require.
                                           Very truly yours,
                                           C. M. Saffer, Jr.
                                           Technical Director
                                           Tel:  212-6H-6438
            CMSrli
            enclosure
WHco Chemical Corporation, 277 Park Avenue, New York. ,Nc* York 10017    Area Code 2l2Telephone 644-630*

-------
                                    A-14
                              Attachment
                    ACTIVATED CARBON QUESTIONNAIRE
                    Company _Witco Chemical Corp.

Do you presently manufacture activated carbon?          Yes_X__    No_
Are you presently a supplier of activated carbon?       Yes X     No_
If you are only a supplier, who manufactures the
 activated carbon?           Company	
                             Address           	      .	
                             Phone
Do you supply an activated carbon suitable for
 hydrocarbon adsorption and desorption?                 Yes X     No	
                                             Identification No. Grade 960
                                             Type Carbon  *xK> Granular
Could                    be                   '*         	
Ie this activated carbon used to control service
 station hydrocarbon emissions?                         Yes X     No	
If yes, where?	
Do  you  regenerate  activated carbon?                     Yes	    No X
If  yes, method used

Do  you  manufacture or supply a service station
 hydrocarbon  emission control system?                   Yes	    No X
If  yes, briefly describe;
Name and  phone number  of  individual  in your  company  to  contact  if
additional  questions arise.
                                   Name      Charles  M.  Saffer,  Jr.
                             Phone   No.      212-644-6438

-------
        CALGON
                             A-15
           « ^
    CORPORATION
    ••-	—
SUBSIDIARY OF MERCK & CO., INC.
ACTIVATED CARBON DIVISION  CALGON CORPORATION  CALGON CENTER  BOX 1346 PITTSBURGH, PA. 15230 (412)923-2345
         p \ V CO
         ^- !
           ju.o                     M*y 25' 1976
         -"
              lecwwtogjf

    Mr.  Michael  Manos
    Scott  Environmental  Technology
    2600 Cajon  Blvd. ,
    San  Bernardino, California 92411

    Dear Mike:

    It  was a pleasure  to meet you and Bob  and  to  discuss  the
    application  of granular carbon  in gasoline  vapor recovery.
    Following our conversation, I have  consulted  our Research
    staff to determine what method  they  use   for  the determination
    of  hydrocarbon heels.  They use  the  two  following methods:

         1 .   The carbon  disulfide technique

             The sample  is Soxhlet  extracted for
             18 hours; the carbon disulfide
             evaporates  at room temperature; the
             sample then is  injected in  the
             chromatograph .   Our Research staff
             which has a lot  of experience with
             the extraction  technique  prefers to
             use carbon  disulfide which they found
             to be the best  solvent for the  operation.
             Other solvents  can also be used.   The
             method is mostly used  with samples
             which contain compounds liquid  at room
             temperature.

         2.  Solid sample  injection technique

             This  method is  used  in particular with
             samples  which contain  low boilers.  It
             requires the  right  GLC equipment.

    Enclosed you  will find the  isotherm data of  the  most  common
    aliphatic and  aromatic hydrocarbons.  The  isotherms  are expressed
    in  percent  loading  on  activated carbon as  a  function  of the
    partial  pressure.  They  were  established at  different temperatures
                              PITTSBURGH ACTIVATED CARBONS

-------
                               A-16
Mr.  Michael Manos
May  25, 1976
Page 2
You will also find enclosed papers describing the general
adsorption correlation for aliphatic organics from which it
is possible to expand the adsorption isotherms to conditions
not shown on the enclosed tables and to calculate the isotherms
of other hydrocarbons.

We hope that our data are useful.  Please advise if there are
any questions or additional ways in which we may assist.

                                Very truly yours,
                                Bernard 1. Grandjacques
                                Development Engineer
BLG:rmk

Enclosures

cc:  Mr. Robert Kinne

-------
                                     A-17

      SCOTT ENVIROMiViEMT&L TECHNOLOGY
       A    SUBSIDIARY    OF    AMERICAN    BIOCULTURE,    INC.
                                                  2600 CAJON BLVD.
                                                  SAN BERNARDINO, CAL. 92411
                                                  (714) 887-2571

                                                  July  13, 1976
Mr. Elaine R. Joyce
Technical Service Director
Activated Carbon Products
Union Carbide Corporation
11709 Madison Avenue
Cleveland, Ohio 44107
Dear Elaine:
     Earlier this year we contacted you to solicit technical  information
on activated carbon products.  Upon review of the data  received we  have
selected one or more of your products, as shown on the  attached form,
for further evaluation.  We will be performing tests  on the activated
carbon to determine the working capacity in  conjunction with  various
loading rates, stripping rates, temperatures, humidities,  gasoline
characteristics, etc.

     Analyses of the results will be  enhanced if v/e  can relate our  findings
to carbon characteristics.  We ask your assistance  in providing the
inspection data, specified on the attached  form,  for the actual batch
of carbon we will be testing.  Additionally, we are  interested in knowing
the relative costs for various carbons based on orders of 100 Ibs.,
F.O.B. your warehouse.
                                        Sincerely,
                                        Michael J.  Manoa
                                        Project Engineer
rs

Attachment
            PLUMSTEADVILLE. PA.  • SAN BERNARDINO. CALIF. .  MADISON HEIGHTS. MICH.

-------
                                       A-18
                      ACTIVATED CARBON INSPECTION DATA FORM
 Brand
 Product
 Type
 Form
 Mesh Size
 Amount Required

*Apparent Density, g/cc
*Hardness
*% Ash
*% Moisture
* Kindling Point,  °c
 Pore Volume, cc/g
 Surface Area, ni2/g
 Iodine No.
 Activity, %CCL4
 Retentivity, %CCL4
 Cost/100 Ibs.
UNION CARBIDE
MBV
Coal
Pell.
12 x 28
25 Ibs,
MBV
Coal
Pell.
4x6
2 Ibs.
MBV
Coal
Pell.
8 x 10
2 Ibs.
*These values may be "typical" numbers

-------
                                     A-19
      SCOTT EiWBROfcSf^E-eOTAL, TECHNOLOGY BBflC
            SUB. SIDIARY   OF    AMERICAN    BIOCULTURE,    INC.
                                                  2600 CAJON BLVD.
                                                  SAN BERNARDINO, CAL. 92411
                                                  (714) 887-2671
                                                  July  13,  1976
Mr. Bernard L. GrandJacques
Activated Carbon Division
Calgon Corporation
Box 1346
Pittsburgh, Pennsylvania 15230

Dear Bernard:

     Earlier this year we contacted you to solicit technical  information
on activated carbon products.  Upon review of the data received  we have
selected one or more of your products, as shown on the attached  form,
for further evaluation.  We will be performing tests  on  the activated
carbon to determine the working capacity in  conjunction  with various
loading rates, stripping rates, temperatures, humidities,  gasoline
characteristics, etc.

     Analyses of the results will  be  enhanced  if  we  can  relate our findings
to carbon characteristics.  We ask your  assistance in providing the
inspection  data, specified  on the  attached  form,  for the actual batch
of carbon we will  be testing.  Additionally, we  are  interested in knowing
the relative  costs for various carbons based on  orders of 100 Ibs.,
F.O.B. your warehouse.

                                        Sincerely,
                                        Michael J. Manos
                                        Project Engineer
 TS

 Attachment
 P. S. Bob Kinne and I enjoyed having the opportunity to meet you during your
 recent trip to the West Coast.  Your assistance has been extremely helpful,
 and the technical information on laboratory test techniques has been beneficial
 in solving some of our analytical problems.
             PLUMSTEADVILLE. PA. • SAN BERNARDINO, CALIF.  • MADISON HEIGHTS, MICH.

-------
                                       A-20
                      ACTIVATED CARBON INSPECTION DATA FORM
  Brand
  Product
  Type
  Form
  Mesh  Size
  Amount Required

 *Apparent Density,  g/cc
     •
 *Hardness
 *l Ash
 *Z Moisture
 ^Kindling Point, °c
  Pore Volume, cc/g
  Surface Area, m2/g
  Iodine No.
  Activity, ZCCLA
  Retentivity, %CCL4
  Cost/100 Ibs.
CALGON
BPX
Coal
Gran.
•'8 x 30
**










$0.71
 *These values may be "typical" numbers
**We currently have a 5-gallon container of BPX 8 x 30 labeled 1-85-931
  (batch no.?) which is sufficient for the scheduled tests

-------
                                     A-21
     SCOTT ENVIRONMENTAL TECHNOLOGY INC.
      A    SUBSIDIARY    OF    AMERICAN   BIOCULTURE.    INC.
                                                 2600 CAJON BLVD.
                                                 SAN BERNARDINO, CAL. 92411
                                                 (7141 887-2671
                                                 July 13, 1976
Mr. John R. Conlisk
Laboratory Supervisor
Product Development Department
ICI United States, Inc.
Wilmington, Delaware 19897

Dear John:

     Earlier this year we contacted you to solicit technical information
on activated carbon products.  Upon review of the data received we have
selected one or more of your products, as shown on the attached form,
for further evaluation.  We will be performing tests on  the activated
carbon to  determine the working capacity in  conjunction  with various
loading rates, stripping rates, temperatures, humidities,  gasoline
characteristics,  etc.

     Analyses of  the results will  be  enhanced  if  we  can  relate our  findings
to carbon  characteristics.  We  ask your  assistance in providing the
inspection data,  specified  on  the  attached  form,  for the actual batch
of carbon  we will be testing.   Additionally, we are interested in knowing
the relative costs for various  carbons based on orders of 100 Ibs.,
F.O.B. your warehouse.

                                        Sincerely,
                                        Michael J. Manos
                                        Project Engineer
 rs

 Attachment
             PLUMSTEADVILLE. PA, • SAN BERNARDINO, CALIF.  • MADISON HEIGHTS. MICH.

-------
                                       A-22
                       ACTIVATED CARBON INSPECTION DATA FORM
  Brand
  Product
  Type
  Form
  Mesh Size
  Amount Required

 * Apparent Density, g/cc
 * Hardness
 * % Ash
 *% Moisture
 * Kindling Point,  °c
  Pore Volume, cc/g
  Surface Area, m2/g
  Iodine No.
  Activity, %CCL4
  Retentivity, %CCL4
  Cost/100 Ibs.
ICI
DXL-0
Li an.
Gran.
12 x 30
**










 *These values may be "typical" numbers
**We currently have 20 Ibs, of DXL-0-6425 on hand

-------
                                     A-23
      SCOTT EE3VBE3OMRflEBflTAi- TECHNOLOGY IMC.
       A    SUBSIDIARY    OF    AMERICAN    BIOCULTURE.    INC.
                                                  2600 CAJON BLVD.
                                                  SAN BERNARDINO, CAl_ 92411
                                                  (714) 887-2571
                                                  July  13,  1976
Mr. E. G. Polito
Environmental Services Engineer
Carbon Department
Chemical Division of Westvaco
Covington, Virginia 24426

Dear Ed:

     Earlier this year we contacted you to solicit technical  information
on activated carbon products.  Upon review of  the data  received we  have
selected  one or more  of your products, as shown  on the  attached  form,
for  further evaluation.  We will be performing tests  on the activated
carbon  to determine  the working capacity  in  conjunction with various
loading rates,  stripping rates, temperatures,  humidities,  gasoline
characteristics,  etc.

      Analyses  of the results will be  enhanced if we can relate our.findings
 to carbon characteristics.   We ask your assistance in providing the
 inspection data, specified on the attached form, for the actual batch
 of carbon we will be testing.  Additionally, we are interested in knowing
 the relative costs for various carbons based on orders of  100 Ibs.,
 F.O.B. your warehouse.

                                        Sincerely,
                                        Michael  J. Manos
                                        Project  Engineer
  rs

  Attachment
 P.  S.  I enjoyed having the opportunity to meet with you at the APCA
 Convention  in Portland and appreciate the assistance you have provided us
 on  this program.
              PLUMSTEADVILLE, PA. • SAN BERNARDINO, CALIF. • MADISON HEIGHTS, MICH.

-------
                                       A-24
                       ACTIVATED CARBON INSPECTION DATA FORM
   Brand
   Product
   Type
   Form
   Mesh Size
   Amount Required

*  Apparent Density,  g/cc
*  Hardness
*  % Ash
*  1 Moisture
*  Kindling Point, °c
   Pore Volume, cc/g
   Surface  Area, m^/g
   Iodine No.
   Activity, %CCL4
   Retentivity, %CCL4
   Cost/100 Ibs.
WESTVA~CO
WV-A
Wood
Gran.
•14 x 35
5 Ibs.










*These values may be "typical" numbers

-------
                                         A-25
ETHYL CORPORATION	
                                                        Petroleum Chemicals Division
                                                       3700 CHERRY AVENUE 'LONG BEACH. CAL 90807
                                                              TELEPHONE 213 GA S-5575
                                                          July 26, 1976
        Scott Research Laboratories,  Inc.
        2600 Cajon Blvd.
        San Bernardino,  CA  92406

        Attention:   Mr.  Dyrle Quick

        Dear Mr.  Quick:

        We have received and tested your three samples of gasoline as you
        requested.   Following are the results of our tests:

        Sample  Identification               1         2         3

        Gravity  °API                      57'3      56'3       *2'7
        Gravity,   API                       .         0>Q17A     Q^Q
        Pb Content, g/gallon               0 Q       a ,        ,00
        **    _.           .                 O • o       O • /        i j * o
        Vapor Pressure, psi

        *Pb  Content by Atomic Absorption

                                         Yours very truly,

                                         B. M. Phillips
                                         Laboratory Supervisor
                                         By:  R. C. Rickhoff
        RCR:dc
        cc:  R. T. Peterson
                                                           DECEIVED
                                                         Scott Environmental Technology

-------
   UNION
 CARBIDE
 I '"t '.'j 141 I'l V 1 ' >

   r  j 'io,x t,OH /

  Kl-jl !. 1)1 »l( ) -14 I I) I
                             A-26

          UNION CARBIDE  CORPORATION
          CARBON  PRODUCTS DIVISION
                                            RECEIVED
                                               SLP - 7 J9/6

                                          Scott Environmental Technology
                  1 I 709 MADISON AVENUE, CLEVELAND, 1>HIO 44107 •  rFLFPHONL: 210
                                      September 1,  1976
Mr. Michael J. Manos
Scott Environmental Technology, Inc.
2600 Cajon Boulevard
San Bernadino, California  92411

Dear Mr. Manos:

     The physical properties of the activated  carbon  samples  you received
for your gasoline vapor adsorption tests  are as  follows:
     Carbon Grade
     Mesh Size

     Activity, %
     Bulk Density,7=
     Hardness, %
     Ash, %
     Screen Analysis, 7»

12/28
58.7
, 0.40
95.4
5.9
MBV
8/10
61.3
0.43
97.8
5.7

4/6
68.9
0.41
98
—
SFV
12/28
66.4
0.48
94.3
6.2
        On
        On
   4 Mesh
   6 Mesh
On 8 Mesh
On 10 Mesh
On 12 Mesh
   14 Mesh
   20 Mesh
   28 Mesh
        On
        On
        On
        On Pan
 0.2
 7.1
55.8
34,8
 2.1
                                  0.4
                                 99.1
                                  0.4
                         0.1
                           0-10
                           90-100
                           0-10
                                                 0.1
 2.7
 8.5
67.4
20.7
 0.7
     I am arranging  to  send  you  the  second  lot of samples as you requested,
If you have any questions, or  if we  can be  of help,  please let us know.

                                          Very truly yours,
                                           Blaine R.  Joyc/
                                           Technical  Service
                                           Activated  Carbon Products
BRJ/jj

-------
                                   A-27

Westvaco
                                      September 26, 1975
                                                               RECEIVED

                                                                  OCT -'4 19/5

Mr. Dyrle Quick                                                           .,,
Scott Environmental Technology                                Scott Env.ronmental
2600 Cajon Boulevard
San Bernadino, California  92411

Dear Mr. Quick:

In reference to our previous phone conversation, I have recently been in
contact with Don Tolles of our Charleston Research Laboratory.  Don has
done quite a bit of work on refueling vapor loss control with activated
carbon and holds several patents.  Two of Don's reports that should be
of particular interest to you are not available for out-of-company
distribution.  However, it is possible to send you some excerpts from
these reports, notably data on working capacity and general regeneration
techniques applied.

Looking at the earliest of the two reports (December 18, 1972) and
referring to figure 3, please note the following conditions:

                influent vapor concentration = 3%

                equilibrium loading = 17.52 (80°F)

                regeneration temp. = 200°F

                regeneration vacuum = 27" Hg

                gives working capacity of 7%

           "bleed" = % residual  butane in air = 0.25$  (point  C)

In the report dated July 18, 1973, note that saturation capacity has been
raised to a maximum of 23.4%, leaving an adjusted working  capacity of
 Chemical Division
 Carbon Technical Center
 Covington, Virginia 24426
 Telephone: 703-962-2111

-------
Mr. Dyrle Quick                    A'28
Page 2
September 26, 1975
greater than 16%.   Note, also, though, that we are using a  thinner bed of
carbon as well  as  higher regeneration temperatures (269-295°F).   The
original report discusses the experimental  results of three methods of
regeneration:  vacuum; air purge at atmospheric pressure; and reduced
pressure air purge, of which vacuum regeneration applies the smallest
load on vapor recovery equipment.   This report also discusses the Model  II
prototype canister dimensions (which is not available for release).
Essentially, this  is a suitcase-type adsorber—portable, and designed to
be used in multiples, connected in series or parallel combinations at
service station sites for adsorption of vented fuel vapors.

As you can see, Westvaco has been very active in this field; and very
comprehensive testing has been done using our carbons.   If  further interest
in controlling  refueling vapor losses arises, I suggest that you get  back
in touch with me,  Mr. Quick.

Look over the enclosed material and product data bulletins  carefully.   I
sincerely hope  that I have successfully covered all of the  points in
question.

                                      Sincerely,
                                      Edward G.  Pol Ho
                                      Environmental  Services Engineer
EGP/ckj

Enclosures

-------
                                          B-l

                          APPENDIX B.  Matrix
Test No:
Carbon:
  (react)
  /diff.N
  \batch/
  {satur)
  (satur)
Canister:
Fuel :
Purge:
Ambient:
1 -
1

W-l UCAR MBV 12 x 28
W-2 UCAR MBV 8 x 10
W-3 UCAR MBV 4x6
W-4 UCAR MBV 12 x 28
W-5 UCAR MBV 12 x 28
W-6 UCAR MBV 12 x 28
X-l CALGON BPX 1 2 x 30
X-2 CALGON BPX 12 x 30
Y-l WESTVACO 14 x 35
Z-l I.C.I. 12 x 30
W-7 UCAR SFV 12 x 28
H x A
2H x A
H x 2A
A 9# RVP Leaded
B 9# RVP Unleaded
C 14# RVP Unleaded
1-1 75° F. Air 4000cc/m
1-3 150° F. Air 4000cc/m
1-4 200° F. Air 4000cc/m
1-5 75° F. Air 20,000cc/m
1-6 75° F. Air 40,000cc/m
2-1 VAC 0 75° F. abs. 100 mm
2-2 VAC @ 150°F. abs. 100 mm
2-3 VAC @ 200° F. abs. 100 mm
2-4 VAC 0 75°F. abs. 25 mm
75° F. Dry Air
50° F. Dry Air
90° F. Dry Air
75° F. 40* R.H,
75° F. 80* R.H.
X








-

X


2
3
4
5
6
7
8
9
10 '
11
12
13|l4
15
16

X










X










X


X











X

X








X




X









FT


X









X



X







X















X

X


X






X






X



X










X







X







X







I
i


X


X


X




X


X










X


X










X



X


X


X


X


x


X


X


X


X


X


X


X


X


X


X


X



X


i
i
1
t
I

1
1
X








X








X








x








X







1
X








X








X








X








X








X








X








X








X








X








X




X




X




X




X




X




xi x








X




X




X




X





X





X





X





X

-------
                                            B-2
Test No:
Carbon:
                                     2-mi
  (react)
  AMff.N
  \batchy
  (satur)
             W-l  UCAR MBV   12 x 28
             W-2  UCAR MBV   8 x 10
             W-3  UCAR MBV   4x6
             W-4  UCAR MBV  K x 28
             W-5  UCAR MBV  12 x 28
             W-6  UCAR MBV  12 x 28
             X-l  CALGON BPX  12  x  30
    (satur)   X-2  CALGON BPX  12  x  30
             Y-l  WESTVACO   14 x 35
             2-1  I.C.I.     12 x 30

 Canister:        H x A
                 2H x A
                  H x 2A

 Fuel:       A  9# RVP Leaded
            B  9# RVP Unleaded
            C 14* RVP Unleaded

   *ge:      1-1 75°  F. Air 4000cc/m
            1-3 150° F. Air  4000cc/m
            1-4 200° F. Air  4000cc/m
            1-5 75°  F. Air 20,000cc/m
            1-6 75°  F. Air 40,000cc/m
            2-1 VAC  9 75°F. abs. 100 mm
            2-2 VAC  
-------
B-3
Test No: 3-12345678

Carbon: U-l UCAR MBV 12x28 X X XJX X X *
W-2 UCAR MBV 8 x 10
W-3 UCAR MBV 4x6
(react) W-4 UCAR MBV 12 x 28
(Plfl\ W-5 UCAR MBV 12 x 28
\batcrv • - ••" •• "
(satur) W-6 UCAR MBV 12 x 28
X-l CALGON BPX 1 2 x 30 X
(satur) X-2 CALGON BPX 12 x 30
Y-l WESTVACO 14 x 35
Z-l I.C.I. 12 x 30

Canister: HxA XXXXXXXX
2H x A
H x 2A

Fuel: A 9# RVP Leaded X X X X
B 9# RVP Unleaded
C 14# RVP Unleaded X X X X

Purge: 1-1 75° F. Air 4000cc/m x
1-3 150° F. Air 4000cc/m
1-4 200° F, Air 4000cc/m
1-5 75° F. Air 20,000cc/m i
1-6 75° F Air 4n,flflOcc/m
2-1 VAC @ 75°F. abs. 100 mm X )(
: x y
2-2 VAC @ 150°F abs. 100 mm'
2-3 VAC @ 200°F. abs. 100 mml 	 x 	
2-4 VAC @ 75°F. abs. 25 mm , 	 x x

Ambient: 75° F. Dry Air 1_ JL J X X Xi X| X
50° F. Dry Air
90° F. Dry Air
75^ F. 40% R.H.
75° F. 80% R.H.
9 10 11 12 13 14 15 16

JLJL





XX X X X X




XXXXXXXX



X X X X X
X
iL x

X





X X
X X x
x x

.JLJL JLJL JLJL JLJL



•

-------
Test No: 4 -

Carbon: W-l UCAR MBV 12 x 28
W-2 UCAR MBV 8 x 10
W-3 UCAR MBV 4x6
(react) W-4 UCAR MBV 12 x 28
(batch) W"5 UCAR MBV 12 x 28
(satur) W-6 UCAR MBV 12 x 28
X-l CALGON BPX 1 2 x 30
(satur) X-2 CALGON BPX 12 x 30
Y-l WESTVACO 14 x 35
Z-l I.C.I. 12 x 30

Canister: H x A
2H x A
H x 2A

Fuel : A 9# RVP Leaded
B 91 RVP Unleaded
C 14# RVP Unleaded

Purge: 1-1 75° F. Air 4000cc/m
1-3 150° F. Air 4000cc/m
1-4 200° F. Air 4000cc/m
1-5 75° F. Air 20,000cc/m
1-6 75° F. Air 40,000cc/m
2-1 VAC 0 75°F. abs. 100 mm
2-2 VAC 0 150°F. abs. 100 mm
2-3 VAC 0 200°F. abs. 100 mm
2-4 VAC 0 75°F. abs. 25 mm

Ambient: • 75° F. Dry Air
50° F. Dry Air
90° F. Dry Air
75* F. 40fc R.H.
75° F. 80* R.H.
1

X








-

X



X



X


I
•
1

1

i
1

X




2

X










X



X



X









X




3

X










X



X




X








X




4

X










X



X





X







X




5

X










X.





X

X









X




6

X










x





X


X








X




7







X




X





X

X









X




8







X




x





X


X








X




9









X


x



X



X









X




10

X










x




X


X









X




11







X




x



X



X









X




12







X




x



X




X








X




13|







X




x



X





X







X




14

X










X



X



X









X




15

X










x



X




X








X




16

X










Y



X





X







X




TEST CELL #4

-------
                                    APPENDIX C—TEST RESULTS
Test
1-1
1-2
1-3
1-4
1-5
(2)1-11
1-7
1-8
1-9
1-10
1-11
1-12
1-13
1-14
1-15
1-16
Date
1976
8/20
9/17
9/02
8/27
8/19
9/15
8/25
9/09
9/13
9/16
9/07
8/26
9/20
9/21
9/14
9/15
Cycles
21
21
21
22
21
21
21
21
21
21
22
21
22
21
21
22
Last
Capacity
gms
10.00
9.25
13.70
6.88
8.70
7.95
5.67
7.30
7.15
10.60
7.45
8.60
7.95
10.40
7.75
9.00
Bed
Size
gms
170
160
120
160
160
160
160**
160**
160
149.3
160
160
160
160
158. 25A
192.95*
Working
Capacity
gms/100 gms
5.88
5.78
11.42
4.30
5.44
4.97
3.54
4.56
4.47
7.10
4.66
5.38
4.97
6.50
4.90
4.66
Vapor
Concentration
%
52
52
55
56
52
53
50
53
52
53
54
57
32
62
52
53
Corrected*
qms/100
5.93
5.83
11.37
4.22
5.49
4.99
3.66
4.58
4.52
7.12
4.64
5.26
5.70
6.21
4.95
4.68
 *Corrected to 53.56% Vapor Concentration
**Moisture equilibrated but not yet vapor saturated
 AMoisture equilibrated @ 40%
 AMoisture equip!ibrated @ 80%

-------
                                                 TEST RESULTS
Test
2-1
2-2
2-3
2-4
2-5
2-6
2-7
thru
2-16
Date
1976
8/20
8/23
8/25
9/16
9/16
9/02
9/03
thru
9/14
Cycles
21
21
21
21
21
21


1023
Last
Capacity
gms
9.42
21.75
18.42
7.90
8.25
9.70


7.00
Bed
Size
gms
160
320
320
160
160
160


160
Working
Capacity
gms /I 00 qms
5.89
6.80
5.76
4.94
5.16
6.06


4.3ft
Vapor
Concentration
%
55
55
53
52
52
50


5-3
Corrected*
gms/100
5.84
6.77
5.77
4.99
5.21
6.18


/i on
                                                                                                                 o
                                                                                                                 I
*Corrected to 53.56% vapor concentration

-------
TEST RESULTS
Test
3-1
3-2
3-3
3-4
3-4
3-4
3-4
3-5
3-6
.3-6
3-7
3-8
3-9
3-10
3-11
3-11
3-11
3-12
3-12
3-12
Date
1976
8/20
8/25
8/27
8/26
9/03
9/07
9/08
9/15
9/14
9/15
9/15
9/16
9/16
8/19
9/03
9/07
9/08
9/03
9/07
9/08
*Corrected to
Corrected to
Cycles
21
21
21
21
1
1
1
1
21
1
1
1
1
21
1
1
1
1
1
1
53.56% vapor
98.0% vapor
Last
Capacity
gms
9.45
0.60
2.10
6.27
6.30
6.10
6.94
11.40
8.35
7.90
8.60
7.85
8.55
8.06
8.35
8.50
9.61
10.35
10.60
11.15
concentration
concentration
Bed
Size
gms
160
160
160
160
160
160
160
160
160
160
160
160
160
160
160
160
160
160
160
160


Working
Capacity
gms/ 100 gms
5.91
0.38
1.31
3.92
3.94
3.81
4.34
7.13
5.22
4.94
5.38
4.91
5.34
5.04
5.22
5.31
6.01
6.47
6.63
6.97


   Vapor
Concentration
     56
     55
     51
     54
     50
     51
     51
     97
     92
     97
     97
     94
     94
     56
     50
   . 51
    51
    50
    51
    51
Corrected*
 qms/100
    5.83
    0.33
    1.40
    3.90
    4.05
    3.90
    4.42
    7.16A
    5.42A
   4.97A
   5.41A
   5.05A
   5.48A
   4.96
   5.33
   5.40
   6.09
   6.58
   6.71
   7,05
o
I
co

-------
                                                 TEST  RESULTS
Test
3-13
3-13
3-13
3-14
3-14
3-14
3-15
3-16
3-16
3-16
Date
1976
9/09
9/10
9/13
9/09
9/10
9/13
9/16
9/09
9/10
9/13
Cycles
1
1
1
1
1
1
1
1
1
1
Last
Capacity
gms
7.30
10.35
8.20
9.80
11.15
9.70
9.30
5.40
5.90
6.50
Bed
Size
gms
160
160
160
160
160
160
160
160
160
160
Working
Capacity
gms/100 qms
4.56
6.47
5.13
6.13
6.97
6.06
5.81
3.38
3.69
4.06
Vapor
Concentration
%
51
50
50
51
50
50
94
51
50
50
Corrected
qms/ 100
4.65
6.58
5.24
6.21
7.08
6.18
5.95A
3.46
3.80
4.18
                                                                                                                o
                                                                                                                -pi
Corrected to 98% vapor concentration

-------
    Test
    4-1
    4-2
    4-3
    4-4
    4-5
   4-6
(3)4-7
   4-8
(2)1-3
   4-10
   4-11
   4-12
   4-13
   4-14
   4-15
   4-16
  Date
  1976
 8/20
 8/23
 9/02
 9/03
 9/15
 9/16
 9/17
 9/17
 9/17
 8/25
 8/19
 9/07
 9/08
 9/13
9/09
9/10
Cycles
 21
 21
 21
 21
 21
 21
 21
 21
 21
 21
 21
 23
 21
 22
 21
 21

Last
Capacity
gms
9.87
9.87
10.10
9.00
10.70
10.15
10.20
10.10
13.60
8.31
8.96
7.50
8.70
10.30
10.05
10.05
TEST
Bed
Size
gms
160
160
160
160
160
160
160
160
120
160
160
160
160
160
160
160
RESULTS
Working
Capacity
gms /I 00 gms
6.17
6.17
6.31
5.63
6.69
6.34
6.38
6.31
11.33
5.19
5.60
4.69
5.44
6.44
6.28
6.28
                                                                         Vapor
                                                                      Concentration
 53
 54
 51
 48
 98
 95
 94
 94
 53
 55
 53
 48
 47
 59
47
59
Corrected*
 gms/100
    6.19
    6.15
    6.40
    5.82
    6.69A
    6.44A
    6.52A
    6.45A
  11.36
   5.15
   5.62
   4.87
   5.65
   6.26
   6.49
   6.10
en
  *Corrected  to  53.56% vapor concentration
  Corrected  to  98.00% vapor concentration

-------
 BRAND

 TYPE
 FORM
 MESH SIZE

 APPARENT DENSITY, g/cc
 HARDNESS
 % ASH
 % MOISTURE
 KINDLING POINT,  <>C
 PORE VOLUME, cc/g
 SURFACE AREA, m2/g
 IODINE NO.
 ACTIVITY, %CCL4
 RETENTIVITY, SCCL4
 COST/100 Ibs.
 DENSITY AS  REC'D./*  H20
DENSITY ACCLM'D/% H20
DENSITY DRY
40% R.H./X H20
80% R.H./% H20
                                                               APPENDIX D
                                                        Carbon  Specifications

                                                    ACTIVATED CARBON INSPECTION DATA FORM
_
7-1
Lign.
Gran.
12 x 30
0.414 (dry)
75 - 80*
16 fi
13.8
433
0.95
625 	
599
__ 42.1 wt.%
1 C O -.«. »

.00
0.448/9.8
OA 1 1 71 c
•411/1. b
0.405

** 	
pica!" numbers

-i
	 	 	 ,
Wood
Gran.
14 x 35
0.25 - .32*
50 - 60*
	 8% MAX*
5-6* MAX*
330° F.*
.8-1.0 *
~-
1200 *
1000 MIN*
60-100 *
— .- —
5-30 *
$55.50
0.310/U,?
0.274/3.5
0.264
- 	
' 	 — — —— ^__

X-l
Coal
Gran.
8 x 30
0.46*
94.2
4.92
0.55
920
960
1045
36.8
27.4
$71.00
0.454/Q.4
0.452/0
	 0.452



W-l
Coal
Gran.
12 x 28
— ,
0.40
95.4
5.9
370
1000
950
58.7
29
$88.00
0.416/11.1
0.380/2.5
0.370
0.411/10.0
0-452/18.1
-, 	
H-2
— • ,1-
Coal
Pellet
3 x 10
OAO
	 97.8
5.7

1000
QCf)
61.3


0.411/0.2
0.412/0.5
0.410



W-3
Coal
Pellet
4x6

.41
98
--

1000
11UO
68.9

$82.00
0.377/3.7
0.388/6.4
0.363



W-5
2nd lot W-l
Coal
Gran.
12 x 28








0.335/3.3
0.338/4.1
0.324



W-7
Gran.
14 x 28

0.48
94.3
6.2

1100
1100
66.4


0.444/3.3
0.447/3.9
0.429



-------
                          	  D-2
                          l™*^—"^^^^^^•"""^••^••^^^••B      __^_^^^^^^^^^^^
                                TECHNICAL REPORT DATA
                          (Please read Inunctions on the reverse befwcompletin*)
EPA-600/2-77-057
                                3. RECIPIENTS ACCESSION NO.
Control Characteristics of Carbon Beds for Gasoline
   Vapor Emissions
                                6. REPORT DATE
                                 February 1977	
                                6. PERFORMING ORGANIZATION CODE
Michael J.  Manos and Warren C. Kelly
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Scott Environmental Technology, Inc.
2600 Cajon Boulevard
San Bernardino, California  92411
12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC 27711
                                                      8. PERFORMING ORGANIZATION REPORT N
                                10. PROGRAM ELEMENT NO.
                                1AB604; RQAP 21AXM
                                11. CONTRACT/GRANT NO.

                                68-02-2140

                                13. TYPE OF REPORT AND PERIOD COVERED
                                Final; 1/76-1/77    ___-- -
1S. SUPPLEMENTARY NOTES TTPOT T3TTJ

Drop 62, 919/541-25T7RL'RTP
                                14. SPONSORING AGENCY CODE

                                  EPA/600/13

                officer for this report is M. Samfield, MaiT
16. ABSTRACT
          The report gives results of a study of the practical working capacity oT
 activated carbon to cyclically adsorb gasoline vapor which would otherwise be lost to
 the atmosphere; e.g. , during gasoline transfer operations at a service station
 Quantitative measurements,  made in the laboratory, were extrapolated to represent
 typical operation of a carbon control system at a service station^mninc 50 000 gal-
 lons of gasoline per month.   Eight types of activated carbon from four manufacturers
 were evaluated to determine  working capacity, basically defined as the amount of
 gasoline vapor which could be cyclically adsorbed per 100 grams of virein activated
 carbon  Tests were conducted at various levels of fuel volatility, lead lontent,
 carbon bed shape, ambient temperature and humidity, purge air flow rate and temper-
 ature,  and  vacuum-stripping  pressure  and temperature.
17.
       KEY WORDS AND DOCUMENT ANALYSIS
a.
                     b.lDENTIFIERS/QPENENDFn
                                                                      COS AT I Fiel
 Air Pollution
 Gasoline
 Vapors
 Transferring
 Activated Carbon
 Adsorption
Measurement
Air Pollution Control
Stationary Sources
Service Stations
                                              13B
                                              21D
                                              07D
                                              14B
                                              11G
18. DISTRIBUTION STATEMENT


 Unlimited


EPA Form 2220-1 (9-73)
                     19.bfcL-UHITY CLASS (This Report/
                     .Unclassified
                     2°f SECURITY CLASS
                     Unclassified
                         21-NO.OFPAUtS

                              113
                         22. PRICE

-------