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
------- |