United States Office of Mobile Source Air Pollution Control EPA 460/3-84-014
Environmental Protection Emission Control Technology Division January 1985
Agency 2565 Plymouth Road
Ann Arbor, Michigan 48105
Air
&EPA The Effect of Methanol on
Evaporative Canister Charcoal
Capacity
Disclaimer
The data presented in this report should be viewed with care,
and any possible conclusions based on these data should be
considered tentative at best due to the inconsistent treatment
of the charcoal samples during the test program. A more
consistently performed extension of this study is underway at
Southwest Research Institute, which may confirm the data in
this report.
-------
EPA 460/3-84-014
The Effect of Methanol on Evaporative-
Canister Charcoal Capacity
by
Mary Ann Warner-Selph
Southwest Research Institute
6220 Culebra Road
San Antonio, Texas 78284
Contract No. 68-03-3162
Work Assignment 12
EPA Project Officers: Robert J. Garbe
Craig A. Harvey
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
Office of Mobile Source Air Pollution Control
Emission Control Technology Division
2565 Plymouth Road
Ann Arbor, Michigan 48105
January 1985
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are available
free of charge to Federal employees, current contractors and grantees, and
nonprofit organizations - in limited quantities - from 2565 Plymouth Road, Ann
Arbor, Michigan 48105.
This report was furnished to the Environmental Protection Agency by Southwest
Research Institute, 6220 Culebra Road, San Antonio, Texas, in fulfillment of
Work Assignment No. 12 of Contract No. 68-03-3162. The contents of this
report are reproduced herein as received from Southwest Research Institute.
The opinions, findings, and conclusions expressed are those of the author and
not necessarily those of the Environmental Protection Agency. Mention of
company or product names is not to be considered as an endorsement by the
Environmental Protection Agency.
Publication No. EPA 460/3-84-014
ii
-------
FOREWORD
This project was conducted for the U.S. Environmental Protection Agency
by the Department of Emissions Research, Southwest Research Institute. The
work was carried out between February and August 1984 under EPA Contract
No. 68-03-3162, Work Assignment 12. It was identified within Southwest
Research Institute as Project 03-7338-012. The EPA Project Officers were Mr.
Robert J. Garbe and Mr. Craig A. Harvey of the Office of Mobile Source Air
Pollution Control, Emission Control Technology Division, Environmental
Protection Agency, 2565 Plymouth Road, Ann Arbor, Michigan. The Southwest
Research Institute Project Manager was Charles T. Hare, and the Project
Leader was Mary Ann Warner-Selph.
iii
-------
ABSTRACT
This program involved the evaluation of four types of untised evaporative
canister charcoal with a hydrocarbon-only blend and a hydrocarbon-methanol
blend. The HC blend consisted of 77% paraffins (butane), 18% olefins
(isobutylene) and 5% aromatics (toluene) by weight. The HC-methanol blend
was composed of 73% butane, 17% isobutylene, 5% toluene, and 5% methanol by
weight. Tests were conducted on a bench-scale apparatus designed to load each
blend onto separate sets of twelve reduced-size mini-canisters, and to
subsequently purge off the hydrocarbons. The charcoals were evaluated by the
measurement of retained charcoal weight gain after purging, time to
hydrocarbon breakthrough, and charcoal working capacity. The mini-canisters
which were loaded with the methanol blend, had shorter breakthrough times,
retained less weight gain after purge, and had lower working capacities than did
mini-canisters tested with the hydrocarbon blend only. These methanol blend
mini-canisters also underwent less simulated aging than the hydrocarbon blend
canisters in this program, since they were only exposed to 40% as much total
vapor.
iv
-------
TABLE OF CONTENTS
Page
FOREWORD iii
ABSTRACT iv
LIST OF FIGURES vi
I. INTRODUCTION 1
II. PROCEDURES AND INSTRUMENTATION 2
A. Development of Procedure and Instrumentation 2
1. Fuel Delivery 2
2. Composition of Hydrocarbon and Methanol
Blends 3
3. Load and Purge Cycles 3
Mini-Canisters 4
B. Description of Charcoal Evaluation Apparatus 5
1. Mini-Canisters 5
2. Charcoal 5
3. Hydrocarbon and Methanol Blend Compositions 9
Hydrocarbon Breakthrough 9
III RESULTS 14
REFERENCES 19
APPENDICES
v
-------
LIST OF FIGURES
Figure Page
1 Several Views of the Charcoal Evaluation Apparatus 6
2 Flow Schematic of Charcoal Evaluation Apparatus 7
3 Purge Cycle Monitored on 10,000 ppmC and 1000 ppmC
Ranges 10
4 Hydrocarbon Breakthrough at 100 ppmC 11
5 Hydrocarbon Breakthrough at 1000 ppmC 12
6 Average Daily Weight Gain After Purge of Four
Types of Evaporative Canister Charcoal 15
7 Average Daily Breakthrough Time of Four Types of
Evaporative Canister Charcoal 16
S Average Daily Working Capacity of Four Types of
Evaporative Canister Charcoal 18
vi
-------
I. INTRODUCTION
Evaporative emissions from gasoline vehicles are controlled by the use of
charcoal canisters. The Sealed Housing for Evaporative Determinations (SHED)
test was developed to measure vehicle evaporative emissions. This test is used
to confirm canister effectiveness in controlling evaporative emissions during
simulated vehicle operation. A study conducted by the Department of Energy
in 1980vD* with 10 percent methanol in gasoline raised concerns over whether
methanol in gasoline reduces canister effectiveness.
The purpose of this program was to design a laboratory bench-scale
apparatus for evaluating the effects of a methanol-hydrocarbon blend on
charcoal from evaporative emissions canisters. Two sets of mini-canisters were
filled with activated charcoal from new evaporative canisters and aged by
repetitively loading and purging with the hydrocarbon and the methanol-
hydrocarbon blends, respectively. Hydrocarbon breakthrough times and
charcoal weight gains (after purging) were monitored throughout the program.
In addition, charcoal working capacity was also measured.
The first part of testing involved operation of the mini-canister apparatus
with a gaseous hydrocarbon blend composed of butane, isobutylene, and toluene
in nitrogen. After breakthrough times and weights were monitored for
repetitive hydrocarbon loading and purging cycles, the mini-canisters were
refilled with fresh charcoal. Methanol in nitrogen was added to the
hydrocarbon blend, and the mini-canisters were once again subjected to
repetitive load and purge cycles. The comparison of hydrocarbon and methanol-
hydrocarbon blends using breakthrough times and charcoal weight gains
provided a preliminary indication of the effect of methanol on canister
performance.
~Numbers in parentheses designate references at the end of this report.
-------
IL PROCEDURES AND INSTRUMENTATION
The work plan called for the development of a bench scale apparatus to
evaluate charcoal from evaporative canisters. The apparatus that was
developed allowed delivery of hydrocarbon vapors to twelve mini-canisters
containing fresh charcoal from standard-size evaporative canisters. After the
HC vapors broke through the charcoal, the canister system was designed to
permit hydrocarbon vapor purge by pulling room air in the reverse direction
through the mini-canisters.
A. Development of Procedure and Instrumentation
The goal of the program was to determine the effect of repetitive
hydrocarbon loading and purging cycles (aging) on four types of canister
charcoal. The variables that were measured during the aging process were;
time to hydrocarbon breakthrough, retained weight gain of the canisters, and
charcoal working capacity. Two feed gases were used for canister loading, a
hydrocarbon blend of butane, isobutylene, and toluene vapors, and a
hydrocarbon-methanol blend composed of the hydrocarbon blend mixed with
methanol vapors.
1. Fuel Delivery
The original work plan called for the development of a fuel
containment vessel equipped with a temperature programmer designed to
deliver unleaded gasoline vapors and gasoline-methanol vapors to twelve mini-
canisters. As a result of discussions with the EPA project officer concerning
safety, the plans for a fuel source were changed from a simulated gas tank
filled with gasoline to hydrocarbons supplied by pressurized cylinders. The new
fuel delivery system still satisfied the requirements for maintaining consistent
fuel vapor composition and vapor volume per cycle while reducing the
possibility of a fire hazard.
The hydrocarbon sources, as mentioned previously, were compressed
gases and vaporized toluene and methanol. Butane and isobutylene were
supplied from pressurized cylinders (both 99.0% pure), and toluene and methanol
were delivered to the mini-canisters by bubbling nitrogen through the
respective liquids. Gas flows were measured with soap bubble meters and
monitored with flowmeters.
Several delivery systems for toluene and methanol were evaluated
before a decision was made to use nitrogen saturated with toluene and methanol
at room temperature. Initially, the liquids were heated to produce pure gas
vapors. However, since the remainder of the canister system was not heated,
the vapors condensed before reaching the canisters. A second method involved
bubbling nitrogen through liquid toluene and methanol in a system heated to
30°C, but the same condensation problem occurred. To avoid vapor
condensation, liquid methanol and toluene were placed in separate containers at
room temperature, and separate flows of nitrogen were bubbled through each
liquid. Toluene made up approximately 1.5% of the toluene-nitrogen flow (by
volume) at a nitrogen flowrate of 38 milliliters /min, and methanol constituted
about 6.8% of the methanol-nitrogen flow by volume with a nitrogen flowrate
of k2 milliliters/min.
2
-------
2. Composition of Hydrocarbon and Methanol Blends
The targeted composition of the simulated fuel was based upon a
hydrocarbon speciation study performed by EPA-RTP in which hot soak
evaporative emissions were measured on forty six 1975 to 1982 model year
gasoline vehicles.^) Overall, the hot-soak emissions were composed of 70
percent paraffins, 21 percent olefins, and 9 percent aromatics by mass. The
flowrates of the fuel gases were set with the goal of achieving these available
hot soak weighting factors. Butane was chosen to represent paraffins,
isobutylene for olefins, and toluene for aromatics. Feed gas flows were
monitored continuously with flowmeters. Heavier hydrocarbons (also
representative of hot-soak emissions) were not included in the simulated fuels
because of the inability to accurately measure vaporization and delivery of such
compounds to the canister system.
The targeted level of methanol in the methanol blend was based on
data from two research projects. A study on the alcohol content of gasoline in
the Houston area showed the fraction of methanol in one brand of methanol-
containing fuel to be about 596 (by volume).^) Another study measured SHED
methanol concentrations from vehicles fueled with methanol-gasoline blends. W
This study indicated that the mass fraction of methanol in hot soak evaporative
emissions was approximately equivalent to the volume percent of methanol in
the fuel. Thus, a 5% methanol concentration by mass was targeted for the
simulated methanol fuel vapors.
3. Load and Purge Cycles
The laboratory procedure was loosely modeled after the Code of
Federal Regulations SHED test, which measures evaporative emissions from
vehicles.'^) The SHED test cycle consists of a diurnal segment during which
the fuel tank is heated from 60-84°F, an FTP driving cycle for purging fuel
vapors from the evaporative canister, and a hot soak segment in which the
carburetor, at temperatures of 150-200°F, emits fuel vapors to the canister.
The laboratory procedure developed for evaluating canister charcoal
combined the two fuel loading segments of the SHED test into a single load
cycle. An approach was to simulate and monitor fuel loading and purging cycles
similar to the SHED test until hydrocarbons broke through the charcoal.
However, the possibility that a SHED-type cycle might not ever show
breakthrough became apparent. Therefore, the load cycle was redefined as the
length of time until hydrocarbons broke through the mini-canisters. This
change was incorporated into the program with the approval of the project
officer. After breakthrough, the canisters were purged until the rate of change
of purge hydrocarbon concentrations became small in relation to that observed
at the start of purging (from over 600ppmC/min to less than 10 ppmC/min).
These revisions in the load and purge cycles still provided the opportunity to
observe the change in canister weight as the charcoal was aged with repetitive
loadings and purgings. In addition, cycle by cycle breakthrough times could be
observed. The hydrocarbon blend was loaded onto each mini-canister for a
cumulative total of about 155 g. This weight is roughly equivalent to the
amount of hydrocarbons that a canister would be subjected to over 69 repetitive
3
-------
SHED tests.a Due to time restrictions, a total of only 68 g of the methanol
blend was loaded onto each mini-canister, an equivalent of about 29 SHED
tests.a At the end of testing, a single load-purge cycle with the HC-only blend
was conducted on the mini-canisters which had been previously exposed to the
HC-methanol blend.
Prior to the decision to alter the load and purge cycles, vapor flow
from the carburetor during the hot soak cycle was measured on two vehicles, a
1980 Mercury Cougar and a 1981 Chevrolet Monte Carlo. Vapor flowrate was
measured using a 10 ml soap bubble meter attached to the carburetor bowl
(where the carburetor vents to the evaporative canister). Vapor volumes
produced by the Cougar were measured and found to be 0.1 ml, 0.3 ml, and 0.75
ml over approximately 10 minutes of the hot soak on three tests. Vapor flow
was measured at 0.3 ml on the Monte Carlo within a period of about 10
minutes.*3 The carburetor bowl was found to produce negligible amounts of
vapor from the carburetor during the hot soak. In addition, canister vacuum
was monitored on the two vehicles during a hot FTP using a vacuum transducer
teed into the line connecting the canister to the engine. The average vacuum
was 4 in. Hg on the Cougar and 2 in. Hg on the Monte Carlo. The disparity in
the amount of vacuum applied to the canisters appeared to be due to the
relative size of the purge ports (Cougar purge port was larger than the Monte
Carlo purge port) and to the engine size. To relate canister vacuum to the
bench-scale apparatus it was necessary to determine purge flowrates. A bench
evaluation of new Ford and GM canisters produced flowrates of ^ ft^/min (113
liters/min) and 2 ft^/min (57 liters/min) at applied vacuums of 4 in. Hg and 2 in.
Hg, respectively. This range of flowrates is consistent with the Ford test
procedures for testing the "Working Capacity" of Evaporative canisters.
According to these procedures, canisters are purged at 2 ft^/min. The total
flowrate of purge air through the mini-canister system was set at about 3.5
ft^/min (99 liters/min). Each mini-canister, which was approximately 1/12 the
size of a standard-size canister, was purged at about 0.3 ft*/min (8.5
liters/min). This rate is within the flow range measured on the Ford and GM
canisters on a proportional volume basis.
4. Mini-Canisters
The bench-scale apparatus was designed to test twelve mini-
canisters. With this setup, multiple positions could be used to evaluate each
type of charcoal. Miniature canisters were used instead of standard size
canisters in an effort to minimize total flow rates and apparatus size. Mini-
canisters were initially made of polypropylene tubing (about 6 in. long) with
plastic caps on each end. The caps were modified to allow flow through the
aBased on two vehicles tested with four fuels, the average increase in canister
weight during each SHED test was 27 g (sum of hot soak plus diurnal).^' The
value is divided by twelve for mini-canister comparisons.
bVapor volume was also measured on a third vehicle, a 1981 Ford Mustang. This
car produced 0.2 ml of vapor over 15 minutes.
4
-------
mini-canisters. Preliminary experiments were performed using only butane and
isobutylene to determine breakthrough times. During initial experimentation
with the hydrocarbon blend (butane, isobutylene, and toluene in nitrogen), the
mini-canisters were found to leak hydrocarbons at the cap. The plastic caps
were apparently not sealing with the additional flow of nitrogen and toluene
through the mini-canisters. A heavier mini-canister design with a threaded
aluminum cap was used in all subsequent testing to prevent leaks. The all-
plastic canisters were selected initially due to their low filled weight (70 g).
The new mini-canisters, though heavier (220-250 g), could still be weighed
accurately after hydrocarbon loading.
B. Description of Charcoal Evaluation Apparatus
The bench scale apparatus for evaluating evaporative canister charcoal is
shown in several views in Figure 1. The mini-canister system is composed of a
hydrocarbon source (liquids and compressed gases); a series of valves,
flowmeters, and tubing to direct equal flows to the mini-canisters; a vacuum
pump for purging; and a hydrocarbon analyzer and recorder. The flow
schematic of the apparatus is shown in Figure 2. The fuel and delivery gases
were set to 20 psig at the cylinder regulator and were individually controlled
with needle valves to achieve the desired proportion of butane, isobutylene,
toluene, and methanol (as needed). Load and purge cycles were controlled by a
timer which automatically switched the purge pump and the fuel solenoid valves
on and off. Total hydrocarbon concentrations could be monitored at the exit to
individual canisters or in the purge manifold before the pump. A sample line to
the HC analyzer allowed sequential hydrocarbon analyses to determine break-
through time for each mini-canister. A second vacuum pump, which was
manually operated, was used to remove hydrocarbons which broke through the
mini-canisters.
Background hydrocarbon levels were monitored in the lab and generally
ranged from about 15 to 25 ppmC. Room temperature was 75°F 2°F, and
relative humidity generally varied from a daily high of about 60 percent to a
daily low of 50 percent.
1. Mini-Canisters
The mini-canisters that were used during experimentation were
made of an acrylic tube (5 3/4 in. long, 1 in. diameter) with a threaded
aluminum cap. The volume of each mini-canister was approximately 74
milliliters. The bottom of the canister was capped by a polypropylene cap with
a large hole cut from the center. A metal screen was inserted into the cap to
retain the charcoal while allowing vapors or air to pass freely. A large hole
(5/8 in.) was drilled into the canister top for a purge outlet, and a smaller hole
(1/16 in.) was drilled in the side of the canister top for fuel delivery. A screen
was placed in the purge opening to prevent charcoal from being pulled off while
under vacuum. In addition, glass wool was used at the purge opening and at the
bottom cap to prevent the loss of charcoal dust.
2. Charcoal
Four types of activated charcoal were evaluated in the mini-
canisters. The charcoals were obtained from new evaporative canisters ordered
5
-------
Measuring Hydrocarbon Breakthrough Toluene and Methanol Delivery System
Figure 1. Several Views of the Charcoal Evaluation Apparatus
-------
OJ
CN
C
CN
53
CU
Z
i—1
—
i—1
a)
O
-------
for four vehicle types. Charcoal weights and volumes contained in the
are listed as follows:
canisters
Typical Standard
Size Canister
Charcoal Weight, g
Approximate
Volume of
Charcoal, ml
Approximate
Density
g/ml
1983 Chrysler Reliant K
344
1270
0.27
1983 Ford Escort
407
1030
0.40
1983 Chevrolet Monte Carlo
438
1500
0.29
1983 Toyota Corolla
362
870
0.42
Initially, charcoal samples were provided by some of the auto manufacturers'
charcoal suppliers for use in the mini-canisters. A visual comparison to actual
canister charcoal, however, showed differences in size and shape of the
charcoal pieces. Because of these variations, the mini-canisters were tested
with charcoal from actual evaporative canisters. The weights of fresh charcoal
used in the mini-canisters are shown below:
Type of
Charcoal
Mini-Canister
Number
Charcoal Weight, g
HC Blend
HC-Methanol Blend
Chrysler
1
19.7
18.9
Chrysler
2
19.7
17.2
Ford
3
27.9
74.5a
Ford
4
27.5
28.4
Ford
5
27.9
31.1
Ford
6
27.9
26.5
GM
7
19.5
22.3
GM
8
18.5
20.5
GM
9
19.5
21.1
Toyota
10
29.1
34.1
Toyota
11
29.0
31.8
Toyota
12
28.8
29.1
aMini-canister number 3 was filled with Teflon chips (when using the
HC-methanol blend) to measure breakthrough time of the mini-canister
system. This breakthrough time was found to be less than one minute.
Charcoals from Ford and Toyota canisters are apparently denser than Chrysler
or GM charcoals, since a greater mass fills the same mini-canister volume. In
addition, Ford and Toyota charcoal particles were generally larger than
Chrysler or GM charcoal particles.
-------
3. Hydrocarbon and Methanol Blend Compositions
The composition of the hydrocarbon and methanol blends, (by volume) was
about 16 percent butane, 4 percent isobutylene, 0.7 percent toluene, and for the
methanol blend, 2 percent methanol. The remainder was nitrogen carrier gas.
The flowrate of hydrocarbon vapors plus nitrogen carriers to the mini-canisters
was on the order of 70 milliliters/min. The actual mass flowrates were
determined using the weight of blend loaded onto the mini-canisters, volume
percentages, molecular weights of the compounds, and the assumption that the
gases obey the ideal gas law. Resulting mass flows are shown below. Due to
the low vapor pressure of toluene, the fraction of toluene in the blends was only
5 percent by mass instead of the 9 percent initially desired.
Mini-Canister Flow, mg/min
Hydrocarbon Component HC HC-Methanol
Butane 31 29
Isobutylene 7 7
Toluene 2 2
Methanol _0 _2
Total 40 40
4. Hydrocarbon Breakthrough
Hydrocarbon vapors were delivered to the canisters at the above
flowrates to establish hydrocarbon breakthrough times. The length of the load
cycle was based on the longest breakthrough time of the four types of charcoal.
The longest average breakthrough time with the HC blend was for Toyota
charcoal, at 110 to 140 minutes. The load cycle for methanol blend testing was
set at 120 minutes. The purge cycle with the HC blend was set with the
intention of reaching a hydrocarbon concentration in the purge manifold of
approximately 300 ppmC. The mini-canisters were purged for 108 to 143
minutes to achieve this level. This level of hydrocarbons represented a drop
from over 10,000 ppmC at the beginning of the purge cycle. The purge cycle
with the HC-methanol blend was set at 110 minutes. The slope of the
logarithmic shaped hydrocarbon purge rate curve leveled out in the 300 ppmC
range, as illustrated in Figure 3.
Initially, breakthrough was defined as the emission of hydrocarbons
from the bottom of the mini-canisters in excess of 100 ppmC for 10 seconds.
The increase in hydrocarbon concentration with time was observed on a
recorder, and it appeared that 100 ppmC HC was an adequate indicator of
breakthrough. During the preliminary evaluations, breakthrough times were
defined at the 100 ppmC level; however, additional experiments were conducted
which indicated that 1000 ppmC was a more appropriate definition of break-
through. Traces of HC breakthrough on the 100 ppmC and 1000 ppmC ranges
are illustrated in Figures 4 and 5, respectively. Comparison of the HC and HC-
9
-------
18 16 14 12 10 8 6 4 2 0
Time, min
Figure 3. Purge cycle monitored on 10,000 ppmC and 1000 ppmC ranges
10
-------
Range 100 ppmC
Figure 4. Hydrocarbon breakthrough at 100 ppmC
11
-------
—1
1
"™ "
1
i
1
T~~
1
1 , ...
1 ' 1
1
I
¦
»
___
1
V
-¦
——
1 1
-
!
L
I
1
T
I
1
L
1
1
L
r
1
v
T
.
l
i
1
i
1
r
1
11
1
¦
,
~
n
m
CO
— 1 i
tinuation
ed load cj
1
Co
ti
1
1
1
or ~
T
—
_
'clez
—a
__
——.
•in——
~l
Pf—
L_
mm.
—
—
—
r~
—
1
—
[I
i i
I
\
i
V
\
!
113
¦
i
\
; .
V —
i
1
\
V-
i
f
|
i
* ¦
X J
n
*
... . \
\J
——.
U
—
—
i
—
ZL
1fcr~
7
- -
—
—
3
^ -
~
pv
Pfcftftr-
1,
—
1 *T|/]
' "
-
w'
•
"
"
-L
_____
u
J
4
i
128 126 124 122 120 118 116
Time, min
Figure 5. Hydrocarbon breakthrough at 1000 ppmC
12
-------
methanol breakthrough times is based on the 1000 ppmC breakthrough
concentration. Breakthrough times measured at 1000 ppmC were considerably
longer than those measured at 100 ppmC as shown in Tables A-l and A-2 of
Appendix A.
13
-------
ffl. RESULTS
The effect of the repetitive loading and purging of a hydrocarbon blend
and a hydrocarbon-methanol blend on evaporative canister charcoal was
measured by canister weight gain and hydrocarbon breakthrough time. The
mini-canisters were weighed after the last load-purge cycle of each day. From
one to three load-purge cycles were accomplished each test day. Canister
weight gain was calculated daily by subtracting the initial clean canister weight
from the weight after the last load-purge cycle. Breakthrough time was
defined as the length of time for hydrocarbons to break through the bottom of
the mini-canisters at 1000 ppmC with a fuel blend flowrate of about 40 mg/min
per canister. This loading was equivalent to approximately 5g of HC or HC-
methanol per mini-canister.
Variations in canister weight gain and breakthrough times were noted
between charcoal types and between fuel blends (HC, HC-methanol). However,
canister weight gains and breakthrough times were relatively constant over the
course of the program for each individual mini-canister. The only major
exception to this finding was the increased weight gain after purge associated
with the 110-minute loading on day 8 with the HC-only blend. The first set of
mini-canisters received a cumulative hydrocarbon loading of about 155 g of
hydrocarbons per canister, and the second set was loaded with about 65 g of the
methanol blend and 23 g of the HC-only blend. Appendix A lists the load-purge
cycles necessary to achieve these loading levels. Average charcoal weight
gains after purge using the hydrocarbon fuel and the hydrocarbon-methanol fuel
are compared in Table B-l of Appendix B and in Figure 6 for charcoal from
Chrysler, Ford, GM, and Toyota mini-canisters. The canisters containing the
smaller mass of lower-density Chrysler and GM charcoals retained greater
amounts of hydrocarbons (and/or methanol) after purging than did the Ford and
Toyota canisters for both fuels. In addition, all four charcoal types retained
more weight after purge when only hydrocarbons were loaded than when the
hydrocarbon-methanol blend was loaded to breakthrough. The effect was
greater in the cases of the Ford and Toyota charcoals, for which the average
weight gain after purge for the hydrocarbon blend was 3 to 7 times that for the
methanol blend. The weight gains measured for the Chrysler and GM charcoals
using the hydrocarbon blend were 54 and 25 percent higher than with the
methanol blend, respectively. Retained weight gains for individual mini-
canisters using the HC and the HC-methanol blend are listed in Tables A-3 and
A-5 of Appendix A.
Hydrocarbon breakthrough times were shorter with the methanol blend
than with the hydrocarbon blend for all charcoal types. This finding is
illustrated in Figure 7 and Table B-2 of Appendix B. Average breakthrough
times for Chrysler, Ford, GM, and Toyota charcoals with the methanol blend
were 44, 22, 29, and 20 percent lower, respectively, than with the hydrocarbon
fuel. The denser Ford and Toyota charcoals had the longer breakthrough times.
Breakthrough times are listed in Tables A-l, A-2 and A-4 of Appendix A for
individual mini-canisters. At the end of testing, a single load-purge cycle with
the HC blend was conducted on the mini-canisters which had been previously
exposed to the HC-methanol blend. Breakthrough times as shown in Table A-4
of Appendix A did not return to the longer times experienced with the HC blend
(working capacities were not measured).
14
-------
Tested with HC Fuel
Tested with HC-Methanol Fuel
t»o 8.Or*
8.0 r
Chrysler Charcoal
Ford Charcoal
oo
8.0
e
« 6.0
o
4J
X
•h 4.0
cu
3
o 2.0
a
u
a)
X!
0
GM Charcoal
Toyota Charcoal
Figure 6. Average daily weight gain after purge of
four types of evaporative canister charcoal
15
-------
-1'
Tested with HC-Methanol Fuel
Figure 7. Average daily breakthrough time of four types
of evaporative canister charcoal
16
-------
The working capacity of each type of charcoal was also measured with the
HC blend (three times) and with the HC-methanol blend (once). Working
capacity is defined as the weight of hydrocarbons that can be purged off after
loading to breakthrough. It was calculated by subtracting the canister weight
after purging from the weight after loading to breakthrough. The results are
shown in Table B-3 in Appendix B and illustrated in Figure 8. The working
capacity of the four charcoal types was reduced from 10 to 52 percent relative
to the hydrocarbon blend when the methanol blend was used. The working
capacities of the Chrysler and GM charcoals were affected to a greater degree
by the presence of methanol. They were 35 to 52 percent lower than with the
hydrocarbon blend. The working capacities of Ford and Toyota charcoals were
10 to 16 percent lower when the methanol blend was used compared to the
hydrocarbon blend. Working capacities for individual mini-canisters are
reported in Appendix A, Table A-6. It should be noted that some of the mini-
canisters were loaded beyond the breakthrough point, since loading times were
determined by the charcoals with the longest breakthrough times.
Another method for measuring working capacity is the Ford butane
working capacity procedure. According to this method, standard size canisters
are first purged at 120°F and 2 ft^/min (57 liters/min) until weight loss is less
than 1.2 g in five minutes. The canisters are then saturated with butane. This
process is repeated two more times. To determine working capacity, the
canisters are weighed after the last butane load cycle, then purged for 20
minutes, and finally weighed again. The difference in these two weights is
called working capacity. Table B-4 in Appendix B compares the working
capacities of new Chrysler, Ford, and GM charcoals measured with this method
and working capacities measured using the SwRI method with mini-canisters.
17
-------
U u - » J
S'J';
Tested with HC Fuel
Tested with HC-Methanol Fuel
8.0
6.0
4.0
2.0
| 1
——— Miifiii
Chrysler Charcoal
8.0 r
6.0
4.0
2.0
Ford Charcoal
8.0 r-
6.0
4.0
2.0
8.0 r-
6.0
4.0
2.0
GM Charcoal
Toyota Charcoal
Figure
8. Average daily working capacity of four types
of evaporative canister charcoal
18
-------
REFERENCES
1. Stamper, Ken R., "Evaporative Emissions from Vehicles Operating on
Methanol/Gasoline Blends", Paper 801360 presented at the SAE Fuels &
Lubricants meeting, Baltimore, Maryland, October 20-23, 1980.
2. Sigsby, I.E., Tejada, S., Ray, W., Lang, J., and Duncan, 3., "Volatile
Organic Compound Emissions from 46 In-Use Passenger Cars." Publication
Preprint.
3. Urban, Charles M., "Volatility of In-Use Gasoline and Gasoline/Methanol
Blends," EPA Report 460/3-84-009, September 1984.
4. Dietzmann, Harry E., "Gasoline Volatility Analysis," Final Data Report,
Task 1 and 2 Work Assignment No. 4, September 1984.
5. U.S. Code of Federal Regulations. Title 40—Protection of Environment
Chapter I—Environmental Protection Agency; Subchapter C—Air
Programs; Part 86~Control of Air Pollution from New Motor Vehicles and
New Motor Engines: Certification and Test Procedures. Federal
Register, Vol. 90, No. 129, pp. 342-569, October 23, 1977.
6. Cantwell, E.N., Letter to EPA with additional information for DuPont's
waiver application for gasoline-alcohol fuels, October 11, 1984.
19
-------
APPENDIX A
-------
APPENDIX A
TABLE A-l. INDIVIDUAL MINI-CANISTER BREAKTHROUGH
100 ppmC BREAKTHROUGH LEVEL
Cumu- Breakthrough Time, min
lative
Chrysler
Ford
GM
Toyota
Day
Cycles
1
2
3
4 5
6
7
8
9
10
11
12
la
5
7.25
6.75
10.75
11_ 13.25
11.25
4.25
4.75
5.25
10
8.75
9
6
6.5
6.5
9.5
11.5 14.25
12.25
4.5
4.5
4.75
10
9
9.5
7
6.25
6.25
10
10.25 13.5
12
4.5
4.5
4.75
9.75
8.75
9.25
8
6
6.25
11.5
10.25 13.5
11.75
5
4.75
5.25
10
6.25
9
9
6.25
6.25
10.5
10.5 14
12
4.75
5.25
5.5
9.5
9
9.25
10
6.5
6.5
11.5
11.75 14.75
14.5
5
5.75
4.75
9.5
9
9
11
8.5
8.5
11
11.25 14.75
13.75
7.25
7.25
7.25
11.5
10.5
10.25
Avg
6.
7
121
5.2
9.4
2a
12
5.25
5.25
8.75
8.75 11
10
3.75
3.75
3.75
3.75
8.5
8.25
13
5.5
6
9.25
10.1 13.8
9
3.5
3.5
3.5
9.25
8.5
8.5
14
5.25
5.1
9
8.75 13.7
13
4
3.8
4.1
9.2
8.2
8.2
15
6.5
6.5
9.6
10.5 12.2
12
5
4.6
5.2
10.2
8.5
9.5
20
6.2
7.1
10.5
10.5 13.1
11.7
6.2
5.7
6.2
11.2
9.4
10.25
22
5.3
5.3
9.5
9.5 12.0
10.2
4.6
4.6
4.8
8.9
7.5
7.5
Avg
5.
8
10.6
4.5
8.9
3a
24
5.2
5.3
10.4
10.2 13.9
13*1
3.9
3.8
4.3
9.2
7.7
7.9
31
6.5
6.6
10.6
10.4 12.4
-11.3
4.4
4.5
4.7
9.8
8.8
8.7
Avg
5.
9
11.5
4.3
8.7
4a
36
6.4
6.4
11.2
9 10.5
11.2
4
3.4
4.7
9.5
9.1
8.5
37
5.4
5.8
11
11.5 14.3
.11*4
3.5
3.2
4
9.6
8
7.7
Avg
6
11.3
3.8
8.7
5a
44
6.7
7.1
10.7
11.4 14.3
10.7
4.1
4.2
5.0
9.8
8.4
8.2
49
6
5.9
12.4
8.9 10.2
.10.1
3.9
3.9
4.
8
7.4
7.2
53
7.1
6.5
12.7
11.2 14.4
11.9
4.3
4.4
5.2
9.9
8.5
9.2
Avg
6.
6
11.6
4.3
8.5
6a
54
5.2
5.4
10.1
8.6 12.7
10.6
3.6
3.7
4
8.9
7.9
8.1
61
5.3
5.4
9.4
8.7 10.8
10.9
4.1
4.2
4.3
8.2
7.5
7.9
64
5.6
6.1
10
8.8 10.4
9.9
4.1
4.4
4.2
8.3
7.8
8
Avg
5.
5
10.1
4.1
8.1
7a
65
5.4
5.2
11.7
10.1 13.1
9.9
4.1
4.2
5
9.3
8.1
8.4
74
4.7
4.4
9
9.3 10
9.6
3.6
3.6
3.8
8.2
7
7.1
81
5.5
5.5
8.5
9.2 8.5
-.8,9
4.6
4.8
5
7.5
6.8
6.2
Avg
5.
1
9.8
4.3
7.6
8a
3
hours
of continuous loading without
purge
to check flow
rates
83
6
6.3
9.3
9.3 9.8
8.6
4.4
4.4
5.4
8.8
7.9
8.8
Avg
6.
2
9.2
4.7
8.5
TIMES WITH HC BLEND
A-2
-------
TABLE A-l (CONT'D)'. INDIVIDUAL MINI-CANISTER BREAKTHROUGH TIMES WITH
HC BLEND 100 ppmC BREAKTHROUGH LEVEL
Cumu- Breakthrough Time, min
lative
Chrysler
Ford
GM
Toyota
Day
Cycles
1 2.
3
4 5
6
7
8
9
10
11
12
9a
85
89
95
4.7 4.6
4.1 4.4
3.8 3.9
8.7
8.7
8.4
8.7 10
9.4 9
8.2 8.2
9.2
8
8.1
3.3
2.8
2.9
3.3
2.9
2.9
4 .
3.8
3
8.2
7.5
6.4
7.4
6.8
6.4
7.6
7
6.3
Avg
4.2
8.7
3.2
7.1
10a
97
107
114
4 4.1
4.2 4.2
4.8 4.7
8.5
8.8
8.7
8.4 10.2
8.9 10.6
7.9 9.8
8.4
9.1
10
2.8
3
2.8
2.9
3
2.8
3.6
3.5
2.8
6.7
7.4
7.2
6.4
6.7
7
6.5
7.5
6.7
Avg
4.3
9.1
3
6.9
lla
116
125
131
4.6 4.5
4.9 4.8
4.1 4.6
8.5
9.3
9.2
8.1 10.7
8.7 12.4
8.7 10
8.8
8.5
8.3
2.5
2.7
2.7
2.6
2.7
2.6
3
3
3
7
7.6
7
6.5
6.6
6.5
6.7
7
6.7
Avg
4.6
9.3
2.8
6.8
12 3
133
144
4.7 4.6
4.6 4.4
8.3
8.5
8.3 11
8.5 11.8
8.9
8.9
2.5
2.7
2.6
2.7
3
3.2
6.9
6.9
6.1
6.3
6.6
6.7
Avg
4.6 .
9.3
2.8
6.6
13a
151
156
4.7 4.7
4.6 4
8.7
7.7
8.7 b
9.3 8
8.6
7.7
2.6
2.7
2.8
2.7
3.5
3
7.8
6.6
7.3
6.1
7.5
6.7
Overall
Avg
4.5
174
8.4
ToTI
2.9
T!
7.0
T7l
a
^Load Cycle
No data
- 15
min, purge
cycle
¦ 25 min
A-3
-------
5-
4>-
TABLE A-2, INDIVIDUAL MINI-CANISTER BREAKTHROUGH TIMES WITH HC BLEND
IOOO ppmC BREAKTHROUGH LEVEL
Breakthrough Time, mln
Chrysler
Ford
GM
Toyota
Dajr
16
Cumulative Cycles
1
2
3
4 5
6
7
8
9
10 11
12
179 cycles w/purge
+280 mln loading, purge
25 min to 3600 ppmC
+216 min loading}purge
124 min to 410 ppmC
17
111 min loading, purge
143 min to 300 ppmC
63
52.5
108
101 100
100
43
42
46.5
109 111
111
Avg
58
102
44
110
18
120 min loading, purge
120 min to 295 ppmC
61
70.5
112
108 104.5
107.5
47
47
49
117.6 116.5
122.5
Avg
66
108
48
119
125 min loading, purge
12:0 min to 190 ppmC
67
71
121
108 107
107
51
51
55
126 122
126
Avg
69
111
52
125
19
140 min loading, purge
108 min to 265 ppmC
75
74
132.5
124 132
127
54
52
61
141 137.5
140
Avg
74
129
56
140
140 min loading, purge
122 min to 330 ppmC
75
75
124
122 121
121
51
47
52
137 135
134
Avg
75
122
50
135
f o
Overall Avg
68
114
50
126
> V,
I
t
*
t
t.
-------
TABLE A-3. INDIVIDIAUL MINI-CANISTER RETAINED WEIGHT GAIN WITH HC BLEND
a
>
i
U1
Retained Weight Gain, g
Chrysler
Ford
GM
Toyota
Day
Cumulative Cycles
1 2
3
4 5
6
7
8
9
10
11
12
lb
11
2.75 2.71
0.41
0.41 0.40
0.42
2.55
2.53
2.58
0.15
0.18
0.13
Avg
2.7
0.4
2.6
0.2
2b
22
3.25 3.17
0.47
0.48 0.47
0.50
3.01
3.00
2.95
0.18
0.23
0.17
Avg
3.2
0.5
3.0
0.2
3b
35
3.45 3.36
0.58
0.59 0.60
0.61
3.18
3.11
3.03
0.23
0.28
0.23
4C
5b
42
53
Avg
3.4
c
0.6
c
3.1
c
0.2
——-.—-J
2.17 2.02
0.32
0.32 0.29
0.31
1.86
1.75
1.67
0.18
0.20
0.17
Avg
2.1
0.3
1.8
0.2
6b
64
2.36 2.18
0.37
0.34 0.34
0.36
1.98
1.85
1.82
0.20
0.23
0.18
Avg
2.3
0.4
1.9
0.2
7b
81
3.35 3.22
0.61
0.62 0.61
0.63
2.86
2.77
2.69
0.28
0.38
0.28
Avg
3.3
0.6
2.8
0.3
8^
3 hours
83
of continuous loading
7.69 7.07
without purge to check flow rates
4.62 4.64 4.56 4.56 7.14 6.90
6.95
4.14
4.19
4.16
Avg
7.38
4.60
7.00
4.16
9b
95
7.90 7.29
4.68
4.69 4.61
4.59
7.22
6.87
6.93
4.16
4.20
4.21
Avg
7.60
4.64
7.01
4.19
10b
114
7.83 7.26
4.67
4.68 4.60
4.60
7.24
6.96
7.00
4.18
4.18
4.22
Avg
7.55
4.64
7.07
4.19
llb
131
8.19 7.58
4.78
4.80 4.74
4.72
7.54
7.15
7.10
4.21
4.21
4.28
Avg
7.89
4.76
7.26
4.23
12b
150
7.28 6.64
4.56
4.56 4.47
4.45
6.54
6.14
6.09
4.13
4.11
4.19
53
6.96 4.51 6.26 4.14
-------
TABLE A-! (CONT'D). INDIVIDUAL MINI-CANISTER RETAINED WEIGHT GAIN WITH KC BLEND
:>
Retained Weight
„ , a
Gain, g
Chrysler
Ford
GM
Toyota
Day
Cumulative Cycles
1 2
3 4 5 §
7 8
9 10 11 12
l3b
160
7.65 7.05
4.69 4.69 4.61 4.60
6.98 6.65
6.68 4.18 4.17 4.24
Avg 7.35 4.65 6.77 4.20
14b c c c
16d Total of 179 cycles
with purge +280 min
loading, purge 25 min
to 3600 ppmC + 216 min
loading, purge 124 min
to 410 ppmC
8.07 7.42
4.88'
4.86 4.79
4.79
7.36
6.97
7.07
4.29
4.26
4.37
Avg
7.75
4.83
7.13
4.31
17d
111 min loading, purge
143 min to 300 ppmC
8.41 7.87
5.27
5.21 5.16
5.13
7.58
7.09
7.47
4.54
4.49
4.61
Avg
8.14
5.19
7.38
4.55
18d
120 min load, purge
120 min to 295 ppmC
+ 125 min load, purge
120 min to 190 ppmC
8.28 7.75
5.11
5.07 5.00
4.99
7.28
6.73
7.18
4.48
4.41
4.53
Avg
8.02
5.04
7.06
4.47
19d
140 min load, purge
108 min to 285 ppmC
+ 140 min load, purge
122 min to 330 ppmC
8.12 7.69
4.96
4.94 -4.88
4.87
7.26
6.75
7.12
4.36
4.29
4.42
Avg
7.91
4.91
7.04
4.36
Overall Avg.
from Day 8 to
Day 19
7.6
4.8
7.0
4.3
d
^Retained weight gain = mini-canister weight after purging - clean mini-canister tare weight
Breakthrough level 100 ppmC, load cycle » 15 min, purge cycle =25 min
^No weight measurement
Breakthrough level = 1000 ppmC
-------
T
Day
4C
TABLE A-4. INDIVIDUAL MINI-CANISTER BREAKTHROUGH TIMES WITH HC-METHANOL BLEND
1000 ppmC BREAKTHROUGH LEVEL
Breakthrough Time, mln
Chrysler
Ford
GM
Toyot<
a.
Cumulative Cycles
1
2
3
4
5
6
7
8
9 10
11
12
HC Blend only
for 2h days
to establish
breakthrough times
107 min load
purge to 270
w/108 min
ppmC
58
64.6
94.9
111.5
89.6
64.2
58.5
64 115.2
106.7
98.2
Avg
61
99
62
107
136 min load
purge to 300
w/110 min
ppmC
b
50
121
139
112
44
48
45 144.5
138
124
Avg
50
124
46
136
118 min load
purge to 300
w/110 min
ppmC
58
58
104
121
103
58
58
58 123
120
110
Avg
58
109
58
118
112 min load
purge to 320.
w/109 min
ppmC
55.5
54
102
116
95.4
53.5
52.3
52.3 121
111
105
Avg
55
104
53
112
140 min load
purge to 300
w/114 min
ppmC
66
59
118
136
114
58
53
61 149
142
130
Avg
62
123
57
140
Overall Avg
57
112
55
122
HC-
-Methanol Blend
46
40
41
38
92.2
91.6
101
100.3
84
79
41
38
41
40
43 110
38 101.3
102
96.3
95.9
95.6
Avg
41
91
40
100
45
41b
92.7
88.8
78.7
107.3
98.2
92
83.4
85
74.8
43
41
45 115
b i fwo -i
106.5
95.4
95.9
101.6
89.9
88.7
42
35
33.9
32.3
33.6 104.5
41
89
38
101
o
-------
TABLE A-4 (CONT'D). INDIVIDUAL MINI-CANISTER BREAKTHROUGH TIMES WITH HC-METHANOL BLEND
1000 ppmC BREAKTHROUGH LEVEL
Breakthrough Time, min
Chrysler
Ford
GM
Toyota
toy
Cumulative Cycles
1 2
3 4 5
6
7
8
9
10
11
12
5C
6
7
8
48 37
41.4 35.2
39.6 33.9
98.5 109.4
86.5 103.3
83.7 115
88.9
81.8
79
39.7
34.9
33.3
39
31.1.
31.3
40.8
32.8
32.7
114.4
113.3
114.3
108.1
91.1
94
103.8
91.1
84.1
Avg
39
94
35
102
6C
9
10
11
46.4 40
28 25
36.7 32.1
101.6 113.8
75.2 87.7
78.7 93.5
93.8
62.9
71.8
40
22
30.3
38
22
30
41.4
24
31.5
127.9
96.3
105.6
117.2
85.7
92.2
108.5
78.8
87.8
Avg
35
87
31
100
7C
12
13
14
43 38.5
40 34.5
36 33
93.5 103.4
82 90.5
79.8 93.8
83.8
73
73.5
39.4
32.5
29.4
37
30.8
28.5
40.9
33.1
31
123.2
112.4
107.9
115
96.6
94.8
104
91.8
86.6
Avg
38
86
34
104
Overall Avg with
HC-Methanol Blend 39
89
36
101
HC-Blend
8
1
39.7 47.5
88.8 111.2
79.6
29
30
33
109.9
104.5
90.9
Avg
44
93
31
101
fMini-canister 3 filled with
No data
cLoad cycle = 120 min, purge
Teflon chips to measure breakthrough time of mini-canister system
cycle = 110 min
-------
TABLE A-5. INDIVIDUAL MINI-CANISTER RETAINED WEIGHT GAIN WITH HC-METHANOL BLEND
Retained Weight Gain. g&
u
>
I
VO
V..
Chrysler Ford GM Toyota
Day Cumulative Cycles 1 2 3 4 5 6 7_ 8 9 10 11 12
HC Blend only for 2 1/2 days to establish breakthrough time
1 1 107 min load w/
108 min purge 2.09 1.90 0.63 0.67 0.56 1.93 1.83 1.85 0.23 0.24 0.22
Avg 2.00 0.62 1.87 0.23
2 2 136 min load w/
110 min purge
3 +118 min load w/
110 min purge
+118 min load w/
110 min purge
4 +112 min load w/
109 min purge
4.19 3.75
0.95
1.04
0.85
4.18
3.88
3.84
0.33
0.35
0.39
Avg
3.97
0.95
3.97
0.36
Overall Avg
Tl)
oTf
I7f
0.3
HC-Methanol Blend
3C
2
5.02 4.53
1.27
,1.37
1.12
5.40
4.98
4.98
0.46
0.49
0.52
Avg
4.78
1.25
5.12
0.49
4C
5
5.08 4.67
1.24
1.33
1.08
5.67
5.17
5.14
0.41
0.51
0.56
Avg
4.88
1.22
5.33
0.49
5C
8
5.25 4.88
1.38
1.44
1.23
5.99
5.57
5.52
0.42
0.60
0.68
Avg
5.07
1.35
5.69
0.57
6°
11
5.20 4.84
1.31
1.43
1.16
5.91
5.42
5.36
0.39
0.54
0.70
Avg
5.02
1.30
5.56
0.54
7C
14
5.54 5.17
1.72
1.76
1.49
6.43
5.91
5.86
0.68
0.84
0.94
Avg
5.36
1.66
6.07
0.82
Overall Avg
53
174
576
0.6
^Retained weight gain = mini-canister weight after purging-clean mini-canister tare weight
cMlni-canister 3 filled with Teflon chips to measure breakthrough time of mini-canister system
Load cycle = 120 min, purge cycle = 110 min
-------
>
I
t—'
o
Day
16
17
18
TABLE A-6. INDIVIDUAL MINI-CANISTER WORKING CAPACITY WITH HC BLEND
a
Cumulative Cycles
Chrysler
Working Capacity, g
Ford
GH
Toyota
8
10 11 12
179 load/purge cycles
(load = 15 min, purge =
25 min)
+280 min load, purge
25 min to 3600 ppmC 2.76 2.82 4.77 4.86 4.90 4 . 76 2 . 05 1.90 2.02 5.00 4 . 65 4 . 77
216 min load, purge
124 min to 410 ppmC 2.10 2.05 4.20 4.34 4.28 3.99 1.59 1.47 1.43 4.46 4'.52 4.35
111 min load, purge
143 min to 300 ppmC 3.32 3.31 5.51 5.52 5.50 5.43 2.66 2.54 2.67 5.61 5.50 5.58
Avg 2.7' 4.8 2.0 4.9
INDIVIDUAL MINI-CANISTER WORKING CAPACITY WITH HC-METHANOL BLEND
Chrysler
Ford
GM
Toyota
10
11 12
14 load/purge cycles
(load = 120 min, purge =
110 min) 1.39 1.29
Avg 1.3
4.29 4.60 3.91 1.33 1.24 1.26 4.39 4.15 3.69
4.3
1.3
4.1
3
Working capacity is defined as the weight of hydrocarbons that can be purged after loading
to breakthrough. It should be noted that due to the procedure used, some canisters were
^loaded beyond the breakthrough point.
Mini—canister 3 filled with Teflon chips to measure breakthrough time to mini-canister system
-------
APPENDIX B
-------
TABLE B—1. AVERAGE DAILY WEIGHT GAIN
HC Blend
HC-Methanol Blend
Type of
Charcoal
Chrysler
Ford
GM
Toyota
Number
of Mini-
banisters
2
4
3
3
Weight
Gain, g
7.7
4.8
7.0
4.3
Weight Gain as
Percentage of
Clean Charcoal
Weight
41%
17%
35%
14%
Number
of Mini~a
canisters
2
3
3
3
Weight
Gain, g
5.0
1.4
5.6
0.6
Weight Gain as
Percentage of
Clean Charcoal
Weight
26%
5%
28%
2%
Percent
Difference
in Weight
Gains
-35%
-71%
-20%
-86%
One mini-canister was filled with Teflon chips to measure breakthrough time of the mini-canister
system
'Percent differences were calculated relative to weight gains using the HC blend
t»
I
NJ
TABLE B-2. AVERAGE DAILY BREAKTHROUGH TIME (minutes)
Type of Charcoal HC Blend HC-Methanol Blend Percent Difference
Chrysler
Ford
GM
Toyota
68
114
50
126
38
89
36
101
-44%
-22%
-28%
-20%
^Loaded at about 40 mg/min with hydrocarbons or methanol blend
Percent of differences were calculated relative to breakthrough
times using the HC blend
-------
TABLE B-3. AVERAGE DAILY WORKING CAPACITY (g)
Type of Charcoal HC Blend HC-Methanol Blend Percent Difference3
Chrysler
2.7
1.3
-52%
Ford
4.8
4.3
-10%
GM
2.0
1.3
-35%
Toyota
4.9
4.1
-16%
Percent differences were calculated relative to working capacities
using the HC blend.
TABLE B—4. WORKING CAPACITIES OF STANDARD SIZE (NEW) CANISTERS USING THE BUTANE
WORKING CAPACITY PROCEDURE AND OF MINI-CANISTERS USING THE SWRI PROCEDURE
1
Charcoal
Auto
HC Blend
HC-Methanol Blend
Manufacturer
Manufacturer
SwRI
SwRI
Specified
Specified
Mini-canisters,
Mini-canisters,
Type of Charcoal
Virgin BWC, g/100 ml
Virgin BWC, g/100 ml
g/100 ml
g/100 ml
Chrysler - 14x35 Westvaco wood
9.0
8.2
3.9
1.8
Ford - 6x16 Calgon coal
6.8
a
6.8
6.1
GM - 10x25 Westvaco wood
8.5
8.5
2.9
1.9
Toyota - description unknown
a
a
6.8
5.7
^ata not available
-------
1. REPORT NO. 2.
EPA 460/3-8A-014
3. RECIPIENT'S ACCESSION'NO.
4. TITLE AND SUBTITLE
The Effect of Methanol on Evaporative Canister
Charcoal Capacity
6. REPORT DATE
January 1985
6. PERFORMING ORGANIZATION COOE
7. AUTHOR(S)
Mary Ann Warner-Selph
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORG \NIZATION NAME AND AOORESS
Southwest Research Institute
Department of Emissions Research
6220 Culebra Road
Ann Arbor, Michigan 48105
10. PROGRAM ELEMENT NO.
11. con+ract/gAant NO,
68-03-3162
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
2565 Plymouth Road
Ann Arbor, Michigan 48105
13. TYPE OF REPORT AND PERIOD COVERED
Final (Feb. 1984 - Aug. 1984)
14. SPONSORING AGENCY COOE
16. SUPPLEMENTARY NOTES
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
This program involved the evaluation of four types of unused evaporative canister
charcoal with a hydrocarbon-only blend and a hydrocarbon-methanol blend. The
HC blend consisted of 77% paraffins (butane), 18% olefins (isobutylene) and 5%
aromatics (toluene) by weight. The HC-methanol blend was composed of 73% butane,
17% isobutylene, 5% toluene, and 5% methanol by weight. Tests were conducted
on a bench-scale apparatus designed to load each blend onto separate sets of
twelve reduced-size mini-canisters, and to subsequently purge off the hydrocarbons.
The charcoals were evaluated by the measurement of retained charcoal weight
gain after purging, time to hydrocarbon breakthrough, and charcoal working
capacity. The mini-canisters which were loaded with the methanol blend, had
shorter breakthrough times, retained less weight gain after purge, and had
lower working capacities than did mini-canisters tested with the hydrocarbon
blend only. These methanol blend mini-canisters also underwent less simulated
aging than the hydrocarbon blend canisters in this program, since they were
only exposed to 40% as much total vapor.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.(OENTIFIERS/OPEN ENDED TERMS
c. cosati Field/Gioup
Air Pollution
Evaporative Canisters
Methanol
Charcoal Evaluation
3. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
36
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
IPA Form 2220-1 (0-73)
------- |