United States
Environmental Protection
Agency
Hazardous Waste Engineering
Research Laboratory
Cincinnati, OH 45268
Research and Development
EPA/600/S2-85/011 Apr. 1985
SER& Project Summary
Measurement of Volatile Organic
Compound Capture Efficiency
D. B. Hunt and J. L. Randall
The U.S. Environmental Protection
Agency (EPA) Office of Air Quality
Planning and Standards (OAQPS) has
issued new source performance stan-
dards regulating the emission of
volatile organic compounds (VOC)
from some surface coating opera-
tions. The regulatory compliance
status in some cases requires deter-
mination of the overall VOC reduction
efficiency based on a knowledge of
the capture efficiency and control
device efficiency. Presently, the only
accepted method for determining cap-
ture efficiency, a gas-phase material
balance, requires installation of an ex-
hausted enclosure to collect and
measure the fugitive VOC emissions.
The study reported here in-
vestigated alternate potential methods
for determining capture efficiency
which might not involve the expense
and inconvenience associated with a
temporary enclosure. Several ap-
proaches were considered, although
the liquid/gas-phase material balance
approach was selected for detailed
testing. The liquid/gas-phase material
balance approach was tested under
laboratory and field conditions to
evaluate the reliability of the available
measurement techniques.
This Project Summary was
developed by EPA's Hazardous Waste
Engineering Research Laboratory, Cin-
cinnati, OH, to announce key findings
of the research project that is fully
documented in a separate report of
the same title (see Project Report
ordering information at back).
Introduction
The EPA Office of Air Quality Planning
and Standards (OAQPS) has issued New
Source Performance Standards (NSPS)
for some industrial surface coating opera-
tions using volatile organic compounds
(VOC). Some operations employing sol-
vent destruction systems must install ex-
hausted total enclosures to demonstrate
compliance through the measurement of
capture efficiency and control device effi-
ciency. The EPA Office of Research and
Development in cooperation with the EPA
OAQPS initiated the work described in
the full report to explore more convenient
and less costly alternatives for measuring
capture efficiency.
The study was designed and conducted
in two phases. Phase I was a review of
existing information and recent studies
and the development and feasibility
evaluation of conceptual alternatives.
Phase II was the detailed testing of the
most feasible method determined from
Phase I.
Alternate measurement concepts
reviewed and considered included:
material balance methods, tracer con-
cepts, modelling, and indirect approxima-
tion techniques. The review included
literature searches, conversations and site
visits with plant representatives of surface
coating operations potentially affected
and review of previous material balance
tests conducted at surface coating plants.
From the review, the liquid/gas-phase
material balance and tracer gas concepts
were found to be the most acceptable ap-
proaches. A comparison of the material
balance and the tracer gas concepts con-
cluded that both methods might be feas-
ible and potentially applicable, but that
the material balance approach is generally
more acceptable. This conclusion was
based on the premise that direct
measurements are more readily accepted
for compliance determination purposes
than indirect determinations, and that the
-------
material balance methods are further
developed than tracer methods.
Therefore, the liquid/gas-phase material
balance methodology was selected for
further testing and evaluation.
The liquid/gas-phase material balance
approach compares the mass of gaseous
VOC captured and sent to the control
device with the mass of liquid VOC used
(vaporized) in the process. Theoretically,
the liquid/gas material balance approach
has only one major disadvantage. That is,
where drying ovens are direct fired, a por-
tion of the captured VOC is destroyed
making a liquid/gas mass balance
technically infeasible.
The experimental testing of the
liquid/gas-phase material balance method
was conducted in two phases: laboratory
testing and field testing. To properly
evaluate the methodology, it was con-
sidered strategically important to begin
testing under the simplest and most con-
trolled conditions and to work up in com-
plexity. The laboratory testing was a
necessary step in this approach, since
plant sites involve too many variables and
potentially unmeasurable streams which
would prevent a complete assessment of
the measurement methods.
The objectives of the laboratory tests
were to: evaluate the measurement
methods for each required parameter,
assess the overall ability to close
liquid/gas-phase material balances under
controlled conditions, and assess the im-
pact of limited field test variables on the
measurements. A systematic approach
was taken in the laboratory tests in order
to effectively evaluate the performance of
the methodology. The test system was
designed as a simple flow-through
evaporation chamber, providing 100%
capture and minimizing the number of
measurement parameters. Therefore, a
known capture efficiency value was
established for comparison with the
measured and calculated values, and
sources of error in measurement were
reduced to the lowest level. The only field
test variables that were incorporated into
the laboratory experimental design were
those that were easily simulated and con-
trolled and that might directly affect the
statistical evaluation of the methodology.
In the laboratory tests, a known mass
of liquid solvent or coating was placed in
a heated evaporation pan located in the
evaporation chamber and mounted on a
balance. Air was pulled through the
system at discrete flow rates to produce a
mass flow of solvent laden air (SLA)
through the measuring duct. While the
known mass of liquid solvent was being
evaporated, the resulting VOC concentra-
tion and the flow rate of the SLA stream
were monitored and recorded for calcula-
tion of the gaseous VOC mass.
Twenty-four experimental runs were
conducted while varying some of the test
conditions in each run. The conditions
were varied systematically through a frac-
tional factorally designed test matrix in
order to assess the impact of the common
field test variables on the measurements.
The variables included the composition of
organics, the mass throughput of
organics, and the gas stream flow rate.
The vast majority of the laboratory tests
involved pure solvents with no solids.
However, in several tests commercial
coating mixes containing solids were
used.
Following the laboratory testing, a field
test was conducted to evaluate the ap-
plicability of the laboratory tested
methods in a field setting. The approach
taken in testing involved selecting a test
site with considerably more complex test
conditions than the laboratory setting, but
less complex than most coating opera-
tions. The selection criteria included a
single coating line with a single applicator
and near steady state operational condi-
tions. The testing approach called for col-
lecting continuous liquid and gas VOC
data to enhance determination of capture
efficiency over any given period. Suffi-
cient data collection was also designed in-
to the testing to allow an error analysis of
the capture efficiency determinations.
The coating line tested was a magnetic
tape coating process operating almost
continually 24 hours per day. The liquid
coating was pumped from the feed tank
and applied directly to the web with no
recirculation between the two points, ex-
cept during splicing of the web rolls.
The test design required measurements
similar to those in the laboratory test. The
major difference between the field test
and the laboratory test measurements was
in the liquid-phase stream. The field test
required determining the liquid mass
through determination of the volume and
density, since direct mass measurements
could not be made.
Results and Conclusions
The laboratory mass balance ex-
periments provided an opportunity to
assess the ability to close liquid/gas-phase
material balances using the available
methodology, while conducting com-
parative testing of three different VOC
analysis methods and testing of a method
for the volatile content of the coating.
Varied solvent amounts, solvent types,
and gas flow rates were systematically in-
troduced into the testing through a frac-
tional factorally designed test matrix in
order to simulate some important field
variables. The test results for all ex-
perimental runs are shown in Table 1.
Test results are also grouped by the gas-
phase measurement method used in Table
2.
Material balance closures by Method
25A gas-phase measurements were by far
the most successful in this set of ex-
periments. Accuracy, bias, and precision
were evaluated for the pure solvent and
the commercial mixture experiments as
separate groups. Accuracy seems more
than adequate for each group since the
mean closure values were, respectively,
99.9% (88.5-110%) and 102.2%
(93.5-119%). Bias for either group was
not statistically significant, since 100%
was included within the 95% confidence
interval in both cases. Precision, or
variability, estimates for the two groups
were excellent, since the coefficients of
variation were 5.9 and 8.9%, respectively.
Test results for the commercial coatings
test runs were less accurate than pure sol-
vent runs, probably due to the smaller
change in liquid mass and lower SLA con-
centrations.
An analysis was conducted of the im-
pact of test variables (e.g. mass
throughput of organics, organic composi-
tion, and gas stream flow rate) on the
Method 25A closure results. No variable
was found to have any impact of practical
significance.
Based on the laboratory test results, the
liquid/gas-phase material balance method
utilizing EPA Method 25A was further
tested under field conditions. Capture effi-
ciency determinations were made on an
hourly basis during each test period using
the gas-phase data collected from the
total hydrocarbon analyzer and the liquid-
phase data from the coating feed tank
measurements. These determinations
were made over a period of 114 hours,
consisting of six discrete batches of liquid
coating material.
Because the hourly capture efficiency
determinations varied significantly, the
data were further averaged over 24 hour
periods and over the different batch
periods to help smooth out the results
and to narrow the confidence limits (see
Table 3). The mean capture efficiency
determination for five 24-hour periods was
106.7% with a coefficient of variation of
7.4%. The capture efficiency determina-
-------
Table 1.
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
26A
22
23
26B
31
Liquid/Gas Mass Balance Closures
Elapsed Solvent Solvent
Time1 Type Amount
(min) (liters)
100
56
140
61
86
51
53
150
70
70
105
35
81
77
56
43
60
175
50
156
190
121
120
241
MEK
Toluene
Toluene
Mixture'1
MEK
MEK
Mixture''
Toluene
Mixture'1
Mixture''
Mixture7
MEK
Mixture1
Toluene
MEK
Mixture7
Mixture7
Mixture7
MEK
Comm 1'
Comm 2"
Comm 2"
Comm 2"
Comm 312
2
0.5
1
1
2
1
0.5
2
0.5
1
1
0.5
2
1
0.5
0.5
2
2
1
1
1
1
1
1
Measured
SLA
THC Closures
Flow Rate2 Aggregate3 Total*
(SCFMI (%) (%)
832
1400
843
1417
1825
1471
1395
1406
844
1753
918
1718
1461
1700
844
1787
1771
859
1834
1411
1769
922
1427
1730
100
101
98.0
107
88.5
109
93.5
98.0
90.1
110
102
98.0
102
96.1
93.9
99.4
98.0
104
100
119
93.5
99.0
104
99.0
100
101
96.1
108
88.5
107
100
99.0
89.3
110
101
100
100
95.2
103
100
102
95.2
107
/VD10
ND
ND
ND
ND
Method 18 Closures
MEK
1%)
73.0
—
—
84.7
103
88.5
103
—
75.8
174
78.7
76.9
84.7
—
92.6
96.1
90.9
76.3
77.5
90.1
117
86.2
98.0
186
Toluene
(%)
—
98.0
88.5
106
—
_
756
119
90.1
546
97.1
—
95.2
87.0
—
35.2
61.3
89.3
-
—
—
—
-
Total*
f%)
73.0
98.0
88.5
95.3
103
88.5
129
119
82.9
360
87.9
76.9
89.9
87.0
92.6
95.6
76.1
82.8
77.5
90.1
117
86.2
98.0
186
NIOSH 127 Closures
MEK
(%)
397
—
—
90.1
117
NM
NM
—
69.0
NM
123
45.2
131
—
NM
50
122
403
NM
NM
NM
118
NM
NM
Toluene
(%)
—
210
235*
216
—
—
NM
NM
155
NM
204
—
350*
111
—
149
290
369
-
_
_
_
-
Totaf
(%)
397
210
235
153
117
NM»
NM
NM
112
NM
163
45.2
240
111
NM
99.5
206
386
NM
NM
NM
118
NM
NM
1 Elapsed time for each analysis varied due to cycle time and sensitivity differences.
2Flow rate measured by EPA Method 2 during each run. (SCFM 47.124 x feet per second. I
'Calculated from (MW x 100/CW) where MW = sum of the balance weight losses and CW = sum of the weight losses from the VOC concentration and the
SLA flow rate. For the commercial mixtures, MW is the difference between the initial and final balance weights.
'Calculated from ITW x 100/CW) where TW is the initial total weight of so/vent placed in the system.
*These data are the arithmetic average of the MEK and toluene closures.
•'Average value of duplicate determinations.
7Mixture of 50% each by volume of MEK and toluene. Individual weights were used in the closure calculations.
'NM = not measured.
'Commercial mixture of rubber in MEK, specified by the manufacturer to contain 31% so/ids by weight.
"VVD = not determined since all of the material was not evaporated.
"Mixture described in 9 above diluted approximately 10:1 with MEK.
^Mixture described in 9 above diluted approximately 2:1 with MEK.
Table 2. Analysis of Mass Balance Test Results
Analysis
Type
EPA Method
25A/Byron
401 THC
Analysis
EPA Method
18/GC-FID
with
Speciation
NIOSH Method
127/Charcoal
Tubes
Data
Type
Pure solvent
(overall)
Commercial
mix (overall)
Pure solvent
(overall)
Commercial
mix (overall)
Pure solvent
(overall)
Mean of 95% Confidence
Closures Interval
99.9 97.1 - 102.7
102.2 89.6-114.9
105.4 75.0- 135.9
115.5 64.4- 166.7
190.4 126.1-254.6
Number of Standard
Observations Deviation
19 5.9
5 9.1
19 63.9
5 41.2
13 106.2
Coefficient
of Variation
5.9
8.9
60.0
35.7
55.8
-------
tions for six batch periods produced a
mean of 103.0% and a 4.6% coefficient
of variation.
In both cases above, the calculated
mean capture efficiencies were higher
than expected, since the maximum ex-
pected capture efficiency would be 100%.
A review of the process streams
measured, the measurement methods,
and quality assurance results indicated no
reason to believe that any source of VOC
went unmeasured or that any measure-
ment bias existed which would cause the
higher measurement values.
In order to estimate the reliability of the
capture efficiency results obtained in the
field test, a complete error analysis was
performed. Estimates for measurement er-
rors were provided by an external audit or
determined from repetitive measurements.
The analysis showed that a single
measurement has a 38% probability of be-
ing within ±5% of its true value. Finally,
if 3 individual sets of closure
measurements are made, the average
value should be within ± 10.7% of the
true value with a 95% confidence limit.
For six measurements, the average value
would be within ±7.6%. Using the data
collected at this site, it appears that the
limits can only be narrowed to approxi-
mately ±4% of the true value.
The confidence interval would be ex-
pected to be much narrower for the same
site using calculations based on a gas-
phase material balance and an exhausted
measurement enclosure. This is primarily
due to the form of the gas-phase material
balance equation. In the equation, the
numerator and denominator are both
composed primarily of the same measure-
ment value, therefore, minimizing the im-
pact of measurement bias or variability on
the results.
Table 3. Field Test Capture Efficiency Results
Test
Period
24 hours
Coating Batch
Number
of Tests
5
6
Range of
Results
100.5 - 120. 1%
96.9- 108.1%
Mean of
Results
106. 7%
103.0%
Coefficient of
Variation
7.4%
4.6%
D. B. Hunt andJ. L. Randall are with Radian Corporation, Austin. TX 78766.
Ronald J. Turner is the EPA Project Officer (see below).
The complete report, entitled "Measurement of Volatile Organic Compound Capture
Efficiency," (Order No. PB 85-173 243/AS; Cost: $13.00. subject to change) wit/be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Hazardous Waste Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati. OH 45268
•t, US GOVERNMENT PRINTING OFFICE 1985-559-016/27039
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
-: PERMIT No.
Official Business
Penalty for Private Use $300
0000329 PS
------- |