SEPA
United States Industrial Environmental Research EPA-600 7-79-152
Environmental Protection Laboratory July 1979
Agency Research Triangle Park NC 2771 1
Procedures for Aerosol
Sizing and H2SO4 Vapor
Measurement at Shawnee
Test Facility
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, 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-600/7-79-152
July 1979
Procedures for Aerosol Sizing and
Vapor Measurement at Shawnee
Test Facility
by
R. F. Maddalone, A. Grant,
D. Luciano, and C. Zee
TRW Defense and Space Systems Group
One Space Park
Redondo Beach, California 90278
Contract No. 68-02-2165
Task No.202
Program Element No. INE624
EPA Project Officer: Robert M. Statnick
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
-------
FOREWORD
This manual has been prepared for the Industrial and Environmental
Research Laboratory of the Environmental Protection Agency, Research
Triangle Park, North Carolina, as part of Task 2 of Contract No. 68-02-2165.
The technical objective of this project was to prepare a series of
procedure documents for sizing dry aerosols and measuring SO, entering and
leaving a Flue Gas Desulfurization (FGD) unit and written for GS-4 per-
sonnel or equivalent. The sizing method for the dry parti oil ate matter
entering the FGD process will be a manual technique utilizing a Brink
Impactor. A manual system for the FGD process effluent was chosen on the
basis of a literature survey, contacts with experts in the field, and an
evaluation of available information. The method chosen was the Meterology
Research Inc. .Cascade Impactor used out of stack. Finally a method for
S03 (H2S04 vapor) was developed based on the Controlled Condensation
(Goksoyr/Ross) method and was successfully tested under laboratory
conditions.
The project was divided into three areas of effort:
1. Aerodynamic Size Distribution
Measurement of Dry Aerosols
2. Procedure for Sampling and Analysis of S03
3. Quality Assurance
Aerodynamic Size Distribution Measurement of Dry Aerosols
Documents were prepared describing the methods for determination of
the size distribution of dry particulate matter at the inlet and outlet of
flue gas desulfurization (FGD) process. The FGD process inlet measure-
ment system was a Brink Impactor while the outlet measurement system,
which must be suitable for extremely low grain loading, was selected from
several candidate systems.
The selection of the outlet impactor system was based on the following
criteria:
Ease of assembly and operation - GS-4 level technicians
should be able to operate the instrument.
m
-------
Ease of sample recovery - Sample removal should be accomplished
under field conditions with minimum of effort.
t Construction material compatibility with sample and sampling
environment - The equipment should not corrode or in any
way contaminate the sample.
t Sampling period required for sample collection - Flow rates
should be maximized to collect adequate amounts of sample for
measurement in a reasonable sampling period under low grain
loading conditions.
Sample capacity - The system should be flexible enough to
accurately size and collect particulate under high and low
grain loadings.
t System design to minimize wall losses and re-entrainment -
All samples should be deposited in collection trays or cups.
Applying these criteria, the MRI Impactor was selected. Procedure
documents describing the operation of the Brink and MRI impactors are
found in Chapters 1 and 2 respectively. These documents include: equip-
ment lists, equipment assembly and preparation, on-site set-up and
operation, sample removal and handling procedures, and sample weighing
procedures. Other than making reference to known procedures (such as
EPA Methods 1 through 4), this document will be designed to stand by itself
and be directed toward GS-4 or equivalent personnel.
Procedure for Sampling and Analysis of SOg
A procedure to sample and analyze for S03 in flue gas prior to and
after FGD process was written. From TRW's knowledge of the S03 sampling
problem, the Controlled Condensation (Goksoyr/Ross Coil), Brink Impactor
and selective liquid impingement appeared to be the methods available. A
literature evaluation of the systems was based on the following criteria:
Sensitivity
Selectivity
Precision
Accuracy
Efficiency
t Ease of Operation
Reliability/Maintainability
Sample Recovery for Analysis
iv
-------
As a result of this evaluation the Controlled Condensation system
was tested in the laboratory simulating the conditions in the FGD unit,
and was found to be precise and accurate. Chapter 3 contains the
document describing this procedure.
Quality Assurance
This effort was devoted to develop methods that will ensure the
overall quality of the data taken in the above procedures. Chapter 4
describes general techniques associated with the dry aerosol sizing and
S03 procedures as well as specific QA activities for each procedure.
Included in the specific QA activities are:
Critical checkpoint lists for each procedure
Data validation procedures
t Maintenance schedules
t Troubleshooting procedures
-------
CONTENTS
Page
Foreword iii
Figures ix
Tables x
Acknowledgement xi
1. PROCEDURE FOR SAMPLING THE INLET OF FLUE GAS DESULFURIZATION
(FGD) UNIT WITH A BRINK IMPACTOR 1
1.1 Documents 1
1.2 Equipment and Materials 1
1.2.1 Sampling Equipment 1
1.2.2 Coating and Weighing Materials 3
1.3 Requirements 3
1.3.1 System Design 3
1.3.2 Sampling Procedure 3
1.3.3 Handling 5
1.3.4 Calibration and Maintenance 5
1.3.5 Cleanliness 5
1.3.6 Safety 5
1.4 Procedure 6
1.4.1 Probe Manufacture 6
1.4.2 Laboratory Preparation of Brink Impactor .... 6
1.4.3 Measurements and Calculations for Isokinetic
Sampling 12
1.4.4 Isokinetic Operation of the Brink Impactor . . 14
1.4.5 Site Equipment Setup and Operation 21
1.5 Data Reduction 24
2. PROCEDURE FOR SAMPLING THE OUTLET OF A FLUE GAS DESULFUR-
IZATION (FGD) UNIT USING A MRI IMPACTOR 32
2.1 Documents 32
2.2 Sampling Equipment 32
2.2.1 Impactors 32
2.2.2 Sampling Probe 32
2.2.3 Aerotherm Sampling Train 35
2.2.4 Tools and Equipment 35
2.2.5 Equipment and Materials for Coating
Collection Discs 35
vi
-------
CONTENTS (Continued)
2.2.6 Probe Construction 36
2.3 Requirements 36
2.3.1 System Design 36
2.3.2 Sampling Procedure 36
2.3.3 Handling of Collection Discs 36
2.3.4 Calibration and Maintenance 37
2.3.5 Cleanliness 37
2.3.6 Safety 37
2.4 Procedure 37
2.4.1 Laboratory Preparation on MRI Cascade Impactor . 39
2.4.2 Measurements and Calculations for Isokinetic . . 42
2.4.3 Equipment Setup 43
2.4.4 Operation of Aerotherm and MRI Impactor
During Sampling . 47
2.4.5 Operation of Aerotherm and MRI Impactor
During Sampling 48
2.5 Data Reduction 51
3. DETERMINATION OF H9SOA VAPOR USING A CONTROLLED CONDENSATION
COIL t 54
3.1 Documents 54
3.2 Equipment and Materials 54
3.2.1 Sampling Materials 54
3.2.2 Reagents and Apparatus for HgSO^ Titration . . 58
3.3 Requirements 60
3.3.1 System Design 60
3.3.2 Sampling 60
3.3.3 Handling of Glassware 60
3.3.4 Calibration and Maintenance 60
3.3.5 Cleanliness 60
3.3.6 Safety 62
3.4 Procedure 62
3.4.1 Probe Manufacture 62
3.4.2 Filter Holder Fabrication 64
3.4.3 Site Equipment Setup and Operation 64
3.4.4 Analysis Procedures 70
4. QUALITY ASSURANCE METHODLOLGY 74
vi 1
-------
CONTENTS (Concluded)
4.1 Laboratory Equipment Care and Technique 75
4.1.1 Analytical Balance 75
4.1.2 pH Meters 77
4.1.3 Laboratory Analytical Glassware 78
4.1.4 Desiccators 80
4.1.5 Dry Test Meters 81
4.1.6 Ovens 82
4.1.7 Reagent Storage 82
4.1.8 Blanks 83
4.1.9 Titrations 84
4.1.10 Handling 86
4.2 Sampling Quality Control 86
4.2.1 Brink Methodology 87
4.2.2 MRI Methodology 97
4.2.3 Goksoyr-Ross Methodology 106
4.2.4 Maintenance Schedules 112
4.2.5 Troubleshooting and Repair Procedures .... 112
4.3 References . 112
Appendices
A. Isokinetic Flow Rate 122
B. Derivation of H^SO. ppm Calculation Equation 139
vm
-------
FIGURES
Page
1. Brink Impactor 4
2. Schematic of Brink Sampling Probe 7
3. Brink 5-Stage Impactor Disassembled 9
4. Brink Laboratory Data Sheet 11
5. Brink Impactor and Filter Assembly ' 13
6. Detail of 8 Point Two Diameter Traverse Pattern 14
7. Brink Field Data Sheet 15
8. Gas Velocity Versus Gas Flow for Several Nozzle Sizes 19
9. Sample Calibration Curve for Brink Impactor 20
10. Upper Stages of MRI Impactor 33
11. Lower Stages of MRI Impactor 33
12. Assembly Drawing of Model 1503 Inertia! Cascade Impactor. ... 34
13. MRI Site Setup 38
14. MRI Laboratory Data Sheet 41
15. MRI Field Data Sheet 45
16. MRI Impactor Stage Cut-Off Diameter (y) 52
17. Vycor Sampling Liner 55
18. Controlled Condensation Coil -57
19. Controlled Condensation System Setup 51
20. Controlled Condensation System Probe Design 63
21. Quartz Filter Holder 55
22. Controlled Condensation Field Data Sheet 68
23. Controlled Condesnation Coil Rinsing Apparatus 69
24. Laboratory Data Sheet 71
25. Control Chart for Controlled Condensation on Measurements of
H2S04 HI
ix
-------
TABLES
Page
1. Location of Traverse Points in Circular Stacks (Percent
of Stack Diameter from Inside Wall to Traverse Point) 17
2. Calculation of Particle Size Cutoffs, Known Data 25
3. Brink Dry Aerosol-Size Distribution (Calculation of Particle
Size Cutoffs, Calculated Data) 26
4. Critical Checkpoints for Brink Dry Aerosol System 89
5. Critical Checkpoints for MRI Dry Aerosol System 99
6. Critical Checkpoints for G/R H2S04 Sampling System 107
7. General Maintenance Schedule 113
8. Troubleshooting and Repair 116
-------
ACKNOWLEDGEMENT
This document describes the procedures developed on Task 02, Lime-
stone Demo Support, on EPA Contract No. 68-02-2165, Sampling and Analysis
of "Reduced" and "Oxidized" Species in Process Streams. The Chemistry
and Materials Laboratory Applied Technology Division was responsible for
the work performed on this task. The work was originally conducted under
the EPA Project Officer Dr. R. M. Statnick, Environmental Research Center,
Research Triangle Park, North Carolina. The current Project Officer is
Mr. Frank Briden. Dr. C. A. Flegal was the Program Manager and the Task
Order Manager was Dr. R. F. Maddalone. Major technical contributions
were provided by Mr. Don Luciani, Ms. Carol Zee, and Mr. Arnie Grant. We
wish to thank Dr. James W. Buchanan, Dr. D. E. Wagoner, and Dr. Douglas
Van Osdell of Research Triangle Institute and Mr. Ray Crote of the EPA
for their review of the Quality Assurance chapter. The comments and sug-
gestions on impactor methodology given by Dr. David S. Ensor of Meteoro-
logy Research Institute and Dr. Kenneth M. Cushing of Southern Research
Institute were extremely helpful in developing the Brink and MRI proce-
dures. The overall review and support during the program from Mr. Richard
G. Rhudy of Bechtel, and Mr. Steven Newton and Mr. John Lawton of the TVA
has been greatly appreciated. Acknowledgement is made to Mr. John Lungren
and Ms. Carmen de la Fuente for their assistance during preparation and
publication of this document.
xi
-------
1. PROCEDURE FOR SAMPLING THE INLET
OF AN FGD UNIT WITH A BRINK
IMPACTOR
This method is for the determination of the aerodynamic size distri-
bution of dry solids prior to the flue gas desulfurization (FGD) units at
the TVA Shawnee Power Plant in Paducah, Kentucky. The Brink impactor with
a specially designed internal cyclone is used to provide aerodynamic size
distribution information between 0.3yn to 10pm in 6 distinct cuts. The
recommended flowrate is 0.01 to 0.08 cfm.
This procedure uses the Brink impactor out of stack to sample the
particulate from the gas stream entering the wet scrubber. After the
large particles have been removed from the gas stream by the internal
Brink cyclone, the remaining particles in the gas stream are then separated
by a Brink Cascade impactor. By weighing each stage of the Brink impactor,
the aerodynamic size distribution can be determined.
1.1 DOCUMENTS
1-1 Federal Register. 36(247):24888-9.
1-2 Brink BMS-11 Instruction Manual. Monsanto, Enviro-Chem
Systems, Inc., St. Louis, Missouri.
1-3 McCain, J.D., A.N. Bird, and K.M. Gushing, "Field Mea-
surements of Particle Size Distribution with Inertial
Sizing Devices. Southern Research Institute, EPA
650/2-73-035, 1973.
1-4 Smith, W.B., K.M., Gushing, G.E. Lacey, and J.D. McCain.
Particle Sizing Techniques for Control Device Evalua-
tion. Southern Research Institute, EPA 650/2-74-102a,
1975.
1.2 EQUIPMENT AND MATERIALS
1.2.1 Sampling Equipment
Brink impactor and 1 cfm pump (obtainable from Monsanto Enviro-
Chem System Inc., St. Louis, Mo.)
Aerotherm Isokinetic Flowrate Calculator (#HVSS-901, Aerotherm
Corp., Mountain View, Ca.)
-------
Assemble a 5-foot probe from the following materials (see 1.4.1):
1) Appropriate size Brink nozzle
2) One Swagelok 1/4-inch SS male elbow (SS-400-2-4)
3) 5-foot x 1/4-inch 304 SS tubing
4) 4-1/2-foot x 1-1/2-inch OD x 0.035 inch wall aluminum tubing
5) Two silicone rubber No. 8 stoppers (A. H. Thomas 8747-E83)
6) Glass tape (Scotch glassfiber electrical heating tape)
7) 50 feet of heater wire (S. Moore Co., Aurora, Ohio,
#1659-40110)
8) Two Omega (Stanford, Conn.) shielded thermocouples (I/C)
(#TJ36-ICSS-18G-12 with a 12-foot lead)
9) Two Omega (Stanford, Conn.) unshielded thermocouples (I/C)
(#IRCO-032 with a 6-foot lead)
10) Five Omega male connectors (ST-IRCO-M)
11) One 6-foot heavy duty (^ 20A) electrical cord with a male plug
12) Two 1-1/2 inch hose clamps
13) 1/2-inch Teflon pipe tape
14) Two square yards of asbestos cloth (VWR, Atlanta, Georgia,
#10930-009)
Stopwatch
t Heating mantle for impactor and filter (Glass-Col, Terre Haute, Ind.)
Two wash bottles, one with distilled H20 and one with acetone
§ A source of 110V electrical power must be provided at the sampling
location
Five-place analytical balance
Gelman, 47 mm inline filter housing (Product No. 2200, Gelman
Inst. Co., Ann Arbor, Michigan)
t Reeve Angle 934-AH, 47 mm glassfiber filter (Reeve Angel Co.)
-------
Onp Dwyer Series 2000 magnehelic differential pressure gauge with
the low temperature option for 0.30 in. H20 (Dwyer Instr. Inc., Michigan
City, Ind.).
One Dwyer Series 2000 magnehelic differential pressure gauge with
the low temperature option for 0-60 in. FLO.
1.2.2 Coating and Weighing Materials
Apiezon H grease.
Drierite, 5 Ib.
Large desiccator, Kimble #21050, 250 mm min. diameter with
porcelain plate.
t Petri dishes (top and bottom), 60 x 15 mm pyrex.
0 Camel hair brush.
Tweezers.
PVC gloves, U.S. Industrial Gloves, Compton, California.
Whatman No. 1 paper sheets (46 x 47 cm).
Kimwipes.
Acetone, reagent grade.
3-inch rubber plug with a 1.5 inch hole drilled in the center.
Toluene, reagent grade.
1.3 REQUIREMENTS
1.3.1 System Design
The Brink sampling train consists of a 5-ft. x 1/4-in. ID 304 stain-
less steel probe, Brink internal cyclone (based on SoRI design), five
impactor stages, a Gelman 47 mm filter holder and a 934-AH Reeve Angel
filter, three impingers, a pump and a calibrated orifice (see Figure 1).
1.3.2 Sampling Procedure
The flow rate through the impactor determines the size cut-off that
each stage will collect. As will be described in Section 1.4, an average
isokinetic sampling rate will be determined. Once the average flow rate
-------
S-PITOT
r
BRINK
NOZZLE
STACK
,RAIL
USE WIRE OR ROPE TO SUPPORT
IMPACTER FROM HOOK
BRINK WET
AEROSOL PROBE
PORT
REDUCER
SS-400-R-4
RUBBER STOPPER
FOR GAS SEAL
UNION
CROSS
IMPINGERS
BRINK CYCLONE
STAGE
HEATING
MANTLE
FILTER
HOLDER
UNION
CROSS
Figure 1. Brink Impactor
4
-------
is established, it must be maintained throughout the run regardless of the
individual velocities at each point in the traverse.
1.3.3 Handling
Care must be taken to limit contact with the stages. At no time after
cleaning should the stages be touched with ungloved hands. All of the
laboratory manipulations are to be done in a clean environment using tweezers
to handle the stages. Remember, several grains of dust could represent the
total weight captured on the stages. Contamination control is essential
during greasing, drying, and weighing.
1.3.4 Calibration and Maintenance
After each run, the probe nozzle, probe, connecting lines, S-pitot
tubes, impactor, and impinger system must be cleaned. After the run, the
probe and connecting lines are rinsed with reagent grade acetone. The
S-pitot tube should be backflushed with a high pressure air line. The
impactor cleaning procedures are detailed in 1.4.4. The impinger system
is flushed out and the proper solvents replaced in the impinger bottles
prior to the next run.
Besides these daily procedures, the S-pitot C and the AH@ of the
flowmeter orifice are determined every 3 months. If any evidence of corro-
sion appears (pitting, scale build-up, etc.), the C and AH@ recalibration
procedure should be repeated as needed or the part returned to the manu-
facturer. The Brink impactor must have an up-to-date AP vs cfm calibration
chart. This chart is supplied by the manufacturer or can be determined
experimentally. See Tables 7 and 8 for further information on equipment
care and maintenance.
1.3.5 Cleanliness
Gas carrying lines should be cleaned weekly. In particular, no partic-
ulate build-up in the pi tot tube can be tolerated. The probe, lines, and
impactor must be completely clean before use.
1.3.6 Safety
OSHA safety requirements as regards to working environment and operator
safety will be met at all times. The reagents mentioned in the procedure are
not extremely toxic, but misuse of any chemicals can be harmful.
-------
1.4 PROCEDURE
1.4.1 Probe Manufacture
Refer to Figure 2. The necessary equipment is listed in paragraph
1.2.1. The following instructions are used for the construction of
a 5-foot probe. At all times follow correct electrical safety procedures.
Be sure that no sharp pieces of metal abrade any of the electrical wires.
a) Cut the 304 SS 1/4-inch tubing to 4.5 ft.
b) Attach the male elbow to one end of the probe.
c) The proper Brink nozzle will be screwed into the male thread of
the elbbw prior to sampling.
d) Align the shielded thermocouple as shown in Figure 2. Using the
glass tape, secure the shielded thermocouple to the probe. Approximately
halfway down the probe from the Swagelok elbow, attach the unshielded
thermocouple. Continue down the probe, securing both thermocouple leads
simultaneously against the tube.
NOTE
Be careful never to kink
thermocouple or thermocouple
leads.
e) Approximately three inches from the end of the tube, place a final
wrapping of glass tape.
f) Take 12 feet of heating wire and fold it in half.
g) Beginning six inches from Swagelok union, wrap the probe with the
doubled up heating wire. Make sure the heating wire is snug up next to»
the probe and secured every six inches with a wrapping of glass tape. Do
not lay the coil of the heating wire on the tip of the unshielded TC.
Simply gauge the wrapping to place the TC in one of the gaps between
coils. Secure the heating coils to either side of the TC with tape to
prevent them from slipping over the TC. Wrap the coils close enough so
that the heating wire is completely used up three inches from the end of
the probe. Secure the end of the heating tape with a final wrap. Wrap
one layer of asbestos cloth around the heating tape.
-------
WIRE
S-PITOT
MALE
ELBOW
1/4" 304 SS
TUBING
HOSE CLAMP
GLASS
TAPE
SILICONS RUBBER
STOPPER'
UNSHIELDED
THERMOCOUPLE
STACK
THERMOCOUPLE
PROBE
THERMOCOUPLE
HEAT TRACE
CONNECTOR
SHIELDED SILICONE
THERMOCOUPLE RUBBER
STOPPER
BRINK NOZZLE
Figure 2. Schematic of Brink Sampling Probe
-------
h) Bore a 5/16-Inch hole into the two No. 8 silicone rubber stoppers,
then cut a slit vertically down one side of the stopper into the 5/16-inch
hole. The slit will allow easy assembly. Also provide a cutout for the
pitot tube along the side.
i) Slide the aluminum sheath over the probe. Avoid scratching the
insulation on the electrical leads. Position the sheath so that the end
near the elbow extends ^ 1 inch past the start of the heating tape. Slide
the pitot tube into the sheath.
j) Spread the stopper open, slip it over the stack end of the probe,
and position it properly over the S-pitot cut-out. Be sure the S-pitot is
positioned parallel to the nozzle. The stopper is then wired to help hold
it in place. Repeat this procedure for the other end, except use a hose
clamp to hold the back stopper in place.
k) After the back stopper is in place, completely wrap the exposed
heating coils with glass tape.
1) Place the male quick connects on the end of the TC leads. The
red TC lead goes to the negative terminal. Connect the heavy duty extension
cord to the heating tape.
m) The probe should be tested in the laboratory to ensure that all
parts are in order. Simply connect the heating wire to the Variac and
allow the probe to heat up. Monitor the temperature to verify the TCs are
functioning.
NOTE
Whenever heating up the probe, start
off with very low power inputs (^ 5%)
until heating starts.
n) The probe is now ready for use.
1.4.2 Laboratory Preparation of Brink Impactor
a) Disassemble the Brink impactor (Figure 3} by unscrewing each stage.
With gloved hands clean each collection plate by wiping the surface with
a Kimwipe wetted with acetone.
8
-------
Figure 3. Brink 5-Stage Impactor Disassembled
-------
NOTE
Residual Apiezon H can be removed using aim NaOH
solution followed by toluene.
Inspect the plate after cleaning for particulate or finger marks on the
collection surface. Inspect interior of the impactor housing for particulate.
Clean the interior of the impactor with a squeeze bottle containing acetone,
and Kimwipes. For hard-to-reach areas, use a camel-hair brush to remove the
particulate.
b) With the tweezers, dip each stage in a 100 ml beaker with ^50 ml
of reagent grade toluene to clean the surface. Remove and hold the stage
in the air until the toluene has dried. Place each stage in a separate
labeled petri dish.
c) With a rubber policeman, carefully apply the Apiezon H grease to
the center of the stage. Should the grease be painted over the edge of
the stage, remove the Apiezon H with toluene and start again. Coat six
stages and use one as a blank.
d) Place the covered petri dishes with the stages into an oven for four
hours at 175°C (347°F). Three glassfiber filters in petri dishes are
conditioned at 287°C (574°F) for four hours.
NOTE
Always handle the filters with a tweezer and avoid
breaking off pieces of the filter.
e) After four hours, remove the petri dishes with stages and filters
and allow them to equilibrate in a desiccator for two hours.
f) Once the stages and filters have been dried and desiccated, they
are weighed on a balance capable of weighing to the nearest 0.01 mg. Remove
the petri dishes from the desiccator just prior to weighing (keep desiccator
closed otherwise). Remove the stages from petri dish with a tweezer being
careful not to touch the greased area with the tweezers, and place them on
the balance. After weighing, record the weight and disc number (on petri
dish) on the laboratory data sheet (Figure 4) and place the coated disc
back in the petri dish, cover, and place the petri dishes near the impactor
in a clean, dust-free area. Weigh a 47 mm filter to the nearest 0.01 mg
and immediately load it into the filter holder,
10
-------
BRINK DRY AEROSOL SIZE DISTRIBUTION
SAMPLE LOCATION .
DATE/TIME
RUN NO.
DATE
STAGE
FILTER
BLANKS
FILTER
BRINK
CYCLONE
TOTAL
DISC*
WEIGHT
FINAL
TARE
GAIN
%
% CUM
MICRONS
DS
% - Weight gain on each stage divided by the total weight gain.
CUM1; - Starting with the filter accumulate each stage to arrive at the
cumulative percent smaller than the previous D,..
* - Disc Code for labeling petri dishes should be the date of run,
stage no. and run letter series (example: 8/27/75, 1A;
8/27/75, 2A; etc.}- Tne letters series represents the sequential
number for each successive run that day: 8/27/75, 1A;
8/27/75, IB would be the next run.
f
** - As corrected by equation (3),
Figure 4. Brink Laboratory Data Sheet
11
-------
NOTE
Avoid breaking off pieces of the filter.
Retain the spare, weighed stages and filters for handling blanks. Store them
in the desiccator until they are needed. (See 4.1.8 for correct weighing
procedures.)
g) Place a greased stage on the Brink #5 collection level (the bottom
section) recording the stage used on the Laboratory Data Sheet (Figure 4).
Repeat this procedure for the rest of the impactor stages (4, 3, 2, 1) until
it is completely loaded. Finally, place the internal cyclone on top of the
first stage. During and after this procedure, the impactor must remain in
an upright position.
h) After the impactor is completely loaded, attach the inlet and outlet
lines including the filter as shown in Figure 5.
i) After the Brink impactor is assembled, it should be leak checked
in the laboratory. Connect a vacuum gauge to the inlet of the impactor,
and attach the outlet of the impactor to the in-house vacuum line.
j) Leave the vacuum on until the gauge indicates 380 torr (15 in. Hg).
k) Close the vacuum line and note any rise in pressure. The vacuum
should not vary over several minutes.
1) If a leak is noted by a decreasing vacuum reading, check the
impactor to verify that all connections are tight and the vacuum gauge is
working. Be sure that all the vacuum lines have tight seal as well. If
these measures do not locate the leak, take the impactor apart and replace
any suspicious gaskets, then repeat the vacuum test.
Once the impactor is leak checked, both ends are sealed to prevent dust
from entering, and the impactor is placed inside of the heating mantle and
taken to the sampling site.
1.4.3 Measurements and Calculations for Isokinetic Sampling
a) The duct geometry must be first considered. For the circular
40-inch diameter ducts at the Shawnee limestone wet scrubber, refer to
Figure 6.
12
-------
SS-400-R-4
REDUCER
PRESSURE TAP
/ 1/4" UNION CROSr
I /4" SWAGELOK CROSS
SS 47 MM
FILTER HOLDER
TO IMPINGERS
1/4" MALE ADAPTER
THERMOCOUPLE TAP
BRINK IMPACT OR
GLASS-COL
HEATING MANTLE
PRESSURE TAP
Figure 5. Brink Impactor and Filter Assembly
13
-------
3" SAMPLING PORT
3"SAMPLING PORT
TRAVERSE POINTS
Figure 6,
Detail of 8 Point Two Diameter
Traverse Pattern
b) Sixteen sample points are selected; eight along one axis across
the duct, and eight along another axis at 90° to the first. The distance
along the probe to mark each sample point location is obtained from Table 1,
c) Test site should consist of a sampling port in the stack with an
opening to allow the easy insertion of the sampling probe, and sealed to
minimize the disturbance of the flow during sampling. Because of the
negative pressure in the stack at the sampling sites, extra care should be
taken in ensuring a good seal around the probe. A poor seal will lead to
low temperatures at the first sampling point. The electrical power
required to operate the equipment must be available is approximately 35
amp/115V.
1.4.4 Isokinetic Operation of the Brink Impactor
a) Prior to initiation of sizing experiments, the S-pitot probe is to
be recalibrated.
(FDS), Figure 7.
The C is noted and entered into the Field Data Sheet
P
14
-------
1. A P AND I, TRAVERSE DATA
SAMPLE
PORT
TRAVERSE
POINT
(JT»2 =
AP
IN. H2O
AVERAGE
NrA P
IN. HjO
S~L~r =
TS,-F
Ts =
2. BRINK OPERATIONAL VARIABLES
VARIABLE
CP
>s
(Jfp)2
TS
%H20
DN
MW2
vs
*pb
APc
VALUE
29.5
Figure 7. Brink Field Data Sheet
-Continued-
15
-------
3. FIELD DATA SHEET
PLANT
DATE/TIME
SAMPLING LOCATION
RUN NUMBER
OPERATOR
AMBIENT TEMPERATURE
BAROMETRIC PRESSURE
STACK PRESSURE, (P$) J
FILTER NUMBER
INLET GAS FLOW
PROBE HEATER SETTING
SYSTEM LEAK RATE
SCHEMATIC OF TRAVERSE POINT LAYOUT
READ AND RECORD ALL DATA EVERY MINUTES
MOISTURj
Imping «n
Final vol.
Initial vol.
Net vol.
Silica Gel
Final wt.
Initial wt.
Net wt.
Total H-O
Impingers
Silica gel
Total
ml
ml
ml
9
9
9
ml
g
9
TIME '
STACK
TEMPERATURE,
-------
TABLE 1. LOCATION OF TRAVERSE POINTS IN CIRCULAR
STACKS (PERCENT OF STACK DIAMETER FROM
INSIDE HALL TO TRAVERSE POINT)
Traverse Point
Number on a
Diameter
1
2
3
4
5
6
7
8
9
10
Number of Traverse Points on a Diameter
5
4.4
14.7
29.5
70.5
85.3
95.6
8
3.3
10.5
19.4
32.3
67.7
80.6
89.5
96.7
10
2.5
8.2
14.6
22.6
34.2
65.8
77.4
85.4
91.8
97.5
b) Using the S-pitot probe and referring to Figure 6 and Table 1,
perform a velocity traverse at the sampling site. At each point, record the
AP and stack temperature (T<.), and AP on the FDS. After the velocity
traverse, be sure to clean the probe of any debris.
c) Determine the average temperature (T
-------
4) Turn over calculator without moving Cursor and set M<. under
Hairline.
5) Read Stack Velocity V$ at stack pressure PS>
6) Record all this data on Table 2 on the FDS (Figure 7).
If the flowrate calculation is off scale on the Aerotherm calculator, then
use this formula:
V, - 85.48 (C) (Jff) /TS * 46° (1)
s p V ps Ms
'e) From the nozzle selection chart (Figure 8) and Vs select a probe
nozzle that will give a- sampling rate (F) between 0.01 and 0.08 cfm. Use
the largest nozzle (DN) possible without exceeding 0.08 cfm. Record the
D., on Table 2 on the FDS.
NOTE
A 1.5 mm nozzle is the recommended size.
f) Using the calibration curve for the Brink Model B cascade impactor
select a AP. corresponding to F (Figure 9).
g) Calculate the AP under stack conditions.
h) Record these values on Table 2 on the FDS (Figure 7).
i) Symbols:
C = Pi tot tube coefficient (in, HJ3).
P *
TS = Average stack temperature (°F)
P_ = Stack pressure (i.n. Hg)
O
PT,. = Static pressure read at inlet to Impactor (normally in.
IM H20)
PT. = Static pressure read at inlet to Impactor converted to
IA absolute pressure
-------
0.20
0 10 20 30 40 50 60 70 80 90 100 110 120
GAS VELOCITY IN FEET PER SECOND
Figure 8. Gas Velocity Versus Gas Flow for Several Nozzle Sizes
-------
u.
U
9
1.0
.8
U.6
0.4
0.2
0.10
0.08
0.06
0.04
0.02
0.0
(
^
^
-*t^5
^
'
-
CALIBRATION CURVE FOR
BRINK, MODEL B, CASCADE
IMPACTOR(BMS 11)
CALIBRATION MADE WITH AIR
AT 14.7 PSIA & 25°C (77°F)
^
^
^
-^
)j 0.2 0.4 0.6 0.8 1.0 2 4
(in. Hq)
^
-
6 8 1(
Figure 9. Sample Calibration Curve for Brink Impactor
20
-------
D., = Nozzle diameter (in.)
AP = Average pitot AP (in. HgO)
AP. = Pressure drop across Brink Impactor at standard conditions,
b in. Hg
AP = Pressure drop across the Brink Impactor corrected to stack
c conditions, in. Hg
AP = Pressure drop across the Brink impactor during the run.
a
M. = Molecular weight of the stack gas
Vs = Average velocity in the stack
F = Brink sampling rate (cfm)
NOTE
Capital subscript (S) refers to stack conditions.
1.4.5 Site Equipment Setup and Operation
a) In the 3-inch port insert a 3-inch plug with a 1.5-inch hole
drilled into the center. Another 3-inch plug is used to prevent gases
from escaping from the wet scrubber prior to the insertion of the probe.
Clean the probe with a brush and rinse with acetone to ensure that all
particulate matter is removed prior to the run.
b) Support the Brink system as shown in Figure 1.
c) Connect the 0-60 in. H^O magnehelic gauge across the impactor.
The 0.30 in. HpO magnehelic gauge is connected to the inlet of the impactor.
Be sure that the high and low pressure taps are connected to the correct
inlets of the gauge.
d) Connect all the thermocouples into the readout.
e) Make sure that all of the connections that have been made are
tight.
f) Put the correct nozzle on the end of the probe.
g) Connect Brink impactor to probe via reducer fitting. If a 1/4-inch
reducer is not available, use a 1/4-inch union and a short (^ 3-inch) piece
of 1/4-inch tubing to go with the union to the union cross.
21
-------
h) Connect the soap bubble flow meter to the vacuum pump exit. Be
sure that the bubble flow meter is vertical. Close off the end of probe
with a stopper and turn on the vacuum pump and adjust the vacuum to read
380 torr (15-inch Hg).
i) Begin measuring the flow rate with the bubble gauge. If the leak
rate is less than 23 mL/min. (0.0008), then the system is ready for use.
If a leak rate greater than 23 mL/min. is found, the system should be
checked for loose joints and connections. The pump should also be checked
and any worn parts replaced. (See Tables 7 and 8 for further information).
j) After the leak check, begin to heat the probe and impactor to the
highest stack temperature plus 12°C (25°F), but not higher than 175°C
(347°F). Do not overheat the impactor or weight loss problems will occur
with the greased stages.
k) When probe and impactor are 12°C (25°F) above stack conditions,
the run can start.
NOTE
Maximum use temperature for the probe is 400°F.
1) Turn the pump on and immediately insert probe into the stack until
it is at the first sampling point. Be sure the S-pitot and nozzle are
parallel to the gas flow. Immediately adjust the AP across impactor at
AP . Record AP , TS, PyM, skin temperature, I/O gas temperatures from
impactor, T , gas meter readings, and AP from the pitot.
Sample at each point for 45 seconds. Be sure to complete all other
information on the FDS.
m) At the end of the run, the probe is removed from the stack and
the pump is shut down. Slowly close the flow control valve to dissipate
any back pressure in the system and thus prevent water from surging forward
in the impinger system. Stop the stopwatch and record final gas meter
reading on field data sheet.
n) Using gloves, remove the Brink Imoactor with heating mantle and
return it immediately to the library. Cover the inlets to the impactor
to prevent particles from entering or leaving.
22
-------
NOTE
During this period, avoid jarring the impactor. Extreme
care should be taken to avoid the addition or loss of
collected particulate. Carry the impactor upright and
do not expose it to dust.
o) Carefully rinse the probe with reagent grade acetone collecting
the rinse until a clean stream of liquid issues from the probe.
NOTE
Take extreme care in performing this task as the small
amounts of particulate matter recovered represent a large
portion of the total particulate aerosol collected.
Both contamination and loss of sample must be avoided.
Any accidents which occur must be recorded on the field
data sheet.
p) Once the impactor is transferred to laboratory, clean the outside
of the impactor of any dust. This should be done in the prep room prior to
entering the clean room.
q) Be sure that the correctly labeled petri dishes are nearby, ready
to accept the collection plates as they are removed.
r) With the impactor in an upright position, begin to remove the
housing starting at the top. Remove the cyclone cup and place it in the
correct petri dish. Inspect the inlet nozzle for any sign, no matter how
little, of particles collected on the walls. Note the presence of the
particulate matter on the back of the laboratory data sheet.
s) If any particles are found, they should be carefully brushed onto
the collection plate below their collection point.
t) Inspect all the nozzles for any sign of pitting or corrosion.
Especially inspect the sides of the nozzle for particles that might have
collected there. If any particles are found, note this fact along with a
description of any patterns formed, color, or quantity obtained on the back
of the laboratory data sheet.
u) Carefully brush these particles onto the collection plate below
the nozzle.
23
-------
v) Repeat these activities for all the stages and the filter. In the
filter's case be sure that all fragments of the filter are removed from the
filter support; even the loss of the smallest fragment can affect the weight
of the filter.
w) Desiccate the collection plates and filter for 2 hours.
x) Rinse the connecting lines from the probe to the impactor with
acetone until a clean stream is obtained. Add this rinse to the probe rinse
y) Rinse the connecting lines from the impactor to the filter housing
with acetone. The particulate weight after evaporation is added to the
filter.
z) Evaporate enough of the acetone from the probe and line rinses so
that all the particulate and the remaining acetone can be quantitatively
transferred to a tared 30 ml beaker. At all times handle the 30 ml beaker
with gloved hands.
aa) Evaporate the bulk of the acetone from the 30 mL beakers on a hot
plate allowing the rest of the acetone to air dry in a clean, dust free
area. Dry the particulate at 110°C for two hours and desiccate with the
rest of the samples for approximately 2 hours.
bb) Weigh the collection plates and filter to the nearest 0.1 mg.
At the same time weigh the balance and sample blanks (see section 4.1.8
for specific procedures to correct for any weighing errors). Weigh the
probe and line rinses. Record the data on the laboratory data sheet
(Figure 4).
1.5 DATA REDUCTION
NOTE
A computer program for this data reduction is
available from Monsanto Enviro-Chem, St. Louis, Mo.
1) Table 2 contains the known data; enter the field data in Table 3.
NOTE
(Lower case (s) subscript refers to an impactor stage.)
24
-------
TABLE 2. CALCULATION OF PARTICLE SIZE CUT-OFFS, KNOWN DATA
Known Variables
Density of Aerosol Particle (g/cc)
Molecular Weight of Sample Gas
Molecular Weight of Calibration Gas
Temperature of Gas at Calibration
Conditions, °K
Static Pressure Under Calibration
Conditions, atm.
Gas Viscosity at Sampling Conditions
(3230F), poises
Stage Jet Diameter, cm
Dimension Conversion Constant
Data
Pp * 1
MW2 = 29.5
MW1 = 29.0
Tj = 298
P = 1.0
y = 2.18 x 10"4
Dr = 0.249
Ll
DC = 0.1775
Dr = 0.1396
L3
DC = 0.0946
Dr = 0.0731
C5
gc = l
25
-------
TABLE 3. BRINK DRY AEROSOL-SIZE DISTRIBUTION
(CALCULATION OF PARTICLE SIZE
CUTOFFS, CALCULATED DATA)
Unknown Variables
Calculated Data
Pressure Drop Across the Impactor During Test
(in. Hg)
Effective Pressure Drop (in. Hg)
Flowrate in Impactor During Sampling
(cc/sec)
Barometric Pressure (in. Hg)
Pressure at Inlet to Brink Impactor (manometer
reading)
Pressure at Inlet to Brink Impactor Corrected
to Absolute (in Hg)
Pressure at Inlet to Impactor Corrected to
Absolute (atm.)
Density of Gas at Inlet Sampling Conditions
(9/cc)
Average Temperature of Gas at Sampling
Conditions (OK)
Pressure (Ps) at Outlet of Each Stage (atm)
Density (Ps) of Gas Out of the Various
Stages (g/cc)
IM
IA
IA
Characteristic Diameter (Dg) (microns)
r5
Pi
P2
P3
P4
p5
Dl
26
Continued
-------
TABLE 3. BRINK DRY AEROSOL-SIZE DISTRIBUTION
(CALCULATION OF PARTICLE SIZE CUTOFFS,
CALCULATED DATA) (Continued)
Unknown Variable
Characteristic Diameter (D ) (microns)
(Continued) s
Cumulative Percentages
Test date/time
Sample Location
Calculated Data
D3 '
D4 '
D5 '
Z6 =
Z3 -
h -
h -
Run Number
27
-------
2) Using the average actual AP, maintained during the run, determine
a
effective pressure drop, AP_, using equation (1-1)
(29.92)
MW T
"W '
P
P
(1-3)
._ o.
where Tj, is the average impactor in/out gas temperature in K.
°
(°K) = 0.55 (Tj (°F) - 32) + 273"]
(1-4)
and PIM is converted to P,. (absolute). For a HgO vacuum gauge:
(1-5)
3) Convert PTfl (in. Hg) to PTfl (atm.):
IA
IA
PIA (in. Hg)
PIA (atm.) 29792
(1-6)
4) Determine pressure at outlet of each stage, P , atm. For stages
1, 2 and 3, PS = PIA (atm.)
For stage 4,
= P
(0.781)APr
- - -
IA 29.92
(1-7)
For stage 5,
AP
-
IA 29.92
(1-8)
28
-------
5) Calculate the density of gas at inlet sampling conditions (g/cc),
= (1.214 x 10"2)
(atm.)
in
(1-9)
Where TI- is the average inlet temperature to the impactor
6) Determine p , the d
stages using equation 1-10
6) Determine p , the density of the gas at the outlet of the various
P,. (atm.)
ps = PI PTfl (atm.)
7) Determine D for each stage from equation 1-11
= -15.3y +
2.05 x 10
+8
'c's's
IA
8) Determine the ratio:
D5X 10
(1-10)
(1-11)
for stage 5. L, the mean free path of gas molecules, may be determined by
the following equations:
I ^
(1-12)
v =
f8 g P,- x 1.013 x 10C
' C j
(1-13)
29
-------
If this ratio:
Dc x 10"
b
is greater than or equal to 2.7, equation 1-11 for DS> the characteristic
diameter, microns, as determined in step (5) is valid. The expression:
x 10
-4
must be greater than or equal to 2.7 for (1-11) to be valid. This ratio is
smallest for Stage 5 and increases for preceding stages. Thus, in step (8).
the ratio, if satisfactory for Stage 5, is also valid for Stages 1, 2, 3,
and 4. Therefore, if
D5xlO
-4
is less than 2.7 for Stage 5, the ratio must be evaluated for Stage 4,
then 3, etc., until the ratio is equal to or greater than 2.7. It should
be noted that L is not the same at each stage. For those stages where
Ds x 10
-4
<2.7,
D may be determined by equations 1-14 and 1-15.
C = 1 +
2L
DS x 10"
(1-14)
1.23 + 0.41e
"°'44
(1-15)
30
-------
Although these equations may be solved explicitly, it is simpler to use a
trial and error solution. To solve by trial and error, first calculate a
c using the D2 obtained from the equation 1-11. Then substitute this C
in equation 1-14 and calculate a new DS< Then calculate a new C using the
last calculated value for D , from this C calculate another D . Compare
the last two DS'S. If they are within 1% of each other, take the last
value as Dg. If they are not, continue the procedure of calculating a C
and then a D until 1% agreement is obtained.
o
9) Express the quantities collected in the cyclone, Stages 1-5, and
filter as percentages of the total amount recovered. Call these r., r2, etc.
10) Calculate the cumulative percentage, 2 , smaller than D for each
stage. These are:
Filter Z; = r?
Stage 5 Eg = Zy + rg
Stage 4 Zg = £g + rg
Stage 3 £4 = Eg + r4
Stage 2 E3 = I4 + r3
Stage 1 S2 = Z3 + r2
Brink Cyclone EI = £2 + ^
11) On log probability paper, plot the cumulative percentages deter-
mined in step (9) against D , D on the log scale (obtain the Dg for cyclones
manufacturer's literature).
31
-------
2. PROCEDURE FOR SAMPLING THE OUTLET OF A
FLUE GAS DESULFURIZATION (FGD) UNIT USING
A MRI IMPACTOR
This method is applicable for determining aerodynamic size distribu-
tion of dry solid particles emitted from the TVA Shawnee flue gas desk-
's
furization (FGD) processes at a mass loading of 0.07g/m (0.03 gr/cfm).
The Meteorology Research Inc. (MRI) Inertial Cascade Impactor is
designed to measure the aerodynamic size distribution between 0.3 and 30 ym
suspended in industrial gas streams at temperatures up to 200°C (392°F)
at a flow rate of 2.4 to 22.7 Lpm (0.1 to 0.8 cfm).
2.1 DOCUMENTS
2-1 Federal Register. 36(247): 24888-9.
2-2 Meteorology Research Inc. Instruction Manual for
Operation, Installation, and Maintenance for the
Inertial Cascade Impactor Model 1502.
2-3 Harris, D. B. Procedures for Cascade Impactor
Calibration and Operation in Process Streams.
EPA-600/2-77-004, January, 1977.
2.2 SAMPLING EQUIPMENT
The MRI Impactor (Model #1503) provides a total of seven cut-off
stages for particulate size determination. The impactor has a collection
disc located below each of the six stages as shown in Figures 10, 11, and
12. Each stage has a sequentially decreasing orifice size until the final,
seventh stage which consists of a filter. The stainless steel collection
plates are doughnut shaped and weigh ^700 mg. The impactor is 2-3/4 inches
in diameter by 11-1/2 inches long with a 1/2-inch NPT pipe fitting in the
outlet section. Both the housing and collection plates are constructed of
stainless steel with Teflon or Viton seals.
2.2.1 Impactors
MRI Model 1503 Cascade Impactor contains six impaction stages with a
seventh stage using a glass fiber filter for total particulate sampling
including:
Light-weight stainless steel collection discs (MRI, Altadena, CA)
Type A 47 mm glassfiber filters (Reeve-Angel 934-AH).
32
-------
Figure 10. Upper Stages
of MRI Impactor
Figure 11. Lower Stages
of MRI Impactor
-------
Nozzle
Jet Plate
Collection
Disc
1st Stage
"O" Ring
Filter
Figure 12. Assembly Drawing of
Model 1503 Inertial
Cascade Impactor
34
-------
a Heating Mantle for MRI Impactor (Special order from Glass-Col
Apparatus Co., Terre Haute, IN).
Out of stack 1/2" NPT connector (MRI, 464 W. Woodbury Road,
Altadena, CA).
2.2.2 Sampling Probe
Aerotherm 1/2-inch OD probe with a range of nozzle sizes.
2.2.3 Aerotherm Sampling Train
e Four impingers.
Ice bath container for impingers.
o Vacuum pump capable of pulling 4 CFM of free air and a vacuum of
12 inches of mercury or more.
@ Dry gas meter.
« Flow monitoring orifice.
2-2.4 Tools and Equipment
t Spanner wrenches supplied with impactor.
Pipe wrenches and/or clamping pliers.
t Gloves.
» Teflon pipe fitting tape.
Stopwatch.
Isokinetic Flowrate Calculator (Aerotherm Corp., Model #HVSS-901).
A source of 110V electrical power must be provided at the sampling
location.
o Suitable platforms must be provided at the sampling location for
Pacing Aerotherm alongside the sampling port and for working space for
the operator.
Five-place analytical balance.
2.2.5 Equipment and Materials for Coating Collection Discs
Grease, Apiezon H grease.
35
-------
Drierite, 5 Ib.
t Large desiccator, Kimble #21050, 250 mm min. diameter with porce-
lain plate.
Petri dishes (top and bottom), 60x15 mm pyrex.
Forceps.
o PVC gloves, U.S. Industrial Gloves, Compton, CA.
Kimwipes.
Acetone, reagent grade.
Dow 111 high vacuum grease.
2.2.6 Probe Connection
Two Swage!ok Quick Disconnect SS-QF8-B-810-VT.
t Two Swagelok Quick Disconnect SS-QF8-S-810.
2.3 REQUIREMENTS
2.3.1 System Design
The MRI system is operated out of stack using the 1/2" Aerotherm probe
to extract sample from the flue gas and the Aerotherm impinger, pump, and
the control unit to measure the gas flowrate. By placing the impactor out
of stack, a larger capacity heating mantle can be used. The higher temper-
atures attainable with this mantle will prevent premature collection of
HpSO. due to condensation on the first several stages.
2.3.2 Sampling Procedure
The flow rate through the impactor will determine the size cut-offs
that each stage will collect. As will be described in Section 2.4, an
average isokinetic sampling rate will be determined. Once the flow rates
are established, they must be maintained throughout the run regardless of
the individual velocities at each point.
2.3.3 Handling of Collection Discs
Care must be taken to limit contact with the discs. At no time should the
discs be touched with ungloved hands. All laboratory manipulations are to
be performed in a clean environment using tweezers to handle the discs.
36
-------
It is important to remember that several grains of dust could represent the
total weight captured on a disc. Contamination control is essential during
greasing, drying, and weighing.
2-3.4 Calibration and Maintenance
After each run, the probe nozzle, probe, connecting lines, S-pitot
tubes, impactor, and impinger system must be cleaned. The probe nozzle,
probe, and connecting lines can be cleaned with a long handle test tube
brush and backflushed with high pressure air. Should further cleaning be
required, deionized water followed by acetone (or isopropyl alcohol) can
be used. The S-pitot tube should be backflushed with a high pressure air
!ine. The impactor cleaning procedures are detailed in Chapter 4. The
impinger system is flushed out and the proper solvents replaced in the
impinger bottles prior to the next run.
In addition to these procedures, the C of the S-pitot and the AH@ of
the flowmeter orifice are determined every two months. If any evidence of
corrosion appears (pitting, scale build-up, etc.), the C and AH@ recali-
bration procedure should be repeated as needed. Tables 7 and 8 in Chapter
4 contain a summary of recommended maintenance and troubleshooting proce-
dures. .
2-3.5 Cleanliness
Gas carrying lines should be cleaned weekly. No particulate build-up
"in the pi tot tube can be tolerated. Impactors must be cleaned completely
after use. Chapter 4 describes the maintenance schedule for the sampling
and analysis equipment.
2-3.6 Safety
OSHA safety requirements with regard to working environment and oper-
ator safety will be met at all times. The reagents mentioned in the pro-
cedure are not extremely toxic but can be harmful if misused.
2-4 PROCEDURE
The MRI impactor system consists of a 1/2-inch Aerotherm probe con-
nected directly to the impactor (Figure 13). The MRI system is used as
Qn out-of-stack extractive sizing method. Using an Aerotherm probe, a
37
-------
O-rO
STACK
AEROTHERM
PROBE
SKIN TEMPERATURE
THERMOCOUPLE
THERMOCOUPLE .
TAP FOR IMPACT OR
GAS TEMPERATURE
\\x\\\\\\\\\\\\\\\\\
AEROTHERM
OVEN AND
CONTROL UNIT
MRI IN STACK
TRANSFORM
QUICK
DISCONNECT
FITTING
GLASS-COL
HEATING MANTLE
AND IMPACT OR
SUPPORT FOR
IMP ACTOR
TO
AEROTHERM
IMPINGER
SYSTEM
GAS FLOW
Figure 13. MRI Site Setup
38
-------
velocity profile for the duct is obtained. The average velocity is calcu-
lated and used to select a nozzle that will sample at the average isokinetic
velocity, but less than 22.6 1pm (0.8 acfm).
Temperature control of the impactor system is maintained by monitoring
the stack and outlet gas temperature from the impactor. The necessary heat
is supplied by a specially designed Glass-Col heating mantle. The gas
flow rate is monitored by measuring the AH across a calibrated orifice
with a magnehelic gauge.
The amount of material collected is determined by weighing the collec-
tion stages before and after the run. Particulate matter collected during
the probe and tubing rinses is added to the weight of the first stage. The
collection plate and filter are thermally conditioned and desiccated prior
to weighing. After sampling, the samples are desiccated to constant mois-
ture content prior to reweighing. Because of the potential for systematic
errors in weighing, blanks consisting of spare collection stages and fil-
ters are conditioned and weighed along with the samples to monitor weighing
errors. If weight changes greater than 0.1 mg occur in the blanks, the
Weight gain or loss is subtracted or added, respectively, to the weight of
the samples (see Section 4.1.8 for further details).
NOTE
The critical checklists (Table 5) for the MRI should
be available to the personnel performing the test run.
This checklist will provide an excellent guideline for
the sampling site and laboratory personnel.
2.4.1 Laboratory Preparation of MRI Cascade Impactor
a) Disassemble the MRI impactor by unscrewing each stage. Figure
12 shows an assembled unit. Inspect the jet plates prior to assembly to
ensure that clogging has not occurred. Clean each collection disc and jet
Plate by wiping the surface with a Kimwipe wetted with acetone. Inspect
the disc and plate after cleaning for particulate or finger marks on the
collection disc and jet plate. Inspect interior of the impactor housing
for particulate matter. Clean the interior of the impactor after removing
the Viton 0-rings with a squeeze bottle containing acetone and Kimwipes.
39
-------
For hard to reach areas, use a camel-hair brush to remove the participate.
The threads on the impactor should be lightly greased with Dow 111 high
vacuum grease.
b) With the tweezers, dip the lightweight SS collection disc into a
250 ml beaker containing 200 ml of reagent grade toluene to clean the sur-
face. Withdraw the disc and hold it in air until the toluene has dried.
Place the discs in separate labeled petri dishes.
c) Using a rubber policeman, spread a thin coating of Apiezon H
around the center of the doughnut-shaped collection disc. Should some of
the Apiezon H be spread over the edge of the disc, clean the disc with
toluene and repeat the procedure at 2.4.1.(b). Prepare eight discs, six
for use and two as spares.
NOTE
At all times these manipulations are to be performed
in a dust-free environment.
d) Place the covered petri dishes with the eight collection discs
into an oven for 4 hours at 175°C (347°F). The filters are heated for 4
hours at 287°C (550°F).
e) After 4 hours, remove the petri dishes with discs and filter, and
allow them to equilibrate in the desiccator for 2 hours.
f) Once the discs and filters have been dried and desiccated, they
are weighed on a balance capable of weighing to the nearest 0.01 mg.
Remove the petri dishes and filters from the desiccator just prior to
weighing (keep desiccator closed otherwise). Remove the discs from petri
dish with a forceps being careful not to touch the greased area and place
them on the balance. After weighing, record the weight and disc number on
the Laboratory Data Sheet (Figure 14). Place the filters and the coated
discs back in the petri dishes, cover them, and place the petri dishes
near the impactor.
g) Using forceps, place a preconditioned 47 mm diameter glass fiber
filter on top of the filter support housing. Then place the locking spacer
on top of the filter.
40
-------
MRI - DRY AEROSOL SIZE DISTRIBUTION
LABORATORY DATA SHEET
SAMPLE LOCATION
DATE/TIME
RUN NUMBER
DATE
STAGE
FILTER
TOTAL
DISC*
WEIGHT
FINAL
TARE
GAIN
%
X CUM
MICRONS
d50
- Weight gain on each stage divided by the total weight gain.
CUM% - Starting with the filter accumulate each stage to arrive at the
cumulative percent smaller than the previous d
* - Disc Code for labeling petri dishes should be the date of run,
stage no. and run letter series (example: 8/27/75, 1A;
8/27/75, 2A; etc.). The letters series represents the sequen-
cial number for each successive run that day. 8/27/75, 1A;
8/27/75, IB would be the next run.
Figure 14. MRI Laboratory Data Sheet
41
-------
h) Screw the body housing into position to receive the collection
disc, and replace the Viton 0-ring. Using forceps, insert collection disc
firmly into the housing groove. Place jet plate on top of disc into groove.
Continue until all stages have been connected and the impactor is completely
assembled as per assembly drawing in Figure 12.
i) After the MRI impactor is assembled, it should be leak checked in
the laboratory. Connect a vacuum gauge to the inlet of the impactor, and
attach the outlet of the impactor to the in-house vacuum line.
j) Leave the vacuum on until the gauge indicates 380 torr (15 in.
Hg).
k) Close the vacuum line and note any rise in pressure. The vacuum
should not vary over several minutes.
l) If a leak is noted by a decreasing vacuum reading, check the
impactor to verify that all connections are tight and the vacuum gauge is
working. Be sure that all the vacuum lines have tight seal as well. If
these measures do not locate the leak, take the impactor apart and replace
any suspicious 0-rings, then repeat the vacuum test.
m) Once the impactor is leak checked, both ends are sealed to pre-
vent dust from entering and the impactor is taken to the sampling site.
2.4.2 Measurements and Calculations for Isokinetic Sampling
a) The duct geometry must first be considered. For circular 40-inch
diameter ducts as encountered on the Shawnee limestone wet scrubber, refer
to Figure 6.
b) Sixteen sample points are selected, eight are along one axis
across the duct, and eight lie along another axis at 90° to the first.
The location of each sample point is obtained from Table 1.
42
-------
c) The test site should consist of a sampling port in the stack with
an opening to allow the easy insertion of the sampling probe. It should
be sealed to minimize the disturbance of the flow during sampling and pro-
tect personnel and equipment from hot exhaust gases. Also, the test site
must have a platform to provide for the safety of personnel and equipment.
The electrical power required to operate the equipment is approximately
35 amp/115V.
2-4.3 Isokinetic Sampling with the Aerotherm System. The Aerotherm
sampler is capable of isokinetic sampling if the nozzle inlet velocity
matches the exhaust stack velocity when a sample is taken. The control
unit contains a set of gauges to measure the pitot pressure (stack veloc-
ity) and the pressure difference across an orifice (sampling rate). By
adjusting the control valve on the pump, the flow rate can be varied there-
by changing the inlet velocity at the nozzle.
Prior to the initiation of the sizing program, the Aerotherm S-pitot
and flow orifice are to be recalibrated. The C and AH@ are measured and
entered into the MRI Field Data Sheet (FDS) (Figure 15).
a) Using the S-pitot attached to Aerotherm probe, perform a
sixteen point two-diameter velocity traverse across the duct.
Refer to Table 1 for the position of the sampling points.
At each point, record the AP, /AT and the stack temperature
(Ts) on the Field Data Sheet.
b) Determine the average stack temperature (Ts) and the average
/AT in inches of H20. Record these values on the AP and Ts
table on the Field Data Sheet.
c) Using the Isokinetic Flow Rate Calculation, complete Table 2
on the FDS.
NOTE
For detailed instructions on how to use the Isokinetic
Flowrate Calculator, see Appendix A.
The following are the condensed instructions found on the back
of the calculators:
43
-------
1.) Set C at AH@.
2) Using hairline, set % H20 at arrow.
3) Read index number at arrow.
4) Set T at index number.
5) Read second index number at T~.
6) Set Ps/Pm at second index number. If the Nozzle Size
(Dn) is known, proceed to Step 9. If not, proceed to
Step 7.
7) Set the average (v/AP) to Reference Arrow C on AH scale.
8) Read exact Nozzle Size at Reference Arrow B on Dn scale.
Select available nozzle that is near this diameter and
suitable for use.
9) Set Nozzle Size (Dn) under Reference Arrow B.
_ _, rt
10) Read AH setting opposite (s/AP) reading using Cursor as
needed.
11) Record the AH as needed on Column 2 in the work sheet.
12) Reset Hairline over Cp.
13) Set TV at Hairline and move Hairline over VI arrow.
14) Set V2 Arrow under Hairline and move Hairline over AP.
15) Turn over calculator without moving Cursor and set MS
under Hairline.
16) Read Stack Velocity (Vs) at Stack Pressure (P~).
17) Record all these data on Table 2 on the FDS.
-------
1. f. r AND TS TRAVERSE DATA
SAMPLE
PORT
TRAVERSE
POINT
ic.t
AVERAGE
2. DATA OF NOZZLE SIZE, £.», AND V$ CALCULATIONS
VARIABLE
tHfc
S
% HjO
Tm
TS
Pm
"s
D
n
(-/T?)2AVG.
AH
MS
vs
VALUE
28.6
Figure 15. MRI Field Data Sheet
45
-Continued-
-------
DATE/TIME _
RUN NUMBER
OPERATOR
FIELD DATA SHEET
AMBIENT TEMPERATURE
BAROMETRIC PRESSURE
STACK PRESSURE, (P$)
INLET GAS FLOW _
REHEATER AIR FLOW _
LEAK RATE
SCHEMATIC OF TRAVERSE POINT LAYOUT
READ AND RECORD ALL DATA EVERY MINUTES
TRAVERSE
POINT
NUMBER
INITIAL
CLOCK TIME
(24 HOUR CLOCK)
GAS METER READING
(CUBIC FEET)
AP
ORIFICE PRESSURE
DIFFERENTIAL
(AH, IN. H2O)
DESIRED
ACTUAL
STACK
TEMPERATURE
'°F
in.
OUTLET
«»^>'t
PUMP
VACUUM
IN. Hg
IMPACT OR
TEMPERATURE,
°F
SKIN
OUT
Figure 15. WRI Field Data Sheet
-------
d) Symbols
AH@ = Orifice calibration coefficient (in. HpO)
C = Pi tot tube coefficient (unitless)
Tm = Temperature of dry gas meter (Average of TJN and TQUT in °F)
Tg = Average stack temperature (°F)
AH = Orifice Pressure Drop (in. H20)
P<; = Stack pressure (in. H00)
o £
T- = Temperature of Stack at specific points (°F)
AP = Pressure drop across pi tot tube (in. H^O)
Vs = Stack Velocity (ft/sec)
P = Meter Outlet pressure (in. HgO)
Dn = Nozzle Diameter (inch)
o
(s/AP) = Average pressure drop across pitot tube (in. H«0)
2<4-4 Equipment Set-Up
a) . Set-Up and Operation of Shawnee Aerotherm Sampler. The probe is
Counted on the side of the sampling oven and is adjusted for height by
of a line cinch attached to the oven support bracket. Connectors
plugs are attached to the cabinet, consisting of:
Pitot lines color coded for proper connection
Power plug for probe heater. The receptacle is located on
the side of the sampling cabinet.
The impinger bottles, two of which are filled with sodium carbonate,
°ne left empty, and one filled with silica gel, are contained in a separate
Tee-cooled box. The cabinet controls consist of:
Magnehelic gauges indicating the pitot pressure corresponding
to stack velocity (V$) and the orifice pressure (AH) drop.
t The multi-point temperature indicator measuring the stack
temperature, (Tc), gas entering and leaving the dry gas meter,
'V-
Gas meter for the total volume of gas sampled.
47
-------
The MRI impactor consists of six impaction stages and a back-up filter.
Using the appropriate connecting tubing, the MRI impactor will be connected
to the probe and mounted in the Aerotherm oven (Figure 13).
n _
1) Using (v/AP) and T~ calculate the correct nozzle to meet the
average isokinetic conditions.
2) Attach this nozzle to the Aerotherm probe.
3) Brush and rinse inside the Aerotherm probe to verify no
particulate remains.
4) Connect the impactor to the probe via a Swagelok 1/2" quick
disconnect with a 1/2" male NPT fitting screwed into the
MRI 1/2" female inlet connector.
5) The first two impingers are filled with saturated
The third is left empty and the fourth impinger is filled
with 250 g of silica gel .
6) Once the vacuum lines are attached, the Aerotherm nozzle is
plugged using a rubber stopper.
7) Open the vacuum pump valve until the gauge indicates 380 torr
(15" Hg vacuum). The flow through the dry gas meter
should be less than 0.02 cubic feet per minute before the
sample can be taken.
8) If the leak rate is less than 0.02 cubic feet. per minute,
close the vacuum valve slowly to prevent a pressure surge
and remove the stopper. If leak rate is greater than 0.02
cfm, tighten all fittings and examine the pump for wear.
Replace worn parts and repeat leak tests. Once acceptable
leak rates are met, the unit is ready for sampling.
2.4.5 Operation of Aerotherm and MRI Impactor During Sampling
Once the AH is determined, the Aerotherm and MRI units are ready to
run. Although a sampling traverse will be performed, the units will be run
at the same AH.
NOTE
Because the volumetric flow determines the particle
size cut-offs for each impactor stage, the flow rate in
the MRI impactor must remain constant throughout the
run regardless of the AP reading at the traverse point.
48
-------
a) Set the probe temperature controller and the MRI gas out temper-
ature at 12°C (25°F) above the highest temperature in the stack, but not
higher than 175°C (347°F).
b) Turn the pump on and immediately insert Aerotherm probe into the
stack until it is at the first sampling point.
c) Adjust AH on the magnehelic gauge for the average setting previ-
ously determined in traverse. Constantly check and record on a field data
sheet the pitot manometer (AP) reading, stack temperature (TS), dry gas
meter temperature (T ), vacuum pump pressure, and the skin and gas out
impactor temperature. Be sure to complete all other Information on the
data-sheet.
d) Sample for 7.5 minutes at each point for a total of two hours
Sampling times should be adjusted to collect a maximum of 10 mg/stage.
NOTE
The precalculated AH must be maintained at each point.
e) At the end of the run, the probe is removed from the stack and
the Aerotherm unit is shut down. Slowly close the flow control valve to
dissipate any back pressure in the system and thus prevent water from
surging forward in the impinger system. Stop the stopwatch and record
final gas meter reading on field data sheet.
f) Using gloves, remove the MRI impactor with heating mantle attached
fl"om the oven and return it immediately to the laboratory. Cover the top
of the impactor to prevent particles from entering.
NOTE
During this period, avoid jarring the impactor. Extreme
care should be taken to avoid the addition or loss of
collected particulate. Carry the impactor upright and
do not expose it to dust.
g) Carefully rinse the probe with reagent grade acetone collecting
rinse until a clean stream of liquid issues from the probe.
49
-------
NOTE
Take extreme care in performing this task as the small
amounts of particulate matter recovered represent a
large portion of the total participate aerosol collected.
Both contamination and loss of sample must be avoided.
Any accidents which occur must be recorded on the field
data sheet.
h) Once the impactor is transferred to laboratory, clean the outside
of the impactor of any dust. This should be done in the prep room prior
to entering the clean room.
i) Be sure that the correctly labeled petri dishes are nearby, ready
to accept the collection plates as they are removed.
j) With the impactor in an upright position, begin to remove the
housing starting at the top. Inspect the inlet nozzle for any sign, no
matter how little, of particles collected on the inlet walls. Note the
presence of the particulate matter oh the back of the laboratory data
sheet.
k) If any particles are found, they should be carefully brushed onto
the collection plate below their collection point.
1) Inspect all the jet nozzle plates for any sign of pitting or cor-
rosion. Especially inspect the underneath of the jet nozzle for particles
that might have collected there. If any particles are found, note this
fact along with a description of any patterns formed, color, or quantity
obtained on the back of the laboratory data sheet.
m) Carefully brush these particles onto the collection plate below
the jet nozzle plate.
n) Repeat these activities for all the stages and the filter. In
the filter's case be sure that all fragments of the filter are removed
from the filter support; even the loss of the smallest fragment can affect
the weight of the filter.
o) Desiccate the collection plates and filter for 2 hours.
p) Evaporate enough of the acetone from the probe rinse so that all
the particulate and the remaining acetone can be quantitatively transferred
to a tared 30 ml beaker. At all times, handle the 30 ml beaker with gloved
hands.
50
-------
q) Evaporate the bulk of the acetone from the 30 ml beaker on
a hot plate allowing the rest of the acetone to air dry in a clean dust-
free area. Dry the particulate at 110°C for two hours and desicate with
the rest of samples for approximately two hours.
r) Weigh the collection plates and filter to the nearest 0.01 mg.
At the same time weigh the balance and sample blanks {see section 4. 1.8. (a)
specific procedures to correct for any weighing errors). Record the
on the laboratory data sheet.
2-5 DATA REDUCTION
The MRI unit is designed to provide a distinct particle size cut-off
*t each stage. Using Figure 16 and the volume flow corrected to standard
impactor conditions, the dgo (cutoffs) for each stage can be determined.
1) Calculate the change in weight for each collection disc, and filter.
2) Add up the differences to get the total particulate weight collected
°n discs and filter.
3) Divide the amount collected on each plate by the total amount col-
Tected to find what percent of the total is impacted on each plate.
4) From the field test log, determine the total volumetric gas flow
in ft3/min. Correct this flow (Qm) to impactor conditions (Q$):
o
Qm = Sampling rate at dry test meter (ft /min.)
o
Qs = Flow rate at impactor conditions (ft /min.)
B = Volume fraction of moisture in gas stream
w
T = Average impactor gas out temperature
(2-1 )
where f. and Tm are in °F.
5) Using Figure 16 with the gas flow rate at Impactor conditions (Q$)
the impactor temperature (T^), determine the d5Q for each stage.
51
-------
100
500° F
300
100
500° F
300
100
STAGE
500° F
300
500° F
300
100
500°F
300
100
0.1 0.2 0.4 0,6 0.81.0
FLOW RATE AT STACK CONDITIONS (0^, FT /MINI)
Figure 16. MRI Impactpr Stage Cut-off
Diameter
52
6
/
-------
6) The results can be plotted on log probability paper with the par-
ti cul ate diameter (d5Q) as the ordinate and cumulative percent by weight
as the abscissa.
7) The cumulative weight percent for Stage 1 is determined by sub-
tracting the weight percentage for stage 1 from 100. For stage 2, the
cumulative weight percentage is found by subtracting the weight percen-
tages from stages 1 and 2 from 100. This process is repeated for all the
stages.
P 8) An alternate approach is to normalize the data. This approach
[(dm/d(log D50)l is not discussed here, but is detailed in Reference 2.3
and is the recommended approach if the Brink and the MRI data are to be compared,
9) The mass loading is found by correcting the volume of air passed
through the dry gas meter to STP (ASTD):
ASTD ' (QJ (120 min.) (i^46o) (202) <2'2>
= /Total weiqht found on stages and filter) (2.
\ STD /
Loading (mg/dcfm) = oa weq oun on sages an er (2.3)
Where Tm is in °F and PS is in inches of
53
-------
3. DETERMINATION OF H2$04 VAPOR USING A CONTROLLED
CONDENSATION COIL
This method was designed to rceasure the vapor phase concentration of
S03 as H2S04 entering the flue gas desulfurization until(FGD) and exiting
from the reheater at the TVA Shawnee Power Plant in Paducah, Kentucky.
This method is specifically designed to operate at temperatures up to
250°C (500°F), 3000 ppm S02 and 8-16% H20. By using a modified Graham
condenser, the gas is cooled to the acid dew point at which the SO, (HJiO^
vapor) condenses. The temperature of the gas is kept above the water dew
point to prevent an interference from S02 while a heated quartz filter
system removes particulate matter. The condensed acid is then titrated
with 0.02 N NaOH using bromophenol blue as the indicator.
3.1 DOCUMENTS
3-1 Federal Register. 36(247): 2488R-9.
3-2 Goksoyr, H. and K. Ross, J. Inst. Fuels, 35, 177
(1962; ~
3-3 Lisle, F.S. and J.D. Sensenbaugh, Combustion, 1,
12 (1965).
3-4 Nacovsky, W., Combustion, 1, 35 (1967).
3-5 Standard Methods for the Examination of Vfater and
Wastewater. 13 Edition, pages 52-56 (1971).
3-6 Maddalone R., C. Zee, and A. Grant, "Procedure
for Titrimetric Determination of Sulfate Using
Sulfonazo III Indicator," TRW Systems, EPA
Contract No. 68-02-1412, Task 6, Feb. 14, 1975.
3.2 EQUIPMENT AND MATERIALS
3.2.1 Sampling Materials
Probe construction materials (including materials for two
3-foot probes and spare).
a) Three Vycor tubes 0.5-inch OD x 36-inch with a 18/9 female
ball-and-socket joint placed on one end (special order
A. H. Thomas or Ace Glass, see Figure 17).
b) Three glass insulated heating tapes - 1/2-inch x 72-1nch;
288 watts (Fisher Sci. Co. #ll-463-50C or equivalent).
54
-------
0.5"
ui
en
T
THERMOCOUPLE WELL
18'
36"
18/9
THERMO-
COUPLE
WELL
Figure 17. Vycor Sampling Liner
-------
c) Three 33-inch x 1-inch x 0.065 inch wall 304 SS tubes used
as probe sheaths.
d) One dozen silicone rubber No. 6 stoppers (A.H. Thomas
#8747-E65).
e) Glass tape (Scotch glass-fiber electrical tape).
f) Four Omega (Stanford, Conn.) shielded thermocouples (I/C),
(TH36-ICSS-18G-12) with 8-foot lead.
g) Four Omega (Stanford, Conn.) unshielded thermocouples (I/C),
(IRCO-032 with 8-foot lead).
h) Six Omega male connectors (ST-IRCO-M).
i) Two six-foot heavy duty (^20A) electrical cords.
j) Two 1-1/2 inch hose clamps.
k) Two square yards of asbestos cloth (VWR, Atlanta, Georgia,
#10930-009).
1) Three adaptors for connecting hoses (Ace Glass, #5216-23).
m) One Teflon Swagelok Union (T-810-6).
Two pumps capable of pulling 1 cfm of free air (Brink impactor
pump may be used).
0 Bath controller-circulator (A.H. Thomas #9840-615 or equivalent).
0 Fifty feet of 1/2-inch x 1/4-inch rubber tubing (A.H. Thomas,
#9544-R57).
Three Graham condensors (controlled condensation coils - CCC)
modified to hold an enclosed 60 mm medium frit (special order from Ace
Glass, Louisville, Ky.; see Figure 18).
Two styrofoam chests capable of holding a 2-gal. bucket.
Three glass insulated heating tapes, 3/8-inch x 24-inch, 96 watts
(A.H. Thomas, #5954-H22 or equivalent).
Four autotransformers, variable, 10 amp. (A.H. Thomas #9461-010
or equivalent).
One hundred Tissuequartz filters, 37 mm diameter (Pall flex Corp-
Kennedy Drive, Putnam, Conn. 06260).
56
-------
THERMOCOUPLE
WELL
18/9
3CM-
60 MM MEDIUM
FRIT
4.0 CM-
23.8 CM
18/9
THERMOCOUPLE
WELL
-4CM-
GAS
FLOW
Figure 18. Controlled Condensation Coil
-------
Eight pinch clamps (A.H. Thomas 2841-21 or equivalent).
Three Greenburg -Smith type impingers or equivalent.
Sodium carbonate, technical grade.
Indicating silica gel.
Stopcock grease (Ace Glass Co., #8229-10).
Three-inch bushing with a 1-1/8 inch hole drilled in the center.
Two RdF digital temperature indicators-series-2000 with iron/con-
stantan sensors.
One vacuum gauge (A.H. Thomas #5654-810).
Two soap bubble flowmeters (Applied Science Laboratory, P.O. Box
440, State College, Penn. 16801, (814)-238-2406.
Glass-Col heating mantle for filter system (Glass-Col, 711 Hulman
St. , Terre Haute, Ind.).
3.2.2 Reagents and Apparatus for H2S04 Titration
Carbon dioxide-free distilled water - Prepare all stock and
solutions, and dilution water for standardization procedure, using distill^
water which has a pH of not less than 6.0. If the water has a lower pH, it
should be freshly boiled for 15 minutes and cooled to room temperature.
NOTE
Deionized water may be substituted for distilled water
provided that it has a conductance of less than 2
microohms/cm and a pH greater than 6.0.
» NaOH pellets - reagent grade.
Stock 1.0 N NaOH - Dissolve 40 g of reagent grade NaOH in 1 liter
of COp free distilled water. Store in a pyrex glass container with a tight
fitting rubber stopper.
0.0200 N NaOH - Dilute 20 mL of 1 N NaOH with C02 free distilled
water to 1 liter. Store in a tightly rubber stoppered pyrex glass bottle
protected from atmospheric C02 by a soda lime tube. For best results,
prepare daily. This solution will be standardized against potassium
biphthalate (see Section 3.4. 3. (b))
58
-------
Potassium biphthalate (KHCgH404)-Anhydrous, reagent grade.
o 0.0200 N potassium biphthalate (KHP) solution - Dissolve 4.085 g
of dry (110°C for 1 hour) KHP into 1 liter of C02 free distilled water.
NOTE
The normality of the KHP solution equals (wt. KHP)/204.2.
t Anhydrous ethyl alcohol - U.S.P. or equivalent.
. Phenolphthalein indicator solution - Dissolve 0.05 g of reagent
grade phenolphthalein in 50 ml ethyl alcohol and dilute to 100 ml with C0?
free water.
Bromophenol blue indicator solution - Dissolve 0.1 g in 7.5 ml of
0.02 N NaOH. Dilute to 250 ml with C02 free distilled water.
, Ten millilHer micro-buret, Kimble 17132F (A.H. Thomas #1993-M-30
°r equivalent).
Desiccator (A.H. Thomas #3751-HI0 with cover and plate to fit).
§ Drierite desiccant - 5 Ib. Dierite (A.H. Thomas #C288-T49).
Four Erlenmeyer flasks with 28/15 ball and socket joint, 125 ml
(Ace Glass Co., Louisville, Ky., #6975 or equivalent).
Four stoppers for 28/15 ball and socket joint (Ace Glass Co.,
#R263-08 or equivalent).
Four 50 mL volumetric flasks.
Dowex 50W-X8 cation exchange resin 20 to 50 mesh.
» Barium perchlorate trihydrate, reagent grade.
0,01 M Ba(C104)2-3 H20 - Transfer approximately 3.9 of reagent
9rade barium perchlorate trihydrate into a one liter reagent bottle. Add
enough D.I. H20 to dissolve the sale and then dilute to the mark.
Sulfonazo III Solution, 0.1% W/V - Transfer 0.025 g of sulfonazo
into a 25 mL bottle, add water to dissolve the indicator and fill to
mark.
59
-------
3.3 REQUIREMENTS
3.3.1 System Design
The S03 (H2S04 vapor) Controlled Condensation System (CCS) consists of
a heated Vycor probe, a modified Graham condenser (condensation coil), a
critical orifice, impingers, and pump (see Figure 19).
3.3.2 Sampling
Since a gas, S03 (HgSO^ vapor), is being sampled, no traverse will be
performed in the stack. The sample probe will be positioned at a point
representative of the stack flow.
Flow control in the CCS is maintained by monitoring the dry test
meter with a stopwatch.
3.3.3 Handling of Glassware
Because of the corrosive nature of S03 (H^SO, vapor), only Vycor and
Pyrex glassware is used. Severe mechanical shocks are to be avoided,
especially when the probe is heated to 250°C (500°F). Never place any
strain on glass ball joints and clean the ball joints of grease and dirt
after each run.
3.3.4 Calibration and Maintenance
After each run the probe, connecting lines, controlled condensation
coil, filter holder, and impinger system must be cleaned. The probe and
connecting lines can be cleaned with a long handle test tube brush and
backflushed with high pressure air. If particulate matter adheres to the
inside of the probe, rinse with deionized water followed by acetone (or
isopropyl alcohol). The impinger system is flushed out and the proper
solvents are then replaced in the impinger bottles prior to the next run.
The filter holder is inspected and cleaned before the next run and the
filter pad is replaced. Table 8 in Chapter 4 details the recommended
maintenance for the CCS by component.
3.3.5 Cleanliness
Contamination of the condensation coil rinse solutions must be avoided
to prevent neutralization of the HgSCh. Keep the rinse solutions in a
covered flask.
60
-------
ADAPTER FOR CONNECTING HOSE
TC WELL
STACK
RECIRCULATOR
THERMOMETER
STYROFOAM
ICE CHEST
Figure 19. Controlled Condensation System Setup
RUBBER VACUUM
HOSE
TEST
METER
ASBESTOS CLOTH
INSULATION
GLASS-COL
HEATING
MANTLE
THREE WAY
VALVE
SILICA GEL
C0
-------
3.3.6 Safety
OSHA safety requirements with regard to working environment and operator
safety must be met at all times. The reagents mentioned in the procedure
are not extremely toxic, but misuse of any chemicals can be harmful.
3.4 PROCEDURE
3.4.1 Probe Manufacture (Refer to Figure 20)
The necessary equipment is listed in Section 3.2.1.(a-m). Follow correc
electrical safety procedures at all times. Be sure that no sharp pieces of
metal abraid any of the electrical wires.
a) Cut the 304 SS one-inch tubing into 32-inch lengths.
b) Align the shielded thermocouple (TC) as shown in Figure 20. Using
the glass tape, secure the shielded thermocouple to the Vycor probe. Place
the unshielded thermocouple in the thermocouple well and secure with the
glass tape. .Continue down the probe, securing both thermocouple leads
simultaneously against the tube.
NOTE
Be careful never to kink the thermocouple or thermocouple
leads.
c) Take the 72 inch glass heating tape and fold it in half.
d) Beginning 5 inches from the probe tip, wrap the probe with the
glass heating tape. Make sure the heating tape is snugly up to the probe
and secured every 6 inches with a wrapping of glass tape. Wrap the coils
close enough so that the heating wire is completely used up 2 inches from
the ball joint. Secure the end of the heating tape with a final wrap.
e) Bore a 9/16-inch hole into two No. 6 silicone rubber stoppers,
then cut a slit vertically down one side of the stopper into the 9/16 inch
hole. The slit will allow easy assembly.
f) Cut a piece of asbestos cloth approximately 30 inches long and wi^e
enough to wrap the probe and heating tape with a 1/2 turn overlap. Tightly
wrap the probe and secure the asbestos cloth with glass tape.
62
-------
PROBE T.C.
18/9
SILICONE
STOPPER
ASBESTOS
CLOTH WRAP
01
CO
SILICONE
STOPPER
GLASS HEATING
TAPE LEAD
STACK
T.C.
VYCOR TUBE
TEFLON UNION
6 MM
SHIELDED
T.C.
l) STOPPERS SHOULD BE AWAY FROM HEATING TAPE
?) ASBESTOS COVER SHOWS SLIGHT OVERLAP
Figure 20. Controlled Condensation System Probe Desi
gn
-------
g) Slide the 304 SS sheath over the Vycor probe. Avoid scratching
the insulation on the electrical leads. Position the sheath so that the
end near the tip extends one inch past the start of the heating tape.
h) Spread the stopper open, slip it over the tip of the probe, and
slide it into the 304 SS sheath. The stopper is then wired to help hold
it in place. Repeat this procedure for the other end, except use a hose
clamp to hold the back stopper in place.
i) Place the male quick connects on the end of the TC leads. The
red TC lead goes to the negative terminal.
j) The probe should be tested in the laboratory to ensure that all
parts are in order. Simply connect the heating wire to the Variac and allow
the probe to heat up. Monitor the temperature to verify the TCs are
functioning.
NOTE
Whenever heating up the probe, start off with very low
power inputs (~5%) until heating starts.
k) The 0.25 inch nozzle and Teflon union (Figure 20) are attached
prior to the test run. The nozzle consists of a 0.5 inch diameter quartz
tube tapered to 0.25 inch at one end and a 90 bend placed in the center
of its 2.5 inch length.
3.4.2 Filter Holder Fabrication
Figure 21 details the recommended design for the quartz filter holder.
This filter holder consists of a modified 40/50 standard taper quartz joint-
The modifications included adding a coarse quartz frit and an extension tub6
to the male joint to act as a pressure seal when the Tissue quartz filter
pad is in place. Ball and socket (18/9) joints are used to connect the
filter holder to the probe and controlled condensation coil.
3.4.3 Site Equipment Setup and Operation
a) In the 3-inch port, insert a 3 inch plug with 1 inch hole.
b) Use a table or another suitable device to support the CCS (see
Figure 19).
64
-------
SPRING
ATTACHMENT
HOOKS
TISSUE QUARTZ
FILTER
THERMOCOUPLE
WELL
18/9
SOCKET
en
tn
STANDARD
TAPER QUARTZ
40/50
SEAL EXTENSION
TO STANDARD
TAPER JOINT
EXJRA CQARSE
QUARTZ FRIT
Figure 21. Quartz Filter Holder
-------
c) Prior to use, be sure the controlled condensation coil (CCC) is
clean and dry. Carry the CCC to the site with each end stoppered. If any
condensation appears because of temperature changes, connect the CCC to the
water bath and start the circulation of the 60°C (140°F) water. This should
evaporate any premature condensate.
d) With the probe still out of the stack, assemble the train as shown
in Figure 19. Be sure that each ball joint is completely clean and free of
dust. Because of,the possibility that the greases will freeze at the temper-
atures employed, it is not recommended that any grease be used. Proper care
of the ground glass fittings will ensure that vacuum seals are maintained,
Should any ground glass fitting not seal vacuum tight, a small amount of
Apiezon H grease may be used for emergency repair. As soon as it is possibl6'
the joint in question should be returned to the glass shop for regrinding
(see Tables 7 and 8 for further suggestions).
e) Connect the soap bubble flowmeter to the vacuum pump exit. Be sure
that the bubble flowmeter is vertical, Close off the end of probe with a
stopper and turn on the vacuum pump and adjust the vacuum to read 380 torr
(15 in. Hg).
f) Begin measuring the flowrate with the bubble gauge, If the leak
rate is less than 85 mL/min (0,003 cfm), then the system is ready for use,
If a leak rate greater than 85 mL/min is found, the system should be checked
for loose joints and connections. The pump should also be checked and any
worn parts replaced. Tables 7 and 8 for further information.
g) Once the vacuum test is completed, slowly turn the three-way valve
to the vent position and allow the air to bleed into the system. This must
be done carefully to prevent a pressure surge from backing up the impingers-
Remove the bubble flowmeters from the system and unstopper the probe,
h) Begin heating the probe and the filter holder to 316°C (600°F) and
288°C (550°F) respectively. The heating bath should already be at 60°C
(140°F). Once the skin temperatures reach these values, the run can commeflc
NOTE
During the course of the run, the filter temperature will
be controlled by the gas out temperature which should be
288°C (550°F).
66
-------
i) After leak testing, the pump is again turned on and the flow
adjusted to 10 Lpm (0.35 cfm). The pump is turned off without read-
Ousting the valve settings.
j) Pinch the hose at the end of the controlled condensation coil and
'insert the heated probe into the duct with the nozzle pointed upstream.
k) Turn on the pump, release the pinched hose, and obtain an initial
dry gas meter reading. Throughout the run, collect the data required (see
Figure 22).
1) Sample for one hour or_ until 1/2 to 2/3 of the length of the coils
are frosted with H2S04>
NOTE
If the coil is operating properly the H2S04 will cover
the inside of the coils as a thin gray-white film. If
large drops of a clear liquid form and begin to block the
coil, then moisture is being condensed. Either the per-
centage moisture has exceeded 16% or the temperature of
the water bath has dropped below 60°C. Abort the run and
check the water bath temperature with a Hg thermometer
and confirm the percentage moisture in the gas stream.
If the water bath is below 6QOC, recalibrate the temper-
ature bath control. For every percent above 16% HgO,
adjust the CCC temperature 2°C upward. Clean and dry
the CCC, and replace the reagents in the impingers prior
to restarting the run.
ro) At the end of the sampling period remove the probe from the duct
ar|cl slowly shut off the pump. After the pressure drops, remove the CCC from
the system without removing the water bath hoses. Carefully connect
(Figure 23) the G/R coil to the Erlenmeyer flask without spilling any
condensate in the tube. In 10 ml increments (up to 30 ml), use
^ionized water to rinse out the CCC. Be careful to avoid introducing
dust or grease into the rinse solution. Take the rinse solution
n the stoppered Erlenmeyer to the laboratory for analysis.
NOTE
Multiple rinses are recommended to ensure a
quantitative rinse of the coil.
67
-------
Sample Locat1on_
Run *
Run Date/Time..
Operator
Flowrate (cfm)
Ambient Pressure (P)
AEROSOL
(CONTROLLED COHDEHSATI05I)
FIELD DATA SHEET
Reheater A1r Flow Rate, acfm_
Inlet Gas Rate, acfm
Sample Location S02 (ppm]_
Boiler Load (mw)
Leak Rate
Time (Min)
AVG.
Temperature (°F)
Stack
Probe
Filter
Skin
Out
Reel re.
Water
Exit
Coil
Dry Gas
Meter
In
Out
Gas Meter
Reading,
01. ft.
Figure 22. Controlled Condensation Field Data Sheet
68
-------
WATER BATH
HOSE
PIPET BULB
ADAPTER
SOLUTION
POSITIONING
DRAIN
STOPCOCK VALVE
125 ML ERLENMEYER FLASK
D.I. H2O
FROM COIL
Figure 23. Controlled Condensation Coil Rinsing Apparatus
69
-------
n) Rinse the probe with 30-40 ml of deionized H20 after it has
cooled. Take this solution back to the laboratory, and filter through a
Whatman number 1 filter dilute to 50 ml.
°) Remove the filter from the filter holder (CAUTION: Wait until the
filter has cooled), and place it into a beaker. Add 30 ml of deionized
H20 and swirl the beaker. Filter the solution through a Whatman number
1 filter into a 50 ml volumetric. Repeat with 10 ml portions of deionized
H^O until the volumetric is filled to the mark.
3.4.4 Analysis Procedures
Two procedures can be used to determine the amount of HpSO^ collected:
1) An acid/base titration using Bromophenol blue indicator or
2) A sulfate titration using Sulfonazo III as the indicator.
Because of the 'simplicity and sensitivity of the acid/base titration, it
is the recommended procedure. The sulfate procedure is included in this
section to act as a backup or total sulfate method if the need arises,
In either case all the titrations should be done in triplicate and the
results recorded on the laboratory data sheet (Figure 24).
a) Sulfate Titration Using Sulfonazo III. This procedure is similar
to the sulfate procedure developed for scrubber liquors (Ref. 3.6). This
procedure may also be used to analyze the water rinse from the filter for
water soluble sulfate.
1) Wash the Dowex 50 W-X8 cation exchange resin with IQ% V/V HC1
Fill a 1/2-inch I,D. ion exchange column to a 3-inch bed depth*
and place glass wool pads at the bottom and top of the bed.
Rinse the column with deionized water until the eluant tests
neutral with pH paper.
2) Transfer 0.025 g of chemically pure Sulfonazo III indicator
[(NaS02)2 CIQ H2(OH)2](N:NC6 H4S03H)2 to a 25 ml bottle, add
water to dissolve the indicator, and fill to the mark.
3) Transfer approximately 3.9 g of reagent grade barium
trihydrate [83(0*04)2 '3^0] into a one liter reagent bottle,
add a small amount of distilled water to dissolve the salts,
and then fill to the mark. Mix the contents of the bottle.
4) Standardize the reagent by titrating sodium sulfate. Dry
the Na2$04 in an oven for two hours at 125°C and allow to cool
to room temperature in a desiccator. Weigh out accurately in
70
-------
AEROSOL S03
(CONTROLLED CONDENSATION)
LABORATORY DATA SHEET
Run
Sample Location.
Run Date/Time.
Analyst
Date Lab Analysis Completed.
Variable
Value
Aliquot Size (A)
Normality of titrant (N)
""I of titrant used to titrate G/R coil
rinses (v)
Blank (equivalent NaOH)
Ngt titration volume (V)
Absolute dry gas meter temperature (Tm)
Volume of gas sampled (V )
Meter Pressure (P )
PPm H2S04 (vol/vol)
-*-
(ml)
(eq/n)
(ml)
(mL)
(mi)
Avq. (ml )
(mi)
(ml)
r°Rl
(ft3)
(in. Ha)
Normality of acid used to titrate blank
Hf used)
PPmH2SO, = 1202.52 X
*
NVT
Figure 24. Laboratory Data Sheet for Acid/Base Titration
71
-------
triplicate 12 to 16 mg of the NagSCty from a weighing bottle into
125 ml Erlenmeyer flasks, dissolve with 10 ml deionized water,
add 10 ml acetone and three drops Sulfonazo III indicator
solution, and titrate with the barium perchlorate.
5) Repeat this procedure in triplicate for the sample and blank D.I.,
(deionized H20):
-
H
(142)va)
Where: M = Molarity of the barium perchlorate solution,
moles/liter
W = Weight of sodium sulfate titrated, mg
V = Average volume of barium perchlorate solution
required for titration of sodium sulfate, ml
v = Average volume of deionized water blank titration
a
6) Take a 10 ml aliquot of the rinse solution and pass it through
the ion exchange column at 3 mL/min. Rinse the column with
30 ml deionized water and collect the eluant and rinsings in
a 50 ml volumetric flask and dilute to the mark with deionized
water.
7) After every tenth use of the column, regenerate it with 100 ml
of 10% W/V HC1 at 3 mL/min flow rate and rinse until the eluant
tests neutral to pH paper. Rinse the column with 50 ml of
deionized water.
8) Add 10 ml acetone and three drops of the Sulfonazo III indicator
to a 10 ml aliquot of the ion exchange eluant.
9) Titrate with 0.01 M Ba(C104)2 using a magnetic stirrer and back
lighting. The color will change from purple to blue. The end
point is the point at which an additional drop of titrant does
not change the color of the solution. The end point should not
fade unless left standing for more than 5-10 minutes. Record
the volume of 0.01 M 83(0104)2 used to reach the end point and
calculate the average titration volume. Titrate a 10 ml aliquot
of deionized water in the same fashion to obtain the titration blank.
b) Acid/Base Titration. The preferred method of analysis is the
acid/base titration using Bromophenol blue indicator. Carefully handle and
store the samples in clean glassware and analyze them as soon as possible.
Record all results on the laboratory data sheet.
72
-------
1) Pipet 10 mL of the 0.0200 N KHP solution into a 125 mL wide-
mouth Erlenmeyer Flask.
2) Add 3 drops of the phenolphthalein indicator. With a wirling
action of the flask, titrate with 0.02 N NaOH solution until
the first pink color stays. Record the volume and repeat from
(1) in triplicate. Repeat this procedure using deioni.::ed H90
(blank). i
3) Average the colume used to titrate the KHP solution. 'he true
normality of the NaOH solution equals:
1.200) (10 mL)
titrant-mL blank) v°"^;
4) To titrate the H2$04 in the condensation coil, probe aod filter
rinses, pi pet 10 mL of one of these solutions into a 1 !5 mL
Erlenmeyer flask. Larger aliquots can be used if the H2S04
is quite low. As a rule of thumb, the aliquot size should be
adjusted to require a minimum of 1 mL of titrant.
NOTE
Be sure to use the same size aliquot
for the blank titration.
c) Calculation of the ppm (v/v) concentration of H2S04 in the Gas
streams. Using either the sulfate or acid/base titration, to analyze
the CCC rinse, the concentration of H2S04 can be calculated.
1) From the Field Data Sheet (Figure 22) obtain the averane dry
test meter temperature, volume of gas sampled and atmo-pheric
pressure. Record these values on the Laboratory Data Sheet.
2) Using the Laboratory Data Sheet, insert the correct nunbers
into the appropriate formula (see Appendix B for the derivation)
For the Add/Base Titration: /NVT \
ppm H2S04 = 1,201.91^ I (3.3)
y m m/
For the Sulfate Titration: /MVT
ppm H2S04 = 12,019 ^-DJ- I (3.4)
m m
The result 1s ppm (v/v) H?S04 at 20°C (68°F) and 1 a tin (29.92
In Hg). £ *
3) Plot this data on the CCS control chart (see Section 4.2.3-b)
and record the results on the Laboratory Data Sheet.
73
-------
4. QUALITY ASSURANCE METHODOLOGY
The intensive test program being carried out at the Shawnee Test
Facility is designed to determine whether scrubber conditions affect the
quantity and quality (as measured by a particle size) of emitted particu-
lates. In addition, a concurrent program of monitoring HpSO. has been
initiated.
To accomplish these tasks, a series of procedures for particle sizing
and HUSO, measurement were written. This quality control document is writ-
ten as a supplement to those procedures to provide guidance to on-site per-
sonnel in controlling the quality of their work.
This document is not designed to be a procedure manual and consequently
does not contain detailed information on the procedure. What is provided
is information on:
Equipment care and usage
Guidelines for laboratory techniques
Specific critical area checklists for each procedure
t Data interpretation aids to monitor the quality of the
results achieved
Specific maintenance and calibration schedules
Troubleshooting and repair schemes.
No matter how good a document is, the degree of implementation will
establish its usefulness. If the spirit of the document is violated, then
the quality of the results will not be improved. Committing a mistake
during the operation of the test equipment is regrettable, but not docu-
menting the problem is unforgivable. The purpose of this document is not
to assign blame, but to provide a basis for understanding and interpreting
program results. Always remember that quality control ultimately rests
with the honesty of the operating personnel, and no piece of paper can
replace a dedicated professional.
74
-------
4.1 LABORATORY EQUIPMENT CARE AND TECHNIQUE
To assure the quality and reliability of all data generated ir. this
Program, it is of utmost importance that all equipment is in proper work-
ing order. This encompasses not only routine periodic maintenance but also
the day-to-day handling and general usage of this equipment in a sefe and
secure manner. The methodologies described in each of the followirg cate-
gories are aimed to help the operator maintain his equipment in an adequate
manner so that it is always ready to operate with a certain measure of
reliability.
The laboratory techniques to be used will have a direct relationship
to the type of end-data which are obtained. To generate better, mere reli-
able data, certain techniques should be used throughout this program. The
specific techniques are tabulated as they relate to the particular appa-
ratus or glassware employed in the performance of that test.
Cleanliness is one of the major factors affecting the quality and
accuracy of data. It is of utmost importance since cross-contaminction
will be minimized just by having glassware and equipment available in a
clean, non-contaminated condition. At the minimum:
An area should be wiped down prior to use of that area
A Whatman No. 1 paper sheet (46 x 47 cm) shall be placed on
the bench top prior to working in that area.
4>1-1 Analytical Balance
a) Equipment Care
Analytical balances are a relatively fragile type of instrument, and
are subject to shock, temperature and humidity changes, general mishandling
and various other potentially injurious occurrences. Some of the ;>recau-
tions to be observed in maintaining and prolonging the dependable -ife of
a balance are as follows:
Analytical balances should be mounted on a heavy shock proof
table, preferably one with adequate working surface and a
suitable drawer for storage of balance accessories.
Balance level should be checked frequently and adjuster] when
necessary.
75
-------
Balances should be located away from laboratory traffic,
protected from sudden drafts and humidity changes.
Balance temperatures should be equilibrated with room tem-
perature; this is especially important if building heat is
shut off or reduced during non-working hours.
t When not in use, the beam should be raised from the knife
edges, all dials set to zero, objects such as weighing
dish removed from the pan, and the sliding door closed.
Never add any weights or samples to the pan unless the
beam is raised from the knife edges (half release position)
Place a petri dish with desiccant in the balance pan area.
Special precautions should be taken to avoid spillage of
corrosive chemicals on the pan or inside the balance case.
The interior of balance housing should be kept scrupulously
clean; a camel hair brush should be used to clean the
balance pan.
Balances should be checked and adjusted periodically by a
company service man or balance consultant. If service is
not available locally, follow the manufacturer's instruc-
tions as closely as possible.
The balance should be operated at all times according to
the manufacturer's instructions, (which are to be posted
on or near the balance).
b) Laboratory Technique
Since the analytical balance is a very sensitive instrument, special
precautions or specialized techniques must be followed:
Only weigh samples which are at room temperature.
0 Never touch weights, pan, samples, etc., with your hands as
they would deposit a thin layer of oil and cause resultant
errors.
Verify the balance is clean and no contaminants are on the
pan. In the event some contaminant, dust, dirt, sample,
etc., is on the pan, remove that contaminant by brushing
it off. If oil or water is on the pan, wipe it off using
an acetone soaked tissue.
Level and zero the analytical balance prior to weighing.
Never put chemicals directly on balance pan.
76
-------
The same person should pre-weigh, manipulate and reweig-i a
given set of filters and liners from an impactor run.
Weigh each sample as quickly as possible (within the con-
straint of good technique) to avoid weight changes due -:o
moisture pickup.
Always keep the balance doors closed while weighing.
0 Always add or subtract weights while the beam is raised
from the knife edges.
Record all weights immediately.
4-1.2 pH Meters
a) Equipment Care
A basic pH meter consists of a voltage source, amplifier, and readout
device, either scale or digital. Certain additional refinements produce
Drying performance characteristics between models. Some models inoorpor-
ate expanded scales for increased readability and solid state circuitry
for operating stability and extreme accuracy. All instruments of recent
design also include temperature adjustment and slope adjustment to correct
asymmetric potential of glass electrodes. Other features are s:ales
facilitate use of selective ion electrodes, recorder output, aid
interfacing with complex data handling systems.
e The pH meter should be kept on top of a counter in a work
area.
Glass electrodes should not be allowed to become dry during
periods of inactivity. When not in use they should be
immersed in distilled water, with the water level being
checked frequently.
The proper KC1 level in the calomel electrode shall be
maintained.
0 Prevent contact of the electrodes with oily substances or
other type of materials which could adhere to the electrode
surfaces. In event contact is made, clean the electrocia
with acetone followed by deionized water. Allow the elec-
trode to equilibrate in a pH 4 buffer until it is stable.
77
-------
b) Laboratory Technique
pH meters are highly dependent on calibration and general use tech-
niques. Essentially, the major area of concern is the proper use of the
electrodes and reference pH solutions. Following are the various tech-
niques which should be adhered to in order to assure good standardization
and pH readings.
0 The first step in standardization of the instrument is done
by immersing the glass and calomel electrodes into a buffer
of known pH, setting the meter scale or needle to the pH of
the buffer and adjusting the proper controls to bring the
circuit into balance. The temperature compensating dial
should be set at the standard solution temperature. The pH
of the standard buffer should be within about two pH units
of the sample.
The instrument should be calibrated against two buffers that
bracket the pH of the samples. If the two standards do not
read accurate values, a troubleshooting mode of operation
should take place to ascertain why not.
A new slope setting must be made whenever electrodes are
either changed, subjected to vigorous cleaning, or refilled
with fresh electrolyte.
t Glass electrodes have a very fast response time in highly
buffered solutions. However, accurate readings are obtained
slowly in poorly buffered samples, and particularly when
changing from buffered to unbuffered samples, as after stan-
dardization. Electrodes, both glass and calomel, should be
well rinsed with distilled water after each reading, and
should be rinsed or dipped several times into the next test
sample before the final reading is taken. Weakly buffered
samples should be stirred during measurement.
4.1.3 Laboratory Analytical Glassware
a) Equipment Care
All glassware is somewhat fragile and requires good handling, storage
and inspection criteria.
Try to avoid bumping glassware or otherwise causing it to
become stressed, cracked, or broken.
t Keep all glassware on shelves with special lips or in drawers;
this will prevent glassware from rolling off the counter.
78
-------
Any cracked or broken glassware should be discarded and
replaced at the direction of the supervisor.
Never place dirty glassware near the cleaned glassware.
Never store dirty glassware. Rinse all glassware with
acetone and water, cleaning thoroughly the next day.
Volumetric flasks should not be dried in an oven, but
rather final rinsed in deionized water and stored with
some deionized water in it.
Grease any vacuum joints using a minimum amount of grease.
b) Laboratory Technique
Always rinse glassware with acetone to remove any organic
substances.
Rinse with tap water.
Clean all glassware with either Alconox or Calgon in water.
Use a brush to scrub all surfaces.
-After either the soap wash or chromic acid treatment,
throughly rinse the glassware with abundant amounts of tap
water.
0 Several separate rinses will be required to remove all soap
and traces of the chromic acid.
Rinse the glassware with deionized water a minimum of 1hree
times. Water should sheet off. Otherwise the glassware is
not clean and the entire process must be repeated.
t Air dry the glassware. CAUTION - do not blow dry the (lass-
ware with lab air as that air will put an oily film balk
onto the glassware.
c) Glassware Usage
Use of glassware essentially revolves around volumetric glassware or
any glassware involved in critical measurements. Guidelines follow which
will ensure better precision and accuracy.
The volumetric apparatus must be read correctly. The bottom
of the meniscus should be tangent to the calibration m;;rk.
t To deliver (TD) volumetric pipets are calibrated to de'.iver
a fixed volume. In emptying volumetric pipets, they should
be held in a vertical position and the outflow should be
79
-------
unrestricted. The tip of the pi pet is kept in contact with
the wall of the receiving vessel for a second or two after
the free flow has stopped. This will remove any hanging
drops. However, do not blow out any remaining solution
from the tip of the pi pet. Do not attempt to dry a pipet
which has been used, simply rinse the pipet with the new
solution to be used several times; increase the number of
rinses if going from a concentrated to a dilute solution.
Burets are used to deliver definite volumes. General rules
in regard to the manipulation of a buret are as follows:
a) Do not attempt to dry a buret which has been cleaned for
use, but rinse it two or three times with a small volume of
the solution with which it is to be filled, b) Do not allow
alkaline solutions to stand in a buret, because the glass
will be attacked, and the stopcock, unless made of Teflon,
will tend to freeze, c) Burets should not be emptied rapidly.
Otherwise too much liquid will adhere to the walls and as
the solution drains down, the meniscus will gradually rise,
giving a high false reading, d) It should be emphasized
that improper use of and/or reading of burets can result in
serious calculation errors.
In the case of all apparatus for delivering liquids, the
glass must be absolutely clean so that the liquid film never
breaks at any point. Careful attention must be paid to this
fact or the required amount of solution will not be delivered.
4.1.4 Desiccators
a) Equipment Care
Desiccators are glassware with a specific use. They should be treated
as glass apparatus with the following added requirements:
Maintain the desiccator in a clean condition. Periodic
cleaning will necessitate removing Drierite, thoroughly
cleaning the unit, then drying it in an oven at 110°C for
a minimum of two hours and putting in fresh desiccant.
Use Drierite or silica gel, and it should be placed at the
bottom of the desiccator under the porcelain plate.
Never allow any desiccant to be placed on top of the por-
celain plate.
Add an indicating Drierite or silica gel to monitor the
effectiveness of the desiccant. If the indicator is blue,
usage may continue; if indicator is pink, the desiccant
must be replaced or regenerated.
80
-------
Always keep a lid on top of desiccator.
Maintain seal between lid and base of desiccator by placing
a small amount of Dow 111 High Vacuum silicone grease on the
mating surface and rotating the lid on the base 360°.
b) Laboratory Technique
Desiccators are used to store samples or glassware whenever moisture
must be eliminated.
Adhere to precautions of desiccator in "equipment care"
section.
Do not allow samples to touch the grease on the mating surface
of the lid and base of desiccator.
Open lid of desiccator by sliding lid sideways until it is
off. Never try to pry open or lift straight up.
0 Keep lid on desiccator except for minimal time needed to
transfer samples in and out of desiccator.
Do not put solvents into dessicator.
Always let samples cool to room temperature inside the
dessicator before removing them for weighing or performing
other temperature and moisture sensitive tests.
4-1.5 Dry Test Meters
a) Equipment Care
Drv test meters are relatively insensitive to handling; however, sev-
eral precautions should be noted and adhered to.
a The meter should not be dropped or handled roughly.
Always maintain the meter in a horizontal position while
flow measurements are made.
The unit must be periodically calibrated for proper gas flow
readings (see maintenance schedule for recommended schedule).
The meter should always be placed downstream of the impingers
and drying train to prevent acid gases and moisture from
reaching the dry test meter.
81
-------
4.1.6 Ovens
a) Equipment Care
Ovens should always be kept clean. Any spills or other
contamination shall result in a thorough cleaning with all
items being removed from the oven.
High outgassing products or solvents should not be placed
in the oven.
Dirty glassware or hardware shall never be placed inside an
oven.
0 Never place tubing or other plastics inside an oven.
Periodic checks on the oven will be required to verify its
temperature readout control is in calibration and that a
uniform temperature exists inside the oven. This task should
be done monthly.
4.1.7 Reagent Storage
It is very important that all reagents, solvents and standard solu-
tions be stored in an appropriate manner to prevent contamination and/or
deterioration of that material prior to their use. All reagents should be
clearly labeled as to the material and concentration as well as the date
standardized and the performing technician.
0 Borosilicate glass bottles with ground glass stoppers are
recommended for most standard solutions and solvents.
0 Plastic containers, e.g., polyethylene, are recommended for
alkaline solutions. Plastic containers must not be used
for reagents or solvents intended for organic analyses.
0 It is important that all containers be properly cleaned and
stored prior to use. (See 4.1.1)
0 Standard reagents, solvents, and other chemicals must always
be stored according to the manufacturer's directions. Rea-
gents or solvents that are sensitive to the light should be
stored in dark bottles and/or stored in a cool, dark place.
0 The analyst should pay particular attention to the stability
of the standard reagents. Reagents should not be kept longer
than recommended by the manufacturer or as normally used in
the method selected.
82
-------
The concentration of the standards will change as a result
of evaporation of solvent. This is especially true of stan-
dards prepared in volatile organic solvents. Therefore, the
reagent bottles should be kept stoppered, except when actu-
ally in use.
Storage should conform to OSHA safety practices.
4.1.8 Blanks
Two types of blanks exist:
Individual blanks - the blank of each chemical such as ace-
tone used to flush the samples from the impactors, the
deionized water and any and all other solvents used.
Method blank - the method blank is determined by following
the procedure step-by-step, including all of the reagents
and solvents, in the quantity required by the method.
a) Laboratory Technique
In general these guidelines should be followed to monitor blanks:
Blanks should be run on each different individual type of
sample and on each batch of samples.
t The conditions for determining the blank must be identical
to those used throughout the analysis, including the detec-
tion system.
i If any individual blanks are found to interfere with the
analysis, the cause of interference will have to either be
determined or a correction factor applied (if it is found
that a bias results).
If the cumulative blank interferes with the determination,
steps must be taken to eliminate or reduce the interference
to a level that will permit this combination of solvents
and reagents to be used. If the blank cannot be eliminated,
the magnitude of the interference must be considered when
calculating the concentration of specific constituents in
the samples being analyzed. Within the program at Shawnee
there are two examples of blanks.
1) Ueighing blanks (see below)
2) Indicator blanks (see Section 4.1.9-b)
Proper implementation of blanks in these two areas will ensure a
successful analysis.
83
-------
b) Weighing Blanks
t Always weigh a fresh set of blank impactor plates and filters
which have been greased and otherwise subjected to the same
conditions as field samples, except that they are not used
to collect samples.
Obtain a known standard weight (about 1 g) and weigh that
standard at least once a day while weighing actual specimens.
Maintain this sample in the desiccator between weighings.
Recprd this weight and maintain it in a separate log book.
This weight should be plotted to ensure that only random
changes are occurring. If positive or negative trends occur,
review the weighing procedures and/or call in the balance
service man for a calibration check.
t If a sample blank varies by more than 0.0001 g, then a cor-
rection must be made on the sample weights. Weight gains
should be subtracted from all the samples in the blank's
group while any weight loss in the blank should be added to
the actual sample.
NOTE
Always inspect the weighing blanks for any obvious sign
of contamination, such as dirt particles or the loss
either of grease or glass fibers due to handling. If
an obvious contamination is noted, do not correct the
samples, but note the reason for the change in weight
of the blank.
4.1.9 Titrations
a) Laboratory Technique
End points for titrations are very color dependent, i.e., the end
point will probably vary slightly for each operator's sense of color. To
obtain the most accurate data, the following techniques should be employed
in all titrations:
Always add the same number of drops of indicator.
t Have the same operator do blank and sample.
Always titrate to the same color intensity.
Avoid parallax errors - keep eyes at the same level as the
liquid meniscus and hold a white piece of paper behind it
with a dark line horizontal to the table top.
84
-------
Remove air bubbles from buret tip prior to use.
Never store reagent in buret. Always rinse out buret with a
slight amount of titrant.
Always record titrant type and volume used.
b) Indicator Blanks
Each indicator will change color over a different pH range. For
example:
Indicator pH range Color change
Bromophenol Blue 3.0-4.6 yellow to blue
Phenolphthalein 8.2-10.0 colorless to pink
The point measured by the indicator is simply the point at which the
color change occurs. The actual end point where exactly the right amount
°f acid and base have reacted (equivalence point) can be close to or far
away from the indicator end point. Thus Bromophenol blue is chosen for the
NaOH + H2SO, titration, since the equivalence point occurs at about pH 3.
phenolphthalien is used for the potassium hydrogen phthalate + NaOH stan-
dardization titration because the equivalence point is near pH 7.
Even though the indicators have been selected to be as close as pos-
sible to the actual end point, a small difference still exists and is
called the indicator blank. The indicator blank for phenolphtalein is the
amount of NaOH required to change a specific amount of water containing a
known number of drops of phenolphtalien pink. This value is subtracted
from the milliliters used to titrate the sample.
The indicator blank for Bromophenol blue is determined in the same
way (known volume and number of drops) except that a standard acid (HpSO.)
"Is used to backtitrate the indicator in distilled water to a yellow color.
The number of mm 1 equivalents used is added to the amount found titrating
the sample.
NOTE
Blanks can vary with sample size and number of drops
of indicator, therefore determine the indicator blank
under the same conditions in which the sample is
titrated.
85
-------
4.1.10 Handling
a) Laboratory Technique
Handling techniques are very critical to the ultimate success of the
program, mostly in terms of obtaining more reliable data, but sometimes
even in terms of getting data.
Care must be taken to limit contact with the impactor discs.
At no time should the discs be touched with ungloved hands.
All of the laboratory manipulations are to be done in a clean
environment using tweezers to handle discs. Remember, sev-
eral grains of dust could represent the total weight cap-
tured on a disc. Contamination control is essential during
greasing, drying and weighing,
0 Handling of glassware is very sensitive and care should be
taken to avoid any shock, bumping or strain of the glassware.
i Do .not touch grease or components with grease on them to
other hardware.
Always use gloves, but be careful that organic solvents do
not come in contact with the gloves, otherwise the gloves
might discolor. Do not use gloves that are powdered by the
manufacturer.
Mechanical shock to hardware should be avoided. This is
especially important in the 6/R system where high temper-
atures reduce the resistance to mechanical shock.
4.2 SAMPLING QUALITY CONTROL
An impactor operates under the principle that if a stream of particle-
laden air is directed at a surface, particles of sufficient inertia will
impact upon the surface while smaller particles will follow the air stream
lines and not be collected. Thus an impactor consists simply of a nozzle
and an impact!on plate. Each stage of an impactor positions the nozzle a
precise distance above the impactor plate. Each successive jet is smaller
in diameter so that the gas velocity increases and smaller particles are
collected. To minimize particles bouncing off of the surface of the col-
lection plate, the impactor surface is coated with a sticky material. The
presence of the sticky material also minimizes re-entrainment of collected
particles by the scouring action of the gas stream. The best approach to
reducing re-entrainment is not to overload the stages with collected
material.
86
-------
4.2.1 Brink Methodology
The impactor selected to measure the particle size distribution enter-
ing the wet scrubber system is the Brink impactor. The Brink impactor
system consists of a 1/4-inch ID, 5-foot probe connected directly to the
impactor (Figure 1). The impactor has been modified by the addition of
a cyclone placed before the first stage. This cyclone is in addition to
the five stages already present in the impactor. A final stage consisting
of a 47 mm filter is attached to the exit of the last impaction stage.
This Brink system will provide aerodynamic size information for particles
from 0.3y to lOy. With Teflon washers and the Apiezon H greased stages,
the maximum operation temperature is 200°C (392°F) at a maximum flow rate
of 0.08 acfm.
The Brink system is an out-of-stack extractive sizing method. Using
an Aerotherm probe, a velocity profile for the duct is obtained. The aver-
age velocity is calculated and used to select a nozzle that will sample at
the average velocity isokinetically at a flow rate of «O.OB acfm.
Temperature control of the impactor and filter system is maintained
°y monitoring the inlet and outlet gas temperature from the impactor. The
necessary heat is supplied by a specially designed Glass-Col heating mantle.
The gas flow rate is monitored by measuring the AP across the impactor
with a magnehelic gauge. The impactor acts as a calibrated orifice and
thus the AP can be related to flow rates by referring to calibration charts
Provided by the manufacturer.
The amount of material collected is determined by weighing the collec-
tion stages before and after the run. Probe and tubing rinses are added
to the cyclone catch. The collection plates and filter are thermally con-
ditioned and desiccated prior to weighing. After sampling, the samples are
desiccated to constant moisture content prior to re-weighing. Because of
the potential for random or systematic errors in weighing, blanks consist-
ing of spare collection stages and filters are conditioned and weighed
along with the samples to monitor weighing errors. Refer to Section 4.1.8
for the proper use of blanks.
87
-------
a) Critical Checkpoints
Table 4 is a checklist of critical items that should be followed
during the test run. These critical items consist of:
Recommended flow rates, temperatures and sampling times
0 Reminders on laboratory and sampling technique
Specific equipment checks.
While this list is provided for review prior to the sampling run, its
best use is an on-site checklist for the supervisor and quality assurance
personnel during the run. During a test audit the supervisor or QA repre-
sentative should initial each item successfully completed. The entire list
should be included with the documentation of that test run. The operating
personnel might also like to have copies of the checklist for reference
during the execution of the test run. Copies can be posted in the labora-
tory and sampling site for this purpose.
b) Data Monitoring Procedures
Dry aerosol sampling procedures cannot be tested in the classical
fashion, i.e., with spiked (standard addition) samples to determine their
reliability. There are simple monitoring activities that can be carried
out on a daily basis. These activities include:
Calculation of percent isokinetic sampling - provides infor-
mation on the quality of sampling,
Comparison of Aerotherm and Brink grain loadings - determines
the efficiency of particle recovery from system.
0 Comparison of Aerotherm and Brink fine particle grain
loadings - indicates the efficiency of sampling by the
impactor.
These procedures are discussed in detail in the following paragraphs.
1. Isokinetic Sampling Tests
Isokinetic sampling is sampling at a rate (measured at the probe
nozzle) equal to the velocity of the gas flowing by the probe. Unless gas
flows are sampled isokinetically, larger or samller particles can be pref*
erentially collected depending on whether the sample rate is less than or
greater than the stream flow rate. The closer to isokinetic conditions the
more representative the gas particle sample will be.
-------
TABLE 4. QRITICAL CHECKPOINTS FOR BRINK DRY AEROSOL SYSTEM
Checkpoint
Initials
Supervisor
QA Inspector
Remarks
I. Conditioning and Preweighing Procedures:
A. Conditioning
- Plates cleaned with toluene prior
to greasing.
- Hands gloved.
- Thin film of grease applied to
plate.
- Clean filter handled with tweezers.
- Clean petri dishes and watch plates
prior to storing impactor plates,
cyclone or filter.
- Plates, cyclone and filter placed
in clean, labeled petri dish and
covered.
- Plates and cyclone conditioned at
175°C for 4 hours.
- Filter conditioned at 290°C
for 4 hours.
- Desiccant fresh (color B/P?)
- Plates, cyclone and filter
desiccated for 2 hours.
B. Weighing
- Balance area clean.
- Balance pan clean.
- Balance leveled.
- Balance zeroed.
- How long were petri dishes and
impactor stages left open to lab
air? ( min?).
,- Weigh impactor plates and filter
blanks for each set of sample
iirtpactor plates and filter.
- Data recorded on correct data
sheets.
- Note condition of plates (color
of grease, thickness of coating,
etc.) on data sheet,
H. Laboratory Impactor Preparation
- Plate, cyclone and filter Identi-
fications and weights recorded.
- Impactor Inspected for wear.
- Impactor cleaned.
- Work area cleaned and bench
covered with Whatman paper.
- Hands gloved.
- Impactor stages loaded, starting
from last stage.
- Cyclone cup in place.
- Backing, filter, and seal washer
placed 1n the filter holder.
- Inlet tubing sealed off.
89
-Continued-
-------
TABLE 4. CRITICAL CHECKPOINTS FOR BRINK DRY AEROSOL SYSTEM (CONTINUED)
Checkpoint
Initials
Supervisor
QA Inspector
Remarks
III. Site Set-Up
Impactor maintained 1n vertical
position during transfer to site.
Brush inside probe prior to run.
Rinse probe with acetone until
rinse solution is clear.
Fresh solutions placed in impingers.
Fresh absorbant replaced in final
impinger.
Leak rate must be less than 0.0008
cfn (0.02 Lpn).
Leak test performed.
Magnehelic gauges zeroed.
Thermocouple leads attached to
impactor.
Skin temperature controlled to
< 375°F (< 191°C).
IV. Sampling Run
Brink gas out temperature main-
tained at highest stack reading
+50° F.
Brink gas out temperature must
never exceed 347°F (175°C).
Check seal between probe and
rubber stopper to prevent any
outside air entering the stack.
Select sampling rates below 0.08
cfm (2 Lpn).
Select sampling time to collect no
more than 10 mg of material on any
stage except the cyclone.
After probe is disconnected, plug
the ends to prevent particle loss
during transfer to lab.
Impactor carried in an upright
position to laboratory.
Support equipment cleaned prior
to next run.
Report any experimental problems
or unusual occurrences on data
sheet.
V. Sample Recovery
- Probe and impactor cooled to
handling temperature.
- Use gloves during removal
procedure .
- Probe and tubing connections Inlet
to the cyclone are brushed and
rinsed with acetone until rinse
stream 1s clean. Rinsings collected
in Erlenmeyer flask and saved for
weighing in tared 50 mL Erlenmeyer
flask.
90
-Continued'
-------
TABLE 4. CRITICAL CHECKPOINTS FOR BRINK DRY AEROSOL SYSTEM (CONTINUED)
Checkpoint
Initials
Supervisor
QA Inspector
Remarks
- Use tweezers to remove impactor
plates.
- Inspect Impactor walls and jet
nozzles for participate matter.
Brush any partlculate matter on
the walls of jet nozzles onto the
next stage.
- Inspect filter holder for shreds
of filter material.
- Collect all pieces of filter
material from filter holder and
place with the intact filter for
weighing.
- Record all data on laboratory
weighing sheet.
- Correct sample weights for any
change in the blank's weight.
- Note any unusual operations.
VI. Data Verification
Plot the daily percentage
isokinetlc for Brink runs (Y-ax1s
for % isokinetlc, X-axis day).
Plot Aerotherm and Brink grain
loading values on a dally basis
{Y-axis for grain loading, X-ax1s
day).
Plot Aerotherm and Brink fine
particle grain loading on a dally
basis (Y-axis fine grain loading,
X-axis day).
Plot sample blank weight change
daily (Y-axis wt., X-axis day).
91
-------
Thus the degree of isokinetic sampling is an expression of the quality
of the sampling run. The normal criteria of acceptability is ±10 percent
of the current isokinetic flowrate.
Because the internal gas velocity determines the size cutoffs for the
collection stages, the impactor must be run at one flow rate is only one
size range of particles is to be deposited on a given stage. Consequently,
the impactor is run at the average isokinetic flow rate.
The procedure for the isokinetic check follows:
a) From the original velocity profile (or actual velocity measured
during the run), compute the average \/AP.
b) Calculate the average stack velocity:
1/2
Vc = (85.48) C
p
T.+460
(4-1)
where
\L = Average stack velocity (ft/sec)
C = Pitot tube coefficient
P
/ l /?
N/AP = Average square root of the velocity head (in. HJ)) '
TS = Average stack temperature (°F)
PS = Absolute stack gas pressure in. Hg)
MS = Stack gas molecular wt (g/m) (29.5 inlet to wet scrubber)
c) Convert APc to AP£
where
APE - Brink pressure drop corrected for impactor conditions (in. Hg)
APQ - Average Brink pressure read during test run (in. Hg)
Tj = Average of the impactor gas in and out temperatures (°F)
92
-------
PIA = Average pressure inlet to Brink impactor (in. Hg absolute)
M = Atmospheric gas molecular wt (g/m) (29.0 at STP)
a
d) Calculate Brink flow rate:
QB = 0.0519 (APE)0'44 (4-3)
where
QD = Brink flowrate (cfm)
b
e) Calculate stack velocity based on Brink flow rate:
bb (5.07 x 10~H)
where
VCD = Brink stack velocity (ft/sec)
OD
DD = Brink nozzle diameter (mm)
D
f) Determine percent isokinetic (I):
« *"*r> »/*
(4-5)
g) Make a continuous plot of I on x-y graph with I on the y-axis
and the day on the x-axis. This daily plot should be kept
as a permanent record of the Brink runs. As the data begins
to accumulate trends will be established. Normally I should
vary randomly between 90 and 110%. Consistently high or low
I values indicate systematic errors in sampling and call for
a review of procedures and equipment.
Errors can be due to:
Equipment - Inaccurate or malfunctioning magnehelic
gauges, dry test meter or thermocouples. Refer to
Tables 7 and 8 for troubleshooting and calibration
procedures.
t Data recording - Wrong numbers taken or misplaced on
the Field Data Sheet. Double check all entries.
93
-------
Calculations - Either the wrong data were input
input or a mathematical error was made. Have
different individuals do the calculation.
2. Stack Mass Loading Evaluation
Since an Aerotherm (Method 5) mass reading (C.) is performed prior to
the Brink run, it is possible to obtain an approximate comparison of Brink
mass loading values (C.J with those obtained from the Aerotherm.
a) Correct the Aerotherm dry meter volume (V ) to standard
conditions: m
where
T = Average meter temperature ( F)
P~ = Average absolute meter pressure (in. Hg)
P _- = Standard pressure (29.92 in. Hg)
b) Determine Aerotherm mass loading (C.) in grains/scf:
WTA (15.43)
CA= VSTD>
where
WT. = Total particle weight recovered from Aerotherm train (grams)
c) Using Brink particle size computer program, calculate the
actual volume of air that passed through the impactor.
d) Correct this value to standard conditions:
VB(STD) ' W) <> - V w P <4-8)
94
-------
where
BW = Volume fraction of O in gas sample (obtained from mass
loading run)
Vn(ACT) = Volume sampled by Brink (acf)
VR(STD) = Volume sampled by Brink (dscf)
e) Calculate mass loading for Brink (CR) in grains/scf:
WTR (15.43)
CR = -^ (4-9)
B VBC(STD)
where
WTR = Weight collected from Brink train (probe, cyclone, stages,
-and filter in grams).
f) Ratio calculated mass loadings
G = CB/CA (4-10]
where
G = correlation variable for Brink and Aerotherm system
g) Plot the daily values and observe the trend. The expected
range for G should be 0.7 to 1.3. Consistently high G values
may indicate contamination or incorrect flow rate calculations.
Verify that the correct values are used in the Brink program.
Also check the calibration of the magnehelic gauges. Con-
sistently low G values are more likely to be found. The most
probable cause for this result in poor overall particulate
matter recovery, but especially poor recovery from the probe.
Review cleaning procedure and make extra effort to clean the
probe and connecting tubing properly.
3. Fine Particle Mass Loading Evaluation
The previous two sections have described evaluation methods that mea-
sure the quality of sampling and the overall sample recovery of the test
crew. The efficiency of the impactor can be monitored by determining the
fine particle mass loading. The fine particle mass loading for the
Aerotherm system is defined as the weight of the particulate matter found
95
-------
on the filter divided by VM/STDX. For the Brink system the fine particle
grain loading is the weight of particulate matter found after the first
stage divided by the VB/cTn\- Comparison of these values will provide an
indication of the operation of the impactor, since all material after the
first stage should be localized on the stages and not require rinsing oper-
ations. The evaluation procedure follows:
a) Determine the Aerotherm fine particle mass loading (CFA)
in grains/scf:
Wc. 15.43
CFA = -p (4-lD
hA VM(STD)
where
WFA = Particulate matter weight on Aerotherm filter (g)
b) Determine Brink fine particle mass loading (CFB) in
grains/scf:
WCD 15.43
C = FB (4-12
UCR \l *
FB VB(STD)
where
W.-r, = Particle weight found in Brink system after the first
FB stage (g)
c) Determine fine particle ratio (GF):
Gp - a (4-13)
h LFA
d) Plog Gp daily and observe any trends. The expected range
for G will be from 0.7 to 1.3. High results (G>1) can be
from :
Low flow rate values for the Brink systems - check
calculations and calibration of magnehelic gauges.
96
-------
Contamination - Review laboratory procedures (Section 4.1)
Low results on (G<1) are the more probable and can be due
to:
Loss of material on transfer - review handling procedures.
High flowrate values - check calculations and magnehelic
gauge calibration.
Grease weight loss - verify impactor temperature was less
than 347°F. Note any amber discolorate on filter signi-
fying grease flow through. Correct these problems by
controlling temperature to <347°F, placing a thin film
of grease on the plates and maintain gas flow to <0.08
acfm.
4.2.2 MRI Methodology
The impactor selected to measure the particle size distribution exit-
ing the wet scrubber system is the MRI impactor. The MRI impactor system
consists of a 1/2-inch Aerotherm probe connected directly to the impactor
(Figure 13). The MRI impactor is designed to measure the aerodynamic size
distribution between 30 and 0.3 microns distributed over six stages and a
final filter. With Viton 0-rings and Apiezon H greased stages, the maxi-
mum operation temperature is 175°C (347°F) at a maximum flowrate of 0.8 acfm.
In this application the MRI system is used as an out-of-stack extrac-
tive sizing method. Using an Aerotherm probe, a velocity profile for the
duct is obtained. The average velocity is calculated and used to select a
nozzle that will sample at the average isokinetic velocity but at or below
a flow rate of 0.8 acfm.
Temperature control of the impactor system is maintained by monitoring
the stack and outlet gas temperature from the impactor. The necessary heat
is supplied by a specially designed Glass-Col heating mantle. The gas
flow rate is monitored by measuring the AH across calibrated orifice with a
magnehelic gauge.
The amount of material collected is determined by weighing the collec-
tion stages before and after the run. Probe rinses are added to the first
stage. The collection plate and filter are thermally conditioned and
desiccated prior to weighing. After sampling, the samples are desiccated
97
-------
to constant moisture content prior to re-weighing. Because of the poten-
tial for systematic errors in weighing, blanks consisting of spare collec-
tion stages and filters are conditioned and weighed along with the samples
to monitor weighing errors (see Section 4.1.8).
a) Critical Checkpoints
Table 5 is a checklist of critical items that should be followed
during the test run. These critical items consist of:
Recommended flow rates, temperatures, and sampling times
0 Reminders on laboratory and sampling technique
Specific equipment checks.
While this list is provided for review prior to the sampling run, its
best use is an an on-site checklist for the supervisor and quality assur-
ance personnel .during the run. During a test audit the supervisor or QA
representative should initial each item successfully completed. The entire
list should be included with the documentation of that test run. The oper-
ating personnel might also like to have copies of the checklist for ref-
erence during the execution of the test run. Copies should be posted in
the laboratory and sampling site for this purpose.
b) Data Monitoring Procedures
Dry aerosol sampling procedures cannot be tested in the classical
fashion, i.e., with spiked samples, to determine their reliability, but
there are simple monitoring activities that can be carried out on a daily
basis. These activities include:
Calculation of percent isokinetic sampling which provides
information on the quality of sampling.
§ Comparison of Aerotherm/MRI grain loadings which determines
the efficiency of particle recovery from system.
The following paragraphs detail these procedures.
1. Isokinetic Sampling Tests
Isokinetic sampling is sampling at a rate (measured at the probe noz-
zle) equal to the velocity of the gas flowing by the probe. Unless gas
flows are sampled isokinetically, larger or smaller particles can be pref-
erentially collected depending on whether the sample rate is less than or
98
-------
TABLE 5. CRITICAL CHECKPOINTS FOR MRI DRY AEROSOL SYSTEM
Checkpoint
Initials
Supervisor
QA Inspector
Remarks
I. Conditioning and Preweighing Procedures:
A. Conditioning
- Plates cleaned with toluene and
NaOH prior to greasing.
- Hands gloved.
- Thin film of grease applied to
plate.
- Handle clean filter with tweezers.
- Clean petri dishes and watch
plates prior to storing Impactor
plates and filter.
- Check labeling system on petri
dishes.
- Plates conditioned at 175°Cfor
2 hours.
- Filter conditioned at 290°C
(±10°C) for 4 hours.
- Deslccant fresh (color B/P?)
- Plates and filter desiccated for
2 hours.
B. Weighing
- Balance area clean.
- Balance pan clean.
- Balance leveled.
- Balance zeroed.
- Weigh standard.
- Weigh Impactor plate and filter
blanks for each set of samples.
- How long were petri dishes and
Impactor stages left open to lab
air ( mln.j?
- Data recorded on correct data
sheets.
- Note condition of plates (color of
grease, thickness of coating, etc.)
on data sheet.
. Laboratory Impactor Preparation
- Plates and filter Identifications
and weights recorded,
- Impactor Inspected for wear
(threads, jet plates, Inlet nozzle,
0-rlngs)
- Impactor cleaned.
1 - Work area cleaned and bench cov-
ered with Whatman paper.
- Hands gloved.
- Impactor stages loaded, starting
from last stage.
99
-Continued-
-------
TABLE 5. CRITICAL CHECKPOINTS FOR MRI DRY AEROSOL SYSTEM (CONTINUED)
Checkpoint
Initials
Supervisor
QA Inspector
Remarks
Backing, filter, stainless steel
seal washer placed in the filter
stage.
Perform impactor leak test.
Inlet tubing sealed off to prevent
particles from entering impactor.
III. Site Set-Up
- Impactor maintained in vertical
position during transfer to site.
- Brush inside probe prior to run.
- Rinse probe with acetone until
rinse solution is clear.
- Fresh solutions placed in impingers.
- Leak test performed and magnehellc
gauges zeroed.
- Leak rate must be less than 0.02
cfm.
- Thermocouple leads attached to
impactor
- Skin temperature controlled to
<347°F.
IV. Sampling Run
- MRI gas out temperature maintained
at highest stack reading +25°F.
- MRI gas out temperature must never
exceed 347°F.
- Select flowrate below 0.8 cfm.
- Select sampling time to collect
no more than 10 mg on any stage.
- Check seal between probe and port
to prevent any outside air enter-
ing the stack.
- Impactor carried in an upright
position to laboratory.
- Support equipment cleaned prior
to next run.
- Report any experimental problems
or unusual occurrances on data
sheet.
V. Sample Recovery
- Keep probe in a horizontal position
prior to particulate matter
recovery.
- Keep Impactor upright while trans-
ferring to lab.
- Particulate matter from probe
rinsed Into 250 tnL Erlenmeyer
flask.
- Probe and tubing connections Inlet
to the~Tipactor are brushed and
rinsed with acetone until rinse
stream Is clean. Rinsings collec-
ted In Erlenmeyer flask and saved
for weighing 1n a tared 50 nL
beaker.
-Continued'
100
-------
TABLE 5. CRITICAL CHECKPOINTS FOR MRI DRY AEROSOL SYSTEM (CONTINUED)
Checkpoint
Initials
Supervisor
QA Inspector
Remarks
- Transfer probe and tubing washes
from the Erlenmeyer to tared 50 ml
beaker.
- Estimate and record any loss of
material during transfer to tared
50 ml Erlenmeyer (% lost ).
- Dry probe washings in oven at
110°C for 1 hour.
- Use gloves during impactor stage
removal procedure.
- Impactor cooled to room
temperature
- Use tweezers to remove inpactor
plates.
- Inspect impactor walls and jet
nozzles for particulate matter.
Brush any paniculate matter
there onto the Impactor plate.
Note presence of particulate
matter on the walls or jet nozzles
on data sheet.
- Inspect filter holder for pieces
of filter material.
- Collect all pieces of filter
material from filter holder and
place them with the Intact filter
for weighing.
Keep all exposed Impactor plates
and samples covered.
- Desiccate all samples 2 hours
prior to weighing.
- Correct all weights for any change
1n Impactor and filter blank
weights.
- Note tackiness of plates.
VI. Data Analysis Verification
- Plot Aerotherm and MRI grain
loading values on a daily basis
(Y-axis for grain loading,
X-axis day).
- Plot the daily percentage 1so-
kinetic for MRI runs (Y-axis for
percent Isoklnetic, X-ax1s day)
- Plot sample blank weight change
(Y-axis wt., X-axis day).
101
-------
greater than the stream flow rate, respectively. The closer to isokinetic
conditions, the more representative the particle sample will be. Thus,
the degree of isokinetic sampling is an expression of the quality of the
sampling runs. The normal criteria of acceptability is ±10 percent of the
correct isokinetic flow rate.
Because the internal gas velocity determines the size cutoffs for the
collection stages, the impactor must be run at one flow rate if only one
size range of particles is to be deposited on a given stage. Consequently
the velocity profile is determined just prior to the test run and the
impactor is run at the average isokinetic flow rate.
The procedure for the isokinetic check follows:
(a) From the actual velocity profile measured during the run com-
pute the average -s/AP.
(b) Calculate the average stack velocity:
/T. +
I -g-^
\ KS
4601/2
= (85.48) Cp JAP I -g-^ - 1 (4-14)
p
Y
where
V<. = Average stack velocity (ft/sec)
Cp = Pi tot tube coefficient
,1/2
= Average square root of the velocity head (in. H«0)
TS = Average stack temperature (°F)
P_ - Absolute stack gas pressure (in. Hg)
NL = Stack gas molecular weight
(c) Correct MRI meter volume (VM ) to flow rate at stack conditions1
102
-------
where
QSM = MRI flow rate (acfto)
VM = MRI dry test meter volume (cf)
Mm
B = Volume fraction of water in gas stream obtained from previous
w mass loading run
T_ = Average stack temperature
T = Average meter temperature
P"M = Average absolute meter pressure (in. Hg)
PS = Absolute stack pressure (in. Hg)
t = Sampling time (min)
(d) Calculate stack velocity based on MRI flow rate:
0.327
(4-16)
where
VCM = MRI stack velocity (ft/sec)
or!
DM = MRI nozzle diameter (in.)
M
(e) Determine percentage isokinetic (I):
o
(100%) (4_17)
(f) Make a continuous plot of I on x-y graph with I on the y-axis
and the day on the x-axis. This daily plot should be kept as
a permanent record with the MRI runs. As the data begins to
accumulate, trends will be established. Normally the points
should vary. Consistently high or low I values indicate sys-
tematic errors in sampling and call for a review of procedures
and equipment. Errors can be due to:
t Equipment - Inaccurate or malfunctioning magnehelic
gauges, dry test meter or thermocouples. Refer to
Tables 7 and 8 for troubleshooting and calibration
procedures.
103
-------
Data recording - Wrong numbers taken or misplaced on
the Field Data Sheet. Double check all entries.
Calculations - Either the wrong data were input or a
mathematical error was made. Have different indi-
viduals do the calculation.
2. Stack Mass Loading Evaluation
Since an Aerotherm (Method 5) mass reading is performed prior to the
MRI run, it is possible to compare MRI mass loading values (CM) with those
M
obtained from the Aerotherm (C.).
(a) Correct the Aerotherm dry test meter volume (V ) to standard
condition:
_\ /
SSJ \
w - v i_ 528
Vm(STD) - Vm \T + 460l \PCTJ (4-18)
where
T = Average meter temperature (°F)
P = Average absolute meter pressure (in. Hg)
PSTD = standard Pressure (29.92 in. Hg)
(b) Determine Aerotherm mass loading (C.) in gr/scf:
WTA (15.43)
CA = -$- (4-19)
rt VM(STD)
where
WTA = Total particle weight recovered from Aerotherm train (g)
(c) Using V|vjm and equation 4-18, calculate the MRI gas volume
sampled at standard conditions
where
VM = Dry test meter reading during MRI run
104
-------
(d) Calculate mass loading for MRI (CM) in gr/scfm:
W (15.43)
CM = -^ (4-20)
M VM(STD)
where
WM = Weight collected from MRI train (probe, stages, and
filter in g)
(e) Ratio calculated mass loadings:
G = CM/CA (4-21)
where
G = Correlation variable for MRI and Aerotherm system
(f) Plot the daily G values and observe the trend. The range of
acceptable agreement is from 0.7 to 1.3. Consistently high
G values indicate contamination or incorrect flow rate calcu-
lations. Verify that the correct values and equations were
used. Also check the calibration of the magnehelic gauges.
Consistently low G values are more likely to be found. The
most probable cause for this result is poor overall parti-
culate matter recovery, but especially poor recovery from
the probe. Review cleaning procedures and make extra effort
to clean the probe and connecting tubing properly.
Other reasons for low G values are:
High flow rate values - check calculation and dry test
meter calibration.
Grease weight loss - verify impactor temperature was
less than 347°F. Note any amber discoloration on filter
signifying grease flow through. Correct these problems
by controlling temperature to <347°F, placing a thin
film of grease on the plates, and maintaining gas flow
at <0.8 acfm.
105
-------
4.2.3 Goksoyr-Ross Methodology
At a given temperature a gas can hold a specific amount of a liquid as
a vapor. As the temperature is lowered less of the liquid can exist as a
gas and condensation begins. The amount of liquid existing as a gas at a
given temperature will be related to the liquid's boiling point. Conse-
quently various liquids will condense at different temperatures. Thus a
flue gas can be conditioned to a specific temperature to separate H?SO.
(b.p. ~300°C) from water (b.p. 100°C).
The G/R system (Figure 19) consists of a quartz probe heated to 316°C
(600°F) to collect the gas from the duct. No attempt is made to sample
isokinetically. The flow rate is controlled during the sampling by moni-
toring the total flow at the dry test meter with a stopwatch. The gas
stream then passes into a heated (288°C-550°F) quartz filter holder which
contains a Tissuequartz filter to remove particulate matter from the gas
stream. The filter temperature must be maintained to ensure quantitative
recovery of HUSCL. The clean flue gas then flows into the water jacketed
coil maintained at 60°C (140°F) to condense and collect any sulfuric acid
vapor that might be present in the gas stream. Temperatures dropping below
this value will condense HpO and S02 and cause positive errors. After a
period of time (1 hour or until 1/2 to 2/3 of the coils are frosted), the
coil is rinsed out and the acid determined by titration with NaOH using
bromophenol blue as the indicator. For a discussion of the proper proce-
dures to be used during this titration refer to 4.1.9.
a) Critical Checkpoints
Table 6 is a checklist of critical items that must be followed dur-
ing the test run. This critical item list consists of:
Recommended flow rates, temperatures and sampling times
Reminders on laboratory and sampling techniques
Specific equipment checks.
While this list is provided for review prior to the sampling run, i'ts
best use is an an on-site checklist for the supervisor and quality assur-
ance personnel during the run. During a test audit the supervisor or QA
representative should initial each item successfully completed. The
106
-------
TABLE 6. CRITICAL CHECKPOINTS FOR G/R H2S04Q SAMPLING SYSTEM
Checkpoint
Initials
Supervisor
QA Inspector
Remarks
I. Laboratory Preparation
- Inspect and clean G/R coil. Both fil-
ter holder and G/R are cleaned with
hot chromic acid solution and D.I. HgO.
- Rinse with acetone and air dry G/R
coil.
- Place Tissuequartz filter in filter
housing.
- Check seal between end of joint and
filter.
- Do not use grease on joints.
- Inspect and clean all glass joints.
II. Site Set-Up
- Rinse the inside of probe prior to
run.
- Rinse probe with acetone until rinse
solution is clear.
- Perform leak test.
- Leak rate must be less than 0.003
cfm or 80 mL/min.
- Zero Magnehelic gauges.
- Thermocouple leads attached to probe
and filter.
- G/R water bath held at 140°F (±2°F)
- Leak test train.
- Probe temperature maintained at 600°F
(+30°F).
- Gas temperature out of filter holder
held at 550°F (±10°F)
- Fresh solutions placed In impingers.
- Fresh absorbent replaced in final
impinger.
- Adjust flowrate in system toio Lpm.
III. Sampling Run
- Turn vacuum pump on just before Insert-
ing probe in the stack.
- Check seal .between probe and port to
prevent any outside air entering the
stack.
- Run test for 1 hour or until colls are
frosted to 1/2 to 2/3 of their length.
- After run cap both ends of the probe
and lay in horizontal position.
- Rinse the G/R coil Into the modified
Erlenmeyer flask with a maximum of
40 mL of D.I. H20.
- Mas any of the solution lost ( mL
estimated)?
- Handle hot glassware carefully to
prevent personnel Injury and damage
to equipment.
107
-Continued-
-------
TABLE 6. CRITICAL CHECKPOINTS FOR G/R H2S04Q SAMPLING SYSTEM (CONTINUED)
Checkpoint
Initials
Supervisor
QA Inspector
Remarks
- After probe has cooled, the probe is
rinsed with a maximum of 40 mL D.I,
H.O into a 125 ml Erlenmeyer.
- Was any solution lost ( ml
estimated)?
- Clean support equipment prior to next
run.
- Save filter for titration.
IV. Laboratory Analysis
- Clean glassware prior to titration.
- Use Bromphenol Blue indicator.
- Is the NaOH buret protected with a
C02 absorbant tube?
- When was NaOH standardized last
(Date )?
- Filter any solution that has suspended
particulate.
- Use same number of indicator drops
for each titration.
- Perform indicator blank on a volume
of D.I. H,0 equal to sample aliquot
used. *
- Indicator blank added to H-SO. milli-
equiva'lents found.
- Perform all analyses in triplicate.
V. Data Analysis Verification
- Obtain and titrate test samples from
main laboratory.
- Plot dally Inlet and outlet ^04
values (Y-axis for ppm H-SO., X-axis
for day). ' £ H
108
-------
list should be included with the documentation of that test run. The oper-
ating personnel might also like to have copies of the checklist for ref-
erence during the execution of the test run. Copies can be posted in the
laboratory and sampling site for this purpose.
b) Data Monitoring Procedures
The data monitoring procedures for the G/R system are devoted mainly
to the acid-base titration performed in the laboratory and to the monitor-
ing of the HpSO. ppm values calculated.
1. Acid Base Titration
In order to check the accuracy of the titrations performed on the G/R
samples, an independent check of the NaOH solution and titration method is
required. From main laboratory or an independent laboratory, a standard-
ized sample of H?SO. approximately 0.01 N should be analyzed by the trailer
Personnel every couple of weeks. Analysis of the sample should be trip-
licate and reported to 3 places (O.X Y Z). Analysis of this sample will
provide information on the precision of the G/R titrations and accuracy of
the results.
The procedure follows:
(a) Take a 10 ml aliquot of the unknown standard
(b) Titrate in triplicate with Srompphenol Blue to the blue
end point and record the number of mil 11 liters used.
(c) Determine the normality of the solution from:
NR VR
NA = -5-B. (4-22)
M
where
N. = Normality of the acid
V. = Volume acid aliquot taken (mL)
NB = Normality of the base
VB = Volume of the base used to titrate the sample (ml)
109
-------
The results of the determination should not differ by more than ±10$
within the triplicate numbers nor should the determined normality be off
by more than ±10%.
If the values differ by more than 10%:
Check the calculations and be sure the correct values have
been used
t Repeat the analysis
If the value is still off, restandardize the NaOH with KHP.
0 Repeat the test.
2. Data Monitoring by Statistical Quality Control
Since there is a direct correlation between the inlet and outlet
values such that the outlet is predictable from the inlet contamination,
a simple control chart using regression analysis can be used Figure 25.
Simply plot all the values obtained on this chart. The region between the
-2o and +2a limits should contain, in the long run, 95% of all future
paired measurements.
The a limits are based on 44 of paired SO., measurements completed at
the Venturi scrubber. It is assumed that similar results will be obtained
on the TCA so that this chart can be used for both systems. The warning
limits will be the (90,90) limits. That is, it is expected that 90% of
the future observations will lie within such limits, 90% of the time. The
rejection limits, or the limits that indicates that the system is out of
control will be the (90,95) limits. That is, it is expected that 95% of
the future observations will be within these limits 90% of the time. As
trends develop, data that is widely outside of the normal range can be
spotted. When those events occur, be sure to:
Record any unusual occurrences during the test on the data
sheets
110
-------
25.0 -
20.0
o_
0.
O
IT)
O
15.0
10.0
5.0
0.0
-5.0
0.00 5.00
J.
10.00 15.00 20.00
INLETH2S04(PPM)
_L
25.00
Figure 25. Control Chart for Controlled Condensation Measurements of H SO
-------
Check with the power plant of scrubber control room to find
out if any mechanical problems occurred during the run.
Verify that all the laboratory numbers are correct and repeat
the analysis if any solution is left over.
4.2.4 Maintenance Schedules
Table 7 details the recommended maintenance schedule. Following of
this schedule is imperative to prevent breakdown and to maintain the high
accuracy required in the program.
4.2.5 Troubleshooting and Repair Procedures
Table 8 lists possible problems that can be encountered with the
equipment used in the test program. This list is only a beginning and
should be updated as new problems are encountered and solved.
4.3 REFERENCES
4-1 Marple, Virgil and Willeke, Klaus. Inertial Impactors: Theory,
Design and Use. From Fine Particle: Aerosol Generation Measure-
ment, Sampling and Analysis, ed B.Y.H. Liu, Academic Press, 1976.
4-2 Federal Register. 41(111): 23082-23083,1976
4-3 Federal Register. 41(11): 23076, 1976
4-4 Aerotherm Isokinetic Flow Rate Calculator Manual, Accurex Corp.
Mountainview, California
112
-------
TABLE 7. GENERAL MAINTENANCE SCHEDULE
Component
S-Pitot nozzles
Probe nozzles
Steel probes
Quartz probe
Impingers
Maintenance Schedule By:
Run
Inspect alignment after
each run from head-on
and side angles.
Blow out parti cul ate mat-
ter after each run.
Inspect nozzle for damage
Brush nozzle before and
after run to remove inside
parti cul ate.
Measure ID with micrometer
before each run.
Check alignment of nozzle
before each run by looking
at nozzle from head-on and
side angles.
Before and after each run,
brush and rinse with rea-
gent grade acetone or
Freon until rinse is
clean.
Before and after each run
brush and rinse with rea-
gent grade acetone or
Freon until rinse is
clean.
Rinse out after each run
with DI water.
Inspect and clean seal
area and 0-rings.
Leak test before each run.
Week
Brush-out inside of pi tot
tubes
Use wire bush to clean
inside surface of the
probe.
Month
Calibrate C every three
months p
Calibration Procedure
Send to Muscle Shoals for C
check and calibration. "
Measure nozzle ID with micro-
meter.
N/A
N/A
N/A
Leak test at 380 torr (15" Hg)
and verify that a leak rate of
<0.3 1pm (0.02 cfn) is maintained.
-Continued -
-------
TABLE 7. GENERAL MAINTENANCE SCHEDULE (CONTINUED)
Component
Maintenance Schedule By:
Run
Week
Month
Calibration Procedure
Pump
Swage!ok fittings
Brink Impactor
MRI Impactor
CCS filter holder
CC coil
HaOH solution
Before each run check leak
rate in pumping system.
Inspect fitting, espe-
cially ferrule and seat
for wear and dirt. Clean
or replace fitting as
required.
Clean Impactor
Check washers for wear.
Inspect jets for blockage.
Clean impactor after each
run. Note: Acetone
should not contact Viton
0-rings.
Check 0-rings for wear
Inspect jet nozzles for
blockage.
Inspect and clean after
each run. Replace filter
after each run.
Inspect and clean after
each run
Inspect vanes on diaphragm.
Inspect and clean motor
brushes.
Every 3 months check impac-
tor flowrate calibration.
Every 6 months check OSQ
calibration.
Every 6 months check D,
calibration.
'50
Depending on the system, a leak
rate musi be less than a certain
value. See specific critical
checkpoint table for information
on recommended maximum leak rates.
N/A
Hook to dry test meter, start
flow, time, read Ap across impac-
tor. Record data. Repeat at
different flowrate. Compare
results to calibration chart.
Send impactor to Southern
Research Institute in Birmingham,
Alabama for calibration.
Send impactor to Meteorology
Research, Inc., Altadena, Cal.
for calibration.
Clean frit each week in hot
chromic acid for 12 hours.
Rinse to neutral pH with
01 H20.
Clean coils and frit each
week in hot chromic acid
for 12 hours. Rinse to
neutral pH with DI H20.
Standardize the NaOH with
KHP weekly.
N/A
N/A
-Continued-
-------
TABLE 7. GENERAL MAINTENANCE SCHEDULE (CONTINUED)
Component
Maintenance Schedule By:
Run
Week
Month
Calibration Procedure
Thermocouples
Temperature
Indicator
Oven or probe
heaters
Connecting
lines
Magnehelic
gauges
Calibrated
orifice
Dry test meter
Inspect lines for wear and
kinks.
Clean readout of all dust.
Clean tips of shielded TCs
Clean connector prongs
with steel wool.
Blow out connecting lines
with air
Visually inspect exterior
for wear. Especially
Inspect hose to fitting
connections.
Check lines to gauges
to ensure there is no
blockage.
Zero gauge before run
with both ports open
to the atmosphere.
Clean exterior
Flush with water and dry
with clean plant air.
Clean and inspect critical
orifice
Calibrate thermocouples
Have electronics shop
remove the back and clean
the inside of the unit.
Check indicated tempera-
ture with calibrated
thermocouple.
Every month check calibra-
tion of gauge versus water
or mercury manometer.
Calibration check
Calibrate versus wet test
meter every 3 months
Calibrate TC at two points (ice-
water and near boiling). Compare
TC readings to mercury thermometer.
Replace TC if agreement is not
within 3°C (6°F).
Perform thermocouple calibration
with readout unit using indepen-
dently calibrated thermocouple.
Check indicated temperature read-
ings with calibrated thermocouple.
N/A
Calibrate versus water or mercury
manometer depending on the range
of the gauge. Connect manometer
and gauge to vacuum or pressure
source simultaneously using a tee.
Check the magnehellc gauge's read-
ings at low, medium and high
points in its range.
Calibration procedure for a criti-
cal orifice is found on page A6
in Appendix A.
Run wet and dry test meters in
series, note temperature and
pressure. If dry test meter is
>3Z off, send to factory for
recalibration.
-------
TABLE 8. TROUBLESHOOTING AND REPAIR
Component
General Remarks
Problem
Repair Sequence
S-Pitot Nozzles
Probe Nozzles
Steel Probe
Quartz Probes
Inpingers
Alignment of pi tot tubes is critical.
The tubes must be facing 180° with
respect to each other and parallel
to gas flow in the duct.
A smooth circular edge is required
for accurate sampling. Alignment
of nozzle face must be perpendicular
to gas flow.
Because the probe contains the S-pitot
nozzles, alignment of the probe must
be checked with a level once the
probe 1s In the stack.
Avoid mechanical shocks especially
when probe 1s hot. Before cleaning
probe with liquids, allow the probe
to cool to air temperature.
Impingers should be cleaned with soap
and water. Deposits should not be
allowed to build up Inside Impinger.
All nozzles should reach to within
±1.3 cm (0.5") of the bottom of the
impinger. To insure good seals,
keep the Impinger seals clean.
Misaligned nozzle
Damaged edge
Nozzle wear or damage
Misalignment
Normal wear and cleanliness
Normal wear and cleanliness
Normal wear and cleanliness
Return S-Pitot tubes to original 180°
alignment.
Align nozzles to be parallel to gas
flow.
Position face of nozzle to be perpen-
dicular to gas flow.
File and buff edge to smooth oval -
repeat alignment checks.
File and buff edge to smooth circle.
Loosen Swagelok fitting and realign
(x-axis) nozzle face to perpendic-
ular to gas flow.
Bend nozzle neck (y-axis) so that
nozzle face is perpendicular to gas
flow.
Pitting on the inside of the probe
should be removed by use of a wire
brush.
Brush and rinse with acetone after
each run (Note: Test brush to insure
it is not dissolved by the acetone).
Rinse out with DI water after each use.
Dry impinger to be used for moisture
trap.
Clean sealing edges and 0-ring of
impinger.
-Continued-
-------
TABLE 8. TROUBLESHOOTING AND REPAIR (CONTINUED)
Component
General Remarks
Problem
Repair Sequence
Impingers
Thermocoupl es
Temperature
Indicator
Oven or probe
heaters
Thermocouple (TC) leads and wire are
fragile and require care in arranging
the equipment set-up to prevent kink-
ing and stripping of leads. Never
pull a TC apart by pulling on the
lead. Verify that the polarity is
not reversed anywhere in the system.
Be sure that the same type of TC
wire and connectors are used in the
system (Iron-constantan or chromel-
alumel). Do not bend casing of
shielded thermocouple.
Store in dust free area
Never exceed maximum temperature as
stated in the manufacturer's manual.
Leakage in impinger system
Temperature indicator fluctuating
over wide range.
Temperature readings fluctuating
on one channel.
No temperature readout or fluc-
tuating temperatures on all the
channels with thermocouples
attached.
No temperature rise with current
on.
Check all Swagelok fittings.
Inspect impinger seal area for dirt
or damage. Clean area if dirt found.
Use larger 0-ring.
If all other measures fail to
locate leak, pressurize and immerse
in water to find leak.
Locate possible short in TC wire or
connectors. Once portion of wire
with short is located, mark and have
the wire replaced.
Have readout checked by electrical
shop if no external short can be
found.
Check thermocouple for short in lead
or connectors.
Return to electronic shop for repair.
Return to manufacture if problem
cannot be found.
Check electrical connections.
Check main power.
Check fuses and circuit breakers.
Verify thermocouple connected.
-Continued-
-------
TABLE 8. TROUBLESHOOTING AND REPAIR (CONTINUED)
Component
General Remarks
Problem
Repair Sequence
Connecting lines
Magnehelic
gauges
CO
Calibrated
Orifice
While these lines are either heavy
vacuum hose or steel braided Teflon
lines, care should be taken to mini-
mize weight supported by the lines
and excessive mechanical abuse. The
Aerotherm lines should never be
kinked to cut off flow.
Magnehelic gauges measure the pres-
sure differences felt by an internal
diaphragm. The diaphragm is mag-
netically linked to the display
needle. These gauges can stand a
certain amount of over-pressure,
but should not be left in that
condition for long. The normal
operating temperature is 30 to
140°F.
A calibrated orifice is a constriction
in a tube in which a gas is flowing
that causes a difference in pressure
between the upstream and downstream
sides of the constriction. This
pressure differential (AH) is
related to the rate of flow.
General maintenance.
Pegged needle
No reading
Erratic Readings
Higher AH values for flowrates
at the same conditions.
Replace any worn or corroded parts.
Remove leads, blow into both sides
and reset zero if necessary.
Check connections to gauge.
Check leads for blockage.
Clean lines if necessary.
Fluctuation in pressure reading
probably due to surges or cycles
in pumping system. Place Swagelok
snubber on the inlet to the gauge.
Recheck AH calculation.
Check lines for particulate matter.
Inspect critical orifice for cor-
rosion or blockage. Clean orifice
with copper wire. Recalibrate
orifice.
-------
TABLE 8. TROUBLESHOOTING AND REPAIR (CONTINUED)
Component
General Remarks
Problem
Repair Sequence
Pump
Care must be taken in shutting the pump
off after a run. Rapid shutdown with-
out bleeding air into the pumping sys-
tem will cause the impingers to back-up
towards the filter.
Leakage (oil-less)
Leakage (diaphragm)
Swagelok fittings
Swagelok fittings are designed to seal
with a minimum of tightening. Exces-
sive torque applied to the fitting
will eventually cause leakage.
Installation
Check all valve and hosing connections
leading to pump.
If the leakage has been isolated in
the pump, disassemble pump and inspect
vanes for wear and replace if necessary.
For leakage or low flow in diaphragm
pumps check the diaphragm cover to
ensure it has not vibrated loose.
Remove face plate and inspect dia-
phragm for signs of wear or pinholes.
Check the diaphragm gasket for wear,
replace if necessary.
Insert the tubing in the service.
Insert the tubing into the Swagelok
tube fitting. Make sure that the
tubing rests firmly on the shoulder
of the fitting and that the nut is
finger-tight.
Due to the variation of tubing dia-
meters, a cannon starting point is
desirable. Therefore, use a wrench
to snug up the nut until the tubing
will not turn (by hand) in the fit-
ting. At this point, scribe the nut
and body at the 6 o'clock position of
the fitting. Now while holding the
fitting body steady with a backup
wrench, tighten the nut one-and-one-
quarter turns. Watching the scribe
mark, make one complete revolution
and continue to the 9:00 o'clock
position.
-Continued-
-------
TABLE 8. TROUBLESHOOTING AND REPAIR
Component
General Remarks
Problem
Repair Sequence
Swage!ok Fitting
Dry Test Meter
Brink impactor
ro
o
MRI impactor
Reinstallation
These meters are very sensitive to
mechanical shock and should be handled
with care. Corrosive gas from the stack
should never be passed through the
meter without prescrubbing.
The Brink impactor operates at a very
low flowrate which requires that very
low leak rates must be maintained.
Any interior part must never be
cleaned with any material that can
scratch the metal.
Incorrect volume readings.
Leakage >80 ml/min
The MRI impactor has an aluminum
housing which requires care to pre-
vent the thread from being stripped.
Since the jet plates are removed
during sample recovery, care must
be taken to ensure that the plates
are not scratched.
Plugged nozzle
Leakage >0.02 cfro
Melted 0-ring
Tubing with preswaged ferrules
inserted into the fitting until front
ferrule seats in fitting. Tighten
nut by hand. Rotate nut about one-
quarter turn with wrench (or to
original one-and-one-quarter tight
position) then snug slightly with
wrench.
Check meter for blockage
Check mechanical linkage for wear
Recalibrate meter
Check all fitting and connections
Tighten impactor housing
Check all Teflon seals. Replace
if necessary
Use copper wire to dislodge
material
After cleaning, check nozzle size.
Check all fittings
Inspect 0-ring seals for damage
or flattening. Replace worn
0-rings
Tighten impactor housing
Replace standard 0-rings with Viton
oversized 0-rings
450°F exceeded during run
-------
TABLE 8. TROUBLESHOOTING AND REPAIR
Component
General Remarks
Problem
Repair Sequence
MRI impactor
CCS filter holder
CC coil
The G/R filter holder is made out of
quartz and especially when it is hot,
mechanical shocks will cause breakage.
The filter holder is designed to
always be run with a filter on the
quartz frit. Because of the high
temperatures employed, greasing the
joints is not recommended.
Melted 0-ring
Plugged jet plates
No seal to filter
Gas leakage
The coil is an especially delicate
piece of equipment. Clear visibility
of the coils is necessary, so main-
tain the water jacket's cleaniness.
Plugged frit
Gas leakage
Viton 0-rings not used
Replace with Viton 0-rings
Use copper wire to clean
Clean in soap and water, and rinse
with DI H20 and blot dry.
Check extension tube. If it is not
making a seal, have the glass blower
repair. As a temporary repair, a
washer out of tissuequartz can be
used to promote a seal.
Check thermocouple well for pinhole
leak.
Check alignment of ball and socket
joints. Try to maintain linearity.
Check seal at joints, clean joints,
and retest.
Check joints for thermal warping.
Replace.
Soak in hot chromic acid cleaning
bath for 12 hours. Rinse with DI
H20 till neutral.
Check thermocouple well for pinhole
leak.
Check alignment of ball and socket
joints. Try to maintain linearity.
Check seal at joints, clean joint,
and retest.
Check joints for thermal warping.
Replace.
Soak in hot chromic acid cleaning
both for 12 hours. Rinse with DI
H20 till neutral.
-Concluded-
-------
APPENDIX A
Isokinetic Flow Rate
Calculator Instructions
122
-------
AEROTHERM
ACUREX Corporation
OPERATING MANUAL
ISOKINETIC FLOW RATE CALCULATOR
P10-01
AEROTHERM DIVISION
ACUREX CORPORATION
485 Clyde Avenue
Mountain View, California 94042
415-964-3200
123
-------
INDEX
1
Introduction ,
Operating Parameters .
Detailed Instructions ,
Additional Thoughts L
1. Rapid Use of Calculator
2. Variable Limits - Resetting Calculator
Worked Examples
Orifice Calibrations ,5
Variable Orifice/High Valume Sampler ^
Variable Molecular Weights
124
-------
iage 2
Consider a source test under the following parameters:
Cp = 0.85 pitot tube coefficient
AH@ = 1.95 orifice calibration factor, supplied
by train manufacturer or determirsd
experimentally ' (page 6)
%H20 =20 Percent water by volume present in
stack gas
Tra =75°F Temperature at the dry gas meter
Ps/Pm =1.1 Ratio of STACK Pressure (Ps) to
Pressure at the meter (Pm)
Tg = 650°P Temperature of stack gas
APavg = 0.2"H20 Average Pitot reading taken en.
preliminary traverse
. ( Dn Nozzle diameter (inches)
fle^e { AH Orifice pressure differential
{ AP Individual Pitot reading
Detailed Instructions
1. Set Cp over AH@
Place the cursor on the correct value of Cp (pitot tube
coefficient) for your train. If Cp has not been jalibrated,
assume 0.85 for an S or reverse type pitot, and 0.99 for a
standard type pitot tube. Almost all EPA type trains employ
an S type pitot tube.
Move the top slide until the correct value of AH@ (orifice
calibration factor) is directly under the Cp valuj. In our
example, 1.95 would be placed under 0.85. If the :rain has
variable orifices or very unusual calibration fac :or, see
page10.
125
-------
Page J
2. Set % H20 to Reference
Move the cursor to the reference arrow. Then move the
second body slide so the correct %H20 is under the ireferen
arrow. (%H20 is either measured directly from a previous
experiment or is estimated. A variation of - 1% is
acceptable, if
In our example, 20% is placed below the arrow.
3&4 Read Index at arrow/set Tm at Index Number
«
To facilitate transferring numbers from one slide to anothe
an index (number line) has been provided. Read the inde
at the reference arrow, then slide Tm to that number. A
faster method is to move the hairline over the reference
arrow and then slide Tm to the hairline. In our exaI"fitoF
the hairline would be moved to the reference arrow and '5
then slid to the hairline.
b&
5&6 Read Second Index Number at TS/set Ps/Pm at Second Index
the
The number line may be read again as in 3 & 4 or using y-
rapid method: Move the hairline to Ts, then set Ps/Pm t°
hairline. Be sure hairline does not move.
7. If Dn is not known, set the average AP reading under the g
reference arrow C on the AH scale. Be sure the B referen
arrow does not move (hold this point with the hairline i*
desired). Read the exact nozzle size under the B arrow.
in our example, this is 0.380 inches. We would select a
3/8" nozzle (0.375 inches) and move the Dn scale until
0.375 is directly under the reference arrow B.
8. Read AH setting opposite AP reading using cursor as neede
The proportional ratio between AH and AP has now been
and any value of AP is now directly under the correct
of AH.
For our example:
AP AH
1 8.8
.5 4.4
.2 1.76
.1 .88
The bottom slide is designed to be tightly held in the calcu
body so that the Ap/AH ratio does not slip during use. sho^n
this loosen through use, a piece of tape on the back side wi
again tighten it.
126
-------
PC ge 4
Additional Thoughts;
Rapid Use of the Calculator
With experience, all the operation and settings car be done
in less than 25 seconds. We suggest the following sequrnce for
maximum speed.
1. Set Cp over AH@
2. Move hairline to arrow, then set %H20 to hairl: ne
3. Move hairline to next arrow
4. Set Tm to hairline
5. Move hairline to Ts
6. Set Ps/Pm to hairline
7. Set hairline to B
8. Set avg.AP to arrow C
9. Select nozzle size and move to hairline
10. Read AP vs. AH
(all that in 25 seconds!)
Resetting calculator
At times, variables may change during the course o: a test
necessitating readjustment of the calculator. The following is
presented as a guideline:
Cp should not change
AH@ should not change
%H20 ± 1%
Tm t 10°F
Ts t 250F
Ps/Pm i 1%
Rapid temperature change: If the stack temperature (Ts)
changes significantly during the course of the test (±2£ - 50°F),
the calculator may be reset rapidly without repeating the entire
calculation as follows:
A. Place the hairline over the new Ts and move old
Ts to the hairline.
B. Move the actual nozzle size to reference Arrow B
if the old nozzle size will still produce reasonable
flow rates. Otherwise, select a new nozzle.
C. Read AH across from AP as before.
127
-------
WORKED EXAMPLE
Page 5
Example 1
Cp
AH@
%H2°
Tm
Ts
Ps/Pm
Dh (exact)
Dn (actual)
Avg. AP
Apl
AH1
AP2
AH-,
.85
1.95
30
100
500
1.1
.383
.375
.2
1.0
8.5
.1
.85
.85
1.95
0
100
500
1.1
.330
.375
.2
.1
1.54
.01
.154
.85
1.95
30
100
1500
1.1
.735
.750
.03
.1
6.67
.01
.667
.85
3
30
100
500
1.1
.230
.250
1.0
1.0
2.59
.1
.259
.85
1.95
30
0
500
1.1
.506
.5
:os
.1
2.21
.01
.221
.99
1.95
30
100
500
1.1
.355
.375
.2
.1
1.15
.01
.115
.85
1.95
30
100
500
.8
.276
.250
1.0
1.0
1.22
.1
.122
128
-------
Cali
ibrating
Orifice Meters
page 6
A constriction in a tuba in which a gas is flowing causes a difference
in pressure between the upstream and downstream sides of tlie ccr.strictlon.
This pressure differential is related to the rate of flow.
1
1
I
AH
t
Figure 1. Oriffce Meter
An orifice meter, Figure I., is a type of constriction wh ch uses
the following relationship between the flow rate and the pressure differential
to measure the rate of flow:
m
tn m
where Qm = volumetric gas flow rate
Tm = gas temperature, absolute
« molecular weight of the gas
Pm - pressure of the gas, absolute
pressure drop across the orifice
K = a proportionality factor
subscript m = refers to the meter
129
-------
page
FLOW DIAGRAM
Fine valve
Pump
Wet test meter
Orifice
Note that although a dry test meter is not normally used as a prainary
standard for calibrating a flow meter, if your train meter has been
calibrated to within 1%, it may be used as a laboratory calibration.
Procedure
1. Level manometer by leveling meter box.
2. Zero orifice leg of the manometer with the manometer bypass
valves "out".
3. Turn on pump.
4. Turn manometer bypass valves "in".
5. Adjust the coarse and fine valves to get a reading of
0.5 inches of water on the manometer leg (AH) .
6. Start stopwatch at same time you read dry test meter
volume (V"i). Let dry test meter rotate at least one
full revolution. (The longer the time, the better
your accuracy.)
7. Stop watch and read the dry test meter (V2) simultaneously.
8. Head t^ and t2.
9. Record AH, V±, V2, 8, tp and t2.
10. Repeat steps 6 through 9, but adjust for new AH of 1.0,
2.0, and 6.0 inches of water on the manometer leg. (With
some units, it may be possible to reach only 3 or 4 inches
of water.)
130
-------
Data
pag.> 8
Table I. Orifice Calibration Data Sheet
Mater Box No.
Am
in. H20
0.5
1,0
2.0
6.0
Vl
cf
V2
cf
e
min
op
-2
op
i
V2 - \,
cf
cfm
. i
s.
Calculations
I. Calculate Qm as follows
V2- V,
460
IiJJl+460
2
2. Calculate ^ for each Am as follows:
3.
^ = MCair> s 29
V = t2 + 450
Calculate the average K,n as follows;
131
-------
Page 9
Using the Average Km, compute AH@ * from the following:
Om = Km
\
Tm AH@
Pm Mm
Where Qm = 0.75 ft.3/min
Tm = 700F = 53QQR
Pm = 29.92 in Hg
Mm = 29.0 Ibs/lb.mole
Km = experimentally determined
Use Qm = 0.75 ft.3/min if this flow is within the measured
range of the orifice. (For example, avg. experimental
Qm - 0.75.) This is a typical EPA type orifice.
If the average value of Qm is significantly different from
0.75 ft.Vmin, then assume a new standard flow rate for
calibration purposes.
For example, a large orifice may have a AH@x of 1.5 inches
of water at 4 scfm dry air.
AH@ is a symbol that identifies the orifice flow
characteristics. It is "defined" as the pressure drop
in inches of H2O across an orifice at standard conditio"
(700p, 1 Atmos.) dry air flowing through at the rate
of 0.75 cfm.
Should any other flow rate or conditions be employed,
this should be clearly indicated.
132
-------
ALTERNATE CALIBRATION PROCEDURES
DRY GAS METER AND ORIFICE METER
Connect the components as shown in Figure 20. The wet tes : meter is a
Ucubic-foot-per-revolution meter with ± 1 percent accuracy. Rui the pump for
about 15 minutes with the orifice manometer set at about 0. 5 inch of water to allow
the pump to warm up and to permit the interior surface of the wet test meter to be
wetted. Then gather the information as requested on the data she :t in Figure 9.
Calculate y, the ratio of accuracy of the wet test meter to the dry test meter, and
AH|. If an average y of 1'. 0^0. 01 is not obtained, the dry gas m. ter should be
Date.
Barometric pressure,
in. Hg
Box No
Dry gas meter Ho.
Orifice
manometer
setting,
AH,
in. H20
0.5
1.0
2.0
4.0
6.0
8.0
Gas volume
wet test
meter
vw>
ft3
5
5
10
10
10
10
Gas volume
dry gas
meter
Vd>
ft3
Temperature
Wet test
Meter
tw,
°F
Dry gas meter
Inlet
trfi.
°F
Outlet
tdo'
°F
Average
*d«
op
Time
0,
in in
Average
Calculations
AH
0.5
1.0
2.0
4.0
6.0
8.0
&n
13.6
0.036S
0.0737
0.147
0.294
0.431
0.588
Y
Vw Pfa (fcd + 46°)
i; fa. a. AH \ A. . AC.n\
Hpb + 13.6/ Vtw 45tV
AH@
0.0317 AH Rtw + 460) e"]2
PK (trl + 460) Vu,
^^^ | - .- --- _ .. - -ir T, _. ^..^ ___ «.
,.,.,.!... ,. ,_
- - -- -
Y = Ratio of accuracy of wet test meter to dry test meter. Tolerai ce - ± 0.01
= Orifice pressure differential that gives 0.75 cfra of ftir at 70' F and 29.92
inches of mercury, in. H20. Tolerance - ± 0.15
Figure 9 Suggested orifice and dry gas meter calibration and calculation orm.
133
-------
0
U-TUBE MANOMETER
GLASS TUBE THERMOMETER
s
1
/N,
UMBILICAL
,^tr-
WET TEST METER
METER BOX
Figure 20. Calibration setup.
adjusted until y meets the specification. This can be accomplished by removing
the plate on top of the gas meter and adjusting the linkages.
134
-------
Pacje 10
High Volume Samplers/Variable Orifice Calibrations
The IFRC is designed for isokinetic sampling trains based on
the original EPA design specifications which limits maximum
sampling rates to aboutl.Scfm. However, certain experimental
conditions (extremely low grain loading, very irregular
operation, etc) may require taking a very large sample or
sampling for only a very short period of time. High volume
samplers have been developed to fill this void. If the
sampler has the basic EPA design but simply has larger pumps,
impingers, and other components, the CSI calculator may be
used directly for isokinetic calculations.
Because of the increased flow rates, these high volume trains
are frequently provided with a series of orifices to monitor
the flow rate out of the gas meter. The only change needed
to use the CSI calculator for any orifice is to obtain a
correct &H@ for each orifice.
1. Obtain £H@ for each orifice. If AH@ is unknown, proceed
to page 6 ; if ^@ is determined at a flowrate other
than 0.75scfm, proceed to #3; if AH@ is determined at
0.75 scfm, proceed to #2.
2. input AH@ ih normal:fashion as described in the condensed
instructions on the calculator body.
3 If AH@ has been obtained at any other flowrate
(flow =^0.75scfm), obtain a new AH@ by using the
following equation.
AH<§' = AH
.75
Where H@x = orifice differential pressure (inches H2O) at
x scfm using dry air
Using this new AH@' proceed as before with normal operation
of the calculator.
If an extremely high flow rate is used and A Hj' does not fall
on the printed scales, the calculator may still be used.
For example, say AH@' is found to be 0 05 which do es not lie
on theAH@ scale, but should lie somewhere to the right of
U, WS-TO.^ are sfts?
o.
2 inches to the left of reference arrow Use this as the
new reference B point and proceed as before. "8J"f the
new reference point automatically takes out the factor of
10 that was introduced when we arbitrarily shifted the
decimal point in the AH@ , but allows the "jjority of
the computations to be performed in the center of the
calculator body.
135
-------
Page 11
EXAMPLES
Example 1:
AH@ is calibrated for a large orifice as 2.0" HO at
3cfm dry air @ STP 70°F, _ z
AH@ = 2
= 0.125
3
Use 0.125 for AH@ when testing with this orifice.
Example 2;
is calibrated for a very large orifice as 1.89" H2° a
6cfm
.89 LiZL- = 0.029
Set 0.29 for AH@ (note decimal shift) but also use the stna
mark to the left of arrow B as reference B.
136
-------
Page 12
VARIABLE MOLECULAR WEIGHTS
The CSI isokinetic flowrate calculator is designed foj use
in systems where Md (the dry molecular weight of the < as)
is approximately 29. Situations may arise, primarily'in a
process stream, where Md will be considerably differei t
from 29. Two corrections will be necessary to adjust this
calculation for use in gases differing significantly from
29 g/mole.
The orifice calibration coefficient (AH@) is normally computed
£or a flow rate of dry air at 0.75 scfm (70°F and 760 mm Hg).
Sampling in a gas stream with AMd ± 29 will, of course, change
this calibration.
The simplest way to determine the new AH@ is to repeat the
calibration as given in section "Calibrating Orifice >eters"
using the actual stack gas as the source and using the correct
Md in equation on page 9. In lieu of this experiment, a new
AH@ may be approximated from the following expression,
A. AH@ (Md = x) = AH@ (Md = 29) i*\
This new AH@ (Md = x) may be used directly in the calculator.
B. A second correction to account for the difference
in Md in the basic isokinetic equations is made through the %H?O
scale. Using Figure Msl find the actual %H2O in the system, then
move up to the curve representing the actual Md. Finally, locate
a new %H2O (Md = x) on the ordinate. Use this new %H2D in the
normal fashion in the calculator.
The isokinetic sampling may now proceed in the normal Eashion.
Example: Md = 20 g/mole AH@ (Md = 29) = 2.0 %H2O = 15%
From equation A, AH@ (M, = 20) = 2.0/i£]= 1.34
a l29/
From Figure Msl, Actual %H2O = 15%
At Md = 20, the "New" % H20 = 17.5%
Continue with computation as in previous examples. ShoaldAH@ (Md = x)
lot fall on scale, see section on variable orifices f
-------
Page 13
Correction Factors
for Variable M
Figure Ml
dry molecular weight
20 25
ACTUAL % H2O
1 1Q
-------
APPENDIX B
Derivation of H2S04 ppm
Calculation Equation
139
-------
Acid/Base Titration
1. Calculate the number of moles titrated
, NV x io"3
moles = - j> -
where N = Normality of base
V = Volume of base used (ml)
2. The number of moles in the original sample are
jQ.
A (2)
moles = (NV x IP"3) 50
= (2.5 x 10"2) NV.
A
where A = aliquot taken from 50 ml sample
3. Volume of acid at 21°C(70°F) and 1 atm (29.92 in Hg)
PV = nRT
u 2.5 x IP" NVRT
V = A RT
>
-
V + = 6.03 x 10
H A
where P = pressure (1 atm)
R = gas constant (0.08205 atm liters/°K -mole)
T = temperature (293°K)
4. Volume of gas sampled at STP (liters):
VG - VS 28.32 (tf^)(^)
= vs501-7
where VQ = liters of gas at 21 °C (70°F) and 1 atm (29.92)
Vs = gas meter volume (cu. ft.)
140
-------
t = dry test meter temperature (°F)
Pjn = meter pressure (in. Hg.)
5- ppm H2S04 (Vol. /Vol.):
v *. Y in+6
ppm H?SO, = V x IJ
v
VG
6- For the sulfate tltration an additional factor of ten is added ;o
equation 5 to correct for the extra dilution due to the ion exc lange
column. Also molarity of the Ba(C104) solution is used in plac-j of
the normality of the NaOH solution. For the sulfate titration:
ppm H2S04 (Vol/Vol) = 12,019 (fi)
141
-------
TITLE AND SUBTITLE
Procedures for Aerosol Sizing and H2SO4 Vapor
Measurement at Shawnee Test Facility
AUTHOR(S)
R.F.Maddalone, A.Grant, D.Luciano, and C. Zee
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/7-79-152
2.
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
July 1979
6. PERFORMING ORGANIZATION CODE
8. PERFORMING
. PERFORMING ORGANIZATION NAME AND ADDRESS
TRW Defense and Space Systems Group
One Space Park
Redondo Beach, California 90278
10. PROGRAM ELEMENT NO.
INE624
11. CONTRACT/GRANT NO.
68-02-2165, Task 202
2. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE
Task
OF REPORT AND PEBIOP C<
Final; 6/76^_2/77
OVER
14. SPONSORING AGENCY CODE
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711 EPA/600/13
5. SUPPLEMENTARY NOTES EPA project officer R.M. Statnick is no longer with IERL-RTP;
for details, contact F.E. Briden, Mail Drop 62, 919/541-2557.
16. ABSTRACT
The report describes a series of procedures for sizing dry aerosols and
measuring H2SO4 entering and leaving the Shawnee flue gas desulfurization (FGD)
prototype units. A Brink impactor was used to size dry particulate matter entering
the FGD process. A manual system for the FGD process effluent was chosen on the
basis of a literature survey, contacts with experts in the field, and an evaluation o
available information. Chosen for the inlet was an FGD Meteorology Research ^nc>,
cascade impactor. Finally, a method for H2SO4 vapor was developed which is base
on the controlled condensation (Goksoyr/Ross) method. In addition to these proce-
dures , a QA program was designed to ensure the overall quality of the data taken
in the above procedures.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Pollution
Aerosols
Size Determination
Sulfuric Acid
Sulfur Trioxide
Vapors
Desulfurization
Dust
Impactors
Condensing
Quality Assurance
13. DISTRIBUTION STATEMENT
Release to Public
b. IDENTIFIERS/OPEN ENDED TERMS
Pollution Control
Stationary Sources
Particulate
Brink Impactor
Goksoyr/Ross Method
19. SECURITY CLASS (ThisReport)
Unclassified
20. SECURITY CLASS (Thispage)
Unclassified
COSATI
13B
07D
14B
07B
Field/Group^.
OTA
11G
131
EPA Form 2220-1 (9-73)
142
------- |