United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
May 1981
Air
Evaluation of Sampling
And Analysis Procedures
For the Secondary
Aluminum Industry
-------
EMB8OSAL1
EVALUATION OF STACK SAMPLING TECHNIQUES
FOR SECONDARY ALUMINUM INDUSTRY
Contract No. 68-02-2815
Work Assignments 44 and 46
Prepared for
U.S. Environmental Protection Agency
Emissions"Measurement Branch
Research Triangle Park, North Carolina
Originally Submitted:
January 1981
Revision Submitted:
September 1981
9103.00/41
Submitted by
Engineering-Science
7903 Westpark Drive
McLean, Virginia 22102
-------
PREFACE
Development of reliable sampling techniques is an iterative process.
Results of the initial field effort provided a basis for investigation of
alternate approaches when certain aspects of the first generation tech-
niques were judged to be inadequate. A separate Chapter 7 has been added
to incorporate appropriate revisions to the methods development program
while at the same time maintaining the time perspective of Chapters 1
through 6.
-------
TABLE OF CONTENTS
CHAPTER 1 INTRODUCTION
CHAPTER 2 DESCRIPTION OF TEST LOCATIONS
Chlorine Scrubber
Charging Well Coated Baghouse
DESCRIPTION OF SAMPLE COLLECTION AND ANALYSIS
PROCEDURES
CHAPTER 3
CHAPTER 4
CHAPTER 5
CHAPTER 6
CHAPTER 7
Chlorine Compounds
Hydrocarbon Compounds
TABULAR SUMMARY OF RESULTS
Chlorine and Chlorides (Scrubber Tests)
Hydrocarbons and Particu'lates (Baghouse Tests)
CRITICAL DISCUSSION OF EXPERIMENTAL RESULTS
Analysis of Free Chlorine
Discussion of Relationship Between Chlorine
and Chloride
Chloride Titrations
Particulate Chlorides
Hydrocarbons Train Front Half
Condensable Hydrocarbons
Noncondensable Hydrocarbons
RECOMMENDATIONS FOR EVALUATION PROCEDURES
Chlorine Compounds - Sample Collection
Chlorine Compound Analyses
Total Chlorides in Solution
Clean Up and Recovery of Back Half Hydrocarbons
REVISION OF SAMPLING STRATEGY
Chlorine Compound Analyses
Particulate and Condensible Organic Matter
1-1
2-1
2-1
2-1
3-1
3-1
3-5
4-1
4-1
4-8
5-1
5-1
5-4
5-5
5-5
5-6
5-7
5-7
6-1
6-1
6-3
6-3
6-4
7-1
7-1
7-4
11
-------
APPENDIX A CHLORINE PROCEDURES
APPENDIX B CHLORIDE PROCEDURES
APPENDIX C METHOD 5A - PARTICULATE AND CONDENSABLE HYDROCARBONS
PROCEDURE
APPENDIX D CHLORINE TEST FIELD DATA SHEETS
APPENDIX E CHLORIDES TITRATION LAB SHEETS
APPENDIX F HYDROCARBONS FIELD DATA SHEETS AND STRIP CHARTS
APPENDIX G HYDROCARBONS LAB DATA
-------
LIST OF FIGURES
2.1 Line Diagram of Chloride Scrubber 2-2
2.2 Schematic (End View) of Coated Baghouse 2-4
3.1 Particulate/Chlorine Compound Sampling Train 3-2
3.2 Particulate and Condensable Hydrocarbons
Sampling Train 3-6
4.1 Non-Condensable Hydrocarbon Peak 4-11
4.2 Impinger Collection Efficiency Experiment 4-16
5.1 Daily Standard Curves Cl Concentration vs.
Absorbance 5-2
6.1 Chlorine Compound Sampling Train 6-2
6.2 Particulate and Condensable Hydrocarbons
Sampling Train 6-5
-------
LIST OF TABLES
4.1 Audit Sample Chlorine Concentrations (mg/1) 4-2
4.2 Audit Sample Chloride Concentrations (mg/1) 4-3
4.3 Summary of Chlorine Compound Train Fractions 4-4
4.4 Tabular Summary of Chlorine and Chloride
Analyses Impinger Trains Only 4-5
4.5 Staged Collection Efficiency of Impingers 4-6
4.6 Chloride Train Fractions as % of Total 4-7
4.7 Chlorine Compound Emission Data 4-9
4.8 Condensable Organics Back Half Data 4-1Q
4.9 Baghouse Total Hydrocarbon Emissions Summary:
Corrected Overall and Periodic Averages of
Test 1 THCA Data 4-12
4.10 Baghouse Total Hydrocarbon Emissions Summary:
Corrected Overall and Periodic Averages of
Test 2 THCA Data 4-13
4.11 Total Hydrocarbon Concentration Ranges and
Peaks for Field Tests on 12/17/80 4-14
4.12 -Hydrocarbon Penetration Through Impingers 4-15
4.13 Hydrocarbon Compound Emission Data 4-17
5.1 Regression Constants for Chlorine Calibration
Curves 5-3
7.1 Initial Evaluation of Arsenite Procedure 7-2
7.2 Summary of Audit Sample Analyses by All Three
Methods 7-2
7.3 Estimated Impinger Concentrations vs. Stack
Gas Concentration 7-3
7.4 Vapor Pressures of AlCl^ at Various Temperatures 7-5
7.5 Results of WYTEX Nylon Brush Tests in Methylene
Chloride 7-7
-------
CHAPTER 1
INTRODUCTION
The U.S. Environmental Protection Agency may propose new source per-
formance standards for the secondary aluminum industry. Compliance with
appropriate emission limiting standards can only be evaluated using refer-
ence air sampling techniques. In association with Contract 68-02-2815,
Work Assignments 44 and 46 directed Engineering-Science (ES) to develop
reliable stack sampling trains necessary to collect accurate samples from
certain secondary aluminum process operations.
EPA work assignment requesters identified chlorine compounds and con-
densable hydrocarbons as the contaminants of particular interest needing
investigation. ES and one of its related companies, Harmon Engineering
and Testing, reviewed established sample collection and analytical tech-
niques for these contaminants. Specific sample trains and methods of
analysis were selected for field evaluation.
During the period December 15 to December 19, ES secured stack emis-
sion samples from processes at a secondary aluminum smelter. The rever-
beratory furnace chlorination scrubber was sampled for total particulates,
particulate chlorides, chlorine gas, and gas phase hydrochloric acid, with
all samples collected at the scrubber outlet. The outlet of the lime coated
baghouse controlling emissions from the furnace scrap charging wells was
sampled for particulates and condensable and noncondensable hydrocarbons.
Some of the sample analyses were completed in the field, whereas the re-
mainder were completed in the McLean and Arcadia laboratories.
This report includes a summary of the collected data and an evalua-
tion of the two sample trains. Operating problems with the sample trains
and the associated analytical techniques are discussed. Final recommen-
dations for train assembly and sample analysis are provided.
1-1
-------
CHAPTER 2
DESCRIPTION OF TEST LOCATIONS
The methods development test program was centered around operation
of the reverberatory furnaces. A hydrocarbons train collected samples
from the outlet of equipment used to abate particulates and hydrocarbons
in off gases from the furnace charging wells. Particulate and gaseous
chlorine compounds contained in furnace chlorination effluent gases were
characterized by collecting samples at the outlet of the process scrubber.
Each site is discussed in more detail in the remainder of this chapter.
CHLORINE SCRUBBER
At the end of each batch aluminum alloy formulation, excess magnesium
is removed by sparging chlorine gas through the melt. The effluent gas
proceeds to emissions control equipment for abatement of aluminum chloride
particulates; its partial hydrolysis product, hydrochloric acid; and unre-
acted chlorine.
A schematic of the scrubbing operation used for emissions abatement
is provided in Figure 2.1. Test ports located between the scrubber and
the fan were used to collect all samples. This location was characterized
by a high negative static pressure and a very high loading of scrubbing
liquor droplets which penetrated the mist eliminator. In fact, the drop-
let concentration was so high that sample train cyclones were used to pre-
vent blinding the filter. The high negative pressure required turning
the sample pump on prior to insertion of the probe in the stack to pre-
vent pulling the sample filter from its holder.
Simultaneous collection of two samples was accomplished by placing
a sample train in each of the two ports. The nozzles were located as
close together as possible at the approximate center of the stack.
Chlorination periods varied, depending upon furnace size and magne-
sium concentration. Emissions tended to increase towards the end of
chlorination, and where possible, stack gas samples were collected dur-
ing the expected maximum emission period.
CHARGING WELL COATED BAGHOUSE
Various types of aluminum alloy scrap, pure alloy components such
as silicon, and flux compounds are fed to an open well at one end of each
2-1
-------
Line Diagram of Chloride Scrubber
ALTERNATE
SAMPLE
PORTS; NO
PLATFORM
EXHAUST
GAS FROM
CHLORINE
LANCING
BUILDING ROOF LEVEL
VENTURI SCRUBBER
SAMPLE PLATFORM
(not to scale)
-------
reverberatory furnace. The exact mixture charged for each furnace run
depends upon the specific desired alloy formulation and characteristics
of available scrap. Exhaust hoods capture volatilized organics and acid
compounds, and the resulting mixture of contaminants and room air pro-
ceeds to a pre-coat type fabric filter collector.
A schematic of the baghouse used to control charging well emissions
is shown in Figure 2.2. Exhaust gas is split by the two fans and dis-
charged through two separate stacks. All samples were collected from
the north stack for the first phase test program.
2-3
-------
FIGURE 2.2
Schematic [end view] of Coated Baghouse
EFFLUENT
GAS
3" SAMPLE PORTS-
SAMPLE PLATFORM -»
COLLECTED
GAS FROM
CHARGING
WELLS
—>C
COLLECTED A f
MATERIAL \ IL
s /
\ /
W
A
/ v
EFFLUENT
GAS '
3' ID DUCT
T
tH-
3" SAMPLE PORTS
ir\
XDUAL
2-4
ENGINEERING-SCIENCE
-------
CHAPTER 3
DESCRIPTION OF SAMPLE COLLECTION
AND ANALYSIS PROCEDURES
The specific air contaminants of interest for the secondary aluminum
industry are gaseous and particulate chlorine compounds and condensable
hydrocarbons. After a review of expected source emission characteristics
and the available characterization methods, specific sample trains, and
analytical methods were selected. Details for each of the two types of
sample trains are discussed separately below. Copies of all reference
procedures are contained in the appendices.
CHLORINE COMPOUNDS
Sample Collection
Exhaust gases from the chlorine lancing operation were expected to
contain particulate chlorides, chlorine gas, and possibly gas phase hy-
drochloric acid. Separate characterization for each type of compound
was desired.
A standard Method 5 sample train was assembled for this purpose,
the only modification involving use of 0.1 normal sodium hydroxide solu-
tion instead of water in the impinger. Because of the uncertain absorp-
tion and reaction characteristics of low concentration of chlorine gas,
three impingers in series were charged with the caustic solution. Tissue
quartz filters were used because of their superior chemical compatibility
with acidic gases as compared to standard glass fiber media. A schematic
of the assembled train is shown in Figure 3.1.
During each chlorination period, simultaneous single point isokine-
tic samples were collected. Sample train operation, including filter
temperature control, was identical to that specified in EPA Method 5.
Inclusion of cyclones upstream of the filters was necessary to remove
entrained scrubbing liquid droplets.
Normal sample train operation according to EPA Method 5 results in a
sample collection rate of approximately 0.75 cfm, meter conditions. The
effluent particles from the chlorine scrubber were present at a moderate
grain loading, but because of their apparent very fine size, the collected
filter cake rapidly produced a high pressure drop. Smaller sample nozzles
were used to effect a lower sample collection rate to solve this problem.
3-1
-------
Particulate/Chlorine Compound Sampling Train
NOTES:
IMPINGERS WILL BE CHARGED WITH 0.1N NaOH SOLUTION
QUARTZ FILTER FOR COLLECT-
ION OF PARTICULATE CHLORIDES
THERMOCOUPLE
THERMOCOUPLE
PROHE
I11 TO I TUBE
STACK WALL
PI TO! MANOMETER
HEAIEI) AREA
THERMOMETERS
ORIFICE
DRY HAS MEIER
IMPINGER TRAIN
FOR ABSORPTION
OF HC1 AND C\2
THERMOCOUPLE
CHECK VALVE
IMP1NHER
ICE BATH
VACUUM LINE
VACUUM GUAfiE
MAIN VALVE
BY-PASS VALVE
AIR-TIGHT PUMP
-------
At the conclusion of each test run, the following separate train
fractions were recovered: probe wash, filter, cyclone, first impinger,
second impinger, and third impinger. Chlorine analyses were performed
immediately after sample collection, whereas the other components were
characterized after samples were returned to the laboratory.
Analytical Methods
Specific characterization methods for particulates and gaseous com-
pounds are discussed in separate sections in the remainder of this sec-
tion. With respect to chlorine compounds, free chlorine was determined
first, followed by analysis of total chlorides.
Particulate Chlorides
A standard gravimetric analysis was completed for all filters. Af-
ter recording the final filter weight, particulate chlorides were deter-
mined by extraction of the filter catch with distilled water followed by
the mercuric titration procedure. (This procedure is discussed in detail
in the subsection for chlorides.)
Additional characterization for aluminum chloride particulate was
considered appropriate. It sublimes completely at 29.92" Hg at 352°F,
and was, therefore, expected to have a substantial vapor pressure at the
Method 5 filtration temperature of 250°F.
Considering the current results, extracting particulates from a stack
environment of approximately 70 CF and 27.4" Hg absolute pressure and heat-
ing then to 250°F and reducing the absolute pressure even further was ex-
pected to result in a phase change of a portion of the aluminum chloride.
Thus, each impinger was analyzed for the presence of aluminum.
Chlorine Gas
In sodium hydroxide solution, chlorine gas reacts as follows:
Clj (gas) "T C12 (aqueous)
C12 (aqueous) + H2O IZIT HC1 + HOCl
HCl H+ + Cl~
HOCl H+ + OC1~
Thus, analysis of the caustic impinger catch for chlorine actually amounts
to analysis for hypochlorous acid or hypochlorite ion, depending upon ac-
tual solution pH.
A colorimetric procedure was selected for analysis of free available
chlorine, defined as the sum of aqueous chlorine, hypochlorous acid, and
hypochlorite ion. The actual method used was 409-D, Stabilized Neutral
Orthotolidene (SNORT), as described in "Standard Methods for Examination
of Water and Wastewater". The chlorine-induced color (turquoise) was quan-
tified with a Bausch and Lomb Spectronic 20 single beam spectrophotometer
at 625 nanometers.
3-3
-------
Prior to field data collection/ several preliminary laboratory pro-
cedures were necessary. Chlorine demand-free water was prepared for use
in all dilutions. Sodium thiosulfate reagent was standardized after the
recommended two-week waiting period.
Fundamental to this procedure is the standardization of the stock
sodium thiosulfate reagent for its subsequent use in preparing a stock
solution of free available chlorine. Standard Methods, Section 409-A
describes the titration employed to determine the above mentioned normal-
ity. The alternative Dichromate method was used over the primary Binio-
date method because of the stability of the stock potassium dichromate
reagent used in the former. A sharp, clear end point was reached with
diluted sodium thiosulfate (0.025N). A diluted commercial solution of
sodium hypochlorite was used to prepare the stock chlorine solution (ap-
proxijmately 50 mg/1). Sodium thiosulfate was then used to determine the
true chlorine concentration of this stock solution. A range of concen-
trations was then made per appropriate dilutions.-
The various dilutions, prepared asynchronously, provided the means to
develop a standard curve. Each dilution was treated and colorimetrically
analyzed as described in the SNORT method. The method requires dilutions
to be in the range of 0.01 mg/1 and 6.0 mg/1 chlorine as defined by detect-
ability and orthotolidine availability, respectively. The same limit ap-
plies to sample analysis.
Upon the addition of the orthotolidine and the buffer stabilizer, the
diluted standard samples developed a turquoise color and were read on the
spectrophotometer at 625 nm. Optical densities (as absorbance or transmit-
tance) were recorded and plotted on a graph. The graph described a line
where an increase in concentration results in an increased absorbance.
Standard curves were prepared daily in concurrence with sample collection
(or audit sample evaluation). Linear regression was used to calculate con-
centration as a function of absorbance.
Impinger samples were analyzed with the SNORT method immediately fol-
lowing their collection in the field. Dilutions were made to a concentra-
tion that indicated 20 to 80% of scale on the spectrophotometer. Where
this was not achieved, the extension of the standard curve was necessary
to locate the corresponding concentration point. As recommended, the
photometer was read within 2 minutes of reagent and sample mixing because
of rapid chlorine degradation to chloride. Interferences were compensated
for by making a blank with sodium arsenite (reduces available chlorine)
and using it as a zero absorbance reference. Due to some observed cloud-
iness, it was evident that this was necessary. Generally, two dilution
aliquots from each sample were analyzed and the average absorbance was
used to identify the chlorine concentration.
Total Chlorides
Two titration procedures were selected for analysis of total chloride
ion. One method followed the EPA procedure for analysis of hydrochloric
acid emissions (mercuric nitrate titration) and the other one was selec-
ted from "Standard Methods", 408A, Argentometric Method (silver nitrate
titration). Both methods were evaluated because of reported difficulties
3-4
-------
with pH control for the mercury titration and with end point detection
for the silver titration.
As discussed in the reference mercury titration procedure, some sam-
ple aliquots were passed through a cation exchanger to investigate the
presence of metal ion interferences.
Total chlorides analysis included contribution by chlorine gas, hy-
drochloric acid, and particulate chlorides. Chlorine poses a problem for
chloride analysis because upon initial absorption it forms an equi-molar
mixture of chloride and hypochlorite ions. Hypochlorite gradually decom-
poses to chloride, so in order to eliminate questions concerning the per-
centage conversion, all field samples were treated to convert hypochlorite
to chloride. This was accomplished through the use of a 3% hydrogen perox-
ide solution added in excess of the molar concentration required to destroy
the chlorine, as measured shortly after sample collection and recovery.
The hydrogen peroxide reacts with hypochlorite in a basic solution
according to the equation:
OH~
0 Cl" + H202 02 + Cl~ + H20
The use of excess hydrogen peroxide has the added advantage of de-
stroying any reducing ions which may be present in the sample and which
interfere in the mercuric nitrate titration procedure.
Thus, the total chloride concentration represents the following frac-
tions:
Cl"
total = HC1 + 2 Cl- + 3 Al Cl,, expressed as mass equivalents
HYDROCARBON COMPOUNDS
Sample Collection
Secondary aluminum smelting operations involve recycling of numerous
types of aluminum scrap. Any cutting or lubricating oils, plastic coatings,
or plastic components are evolved from furnace charging wells, chip dryers,
and sweat furnaces. As a first cut type of analysis, separation of these
hydrocarbons into particulate (at 250°F) and condensable (liquid phase be-
tween 250°F and 70°F) fractions are desired. Condensable hydrocarbons are
of interest because of their formation of droplets when stack gas cools
while mixing with ambient air. The noncondensable sample train fraction
is unimportant in this case.
A standard Method 5 sample train was selected for hydrocarbons char-
acterization. As indicated in Figure 3.2, a portion of the gas stream was
removed from the U-tube connecting the empty third impinger and the silica
gel impinger, and it was pumped to a flame ionization detector (FID) based
total hydrocarbon analyzer (THCA). This fraction of hydrocarbons, defined
as "noncondensable," includes any water insoluble organics with a vapor
pressure greater than a few ppm by volume at the actual impinger tempera-
ture.
3-5
-------
Partlculate and Condensable Hydrocarbons Sampling Train
THERMOCOUPLE
IMPINGER TRAIN
SLIP STREAM @ 6 SCFIT
INTERNAL PUMP OF THCA
_N£T > TO FID
THERMOCOUPLE
CHECK VALVE
THERMOCOUPLE
PROBE
PHOT TUBE
SIAC.K WALL
PI TOT MANOMETER
HEATED ARF.A
THERMOMETERS
OKIPICE
DRY GAS METER
FILTER HOLDER W/
—QUARTZ FILTER
IMPINGER
ICE BATH
VACUUM LINE
VACUUM GIIARE
MAIN VALVE
BY-PASS VALVE
AIR-TIGHT PUMP
-------
The lime coated baghouse which abates furnace charging well emissions
was selected as the source for evaluation of the hydrocarbons train. At
least one and sometimes two furnaces were being charged during sample col-
lection periods.
Removal of the slipstream to the FID caused a difference between gas
volume entering the sample nozzle and volume registered by the dry gas meter,
so a correction factor to the desired metering system orifice pressure drop
was required to maintain isokinetic sampling at the nozzle. The particular
hydrocarbon analyzer used had an internal pump which removed approximately
6 scfh from the Method 5 gas sample. The H correction factor was estimated
by noting the actual meter box H when the sampling rate was approximately
0.75 cfm and when it was approximatley 0.65 cfm. The difference was 0.5"
w.c., such that nomograph values were decreased by this amount when setting
individual traverse point sampling rates.
The magnitude of the H correction factor was high enough for some
sample points that it represented 1/2 of the desired H. For example, a
certain sample point velocity head was such that the orifice pressure drop
was 1" w.c., or 0.5" w.c. with the correction factor deducted. A sample
nozzle larger than normal was used to lower the % reduction due to the FID
slipstream.
Prior to charging the train, all glassware was cleaned with chromic
acid solution, as specified in Method 5A. Sample train operation for sam-
ple collection was identical to that described in the reference procedure.
At the conclusion of each run, the following separate fractions were
recovered: probe wash, filter, back half rinse including filter holder,
and combined impingers. The probe wash and initial back half rinse were
completed with acetone, and a second back half rinse was performed with
trichloroethylene. U-tube connections were wiped with trichloroethylene
prior to back half recovery to control collection of silicone grease.
Analytical Methods
Total Particulates
A standard gravimetric analysis was completed for the sample train
quartz filters and acetone probe wash.
Condensable Organics
Dual back half rinses were performed because of the unknown recovery
characteristics of acetone. A standard gravimetric analysis was completed
for the acetone and 1,1,1 trichloroethane back half rinses.
The chloroform-ether extraction procedure described in Method 5A was
followed for the combined impinger catch. For each run, the initial solu-
tion pH was approximately 2, such that neutralization with caustic was
necessary prior to the initial extraction. Because some organics are more
completely extracted at low pH that at neutral pH, a second stage extrac-
tion was completed after readjusting to a pH of 1 with hydrochloric acid.
A standard gravimetric analysis was completed for all chloroform-ether ex-
tracts.
3-7
-------
Noncondensable Organics
Hydrocarbons that were not successfully collected in the impinger
train were determined by diverting part of the "back half" Method 5 gas
stream to a flame ionization detector (FID) in a total hydrocarbon anal-
yzer (THCA).
A Scott Model 215 THCA was utilized in the field. It was located
and operated on the ground beneath the sampling ports. Data were re-
corded on a Linear Model 555 flat bed recorder.
The THCA was zeroed with zero air (THC <0.1 ppm) and calibrated with
methane span gases. Span gas calibrations should be made in a concentra-
tion range that will approximate test emissions. Therefore, methane span
gases of 31.6 ppm and 81.7 ppm were used in this test anticipating low
hydrocarbon loadings.
Unfortunately, span gas calibration gases, even at full range adjust-
ments, failed to respond to their true value. Methane span gases of 31.6
ppm and 81.7 ppm corresponded to 27.1 and 70.3% chart responses on a 0-100
range. The most probable cause of this anomaly was operating below the re-
commended ambient temperature limit of the instrument. Scott THCA Model
215 operating manual specifications indicate an operation range between
32°F and 130°F. Field test conditions remained around 25°F. Later, at
the ES McLean office, under room conditions, THCA span gas calibrations
yielded accurate linear responses.
The low span gas values in the field did not discount THC concentra-
tion responses, however, even though the actual concentration was not in-
dicated, the span gases retained linearity. Thus, a simple correction was
made for true hydrocarbon emissions by multiplying observed responses, by
the ratio of the true span value and the observed span response (e.g.
31.6 ppm
27. 1 ppm CH4
Prior to sample collection, a leak check was conducted through the
teflon line and analyzer by capping off the sample inlet adjacent to the
U-tube assembly. With the pump on, the bypass flow meter dropped down
to zero, indicating a leak tight sample line and instrument.
To insure simultaneous sampling with the Method 5 run, it was neces-
sary to turn off the pump while connecting the sample line to the U-tube
assembly. Consequently, earlier leak test procedures excluded the sample
line/U-tube assembly connection. Problems with simultaneous leak checks
(i.e., meter box pump override of the THCA pump) justified this particu-
lar procedure.
The THCA and Method 5 train started sampling simultaneously. The
flow rate of the gas stream diverted to the THCA was controlled by an-in-
ternal sample pressure regulator (i.e., only accessible by removing the
top panel) and monitored by a bypass flow meter. The bypass flow typi-
cally registered 5 to 7 scfh, rotameter scale. From this gas stream,
approximately 6 ml/min was directed to the FID.
3-8
-------
Sample pressures of 1.0 psi and bypass flow rates of 6 scfh approxi-
mated starting conditions. However, particulate loading on the filter in
the Method 5 sample box produced a proportional vacuum. Hence, without
a compensating adjustment to the THCA sample pressure, the actual sample
diversion slowly dropped to about 4 scfh by the termination of the tests.
At the conclusion of each test run, the piece of unheated teflon con-
necting the.THCA to the Method 5 sample box was purged with ambient air to
insure constant background sampling conditions.
3-9
-------
CHAPTER 4
TABULAR SUMMARY OF RESULTS
During the period of December 15 through December 19, a total of six
chlorine compound and two hydrocarbon compound samples were collected us-
ing minor modifications of the EPA Method 5 manual stack sampling train.
The two complete hydrocarbon samples were supplemented with a brief ex-
periment involving impinger collection efficiency. Results of all field
samples are summarized in tabular form in this chapter.
On December 16, EPA representatives delivered quality assurance sam-
ples for chlorine and chlorides. Initial chlorine analyses were completed
on site, and chloride samples were titrated at a later date. Results of
these analyses are also included in this chapter.
Separate tables are presented where possible, but because of the re-
lationship between chlorine and chloride analyses, some combined tables
are noted.
CHLORINE AND CHLORIDES (SCRUBBER TESTS)
Results for EPA audit samples are listed in Tables 4.1 and 4.2. The
EPA reference values for chlorine are not included pending re-analysis of
some of the samples.
For the total of six chlorine compound runs, three repetitions each
with the two simultaneously operating trains, overall results are pro-
vided in Table 4.3. This table includes total filter weight and extract-
able chlorides from the filter as well as separate chlorine and chloride
analyses for the probe wash, cyclone, and first, second, and third impin-
gers. A comparison of impinger catches only is provided in Table 4.4.
Evaluations of individual impinger chlorine collection efficiency
and of particulate vs. gaseous chloride fractionation are provided in
Tables 4.5 and 4.6.
Chemical reactions of molecular chlorine in a basic, aqueous solu-
tion result in formation of both chloride and hypochlorite ions. Sub-
sequent to chlorine (hypochlorite) determination and prior to chloride
determination, hypochlorite must be converted to chloride by an oxida-
tion-reduction reaction. Analyses for chlorides reported in Tables 4.3,
4.4, and 4.6 include the amount formed by conversion of hypochlorite to
chloride.
4-1
-------
TABLE 4.1
AUDIT SAMPLE CHLORINE CONCENTRATIONS (mg/1)
Samole f
2256
2285
3071
3087
4050
12/16
98.5
88.4
188.2
170.4
_..
Date Analyzed
12/18 12/19
136.8
—
287.2
—
292.6
1/7
124.6
116.7
224.4
263.8
366.5
1/9
—
127.0
—
218.4
292.1
EPA
Value
153
153
255
255
357
4-2
-------
TABLE 4.2
AUDIT SAMPLE CHLORIDE CONCENTRATIONS (mg/1)
CO
Sample #
2210
3021
4020
5231
Overall
Concentration
by Mercury
Method
3042
5107
7103
9145
Average Accuracy:
Concentration
by Silver
Method
2902
4793
6770
8727
Mercury
Silver
EPA
Value
3000
5000
7000
9000
Relative
Accuracy by
Mercury
Method, %
+ 1.40
+2.14
+ 1.47
+ 1.61
+1.66%
-3.43%
Relative
Accuracy
Silver
Method,
-3.27
-4.14
-3.29
-3.03
by
%
-------
TABLE 4.3
SUMMARY OF CHLORINE COMPOUND TRAIN FRACTIONS3'15
Filter Probe Cyclone 1st Impinger
Run
2A
2B
3A
3B
4Ae
4B
a
b
c
d
e
mg(Tot)
470.9
193.2
166.7
160.5
512.6
499.0
mgCl" mgCl"
62.0
15.8
11.5
11.2
49.6
55.8
1.0
1.2
0.2
0.3
0.7
0.6
mgCl2 mgCl mgCl0
1.0
1.5
1.1
1.4
0.3
ND
36.3 267.0
19.6 305.4
10.4 21.3
17.3 14.6
15.4 99.8
18.1 101.2
mgCl
312.0
446.0
45.8
29.8
234.9
216.2
2nd
mgCl
45.
169.
1.
0.
5.
7.
Impinger 3rd Impinger Totals'3
2 mgCl mgCl0
0
8 1
7
9
6
0
46.
19.
4.
1 .
9.
7.
9 2.7
5 16.0
0 NDd
5 ND
3 ND
9 ND
mgCl mgCl0 mgCl
2.6 314.7 423.5
8.3 491.2 589.6
3.6 23.0 64.9
0.4 15.5 42.9
0.7 105.4 294.5
0.6 108.2 280.5
All weights in milligrams. Cl~ values include CC1~ converted to Cl~.
Chlorine reported in atomic form, equivalent weight 35.45.
Cyclone and probe wash fractions not included in totals.
ND means less than 0.1 milligram.
The final test run incorporated two separate furnace clorinations.
-------
TABLE 4.4
TABULAR SUMMARY OF CHLORINE AND CHLORIDE ANALYSES
IMPINGER TOTALS ONLY
Test
Run
2A
2B
3A
3B
4A
4B
Total
Milligrams
Chlorine
314.7
491.2
23.0
15.5
105.4
108.2
Total
Milligrams
Chloride
361.5
573.8
53.4
31.7
244.9
224.7
Note: Chlorine reported in atomic form, equivalent
weight 35.45. Cl~ values include OC1~ con-
verted to Cl~.
4-5
-------
TABLE 4.5
STAGED COLLECTION EFFICIENCY OF IMPINGERS
Total % Collected of Total
Chlorine First Second Third
Run (mg) Impinger Impinger Impinger
2A 314.7 84.8 14.3 0.9
2B 491.2 62.2 34.6 3.2
3A 23.0 92.6 7.4 0
3B 15.5 94.2 5.8 0
4A 105.4 . 94.7 5.3 0
4B 108.2 93.5 6.5 0
4-6
-------
TABLE 4.6
CHLORIDE TRAIN FRACTIONS AS % OF TOTAL
Run
2A
2B
3A
3B
4A
4B
Total
Chloride
(mg)
423.5
589.6
64.9
42.9
294.5
280.5
% Contribution
Filter
14.6
2.7
17.7
26.1
16.8
19.9
First
Impinger
. 73.7
75.6
70.6
69.5
79.8
77.1
Second
Impinger
11.1
20.3
6.2
3.5
3.2
2.8
Third
Impinger
0.6
1.4
5.5
0.9
0.2
0.2
Note: Cl~ values include OC1~ converted to Cl~.
4-7
-------
Calculated values for particulate and gaseous loadings in terms of
concentration and mass emission rate are contained in Table 4.7. Values
for hydrochloric acid emissions were not calculated because of uncertainty
in the analysis for aluminum in the impinger catch, as discussed more com-
pletely in Chapter 5.
HYDROCARBONS AND PARTICULARS (BAGHOUSE TESTS)
Back half extraction and rinse data are summarized in Table 4.8. Sam-
ple filter weight gains were less than 0.1 milligram and are not reported.
Acetone probe rinse catches for the'two runs were 3.7 and 22.6 milligrams,
respectively.
Hydrocarbon analyzer data indicated widely varying concentrations.
Typically, there was a certain background concentration, and spikes above
background were observed soon after a batch of scrap was charged. A por-
tion of a strip chart containing one peak is reproduced as Figure 4.1.
Strip chart data reduction was based on a manual calculation of ten-
minute averages. Overall test period concentrations were then calculated
from the ten-minute averages. Results of these calculations are provided
in Tables 4.9 and 4.10. In order to compare average concentrations to cal-
culated versus actual observed values, some typical concentration ranges
and peak values are listed in Table 4.11. Copies of the strip charts are
provided in the appendices.
Completion of the two normal test runs confirmed that some hydro-
carbons were.not collected by the impingers. A brief experiment was de-
vised to evaluate impinger collection efficiency at this relatively low
hydrocarbon concentration. The results are provided in Table 4.12 and
Figure 4.2.
Calculated results for the front hal and back half emissions are
listed in Table 4.13. Tabulated values for the FID "catch" are based on
the average concentrations contained in Tables 4.9 and 4.10.
4-8
-------
TABLE 4.7
CHLORINE COMPOUND EMISSION DATA
Run
Number
Total Particulates
Emission Grain
Rate Loading
(Ib/hr) (gr/dscf)
Chlorine
Chloride Particulates Gas
Emission
Rate
Emission
Rate
(Ib/hr)
Grain
Loading Rate Cone.
(gr/dscf) (Ib/hr) (ppm)
2A
2B
3A
3B
4A
4B
18.0
7.3
8.6
7.7
27.4
25.6
0.238
0.097
0.103
0.092
0.315
0.305
2.4
0.6
0.6
0.5
2.6
2.9
0.031
0.008
0.007
0.006
0.030
0.034
12.1
18.7
1.2
0.7
5.7
5.6
247
383
22
14
101
103
4-9
-------
TABLE 4.8
CONDENSABLE ORGANICS BACK HALF DATA
(ALL WEIGHTS IN MILLIGRAMS)
Run 1 Run 2
Acetone Rinse 14.3 7.3
Trichloroethylene Rinse 4.3 10.4
Neutral Extraction 2.7
Acid Extraction — 1.1
Total 21.3 18.8
4-10
-------
FIGURE 4.1
Non-Condensable Hydrocarbon Peak
4-11
ENGINEERING-SCIENCE
-------
TABLE 4.9
BAGHOUSE TOTAL HYDROCARBON EMISSIONS SUMMARY:
CORRECTED OVERALL AND PERIODIC AVERAGES
OF TEST 1 THCA DATA
Test #1 (12/17/80)
Port A
Overall Port A
Port B
Overall Port B
Overall Test Average
Period
(Test Minutes)
0-10
10-20
20-30
30-39
0-39
0-10
10-20
20-30
30-35
0-35
0-74
Average
(ppm)
41.6
19.9
13.5
22.8
24.5
33.3
16.5
11.7
12.3
16.7
22.0
4-12
-------
TABLE 4.10
BAGHOUSE TOTAL HYDROCARBON EMISSIONS SUMMARY:
CORRECTED OVERALL AND PERIODIC AVERAGES
OF TEST 2 THCA DATA
Test #2
Port B
Period
(Test Minutes)
0-12
12-22
22-32
32-36
Average
(ppm)
84.9
127.0
106.9
69.2
Overall Port B 0-36 101.0
Port A 0-10 49.5
10-20 57.2
20-30 72.5
30-36 . 46.5
Port A Average 0-36 57.5
Overall Test Average 0-72 79'. 2
4-13
-------
TABLE 4.11
TOTAL HYDROCARBON CONCENTRATION RANGES
AND PEAKS FOR FIELD TESTS ON 12/17/80
Time Run
1005 1
1100 1
1501 2
1613 2
Range
Port (ppm)
A 11.2. to >104.4a
B 10.2 to 52.9
B 66,2 to 157.3
A 5.8 to 103.0
31.2,
52.8,
>104.4,a
99.6,
81.5,
Peaks
(ppm)
47.7,
46.0,
82.0,
73.8,
147.0
>104.4a
15.3
129.5
75.6,
a Peak went off scale, 104.4 represents the minimum corrected con-
centration.
4-14
-------
TABLE 4.12
HYDROCARBON PENETRATION THROUGH IMPINGERS
Time
1047
1050
1053
1056
1059
1102
1105
1108
Modea
B
I
B
I
B
I
B
I
Range
(ppm)b
12.8 to 16.2
19.2 to 20.2
31.0 to 57.2
14.0 to 27.4
14.5 to 23.6
14.5 to 24.1
12.3 to 13.6
10.9 to 16.8
Average
(ppm)
15.1
19.9
52.5
23.4
20.1
18.4
12.7
13.3
a B denotes bypass of impingers or stack gas concentration
whereas I denotes hydrocarbon concentration after sample
was pulled through two impingers.
k Total hydrocarbon concentration as methane.
Overal . average concentration with impingers: 18.8
' Overal . average concentration without impingers: 25.1
Imping ir collection efficiency: 25%
4-15
-------
FIGURE 4.2
Impinger Collection Efficiency Experiment
4-16
ENGINEERING-SCIENCE
-------
TABLE 4.13
HYDROCARBON COMPOUND EMISSION DATA
Run Number 1
Run Number 2
Filter and Probe Catch (mg)
Impinger Catch (mg)
FID Catch (mg)
Total Mass (mg)
Filter and Probe Catch
Concentration (gr/dscf)
Impinger Catch Concentration
(gr/dscf)
FID Catch Concentration
(gr/dscf)
Total Mass Concentration
{gr/dscf)
Mass Emission Rate, Filter
and Probe Catch (gr/dscf)
Mass Emission Rate,
Impinger Catch (Ib/hr)
Mass Emission Rate,
FID Catch (Ib/hr)
Total Mass Emission Rate
(Ib/hr)
3.7
21.3
37.9
62.9
0.00062
0.0036
0.0064
0.011
0.063
0.36
0.65
1.07
22.6
18.8
102.0
143.4
0.0051
0.0042
0.023
0.032
0.56
0.46
2.51
3.53
4-17
-------
CHAPTER 5
CRITICAL DISCUSSION OF EXPERIMENTAL RESULTS
Some problems were experienced with both sample collection and anal-
ytical procedures used for the Wilmington tests. There is every indica-
tion, however, that conventional stack sampling trains and routine labora-
tory procedures can be used for satisfactory characterization of chlorine
and hydrocarbon compound emissions. Prior to recommending the final pro-
cedures, problems with the developmental procedures are discussed using
the current test data.
ANALYSIS OF FREE CHLORINE
As indicated in Table 4.1, repeated analyses for the same quality
assurance samples contained an unacceptable degree of imprecision. Also,
some of the individual impinger analyses reported in Table 4.2 do not in-
dicate the minimum expected chloride to chlorine ratio of 2:1. Specific
reasons for the analytical errors are currently unknown.
The fact that chlorine analyses by 409-D relies on the intermediate
spectrophotometer calibration curve places crucial importance on that
curve. As shown in Figure 5.1 and Table 5.1, the calibration curve was
not constant from day to day. It is true that some inherent scatter is
present for all chlorine analyses (published results by American Public
Health Association attribute a relative accuracy of 12.8% and relative
standard deviation of 34.7% for results of 15 different laboratory anal-
yses for 0.8 mg/1 Cl2 by 409-D), but a standard, repeatable calibration
curve as developed by the same individual is essential. Our observed
scatter in the calibration curve no doubt was responsible for the lar-
gest portion of the differences in the three reported concentrations
for each audit sample.
Substantial dilution of commercial sodium hypochlorite solution was
required to produce a low enough concentration for calibration curve de-
velopment, and usually, dilutions of 20:1 to 1000:1 were necessary for
sample adjustment. Errors introduced during dilution, including use of
dirty glassware, may well have contributed to variations in the analyses.
Chlorine demand free water, prepared as directed in 409-D, was used for
all dilutions.
The instability of the turquoise color also introduces variations due
to individual field technician analysis. A slight difference in analysis
speed could introduce significant error considering the deterioration of
chlorine.
5-1
-------
FIGURE 5.1
Daily Standard Curves
C1 Concentration vs. Absorbance
^ 12/16/80
0 12/18/80
H 12/19/80
A 01/06/81
O 01/07/81
V 01/09/81
0 =
0.1 0.2 0.3 0.4 0.5 0.6. 0.7 0.8 0.9
ABSORBANCE
5-2
ENGINEERING-SCIENCE
-------
TABLE 5. 1
REGRESSION CONSTANTS FOR CHLORINE CALIBRATION CURVES
Date
12/12
12/16
12/18
12/19
1/7
1/9
Number of
Points on
Line
5
5
4
3
5
4
a
al
2.379
2.098
2.667
2.8483
2.625
2.467
a
ao
0.0160
0.0154
0.0988
0.0447
0.211
-0.003
b
r2
0.999
0.999
0.996
1.000
0.999
0.996
a These are constants for the linear correlation.
C = a0 + &i A
C = chlorine concentration, mg per liter, expressed as Cl
A = absorbance (optical density) of sample as determined by
spectrophotometer
r2 is the coefficient of determination.
5-3
-------
The standardizations and standard curve preparation require a signi-
ficant allotment of time. The deterioration of chlorine in the standard
dilutions requires that new dilutions be prepared either during or between
sample runs rather than relying on the original. Also, as mentioned ear-
lier, dilutions have to be made and analyzed one at a time and, therefore,
delegate the method to a. tedious slow pace. Overall, the time require-
ment of this method seems to limit its field application. Use of an al-
ternate method is desirable.
Initially, it was supposed that variations in spectrophotometer op-
erations caused at least some of the analytical problems. Subsequently,
daily checks with stock cobalt chloride solutions indicated steady in-
strument operation.
DISCUSSION OF RELATIONSHIP BETWEEN CHLORINE AND CHLORIDE
Inspection of results in Table 4.2 from the standpoint of chlorine
vs. chloride indicates some expected trends and some unexpected. Based
on our accurate, precise analysis of EPA quality assurance samples for
chlorides, a relatively high degree of confidence is placed on the chlo-
ride numbers. The expected minimum ratio Cl'tCl- of 2:1 was observed
only for the first impingers of Runs 3A, 3B, 4A, and 4B, and for the
second impinger of Run 3A. Discrepancies are attributed to errors in
the chlorine analysis.
Operating problems were experienced during Runs 2 and 3. During
Run 2, at a normal sampling flow rate, a high pressure drop across the
filter caused the screen type filter supports to detach from the filter
holder, permitting some particulate to proceed directly to the impingers.
The post-test leak check for Run 3B was very poor. These problems would
certainly negate use of the samples for compliance purposes, but A vs.
B results should compare better for Runs 2 and 3. (Pirticulates, both
total and chloride fractions, compared very well for A and 3B in spite
of the leak, but the gaseous components do not compare.) Only for Test
Runs 4A and 4B were both particulate and gaseous results comparable.
Probe wash and cyclone fractions were not included in the reported
results because they represented mainly entrained scrubbing liquid drop-
lets. The high ratio of C1~:C12 for the cyclone fraction could possibly
be due to thermal decomposition of hypochlorite ion in the cyclone flask
contained in the heated front half compartment.
Although problems were apparent with the chlorine analyses, a defi-
nite statement can be made concerning the number of impingers, based on
results contained in Table 4.5. If a nozzle size is selected to sample
at about 0.75 cfm, three impingers must be charged with absorbing solu-
tion. If the next smaller standard nozzle size is used, two impingers
will suffice. The latter case would, of course, extend the sampling
period required to collect a minimum gas volume, so for routine sampling
three impingers should be used. Each impinger should be analyzed sepa-
rately to prevent excessive dilution of the first impinger, which should
always capture at least 60% of the total chlorine.
5-4
-------
CHLORIDE TITRATIONS
Both the mercuric nitrate and silver nitrate titration procedures
were evaluated. These procedures were initially investigated by ES using
EPA audit samples (Table 4.2). The mercuric nitrate procedure was consis-
tently more accurate, but no problems were experienced with either method.
The mercury method was used for analysis of all field samples.
As a general rule, no difficulty was experienced in use of the mer-
curic nitrate titration procedure. Concerns were expressed in the Work
Plan regrading the final pH of the sample prior to titration. Several
samples were checked for pH prior to analysis using narrow range pH pa-
per, and all were found to fall within the required range. Sample pH
adjustments were accomplished as described in the method.
Several sample aliquots were passed through an ion exchange column
of Rexyn 101H as a check for metal ion interference. For the specific
case of secondary aluminum reverberatory furnace exhaust gas sampling,
the use of this ion exchange column was not required, since passage of
a sample appeared to have no effect on the results.
Note that the criterion for establishing the presence of metal ion
interference, as explained in the modified method presented in the ap-
pendix, was to pass a sample aliquot through the ion exchange column,
and compare the titration results with an untreated sample's results.
If these results differ by more than 1%, the sample is said to contain
interferring ions. This is too stringent a criterion, since it is dif-
ficult to obtain that degree-of--reproducibility with a standard sodium
chloride solution. Perhaps a more realistic value for titration dif-
ferences would be 3%. Differences of this magnitude would definitely
indicate possible interferences and not just variations inherent to the
analytical procedure.
In summary, both the silver and mercury titration procedures pro-
duced accurate analyses. The silver procedure was actually less com-
plicated because it involved fewer steps, but the mercury method was
more accurate. It should be noted, however, that the end point for the
Ag NC>3 titration was not: difficult to detect, contrary to comments re-
ceived by ES regarding this procedure.
PARTICULATE CHLORIDES
The most surprising result of the scrubber sample filter analyses
was that chlorine, reported as chloride, contributed only about 15-20%
of the total filter weight. Aluminum chloride must have been the pre-
dominant particulate compound (one method for production of aluminum
chloride for use as a catalyst is similar to that by which it is formed
during demagging), but it has a chloride weight fraction of 80% in anhy-
drous form. Handbook data indicates Al CI^ to be very soluble in water,
however, the filter extractions produced a colloidal material that re-
quired filtering prior to titration. It is unknown whether this contri-
buted to the low chloride fraction.
5-5
-------
The collected particles were white and appeared to have a very fine
size. Some alloy compound chlorides like copper and manganese are non-
volatile at the temperature of molten aluminum (greater than about 1,220°F).
Only zinc and silicon chlorides among common alloy constituents are vola-
tile under these conditions, and they should be present in far lower con-
centrations than aluminum chlorides. All of these metal chlorides are
white solids. At this time, it is impossible to state the exact chemical
composition of the particulate effluent. The only firm characterization
is that approximately 15 to 20% are water soluble chlorides.
Chlorine compound impinger catches were analyzed for the presence
of aluminum because of the anticipated decomposition of Al 0.3. Atomic
absorption was the selected analytical techniques for undiluted aliquot
samples from each impinger. Aluminum has a relatively high concentration
threshold for AA analysis, and our results were typically very close to
our detectability limit of 15 milligrams per liter. Qualitatively, it
can be stated that aluminum was present in at least some of the impingers
for all test runs. Quantitatively, it can only be stated that as much as
a few milligrams were present in some of the impingers.
Some of the impingers appeared slightly cloudy immediately after sam-
ple collection, but many later developed a definite white precipitate.
This was probably due to initial formation of aluminate ion in the caus-
tic solution followed by gradual conversion to insoluble aluminum hydrox-
ide. Certainly these precipitates contributed to the low aluminum con-
centration in solution.
Aliquots of the fil-ter--extract were also -analyzed by -AA.- —Again-,- --the- •=—•
results were inconclusive. The highest reported value was only about 6%
of the total filter weight.
Although the impinger catch aluminum analyses were semi-quantitative,
they did confirm the presence of aluminum. The quartz filters used (Pall
Corporation Type QAST) met the Method 5 filtration efficiency specification,
and it would not have been possible for any particle size to penetrate the
filter in milligram quantities during the course of a typical sample run.
The aluminum must have reached the impingers in vapor form, vaporization
possibly resulting from chemical reaction with collected material on the
heated filter.
Operation of the sample train filter at the standard Method 5 temper-
ature of about 250°F thus biases the reported mass loading towards lower
values. Careful temperature control high enough only to prevent water
condensation should be used for more accurate particulate matter charac-
terization.
HYDROCARBONS TRAIN FRONT HALF
Quartz filters for the two baghouse runs had a negative weight change.
No doubt this was due to the poor mechanical properties of the selected
media. The filters were very difficult to handle at the end of a test,
and some pieces were separated from the main filters. They had developed
an electrostatic charge, and some small pieces were impossible to recover.
5-6
-------
There was no apparent particle loading on the filters. During the
course of both sample runs/ the baghouse stacks were optically clear, so
it is speculated that the filter loading would have been very low in any
case.
Substantial acetone probe wash weights were recorded for the two com-
plete test runs. The particulate matter appeared to have included rela-
tively large lime dust particles which must have seeped thru the bags.
Some oily material was also present.
CONDENSABLE HYDROCARBONS
When inspecting the back half data listed in Table 4.8, it is noted
that the glassware rinses produced a higher mass loading than the impinger
extractions. This was probably due to the fact that the condensed organics
were very insoluble in water and preferentially adhered to the glass walls.
Collection of silicone grease is a possibility, but all connecting joints
were wiped off with trichloroethylene prior to back half recovery.
Trichloroethylene appears to be a better solvent for the condensed or-
ganics than acetone. The chlorinated hydrocarbons, however, have undesir-
able properties, and probably should not be recommended for routine field
use. Use of acetone only will cause incomplete recovery of some hydrocar-
bons .
With respect to the extraction procedure specified in Method 5A (Ap-
pendix C), firm conclusions-are-difficult because there^was:-onlyi=a-very--.-^ ..
small mass loading to work with. It seems that acid extractions are nec-
essary because of the acidic nature of the partially oxygenated hydrocar-
bons. This comment is specific to the charging well hydrocarbons, the
exact composition of which are unknown. Other hydrocarbon classifica-
tions could well respond differently.
NONCONDENSABLE HYDROCARBONS
When interpreting THCA data, it is important to remember that the FID
reports responses equivalent to methane. Thus, 10 ppm of n-octane would
have a response roughly equivalent to 80 ppm methane. The following dis-
cussion contains references to concentrations greater than 100 ppm, but
without knowing exact chemical composition and individual response fac-
tors, the true hydrocarbon concentration cannot be determined.
During test runs, insignificant zero and span gas drifts occurred
from post-test checks. Concentration values were corrected for a temper-
ature related reduction in instrument response, as discussed in the pre-
vious chapter. Ambient background concentrations ranged from 2.3 to 5.5
ppm, and were typically low compared to the FID results.
The second run maintained a running higher concentration level through-
out the duration of the test. Plant operators had indicated that their
"dirtiest" scrap was processed during that general period of time.
5-7
-------
As indicated in Chapter 4, there were periods during both test runs
where the FID peaks were greater than 100 ppm. Overall, both tests indi-
cated an average hydrocarbon concentration less than 100 ppm. Whether
or not this is significant depends on the source classification and the
levels of interest.
A brief experiment was completed using a modified impinger and a re-
gular Greenburg-Smith impinger. Absorption and condensation conditions
were much better than during a regular Method 5 run because of the lower
sample flow rate (0.1 cfm using the internal THCA pump only). Even so,
the impinger collection efficiency was only about 25%. Ordinarily, though,
these concentrations would not be of interest.
It was discovered later that the impinger water from the Method 5A
samples had experienced a reduction of their pH's to 2. Therefore, the
acidity may have retarded the ability of the water to remove condensable
hydrocarbons. In light of this, representative collection of condensable
hydrocarbons may be more successfully retained by a caustic or strong buf-
fer solution.
As indicated in Table 4.13, for both test runs, the mass emission
rate of the noncondensable hydrocarbons expressed as methane exceeded the
sample train catch. Strict comparison of the two mass fractions is impos-
sible because of the unknown composition of the hydrocarbons.
5-8
-------
CHAPTER 6
RECOMMENDATIONS FOR EVALUATION PROCEDURES
Our experience to date with chlorine and hydrocarbon compound eval-
uation techniques does not yet provide a firm data base for potential re-
ference procedures. Analysis of the test data does, in fact, answer some
important questions concerning sample train configuration and collected
sample analytical procedures; however, some aspects must be viewed as de-
velopmental. The particular area of concern at this point is the analysis
of chlorine. A recommendation for future studies for this analysis and
all others of interest for the secondary aluminum industry are discussed
in this chapter.
CHLORINE COMPOUNDS - SAMPLE COLLECTION
Complete recovery of chlorine concentrations as low as approximately
20 ppm can be achieved by using three impingers in series charged with 100
ml of 0.1N sodium hydroxide solution. At relatively low sample flow rates,
all chlorine is recovered by the first two impingers. A third impinger in
series provides a safety factor when low flow rate sampling (^0.3 cfm) is
impractical. Although hydrochloric acid was not absolutely identified in
the present results, complete recovery with two impingers should be ex-
pected, regardless of flow rate up to about 1 cfm.
Particulate chlorides are retained on the sample filter only as long
as high temperature does; not cause sublimation or decomposition. The stan-
dard Method 5 temperature of 250°F is too high for accurate characteriza-
tion of aluminum chloride. Based upon a preliminary pre-test stack tem-
perature measurement, the sample filter should be heated only about 20°F
above stack temperature.. This should be high enough to prevent both mois-
ture condensation and significant sample deterioration.
Quartz filters have superior chemical properties, but their mechani-
cal properties leave much to be desired. Standard Reeve-Angel glass fiber
filters should be used for all future samples.
The recommended sample train is diagrammed in Figure 6.1. Operation
is identical to EPA Method 5, except that filter temperature control is
based on actual stack temperature instead of the conventional value of
250°F.
6-1
-------
Chlorine Compound Sampling Train
NOTES:
IMPINGERS SHOULD BE CHARGED WITH 0.1 N NaOH SOLUTION
BECAUSE OF VOLATILITY OF PARTICULATE A1C13,
FILTER SHOULD BE HEATED TO STACK TEMPERATURE
PLUS 20 DEGREES F
GLASS FIBER FILTER FOR COLLECT-
ION OF PARTICULATE CLORIDES
THERMOCOUPLE
THERMOCOUPLE
PROBE
PI TOT TUBE
STACK WALL
PI TOT MANOMETER
HEATED AREA
THERMOMETERS
ORIFICE
DUY r,AS Mf.ir.R
IMPINGER TRAIN
FOR ABSORPTION
OF HC1 AND Cl2
.THERMOCOUPLE
CHECK VALVE
IMP INKER
ICE BATH
VACUUM LINE
VACUUM GUAOE
MAIN VALVE
BY-PASS VALVE
AIR-TIGHT PUMP
O
-------
CHLORINE COMPOUND ANALYSES
Chlorine Method 409-D, and all other methods designed for evaluation
of potable or wastewater, should be rejected for stack sampling applica-
tion because they were designed for very low chlorine concentration and
are time consuming. Extensive dilution is normally required to lower im-
pinger sample concentration to the range of 1 mg/1. Much higher dilu-
tions would be necessary for samples collected at control device inlets.
Errors are possible in dilution, and once the chlorine concentration has
been reduced, sample stability is a problem. A more rugged technique
suitable for routine field use is desired.
The alkaline arsenite procedure (included in the appendices) was ini-
tially considered and rejected as being too insensitive. This technique
can be used for simultaneous evaluation of chlorine, hydrochloric acid,
and particulate chlorides. The reported lower limit for application was
10 ppm by volume of C12 or HCl, which is equivalent to a chlorine or chlo-
ride concentration of about 40 mg/1.
Briefly, the arsenite technique is used to convert chlorine to chlo-
ride. Chlorine gas is absorbed in the impinger solution of sodium hydrox-
ide and sodium arsenite and is immediately reduced. A back titration of
spent absorbing solution determines the amount of arsenic consumed which
indicates how much chlorine reacted. The difference in total chlorides
(titrated using the mercury procedure) from the amount attributable to
chlorine conversion provides the analysis for hydrochloric acid or other
chloride compounds. Only titration procedures are required, and sample
dilution is unnecessary. In fact, the techniques should be more accurate
at higher concentrations.
One method of extending the useful lower range of the arsenite pro-
cedure below 40 mg/1 is to use lower normality titrating solutions. This
and other improvements will be considered prior to the next field test
program.
Analysis of chlorine samples should be completed in the field within
two hours after sample collection. Samples should be kept on ice until
analysis. Impingers should be recovered and analyzed separately, as mix-
ing the contents would result in dilution of the first impinger catch.
Based upon the samples, the predominant gaseous chlorine compound is
molecular chlorine and riot hydrogen chloride.
TOTAL CHLORIDES IN SOLUTION
The mercuric nitrate titration procedure is recommended. Our exper-
ience confirmed it to be reliable and accurate. Narrow range pH paper
is a satisfactory indicator for the pH dependent titration end point de-
tection.
Sample aliquot volumes should be selected so as to result in a mini-
mum titration volume on the order of one milliliter.
6-3
-------
Accurate determination of total chlorides is dependent upon conver-
sion of chlorine in hypochlorous form to chloride ion. A direct, simple
approach is to add a stoichiometric excess of hydrogen peroxide. Solu-
tion pH must be at a pH greater than 7. After the chlorine concentrations
have been determined, the amount of peroxide necessary for a 10 to 20% ex-
cess may be calculated.
Chlorination Scrubber Effluent Particulates
A certain percentage of the filter catch is water soluble chlorides,
but the largest portion is not. Both total filter weight and chloride
fraction should be reported.
HYDROCARBON COMPOUNDS - SAMPLE COLLECTION
The recommended layout for the hydrocarbons sample train is pictured
in Figure 6.2. Operation is identical to the standard Method 5 train.
Similar to the scrubber train, Reeve-Angel filters should be used.
Quartz filters would be suitable if only particulate chloride analysis
instead of total weight were desired.
Impingers should be charged with 0.1N potassium or sodium hydroxide
instead of distilled water. Some organics are more soluble at high pH,
and use of a caustic impinger solution would promote sample recovery and
would prevent a large portion of organics adhering to the impinger walls.
Use of a caustic solution would also help prevent formation of an acidic
medium due to absorption of acid gases and would, therefore, again result
in more complete recovery.
Use of a hydrocarbon analyzer should not be necessary. The amount
of material recoverable by physical condensation is proportional to the
difference in hydrocarbon compound partial pressure at about 250°F and
about 70°F. If the material does not condense out at the impinger tem-
perature, it may cause some type of hydrocarbon emission problem, but
it should not result in formation of a particulate matter plume at the
stack exit.
CLEAN UP AND RECOVERY OF BACK HALF HYDROCARBONS
Use of all conventional hydrocarbon solvents in the field involves
potential exposure to hazardous conditions. Rinsing the probe and impin-
ger glassware in a laboratory hood would be much safer, but would be to-
tally impractical. Thus, desirable solvents such as chloroform (or any
of the chlorinated or fluorinated hydrocarbons), ethyl ether, and benzene
are eliminated from consideration. Presently, acetone seems to be an ac-
ceptable compromise. Some organics adhering to the probe liner or other
glass surfaces will not be completely recovered by acetone, but acetone
is currently believed to have fewer hazardous properties than other suit-
able solvents.
Some hydrocarbon compounds are not completely recoverable from a neu-
tral solution. Impinger contents should be acidified to a pH less than 2
6-4
-------
Particulate and Condensable Hydrocarbons Sampling Train
NOTE: Impingers should be charged
with 0.1 N NaOH or KOH
FILTER HOLDER U/REEVE ANGEL FILTER
THERMOCOUPLE
IMPINGER TRAIN
THERMOCOUPLE
PRONE
PHOT rilUF.
SIACK WAI.
I'll Of MANOMETER
HEAIED AREA
IHERMOMETERS
ORIFICE
THERMOCOUPLE
CHECK VALVE
DRY GAS MEIER
IMPINGER
ICE RAIN
VACUUM LINE
VACUUM GIIAGE
MAIN VALVE
BY-PASS VALVE
AIR-TIGHT PUMP
-------
with hydrochloric acid prior to extraction. The necessity of using an
acid extraction vs. a neutral extraction could be determined on a case-
by-case basis for each source/control equipment classification, but it
is more straightforward to rely upon the acid extraction for general
application. Impinger contents should be acidified to a pH less than
2 with hydrochloric acid prior to extraction.
6-6
-------
CHAPTER 7
REVISION .OF SAMPLING STRATEGY
At the conclusion of the initial field effort, it was apparent that
modification of the first generation sample and analysis procedures was
in order. Further investigations resulted in development of second gen-
eration techniques for use in subsequent field studies. These revised
techniques are discussed in the remainder of this chapter.
CHLORINE COMPOUND ANALYSES
Experience with Method 409-D proved it to be a laborious procedure
for field application. Additionally, as indicated in Table 4.1, repeated
analysis of the same EPA audit sample produced results with an unaccept-
able degree of imprecision. The alkaline arsenite titration procedure
was selected for investigation using a second set of EPA audit samples.
Results of the laboratory work performed in February 1981 are summarized
in Table 7.1.
Analysis of the results included in Table 7.1 again indicated an
unacceptable degree of imprecision am inaccuracy. The arsenite method
'analysis for the lowest audit sample oncentration (51 mg/1) was very
poor. Significant improvements in the arsenite method of analysis were
projected based on using large sample aliquots and using a more dilute
iodine solution for titration of residual arsenite.
At this point in time, neither the arsenite nor the Method 409-D
procedure was considered satisfactory as a reference method compliance
procedure. A relatively straightforward, accurate technique for low
chlorine procedure, 409-E DPD Ferrous Titrimetric Method, from Standard
Methods for the Examination of Water and Wastewater, appeared to satis-
fy the low concentration accuracy criterion, and it was considered more
amendable for field use than Method 409-D.
A third series of audit sample analyses using the arsenite, 409-D
and 409-E procedures, WEIS completed in March. Results are summarized
in Table 7.2. For reference purposes, estimated chlorine impinger con-
centrations as a function of stack gas concentration were calculated.
Estimated values and the assumptions made for the calculations are pro-
vided in Table 7.3.
Across the board, all three techniques produced results with an av-
erage error of approximately -26%. Two factors could have contributed
7-1
-------
TABLE 7.1
INITIAL EVALUATION OF ARSENITE PROCEDURE
Sample Chlorine
Number Concentration (mg/1!
1038
2260
2292
3053
3099
4057
4081
5264
5270
2
115 •
113
184
161
208
277
361
375
EPA Value
) (mg/1)
51
153
153
255
255
357
357
459
459
Relative
Error ( % )
96.08
24.84
26.14
27.84
36.86
41.74
22.41
21.35
18.30
Average Relative Error: 35.06%
Standard Deviation of Relative Error:
24.07%
TABLE 7.2
SUMMARY OF AUDIT SAMPLE ANALYSES BY ALL THREE METHODS
(All Chlorine Concentrations Expressed in mg/1)
Sample EPA Relative Relative Relative
Number Value Arsenite Error (%) 409-D Error (%) 409-E Error (%)
1028
2280
3072
4061
5277
Average
Standard
Relative
51
153
255
357
459
Relative
32
124
189
267
331
Error:
37
18
25
25
27
27
.25
.95
.88
.21
.89
.04
38
106
193
262
348
25.
30.
24.
26.
24.
26.
49
72
31
61
18
26
38
110
172
275
360
25.
28.
32.
22.
21.
26.
49
10
55
97
57
14
Deviation of
Error:
6
.62
2.
68
4.
37
7-2
-------
TABLE 7.3
ESTIMATED IMPINGER CONCENTRATIONS
VS. STACK GAS CONCENTRATION
Stack Gas
Cl- Concentration
fppm)
1
5
10
20
First Impinger
OC1~ Concentration
(mg/1)
8.2
41
82
164
Combined Second
and Third Impingers
Concentration (mg/1)
0.46
2.3
4.6
9.2
Assumptions: 20 ft-* gas registered on meter, standard pressure, 85°F
meter temperature, 100 ml absorbing solution in each im-
pinger, 90% collection efficiency in each impinger, stack
gas saturated with water at 120°F.
7-3
-------
to the inaccuracy. The audit sample volumes were relatively small, ap-
proximately 20 ml, such that only one analysis for each sample by each
method was usually possible. Also, the audit samples were originally
prepared in November 1980, and gradual decomposition of hypochlorite
ion to chloride ion probably caused the consistent negative error trend.
For the lowest audit sample concentration, the arsenite procedure
was less accurate than either 409-D or 409-E. At higher concentrations,
the arsenite procedure produced comparable results to the other two tech-
niques. Method 409-D was the most precise of the three techniques.
In the interest of streamlining field test protocol, use of two dif-
ferent procedures contingent upon expected stack gas chlorine concentra-
tion is recommended. The simplicity of the arsenite procedure is appeal-
ing, and it can be used whenever the expected chlorine stack gas concen-
tration is greater than about 10 ppm. Potential applications include
sampling locations at the inlet of process exhaust gas emissions control
equipment and some locations at the outlet for control devices. If very
low chlorine concentrations are anticipated, Method 409-E should be used.
Method 409-E is recommended instead of 409-D because the former procedure
involves a briefer analysis procedure with fewer steps and is, therefore,
more suited for field use. Method 409-E in Federal Register format for
Reference Test Methods is attached at the end of Appendix A. This pro-
posed Federal Register format also includes the chloride procedure.
In particulate-chlorine-chloride sample collection, much of
the particulate may be in the form of Aid 3. On the downstream side of
a scrubber, less of the particulate may be A1C13 than on the scrubber
inlet because AlCl^ reacts with water to form aluminum oxides and hydro-
chloric acid. Investigations of A1C13 vapor pressure at various tempera-
tures showed vapor pressure to increase significantly at temperatures
above 93°C (200°F). See Table 7.4. It is therefore recommended that
temperatures of the filter and sample probe should under no circumstar :es
be allowed to exceed 93"C (200°F), and it is preferable that temperatures
be maintained at approximately 28°C (50°F) above the temperature of the
specific stack being sampled.
To measure the water soluble chloride caught on the filter, the
filter weighing and extraction of the filter with water should be made
as soon as possible after sampling to prevent decomposition of the AlCl^
to A12C>3 and HCl. If the decomposition is allowed, the HCl may evaporate
off the filter and not be weighed as particulate, or show up in the chlo-
ride analysis. Careful filter handling and immediate filter desiccation
should minimize this A1C13 decomposition.
PARTICULATE AND CONDENSIBLE ORGANIC MATTER
EPA recommended that in the cleanup procedure methylene chloride be
used instead of acetone as acetone may be an inadequate solvent to remove
hydrocarbons from the probe and glassware. The question of possible meth-
ylene chloride reaction with the nylon in the probe cleaning brushes was
raised. ES subsequently conducted an evaluation of the reaction of methy-
lene chloride on ordinary nylon brushes and concluded the brushes would
7-4
-------
TABLE 7.4
VAPOR PRESSURES OF A1C13 AT
VARIOUS TEMPERATURES3
Temperature Vapor Pressure of
(°F) A1C13, mm Hg
175 0.1
200 ' 0.6
250 9.5
309 95.0
350 650.0
J. S. Sconce, "Chlorine, its Manufacture, Properties, and Uses",
Robert E. Krieger Publishing Co.
7-5
-------
not hold up under methylene chloride contact. The evaluation consisted
of tare weighing beakers, adding methylene chloride and soaking brushes
in different beakers for 1, 5, 10, and 15 minute periods, and drying-down
and dessicating the beakers to obtain final weights. After subtracting
the methylene chloride blank value it appeared that an average probe
brushing would cause a sample weight gain of at least 2 mg, an amount
judged unacceptable in control device outlet samples. Also, the brushes
showed signs of physical deterioration following methylene chloride con-
tact. ES submitted the results to EPA in a letter dated May 27, 1981,
and continued investigation of other brush types. In the interim ES
recommended that both methylene chloride and acetone be used for probe
washing, with the methylene chloride used first without a brush, fol-
lowed by brushing with tin acetone rinse, and a final non-brushing meth-
ylene chloride rinse.
ES obtained natural, fiber brushes for evaluation, and special-or-
dered Teflon brushes. The Teflon brushes required special manufacture
and long delivery times, and as of the end of August 1981 had not yet
been delivered. The natural fiber evaluations showed the brushes se-
verely deteriorated upon methylene chloride contact. ES also obtained
brushes made of WYTEX, ai type of nylon, for evaluation. Table 7.5 shows
the evaluation results and the methodology used. It appears that most
of the residue extraction from the brushes into the methylene chloride
occurred during the 5 and 10 minute soakings, leaving little extractable
material for the 15 minute soaking. If brushes were preconditioned
before first use by soaking for approximately 20 minutes in methylene
chloride, the residue contributed by the brushes in a probe cleaning
probably would not exceed 0.5 milligrams. As a comparison, particulate
obtained in the probe Wcish at NASCO was 8 and 29 milligrams. From the
evaluation there was not apparent brush deterioration following the soak-
ings. It is therefore concluded that the WYTEX nylon brush material is
suitable for use with methylene chloride, although before first use the
brushes should be presoaked.
Silicone grease on sampling train connecting points is prohibited
because of the possibility of contaminating the sample. Without sili-
cone grease it becomes much more difficult to achieve a leak-tight sam-
pling train. ES looked into various other forms of connectors, but fi-
nally decided to use standard ball and socket joint connectors that are
new and leak-tight. The glassware is marked for assembly the same way
each time.
It was proposed earlier that a caustic solution be used in the im-
pingers, as the pH of the solution after the early test runs was found
to be low, and it was expected the use of caustic would reduce hydrocar-
bon loss. It has since been determined to stay with sample collection
in distilled deionized water for two reasons: (1) it is simpler, and
(2) because extraction of the hydrocarbons take place at a pH of 2, any
loss due to acidification may occur anyway.
7-6
-------
TABLE 7.5
RESULTS OF WYTEX NYLON BRUSH TESTS
IN METHYLENE CHLORIDE
Brush/Beaker
Blank
Brush "A"
Beaker "A^"
Beaker "A2"
Beaker "A3"
Brush "B"
Beaker "B-|"
Beaker "B2"
Beaker "83"
Brush "C"
Beaker "C-|"
Beaker "C2"
Beaker "C3n
Tare Weight
(g)
96.4795
5.6696
96.8830
104.3376
100.8572
5.6751
98.2901
99.4174
94.9413
5.6124
100.2728
100.9245
100.0266
Final Weight
(g)
96.4748
5.6743
96.8849
104.3386
100.8577
5.6805
98.2915
99.4185
94.9421
5.6176
100.2742
100.9260
100.0277
Weight Minus Blank
(g)
+.0003
+.0044
+.0016
+.0007
+.0002
+.0051
+.0011
+.0008
+.0005
+.0049
+.0011
+.0012
+.0008
Evaluation Method
1. A total of three brushes; Justman Special B.C. WYTEX were labeled "A",
"B", and "C" i
5 procedures.
"B", and "C" and dessicated and tare weighed according to EPA Method
2. A total of ten beakers labeled (A-|, A2, A^) (B-j, B2, 63) (C-|, C2, C3)
and blank were dessicated and tare weighed according to EPA Method 5
procedures.
3. Each brush was soaked in its appropriate beaker for a consecutive per-
iod of 5, 10/ and 15 minutes in a bath of 2000 ml methylene chloride
and rinsed in the same beakers with an additional 50 ml of methylene
chloride. The blank beaker contained 250 ml methylene chloride.
4. Brushes and beakers were allowed to dry-down under a laboratory hood
and constantly weighed by dessication after dry-down had occurred.
7-7
-------
APPENDIX A
CHLORINE PROCEDURES
-------
For the Examination of
Water and Wastewater
FOURTEENTH EDITION
Prepared and published jointly by:
AMERICAN PUBLIC HEALTH ASSOCIATION
AMERICAN WATER WORKS ASSOCIATION
WATER POLLUTION CONTROL FEDERATION
Joint Editorial Board
M.C RAND, WPCF, Chairman
ARNOLD E. GREENBERG. APHA
MICHAEL J. TARAS, AWWA
MARY ANN FRANSON
Managing Editor
Publication Office: •.
American Public Health Association
1015 Eighteenth Strew NVV
Washington, DC 200 3 6
-------
409 CHLORINE (RESIDUAL)
The chlorination of water supplies
and polluted waters serves primarily to
destroy or deactivate disease-producing
microorganisms. A secondary benefit is
the overall improvement in water qual-
ity resulting from the reaction of chlo-
rine with ammonia, iron, manganese,
sulfide, and some organic substance?.
Chlorination may produce adverse ef-
fects by intensifying the taste and odor
characteristics of phenols and other or-
ganic compounds present in a water
supply. Combined chlorine formed on
chiorination of ammonia- or amine-
bearing waters affects some forms of
aquatic life adversely. To promote the
primary purpose of chlorination and to
minimize any adverse effects, it is essen-
tial that proper testing procedures be
used with a fore-knowledge of the limi-
tations of the analytical determination.
Chlorine applied to water in its ele-
mental or hypochlorite form initially
undergoes hydrolysis to form free, avail-
able chlorine consisting of aqueous mo-
lecular chlorine, hypochlorous acid, and
hypochlorite ion. The relative propor-
tion of these free chlorine forms is pH-
dependent. and at the pH of most wa-
ters the hypochlorous acid and hy-
pochlorite ion will predominate.
Free chlorine reacts readily with am-
monia and certain nitrogenous com-
pounds to form combined available
chlorine. With ammonia, chlorine re-
acts to form the chloramines: mono-
chloramine, dichioramine. and nitrogen
trichloride. The presence and concentra-
tions of these combined forms depend on
many conditions, chiefly pH, temper-
ature, and the initial chlorine-to-nitro-
gcn ratio. Both free and combined chlo-
rine may be present simultaneously.
Combined chlorine in water supplies
may be formed in the treatment of raw
waters containing ammonia, by the ad-
dition of ammonium salts in pre-chlori-
nation, or by producing a combined
chlorine residual in the distribution sys-
tem. Chlorinated wastewater effluents.
as well as certain chlorinated industrial
effluents, normally contain only com-
bined chlormTforms. Historically, the
principal analytical problem has been to
distinguish between free and combined
forms.
In two separate but related studies,
designated Study No. 1 and 2, respec-
tively, samples were prepared and dis-
tributed to participating laboratories for
the purpose of evaluating the residual
chlorine methods.
In Study No. 1, three solid synthetic
unknowns were prepared: One pow-
dered unknown was compounded of
70% calcium hypochlorite* and NaCl
filler to yield a free available chlorine
concentration of 800 jjg/1 upon dis-
solution in chlorine-demand-free dis-
tilled water; the second was com-
pounded of 70% calcium hypochlorite*
and NaCl filler to yield a total available
chlorine concentration of 640 ^g/1
upon dissolution and mixing with an
aqueous ammonium buffer solution in
chlorine-demand-free distilled water;
and the third was prepared from Hala-
zone.t p-(N.N-dichlorosulfamyl) ben-
zoic acid, to yield a total available chlo-
rine concentration of 1.830 /xg/I upon
dissolution in chlorine-demand-free dis-
tilled water. The results obtained by the
participating laboratories are summa-
rized in Table 409:1.
-------
In Study No. 2. each participating
laboratory.received four sealed glass am-
poules of concentrated solution (three
hypochlorite solutions of different con-
centrations and one ammonium chloride
borate buffer solution) which, when
diluted according to instructions, pro-
vided two samples containing free chlo-
rine and one containing combined chlo-
i^
rine. Sample No. 1 contained 440 jig/1
free chlorine, a concentration likely
to be encountered in analysis of treated
potable water. Sample No. 2 contained
980 Mg/' free chlorine, the maiximum
concentration likely to be encountered in
analysis of treated potable water.
Sample No. 3 contained only combined
chlorine at a concentration of 6<>0 >ig/l
to simulate an insufficiently chlorinated
water having no free chlorine residual.
To facilitate statistical computations, a
value of 50 ^g/l free chlorine was se-
lected as the "true" value rather than
the theoretical 0.0 jig/l. In Study No.
2, the data were treated statistically ac-
cording to a total error term defined as-
of reproducibility) inherent in a bleach-
ing technic in which operator skill and
environmental factors such as temper-
ature are critical.
The results presented in Tables 409:1
through IV are valuable only for com-
parison of the methods tested. Many fac-
tors, such as analytical skill, recognition
of known interferences, and inherent
limitations, determine the reliability of
any given method.
Some oxidizing agents, including free
halogens other than chlorine, will ap-
pear quantitatively as free chlorine; this
is also true of chlorine dioxide. Some ni-
trogen trichloride may be measured as
free chlorine. The actions of interfering
substances should be familiar to the ana-
lyst because they affect a particular
method.
Total error » 100
Absolute value of mean error +2 (Srd. Dev.)
True value
The results obtained by the participating
laboratories are summarized in Tables
409:11 through IV. In the last column of
these tables all data on Sample No. 1
are omitted because of sample instabil-
ity, and therefore the results can be used
only for comparative purposes and not
as a measure of the overall precision or
accuracy. Free chlorine data also were
omitted in Column J because the stated
free chlorine content is an artifact and
actually was zero.
. Because of poor accuracy and pre-
' cision and a high overall (average) total
error in comparison with other available
methods, the orthotolidine procedures
that have been so widely used have been
deleted as standard methods, The
methyl orange technic also has been de-
leted because of the poor precision (lack
* Olin \Uthnon ChemieaJ Corp.. HTH (granu-
lar).
' Abboct Lzboratonn.
-------
TABLE 409:1. PRECISION AND ACCURACY DATA FOR RESIDUAL CHLORINE METHODS IN STUDY No. I
Method
Ticrimetric (iodine)
Amperometric
Onhotolidine
Orthotolidinc-anenite
Stabilized neutral
onhotolidine
Ferrous DPD
DPD Colorimetric
Leuco crystal violet
Methyl orange
Residual Chlorine
Concentration
Free
/£//
800
„ 800—-
800
800
800
980
800
800
Total
*£//
840
640
1,830
640
1,830
__lir_u. ___ _.,
640
1.830
640
1.830
640
1.830
640
1.830
660
640
1.830
640
1.830
Number of
Laboratories
32
30
32
23
24
24
- -15
17
18
20
21
23
15
16
17
19
19
19
26
25
17
17
18
26
26
26
Relative
Standard
Deviation
%
27.0
32.4
23.6
42.3
' 24.8 .
12.5
64.6
37.3
31.9
52.4
28.0
35.0
34.7
8.0
26.1
39.8
19.2
9.4
20.7
27.6
32.7
34.4
32.4
45.0
30.1
19.9
Relative
Error
%
23.6
18.5
16.7
25.0
8.5
8.8
42.5
20.2
41.4
42.3
14.2
49.6
T2.8
2.0
12.4
19.8
8.1
4.3
15.6
15.6
7.1
0.9
18.6
22.0
14.2
7.2
-------
TABLE 409:11. SUMMARY OF OVERALL ACCURACY IN STUDY No. 2 ( AVERAGE MEAN ERROR IN Mu/L)
Method
Methyl orange
Lcuco crystal violet
SNORT
DPD-titrimetric
DPD-colorimetric
Amperometric
OTA
Sample 1
Free
-O.219
-0.250
-O.241
-0.259
-0.263
-0.24!
-0.282
Total
-0.171
-0.209
-0.198
-0.198
-0.213
-0.189
-0.253
Sample 1
Free
-0.044
-0.08 5
-0.113
-0.192
-0.153
-0.230
-0.198
Total
-0.006
-0.070
-0.107
-0.059
-0.097
-O.119
-0.102
Sample 3
Free
-O.006
-0.050
-0.048
-0.0} 8
-0.014
-0.012
+0.114
Total
4>0.029
-0.007
-O.032
-0.031
4-0.103
-0.108
-0.092
Over-
all
Aver-
age
0.079
0.112
0.123
0.130
0.140
0.150
0.174
Omitting all
Data on
Sample I and
Free on
Sample 3-
0.026
0.054
0.084
0.094
0.117
0.152
0.130
TABLE 405>:HI. SUMMARY OF OVERALL PRECISION IN STUDY No. 2
(AVERAGE STANDARD DEVIATION IN Mc/L)
Method
Leuco crystal
violet
SNORT
Amperometric
DPD-colorimetric
DPD-titrimetric
Methvl orange
OTA
Sample 1
Free
0.085
0.093
0.106
0.102
0.110
0.143
0.090
Total
0.055
0.092
0.072
0.100
0.103
0.162
0.098
Sample 2
Free
0.042
0.120
0.206
0.171
0.298
0.315
0.335
Total
0.015
0.142
0.137
0.152
0.205
0.301
0.325
Sample 3
Free
0.000
0.004
0.040
0.057
0.019
0.055
0.195
Total
0.089
0.110
0.171
0.210
0.121
0.143
0.218
Over-
all
Aver-
age
0.048
0.094
0.122
0.132
0.143
0.187
0.210
Omitting all
Data on
Sample 1 and
Free on
Sample 3
0.049
0.124
0.171
0.177
0.208
0.253
0.293
TABLE 409:IV. SUMMARY OF OVERALL (AVERAGE) TOTAL ERROR IN STUDY No. 2
Method
Leuco crystal
violet
SNORT
DPD-titrimetric
Amperometnc
DPD-colorimetric
Methvl orange
OTA'
Sample 1
Free
95.31
97.04
108.86
102.95
106.18
114.86
104.95
Total
72.59
87.00
91.77
75.63
94.00
112.59
101.90
Sample 2
Free
17.24
35.95
80.51
65.46
50.57
68.83
88.57
Total
10.20
39.97
47.89
40.14
40.83
62.12
76.81
Sample 3
Free
100.00
112.40
153.40
182.80
256.40
232.00
1007.80
Tocal
28.00
38.15
41.33
68.21
79.33
47.75
79.95
Over-
all
Aver-
age
53.90
68.42
87.29
89.20
104.55
106.56
243.33
Omitting all
Data on
Sample 1 and
Free on
Sample 3
18.52
38.02
56.64
57.94
56.90
59.56
81.77
-------
1. Selection of Method
a. Natural and treated n-aterj: The
iodometric methods (A and B) are suit-
able for measuring chlorine concentra-
tions greater than 1 mg/1. but are not
accurate at lower concentrations or in
the presence of interferences.
The amperometric titration method
(Q is a standard of comparison for the
determination of free or combined chlo-
rine. It is affected little by common oxi-
dizing agents, temperature variations,
turbidity, and color. The method is not
as simple as the colorimetric methods
and requires greater operator skill to ob-
tain the best reliability.
The ferrous DPD method (E) pro-
vides a titrimetric procedure for deter-
mining free available chlorine and for
estimating free and combined chlorine
fractions present together.
The stabilized neutral orthotolidine
(SNORT) and the DPD colorimetric
methods (Methods D and F, respec-
tively) are applicable to the determina-~
tion of free available chlorine. Proce-
dures are given for estimating the
combined fractions. Increasing concen-
trations of monochloramine are likely to
produce an increased interference with
the free chlorine determination. In addi-
tion, the SNORT and DPD methods
are subject to interference by oxidized
forms of manganese.
The leuco crystal violet (LCV)
method (G) makes possible the determi-
nation of free available chlorine, total
chlorine, and combined chlorine by
difference.- The LCV method exhibits a
relatively minimal interference as mono-
chloramine concentrations are increased
in the determination of free available
chlorine-, however, nitrite and mono-
chloramine in combination, as well as
oxidized forms of manganese, will pro-
duce interference in determining free
available chlorine.
The amperometric. LCV. DPD, and
SNORT methods are unaffected by
dichloramine concentrations in the
range of 0 to 9 mg/1 (as Ch) in the de-
termination of free chlorine. Nitrogen
trichloride, if present, reacts partially as
free available chlorine in the ampero-
metric. DPD. and SNORT methods.
Nitrogen trichloride does not interfere
with the LCV procedure for free chlo-
rine.
The free available chlorine test, syr-
ingaldazine (Method H. Tentative)
was developed as a procedure spe-
cific for free available chlorine. It is
.unaffected by significant concentrations
of monochloramine, dichloramine, ni-
trate, nitrite, and oxidized forms of
manganese.
Sample color and turbidity may in-
terfere in all colorimetric procedures un-
less they are compensated for.
Organic contaminants may produce a
false free chlorine reading in most cdlor-
imerric methods (see fi 1 b below).
b. Polluted waters: The determina-
tion of residual chlorine in samples con-
taining organic matter presents special
problems. Because of the presence of or-
ganic compounds, particularly organic
nitrogen, the residual chlorine exists in a
combined state. A considerable residual
may exist in this form, but at the same
rime there may be appreciable unsatis-
fied chlorine demand. The addition of
the reagents in the determination may
change these relationships so that resid-
ual chlorine is lost during the analysis.
In wastewater, the differentiation be-
tween free available chlorine and com-
bined available chlorine is not ordinarily
made because wastewater chlorination is
seldom carried far enough to produce
free available chlorine.
-------
The determination of residual chlo-
rine in industrial wastes is similar to that
in sewage when the waste contains or-
ganic matter, but may be similar to the
determination in water when the waste
is low in organic matter.
Although the methods given below
are useful for the determination of resid-
ual chlorine in wastewaters and treated
effluents, selection in accordance with
the composition of the sample being
tested is necessary. Some industrial
wastes, or mixtures of wastes with do-
mestic wastewater. may require special
precautions and modifications to obtain
satisfactory results.
Free chlorine in a wastewater can be
determined by any of the methods
presented in Sections 409 A through
409 H, provided that known interfering
substances are absent or compensated
for. The amperometric method (C) is
the method of choice because it is not
subject to interference from color, tur-
bidity, iron, manganese, or nitrite nitro-
gen. The DPD and -SNORT- methods—
are subject to interference from high
concentrations of monochloramine un-
less these arc compensated for by addi-
tion of arsenite immediately after re-
agent adr'tion. The LCV method (G) is
significani y less affected by high mono-
chloramine concentrations and an arse-
nite addition will produce a minimum
interference. The presence of oxidized
forms of manganese will interfere in
most colorimetric procedures and. in ad-
dition, the combination of mono-
chloramine and nitrite will seriously in-
terfere with the LCV procedure.
The tentative syringaldazine method
(H) is unaffected by concentrations of
monochloramme, dichloramine. nitrite.
nitrogen, iron, manganese, and other in-
- terfering compounds normally found in
domestic wastewaters.
For total available chlorine in sam-
ples containing significant amounts of
organic matter, the iodomerric back ti-
tration method (B) should be used to
prevent contact between the full concen-
tration of liberated iodine and the
sample. Either the amperometric or the
starch-iodide end point may be used. In
the absence of the interference, the two
modifications give concordant results.
The amperomerric end point is inher-
ently more accurate and is free of inter-
ference from color and turbidity, which
can cause difficulty with the starch-io-
dide end point. On the other hand, cer-
tain metals and complex anions in some
industrial wastes interfere in the am-
perometric titration and indicate the use
of another method for such wastewaters.
Silver in the form of soluble silver cy-
anide complex, in concentrations as low
as 1.0 mg/1 silver, poisons the cell at pH
4.0 but not at 7.0. The silver ion, in the
absence of the cyanide complex, gives
extensive response in the current at pH
_4.0 and gradually poisons the cell at all
pH levels. Cuprous copper in the sol-
uble copper cyanide ion, in concentra-
tions as low as 5 mg/1 copper or less,
poisons the cell at pH 4.0 and 7.0. Al-
though manganese, iron, and nitrite
may interfere with this method, the in-
terference is minimized by buffering to
pH 4.0 before addition of KI. An un-
usually high content of organic matter
may cause some uncertainty in the end
point. Whenever manganese, iron, and
nitrites are definitely absent, this uncer-
tainty can be reduced and precision im-
proved by buffering below pH 4.0—
.even as low as pH 3.0.
Regardless of the method of endpoint
.detection, either phenylarsine oxide or
thiosulfate may be used as the standard
reducing reagent. The former is more
stable and is preferred.
The SNORT colorimetric. DPD ti-
trimetric and colorimetric, and the LCV
-------
methods (D, E. F, and G, respectively)
are applicable to the determination of
total available chlorine in polluted wa-
ters. In addition, both DPD procedures
and the amperomerric titration and
SNORT methods allow for the estima-
tion of the monochloramine and di-
chloramine fractions. Since these meth-
ods depend on the stoichiometric
production of iodine, polluted waters
containing iodine-reducing substances
may not be analyzed accurately by these
methods.
In all of the colorimetric procedures,
compensate for color and turbidity by
use of color and turbidity "blanks" in vi-
sual or spectrophotometric determina-
tions.
2. Sampling and Storage
Chlorine in aqueous solution is not
stable, and the chlorine content of sam-
ples or solutions, particularly weak solu-
tions, will decrease rapidly. Exposure to
sunlight or other strong light or agita-
tion will accelerate the reduction of chlo-
rine. Therefore, start chlorine determi-
nations immediately after sampling,
avoiding excessive light and agitation.
Do not store samples to be analyzed for
chlorine.
409 D.
Stabilized Neutral Orthotolidine (SNORT)
Method
1. General Discussion
a. Principle: Orthotolidine is quite
stable in the reduced form when stored
in brown bottles in the presence of hy-
drochloric acid. However, the stability
of oxidized Orthotolidine decreases as the
pH increases. For this reason, ortho-
tolidine usually has been used at a pH of
1.3 or less. As the pH increases, the rate
of reaction of Orthotolidine with com-
bined chlorine, iron, and nitrite becomes
slower and their interference essentially
disappears at pH 7. Anionic surface-ac-
tive agents stabilize the color developed
by free chlorine and Orthotolidine at pH
7.0. "Aerosol OT."* sodium di(2-
ethyl-hexy!) sulfosuccinate. is the best
stabilizing reagent. The optimum con-
centration of stabilizer is 40 mg for each
100 ml of sample plus reagents.
* A trademark of the American Cvanamid Cu.
-------
The ratio by weighc of orthocolidine
dihydrochloride to chlorine must be at
least 8 to 1. With the concentration of
orthotolidine recommended in the pro-
cedure, the chlorine concentration must
not exceed 6 mg/1.
The pH of the final solution must be
between 6.5 and 7.5 to minimize low-
pH interference and high-pH fading. If
the pH of the sample is less than 5 or
greater than 9, and the alkalinity is
greater than I 50 or the acidity greater
than 200 mg/1. check the final pH of
the solution. If the alkalinity is high and
the final pH does not lie within the
range of 6.5 to 7.5, adjust the sample
pH to this range before analysis.
To insure correct color development,
minimum interference, a pH of 6.5 to
7.5, and a ratio of orthotolidine to free
chlorine of at least 8 to 1, the sample
must be added to the reagents.
The reaction time and temperature
are much less important in this method
than in other colorimetric methods for
chlorine determination.-- Nevertheless;-
for extremely large combined chlorine to
free chlorine ratios, high temperature
and long waiting times are undesirable.
At J 5 C a 1-mg/I monochloramine solu-
tion produces a false free chlorine resid-
ual of 0.01 mg/I per min. At high tem-
perature and for long waiting rimes,
color fading may become important, es-
pecially at levels below 0.1 mg/1 of free
chlorine. A 1-mg/l solution of free chlo-
rine fades at the rate of 0.005 mg/1 per
min at J5 C.
Iodide can be added in neutral solu-
tion to measure monochloramine and in
acidic solution to measure dichloramine.
The reaction of iodide and chloramine
yields a concentration of iodine equiva-
lent to the chloramine. In the color-
imetric procedure, orthotolidine is
present with the chloramine when io-
dide is added and the iodine produced by
the chloramine is immediately reduced
back to iodide and acts as a catalyst in
generating an amount of blue ortho-
tolidine equivalent to the original
chloramine present. Because of this cat-
alytic effect, lesser amounts of iodide are
required than in amperometric titra-
tion; this improves the separation of the
monochloramine and dichloramine
fractions.
b. Interference: When orthotoiidine
or any other chromogenic reagent is
used to measure residual chlorine,
strong oxidizing agents of any kind in-
terfere. Such interferences include bro-
mine, chlorine dioxide, iodine, man-
ganic compounds, and ozone. However,
the reduced forms of these compounds—
bromide, chloride, iodide, manganous
ion, and oxygen—do not interfere. Re-
ducing agents such as ferrous com-
pounds, hydrogen sulfide, and oxidiz-
- able organic-matter do 7ioHnterfere-
-------
using a blank. In the presence of more
than 10 Mg/1 manganic manganese, a
blank is prepared by adding 5 ml so-
dium arsenite to a 100-ml sample. This
sample is added to the reagents, as
usual, and this blank is used as a refer-
ence in measuring the free chlorine
present, either by zeroing the photome-
ter with this blank or by using the blank
as a reference when making color com-
parison.
If nitrogen trichloride is present, half
reacts as free available chlorine but the
remainder does not interfere in the
monochioramine and dichloramine
measurements. Many different organic
chloramines are possible. The extent to
which these organic chloramines inter-
fere in the monochioramine or dichlora-
mine steps depends on the nature of the
organic compound; they may appear in
either or both fractions.
c. Minimum detectable concentra-
tion: Approximately 10 /ig/1 free chlo-
rine.
2. Apparatus.
ColorJmctric equipment: One of the
following is required:
a. Filter photometer, providing a
light path of 1 cm or longer for < 1 mg/
1 free chlorine residual, or a light path
from I to 10 mm for free chlorine resid-
ual > 1.5; also equipped with a red filter
having maximum transmission in the
range of 600 to 650 nm.
b. Spectropbotometer, for use at 625
nm, providing a light path noted in the
paragraph above.
2, Reagents
a. Cblorine-demand-jree distilled ti-a-
ter: Add sufficient chlorine to distilled
water to destroy the ammonia and ni-
trite. The amount of chlorine required
will be about 10 rimes the amount of
ammonia nitrogen present; produce an
initial residual of more than 1.0 mg/1
free chlorine. Ler the chlorinated dis-
tilled water stand overnight or longer-.
then expose to direct sunlight until all
residual chlorine is discharged.
b. Neutral onbotolidine reagent: Add
5 ml cone HQ to 100 ml chlorine-de-
mand-free distilled water. Add 10 ml of
this acid solution, 20 mg mercuric chlo-
ride, HgQ:. 30 mg disodium ethyl-
enediamine tetraacetate dihydrate, also
called (ethylenedinitrilo)-tetraacetic acid
sodium salt, and 1.5 g orthotolidine
dihydrochloride to chlorine-demand-
free distilled water and dilute to 1 1.
Store in a brown bottle or in the dark at
room temperature. Protect at all times
from direct sunlight. Use no longer than
6 months. Avoid contact with rubber.
Do not let the temperature fall below 0
C because the resulting crystallization of
orthotolidine-can lead-to-deficient-sub--*
sequent color development.
CAUTION: Handle this chemical ~ith
extreme care. Never use a mouth pipet
for dispensing this reagent, but rely on
an automatic dropping or safety pipet to
measure the necessary volumes. Avoid
inhalation or exposure to the skin.
c. Buffer-stabilizer reagent: Dissolve
34.4 g dipotassium hydrogen phos-
phate. KiHPO*, 12.6 g potassium di-
hydrogen phosphate, KH:PO<, and 8.0
g'"Aerosol OT," 100% solid Ji(2-ethyl-
hexyDsulfosuccinate, in a solution of 500
ml chlorine-demand-free water and 200
ml diethylene glycol monobutyl ether.
Dilute to 1 1 with chlorine-demand-free
water.
d. Potassium iodide solution: Dissolve
-------
in the photometer tube or a 250-ml
beaker on a magnetic srirrer. Mix the
reagent slightly and add the sample to
the reagents with gentle stirring. Meas-
ure the percent transmittance and con-
vert to absorbance at 625 nm. The
value obtained (A) from the calibration
curve represents the free chlorine resid-
ual. To minimize possible interference
from high concentrations of combined
chlorine and high-temperature fading,
complete mixing of the sample with the
reagents and reading on the photometer
within approximately 2 min.
c. Monocbloramine: Return any por-
tion used for measuring free chlorine in
1f4£ to the sample. Add, with stirring.
0.5 ml KI solution to each 100-ml
sample, or a similar ratio for other
sample volumes. Again measure the
transmittance of the residual free chlo-
rine plus the monochloramine and ob-
tain the value (fi) from the calibration
curve.
d. Dicbioramine: Return any portion
used for measuring the-monochloramine-"
in If4c to the sample. Add. with stirring,
1 ml H:SO4 solution to each 100-ml
sample, or a similar ratio for other
sample volumes. After 30 sec for color
development add 1 ml sodium carbo-
nate solution slowly with stirring or un-
til a pure blue solution returns. Measure
the transmittance of the total residual
chlorine—free chlorine, monochlora-
mine, and dichloramine—and obtain
the value (C) from the calibration curve
with a slight dilution correction.
e. Compensation for interferences:
Compensate for the presence of natural
color or turbidity as well as manganic
compounds by adding 5 ml arsenite to
100 ml sample. Add this blank sample
to the reagents as above. Use the color of
0.4 g KI in chlorine-demand-free dis-
tilled water and dilute to 100 ml. Store
in a brown glass-stoppered bottle, pref-
erably in a refrigerator. Discard when a
yellow color develops.
e. Sulfuric acid solution: Cautiously
add 4 ml cone H:SO4 to chlorine-de-
mand-free distilled water and dilute to
100ml.
f. Sodium carbonate solution: Dis-
solve 5 g Na:CC>3 in chlorine-demand-
free distilled water and dilute to 100 ml.
g. Sodium arsenite solution: Dissolve
5.0 g NaAsO: in distilled water and di-
lute to 1 1. (CAUTION: Toxic—take care
to avoid ingestion.)
4. Procedure
j. Calibration of photometer: Con-
struct a calibration curve by making di-
lutions of standardized hypochlorite so-
lution prepared as directed under
Chlorine Demand, Section 410A.3a.
Take special precautions when diluting
to low concentrationS'because of possible *
consumption of small amounts of chlo-
rine by trace impurities. Use chlorine-
demand-free water in making the dilu-
tions. Expose all glassware to be used in
the dilutions to water containing at least
10 mg/1 of chlorine and leave it in con-
tact for a few hours. Rinse with chlo-
rine-demand-free water. Develop and
measure the colors as described below
for the sample.
b. Color development of free chlo-
rine: Use 0.5 ml neutral orthotolidine
and 0.5 ml stabilizer-buffer reagent
with 10-ml samples; 5 ml neutral or-
thotolidine and 5 ml stabilizer-buffer
reagent with 100 m!; and the same ratio
for other volumes. Place the neutral or-
thotolidine and stabilizer-buffer mixture
-------
che blank to set 100% transmirtance or
zero absorbance on the photometer.
Measure all samples in relation to this
blank. Read from the calibration curve
the concentrations of chlorine present in
the sample.
5. Calculation
mg/1 free residual chlorine
=/4, including Vi trichloraminc if present
me/1 monochloraminc
mg/1 Jichloramine
me/I ratal chlorine
= 5-/l.asme/lCl
• l.OJ C-fl.asmg/ICl
= 1.05 C. as me/I Q
6. Precision and Accuracy
See Tables 409: I through IV pre-
ceding and the general introduction to
Section 409.
-------
INORGANIC NON-METALS (400)
409 A. lodometric Method !
1. General Discussion
• a. Principle: Chlorine will liberate
free iodine from potassium iodide solu-
tions at pH 8 or less. The liberated io-
dine is titrated with a standard sol tion
of sodium thiosulfate, with starch as the
indicator. The reaction is preferably car-
ried out at pH 3 to 4.
b. Interference-. Although the neutral
titration minimizes the interfering effect
of ferric, manganic, and nitrite ions, the
acid titration is preferred; it is most ac-
curate for determination of total avail-
able residual chlorine. Use acetic acid for
the acid titration; use sulfuric acid only
when interfering substances are absent;
never use hydrochloric acid.
c. Minimum delectable concentra-
tion: The minimum detectable concen-
tration approximates 40 Mg/l Cl if
0.0IN sodium thiosulfate is used with a
500-ml sample.
2. Reagents
a. Acetic acid, cone (glacial).
b. Potassium iodide, KI, crystals.
c. Standard sodium tbiosulfate, O.lN:
Dissolve 25 g NaiS^Cb-JHiO in 1 1
freshly boiled distilled water and stand-
ardize the solution against potassium
biniodate or potassium dichromate after
at least 2 wk storage,. Use boiled distilled
water and add a few milliliters CHCb
to minimize bacterial decomposition of
the thiosulfate solution.
Standardize the 0. IN sodium thiosul-
fate by one of the following procedures:
1) Biniodate method—Dissolve
3.249 g anhydrous potassium biniodate,
, of primary standard qual-
ity,* in distilled water and dilute to
1,000 ml to yield a 0.1OOON solution.
Store in a glass-stoppered bottle.
To 80 ml distilled water, add, with
constant stirring, I ml cone H:SO«,
10.00 ml 0.1000N KH(lCb): and 1 g
KI. Titrate immediately with O.lN
NazSiOs titrant until the yellow color of
the liberated iodine is almost discharged.
Add 1 ml starch indicator solution and
continue titrating until the blue color
disappears.
2) Dichromate method—Dissolve
4.904 g anhydrous potassium dichro-
mate, KjCnO?, of primary standard
quality, in distilled water and dilute to
1,000 ml to yield a 0.1 OOON solution.
Store in a glass-stoppered bottle.
Proceed as in the biniodate method,
with the following exceptions:. Sub-
stitute 10.00 ml 0.1 OOON KjCnOi for
the KH(IOj)j and let the reaction mix-
ture stand 6 min in the dark before ti-
trating with the 0. IN NaiSzOj titrant.
Normality NaiS:Oj
1
ml Na:SiO> consumed
C d. Standard sodium tbiosulfate ti-
trant, 0.01 N or 0.025N: Improve the
stability of 0.01 N or 0.025N NazS:Oj
by diluting an aged 0.IN solution, made
as directed above, with freshly boiled
distilled water. Add a few milliliters
CHCb or 0.4 g sodium borate and 10
mg mercuric iodide/1 solution. For ac-
curate work, standardize this solution
daily in accordance with the directions
given above, using 0.0IN or 0.025N
KHUOih or KiCnCK To speed up
• G. F. Smith Chemical Company. Cotumbiu.-
Ohio. or equivalent.
-------
CHLORINE (RESIDUALJ/lodornetric Method 1
317
operations where many samples must be
titrated use an automatic buret of a type
in which rubber does not come in con-
tact with the solution. Standard sodium
thiosulfate titrants, O.OIOO/V and
0.0250A/, are equivalent, respectively,
to 3 54.5 M§ and 886. J M§ available Cl/
1.00 ml.
e. Starch indicator solution: To J g
starch (potato, arrowroot, or soluble),
add a little cold water and grind in a
mortar to a thin paste. Pour into 1 t of
boiling distilled water, stir, and let settle
overnight. Use the clear supernate. Pre-
serve with 1.25 g salicylic acid, 4 g zinc
chloride, or a combination of 4 g sodium
propionate and 2 g sodium azide/I
starch solution. Some commercial starch
substitutes are satisfactory.
/ Standard iodine, 0.1N: Refer to
Method C,1fJa2).
g. Dilute standard iodine, 0.02 8 2/V:
Refer to Method C, fl JaJ).
3. Procedure
a. Volume of sample: Select a sample
volume that will require no more than
20 ml O.OI/V Na:S:Oj. Thus, for resid-
ual chlorine concentrations of I mg/1 or
less, take a 1,000-ml sample; for a chlo-
rine range of 1 to 10 mg/1, a 500-ml
sample; and above 10 mg/1, propor-
tionately less sample.
b. Preparation for titradon: Place 5
ml acetic add, or enough to reduce the
pH to between 3.0 and 4.0, in a flask or
white porcelain casserole. Add about 1 g
KI estimated on a spatula. Pour in the
sample and mix with a stirring rod. Add
chlorine-demand-free distilled water if a
larger volume is preferred for titration.
c. Titration: Titrate away from direct
sunlight. Add 0.025N or 0.0IN thiosul-
fate from a buret until the yellow color
of the liberated iodine is almost dis-
charged. Add 1 ml starch solution and
titrate until the blue color is discharged.
If the titration is made with 0.02 5iV
thiosulfate instead of 0.01, then, with a
I-I sample, 1 drop is equivalent to about
50 Mg/1- It is not possible to discern the
end point with greater accuracy. If a
500-ml sample is titrated, 1 drop will
correspond to about 100 Mg/1. which is
within the limit of sensitivity. Hence.
use of 0.02 5N solution is acceptable.
(Many laboratories have this on hand.
d. Blank titration: Correct the result
of the sample titration by determining
the blank contributed by such reagent
impurities as: (a) the free iodine or io-
date in the potassium iodide that liber-
ates extra iodine;' or (b) the traces of re-
ducing agents that might reduce some of
th'elodine liberated.
Take a volume of distilled water cor-
responding to the sample used for titra-
tion in 1fs Ja-c, add 5 ml acetic add. 1 g
KI, and 1 ml starch solution. Perform
either Blank Titration A* or B, which-
ever applies.
1) Blank titration A—If a blue color
develops, titrate with 0.01N or 0.02 5N
sodium thiosulfate to the disappearance
of the blue and record the result.
2) Blank Titration B—If no blue
color occurs, titrate with 0.0282N io-
dine solution until a blue color appears.
Back-titrate with 0.0 hV or 0.02 5;V so-
dium thiosulfate and record the differ-
ence as Titration B.
Before calculating the chlorine, sub-
tract Blank Titration A from the sample
titration; or,- if-necessary, add the net -
equivalent value of Blank Titration B.
-------
Ferric Alum Indicator
Dissolve 28.0 g of ferric ammonium sulfatc FeNhU ( S04 )2 • 12 H70 in 70
ml of hoi water. Cool, filter, add 10 nd of concentrated nitric acid( HN03 ),
and dilute to 100 ml in a volumetric flask.
Nitric Acid (8N)
Prepare N0x-frec nitric acid by adding 100 ml of HNOs to 100 ml of water
and boiling in a flask until the solution is colorless. Store in 3 glass reagent
bottle.
Nitrobenzene
Reagent grade.
Sodium Chloride (Primary Standards)
1. NaCl (0.1 N) - Dissolve 5.846 g of dried sodium chloride (NaCI) in
water and dilute to 1 b'tcr in a volumetric flask.
2. Nad ( 0.01 N ) - Dissolve 0.5846 g of dried NaCl in water and dilute to
1 liter in a volumetric flask, or dilute 100 ml of 0.1 N NaCI to 1 liter.
Ammonium Thiocyanate
1. Nrt,CNS (0.1 N) - Dissolve 8 g of NH«CNS in water and dilute to 1
liter in a volumetric flask.
2. NH«CNS ( 0.01 N) - Dilute 100 ml of 0.1 N NH.CNS to 1 liter in a
volumetric flask, or dissolve 0.8 g Nrl,CNS in 1 liter of distilled water.
Silver Nitrate
1. AgNOj (0.1 N) - Dissolve 17.0 g of silver nitrate ( AgN03 ) in water
and dilute to 1 liter in a volumetric flask. Transfer to an amber reagent
bottle. Standardize this solution against 0.1 .N.NaG solution,, according
to the Vplhard titration.9
2. AgN03 'C0.01 N ) - Dissolve 1.7 g of AgN03 in water and dilute to 1
liter in a volumetric flask. Transfer to an amber reagent bottle.
Standardize this solution against standard 0.01 N Nad solution,
according to the Volhard titration.
Starch Solution (Iodine Indicator), 1.0%
Make a thin paste of 1 g of soluble starch in cold water and pour into 100
ml of boiling water while stirring. Boil for a few minutes. Store in a
glass-stoppered bottle.
Standard Iodine Solution (0.1 N )
Dissolve 12.69 g of resublimed iodine ( I2 ) in 25 ml of a solution
containing 15 g of iodate-frce potassium iodine (KI); dilute to 1 liter in a
volumetric flask. Standardize this solution against a standard tluosulfate
solution using starch as an indicator. This solution should be stored in an
amber reagent bottle and refrigerated when not in use.
Apparatus
(See Figures A-1 andA-2)
Appendix A 35
-------
409 E. DPD Ferrous Titrimetric Method
1. General Discussion
a. Principle: N,N-diethyl-p-phenyl-
enediamine (DPD) is superior to neu-
tral orthotolidine as an indicator in the
ferrous method. The colors produced are
more stable, fewer reagents are re-
quired, and a full response in neutral so-
lution is obtained from dichloramine. In
the titrimetric procedure, decolorization
by standard ferrous ammonium sulfate
(FAS) titrant is instantaneous, thereby
enabling each step to be performed more
rapidly. Where complete differentiation
is not required, the procedure may be
simplified further to give only free and
combined available chlorine or total re-
sidual available chlorine.
In the absence of iodide ion, free
available chlorine reacts instantly with
the N,N-diethyl-p-phenylenediamine
(DPD) indicator to produce a red color.
Subsequent addition of a small amount
of iodide ion acts catalytically to cause
monochloramine to produce color. Fur-
ther addition of iodide ion to excess
evokes a rapid response from dichlora-
mine. Unlike the reaction with neutral
orthotolidine, any nitrogen trichloride
present no longer 'displays color with
free available chlorine but is included
with dichloramine. However, if iodide
ion is added before DPD, a proportion
of the nitrogen trichloride appears with
free available chlorine. A supplemen-
tary procedure based on this alteration
of the order of adding the reagents thus
permits the estimation of nitrogen tri-
chloride.
Chlorine dioxide appears, to the ex-
tent of one-fifth of its total available
chlorine content, with free available
chlorine. A full response from chlorine
dioxide, corresponding to its total avail-
able chlorine content, may be obtained if
the sample first-is acidified in the pres-
ence of iodide ion and subsequently is
brought back to an approximately neu-
tral pH by the addition of bicarbonate
ion. Bromine, bromamine, and iodine
react with DPD indicator and appear
with free available chlorine. DPD pro-
cedures for the determination of these
halogens and related compounds have
been developed.
b. pH control: For accurate results
careful pH control is essential. At the
proper pH of 6.2 to 6.5, the red colors
produced may be titrated to sharp color-
less end points. Carry out the titration as
-------
soon as the red color is formed in each
step. Too low a pH in the first step will
tend to make the monochloramine show
in the free-chlorine step and the di-
chloramine in the monochloramine step.
Too high a pH may cause dissolved oxy-
gen to give a color.
c. Temperature control: In all meth-
ods for differentiating free chlorine from
chloramines, the higher the temperature
the greater the tendency for the chlora-
mines to react with the reagents and
thus lead to increased apparent free-
chlorine results after a fixed time inter-
val. Exceptions to this are the titration
methods, probably because of the speed
with which the titration is completed
compared with the 2 to 3 min required
for the colorimetric measurement to be
made. The DPD methods are among
those least affected by temperature.
d. Interference: The only interfering
substance likely to be encountered in
water is oxidized manganese. To correct
for this, place 5 ml buffer solution, one
small crystal of potassium iodide, and
0.5 ml sodium arsenite solution (500
mg NaAsO: plus 100 ml distilled wa-
ter) in the titration flask. Add 100 ml
sample and mix. Add 5 ml DPD in-
dicator solution, mix, and titrate with
standard ferrous ammonium sulfate ti-
trant until any red color is discharged.
Subtract the reading from reading A ob-
tained by the normal procedure as de-
scribed in 1f3al) of this method or from
the total available chlorine reading ob-
tained in the simplified procedure as
given in If 3a4). If the combined reagent
in powder form (see below) is used, add
the potassium iodide and arsenite first to
the sample and mix, then add the com-
bined buffer-indicator reagent after-
wards.
Interference by copper up to approxi-
mately 10 mg/1 copper is overcome by
the EDTA incorporated in the reagents.
The presence of EDTA enhances the
stability of the DPD indicator solution
by retarding deterioration due to oxida-
tion, and in the test itself provides virtu-
ally complete suppression of dissolved
oxygen errors by prevention of trace
metal catalysis.
2. Reagents
a. Phosphate buffer solution: Dissolve
24 g anhydrous disodium hydrogen
phosphate, NajHPCh, and 46 g an-
hydrous potassium dihydrogen phos-
phate, KH:PO4, in distilled water.
Combine with 100 ml distilled water in
which 800 mg disodium ethyl-
enediamine tetraacetate dihydrate, also
called (ethylenedinitrilo) tetraacetic acid
sodium salt, have been dissolved. Dilute
to 1 1 with distilled water and add 20
mg HgCb to prevent mold growth and
to prevent interference in the free avail-
able chlorine test caused by any trace
amounts of iodide in the reagents:
b. N,N-Diethyl-p-phenylenediamine
(DPD) indicator solution: Dissolve 1 g
DPD Oxalate,* or 1.5 gp-amino-N:N-
diethylaniline sulfate.t in chlorine-free
distilled water containing 8 ml 1 + 3
HzSO4-and 200 mg disodium ethyl-
enediamine tetraacetate dihydrate, al-
so called (ethylenedinitrilo)tetraacetic
acid sodium salt. Make up to 1 1, store
in a brown glass-stoppered horde, and
discard when discolored. (The buffer
and indicator sulfate are commercially
* Eastman chemical No. 7102, or equivalent.
t British Drug House chemical available from Gal-
lard- Schlesinger Chemical Mfg. Corp.. J84 Mineola
Avenue, Carle Place, N.Y. 11514.
-------
available as a combined reagent in
stable powder form.) CAUTION: The
oxalate is toxic—take care to avoid in-
gestion.
c. Standard ferrous ammonium mi-
fate (FAS) titrant: Dissolve 1.106 g
Mohr's salt, FeWH^SChh^hhO in
distilled water containing 1 ml of 1 + 3
H2.SO4 and make up to 1 1 with freshly
boiled and cooled distilled water. This
primary standard may be used for 1
month, and the titer checked by potas-
sium dichromate. .The FAS titrant is
equivalent to 100 jig Cl/1.00 ml.
d. Potassium iodide, KI, crystals.
e. Potassium iodide solution: Dissolve
500 mg KI and dilute to 100 ml, using
freshly boiled and cooled distilled water.
Store in a brown glass-stoppered bottle,
preferably in a refrigerator. Discard the
solution when a yellow color develops.
3. Procedure
The quantities given below are suit-
able for concentrations of total available
chlorine up to 4 mg/1. Where the total
chlorine exceeds 4 mg/1, use a smaller
sample and dilute to a total volume of
100 ml. Mix the usual volumes of buffer
reagent and DPD indicator solution, or
the usual amount of DPD powder, with
distilled water before adding sufficient
sample to bring the total volume to 100
ml.
a. Free available chlorine or chlora-
mine: Place 5 ml each of buffer reagent
and DPD indicator solution in the titra-
tion flask and mix (or use about 500 mg
of DPD powderX Add 100 ml sample
and mix.
1) Free available chlorine—Titrate
rapidly with standard FAS titrant until
the red color is discharged (reading A).
2) Monochloramine—Add one very
small crystal of KI and mix; or if the
dichloramine concentration is expected
to be high, add 0.1 ml (2 drops) KI so-
lution and mix. Continue titrating until
the red color is again discharged (read-
ing 5).
3) Dichloramine—Add several crys-
tals KI (about 1 g) and mix to dissolve.
Let stand for 2 min and continue titrat-
ing until the red color is again dis-
charged (reading C). In the case of very
high dichloramine concentrations, let
stand 2 min more if color driftback in-
dicates slightly incomplete reaction.
When dichloramine concentrations are
not expected to be high, use half the
specified amount of KI.
4) Simplified procedure for free and
combined available chlorine or total
available chlorine—Omit step 2) above
in order to obtain monochloramine and
dichloramine together as combined
available chlorine. To obtain total avail-
able chlorine in one reading, add the full
amount of KI at the start, with the spec-
ified amounts of buffer reagent and
.DPD indicator, and titrate after 2 min
standing.
b. Nitrogen trichloride: The absence
of color in the first step indicates the ab-
sence of nitrogen trichloride (and of
chlorine dioxide). Nitrogen trichloride,
readily identified by its distinctive odor,
may be estimated by the following pro-
cedure.- Place a small crystal of KI in a
titration flask. Add 100 ml sample and
mix. Then add the contents to a second
flask containing 5 ml each of buffer re-
agent and DPD indicator solution (or
about 500 mg DPD powder direct to
the first flask). Titrate rapidly with
standard FAS titrant until the red color
is discharged (reading D).
-------
Monochloramine is unlikely to be
present with nitrogen trichloride. If high
concentrations of dichloramine are
present, use KI solution as in 1J3a2) in
place of a KI crystal.
4. Calculation
For a 100-ml sample, 1.00 ml stand-
ard FAS titram=1.00 mg/I available
residual chlorine.
Reading NCIi Absent NCb Present
A
B-A
C-B
D
2(D-A)
C-D
free Cl
NHjCl
NHCh
—
—
—
freed
NH:C1
NHCh-f
WNCh
freeQ+
WNCh
NCb
NHCh
Should monochloramine be present
with nitrogen trichloride, which is un-
likely, it will be included in reading D,
in which case NCb is obtained from
2(D-5).
Chlorine dioxide, if present, is in-
cluded in reading A to the extent of one-
fifth of its total available chlorine con-
tent.
In the simplified procedure for free
and combined available chlorine, only
reading .4 (free Cl) and reading C-(total
CD are required. Combined available
chlorine is obtained from C-A.
The result obtained in the simplified
total available chlorine procedure corre-
sponds to reading C.
5. Precision and Accuracy
See Tables 409:1 through IV preced-
ing and the general introduction to Sec-
tion 409.
-------
ATMOSPHERIC EMISSIONS
FROM HYDROCHLORIC ACID
MANUFACTURING PROCESSES
Cooperative Study Project
Manufacturing Chemists' Association, Inc.
and
Public Health Service
U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
Public Health Service
Consumer Protection and Environmental Health Service
National Air Pollution Control Administration
Durham, North Carolina
September 1969
For sale by the Superintendent of Documents, T/.S. Government Printing Office
Washington. D.C. 20402 - Price 35 cents
-------
APPENDIX A. SAMPLING AND ANALYTICAL
TECHNIQUES
DETERMINATION OF HYDROGEN CHLORIDE AND
CHLORINE IN STACK GAS.
The method discussed herein is used to determine the presence of hydrogen
chloride in the stack gases from hydrochloric acid manufacturing processes.
Samples are collected from the gas stream in either midget impingcrs or a
grab-sampling bottle containing sodium hydroxide, and hydrogen chloride is
determined by the Volhard titration.9 ff chlorine is suspected to be present in
the stack emission, another sample is collected in a similar manner, except that
a known quantity of alkaline arsenite absorbing reagent is substituted for
sodium hydroxide. Chlorine is reduced to chloride by arscnite, and is measured
by titration of the unconsumed arscnite with standard iodine solution. Total
chloride content of the sample is determined by the Volhard titration.
Hydrogen chloride concentration is calculated by subtraction of the chlorine
concentration from total chloride concentration. This method is applicable in
determining hydrogen chloride and chlorine concentrations ranging from 10
ppm to percentage quantities.
Reagents
All chemicals must be ACS analytical reagent-grade.
Water
Deionized or distilled water.
Absorbing Reagents:
1. Sodium hydroxide (IN)- Dissolve 40 g-of sodium hydroxide in water
and dilute to 1 liter. This reagent is used when the stack gas
concentration of hydrogen chloride is suspected to be less than 1,000
ppm.
'L Sodium hydroxide ( 2.5 H) — Dissolve 100 g of sodium hydroxide in
water and dilute to 1 liter. This reagent-is used when the stack gas
concentration of hydrogen chloride is suspected to be greater than 1,000
ppm,
3. Alkaline arsenite ( 1 N NaOH and 0.1 N NaAsO2 ) Dissolve 40 g of
NaOH and 6.5 g of sodium arsenite ( NaAs02 ) in water and dilute to 1
liter in a volumetric flask. This reagent is used when the stack gas
concentration of hydrogen chloride and G? is suspected to be less than
1,000 ppm.
4. Alkaline arsenite ( 2.5 N NaOH and 0.5 N NaAsOj ) Dissolve 100 g of
NaOH and 32.5 g of NaAsO? in water and dilute to 1 liter in a
volumetric flask. This reagent is used when the stack gas concentration
of hydrogen chloride and Cla is suspected to be greater than 1,000 ppm.
34 HYDROCHLORIC ACID EMISSIONS
-------
Sampling Probe
10-mm Pyrex® glass tubing of any convenient length.
Filter
A small fiberglass filter may be fitted to the probe inlet when paniculate
matter is present in Ihc gas stream being sampled.
Healing Tape
Used to heat probe.
Variable Voltage Regulator
Used to regulate probe heating.
Dry Gas Meter
Readable to the nearest 0.01 cubic foot.
Vacuum Pump
A diaphragm-type pump rated at 15 liters per minute.
Absorbers
1. Midget impinger — An all-glass midget impingcr sampling train capable of
removing hydrogen chloride and chlorine from a gas sample may be used
when sampling stack gas suspected of containing less than 0.1 percent
hydrogen chloride or chlorine.
The sampling train should consist of a probe, four midget impingers, a
gas drying tube, a vacuum pump, and a flow meter, as illustrated in
Figure A-l.
PAOII
WITH HBATING
CLEMENT
ICZ UTH I
MCTCR
PUMF
Figure A-l. Impinger gas sampling train.
A Pyrex® glass tube serves as the probe. It should have a ban joint on the
outlet to which other glassware can be easily connected. The probe
should be wrapped with nickel-chromium heating wire, insulated with
glass wool and installed into a protective 1-inch-diameter stainless steel
36
HYDROCHLORIC ACID E.MISSIONS
-------
tube. During sampling, the heating wire is connected to a calibrated,
variable transformer to maintain a temperature of up to 250 F on the
probe wall. A heating tape may also be used to heat the probe.
Four midget glass impingcrs are connected in series to the probe outlet
with glass ball joints. The first three impingers each contain 15 ml of
absorbing reagent. The fourth impingcr is left dry to catch any material
that is carried over from the other impingcrs. All four impingcrs arc
cooled in an iccwatcr bath.
Gases leaving the impingcrs are dried as they pass through a tube of silica
gel They then are pumped to tiie airtight vacuum pump and gas meter.
Sampling rates are regulated by using a valve to adjust the flow through
the pump.
2. Grab-sampling bottle An accurately calibrated 2-liter glass boi !c
equipped with Teflon® stopcocks (Figure A-2) should be used when
•sampling percentage quantities of hydrogen chloride or chlorine.
Absorbing reagent is added to the grab-sample bottle after the sample has
been collected.
2-UTER FLASK
ATTACH "•«*- -^ TO
PROBE VACUUM
Figure A-2. Grab sample bottle.
Dispenser (Absorbing Reagent)
A 100-milliliter round-bottom flask, modified with a Teflon© stopcock and
ball joint extension (see Figure A-3) is used as a dispenser for absorbing
reagent. It is used to add jcagcntao the-grab-sampling buttle-after-the sample
has been collected.
Analytical Procedure
Collection of Samples
1. Midget impinger train:
Pipet 15 nil of absorbing reagent into each of the three midget impingcrs.
Heat probe to prevent moisture condensation. Start pump and sample at
a rate of 1 to 3 liters per minute for'at least 60 minutes.
2. Grab sampling:
Rush probe and draw about 20 liters of gas through the sample bottle.
Pipct 25 ml of absorbing reagent into the dispenser and add lo the
grab-sample bottle following collection of the sample. Shake bottle
thoroughly for about 2 minutes to insure complete absorption.
3. Transfer the contents of the impingers or grab-sampling bottle to a
sample container, such as a polyethylene bottle, containing deioni/cd or
distilled water.
Appendix A
-------
C = HC1, mg = net ml AgN03 X T X F
T = hydrogen chloride equivalent of standard AgN03
( T = 3.65 mg HQ/ml for 0.1 N AgN03 )
( T = 0.365 mg HCl/ml for 0.01 N AgN03 )
_ _ sample volume, ml
aliquot volume, ml
Convert the volume of gas sampled to the volume at standard conditions of
70° F and 29.92 in. Hg.
V = VX— £-X 530°R
5 29.92 ( t + 460° R)
V = volume of gas sampled, as measured on dry gas meter or equal to
volume of grab sample bottle-liters
P = barometric pressure (in. Hg) or absolute pressure at gas meter
t - average temperature of gas sampled, ° R
Determine the concentration of hydrogen chloride in the gas sample by the
following formula:
ppm
Vs
662 = M/mg of HC1 at 70° F and 29.92 in. Hg
C = concentration of HC1, mg
V , = volume of gas sampled in liters at 70° F and 29.92 in. Hg
Procedure R
Analysis of Hydrogen Chloride in the Presence of Chlorine
Pipct an aliquot of the sample into a 250-ml Erlenmeycy flask and proceed
with the Volhard titration for total chlorides as described under Procedure A.
A blank determination for chloride in the absorbing reagent (alkalinc-arscnitc
reagent) should be run simultaneously and subtracted from the sample results.
Pipet another aliquot of the sample into a 250-ml Erlenmeyer flask. Add a
few drops of phenolphthalein indicator, neutralize carefully with concentrated
hydrochloric acid, and cooL Add sufficient solid sodium bicarbonate
( NaHCOj ) to neutralize any excess hydrochloric acid, then add 2 to 3 g more.
Add 2 ml of starch indicator and titrate with 0.1 N iodine solution to the blue
cndpoint. For the reagent blank, determine the number of ml of 0.1 N 12
required to Citrate 25 ml of alkalinc-arscnitc absorbing reagent, as described
above.
Appendix A 39
-------
Calculations
Determine the number of milliliters of 0.1 N Ij required to titrate the entire
sample by the following formula:
Sample ( ml 0.1 N I2 ) = ( ml of 0.1 N 12 for aliquot) X F
_ volume of sample, ml
F s E
volume of aliquot, ml
Clj, mg = Blank ( ml 0.1 N Ia ) - Sample ( ml 0.1 N I, ) X 3.546
3.546 = chlorine equivalent of 0.1 N 12, mg
Convert the volume of gas sampled at standard conditions of 70° F, 2^.92
i. Hg, using the formula in Procedure A. Calculate the concentration of Clj in
the sample using the following formula:
ppm G2 by volume = ——-
" s
340 = 1/mg of Q7 at 70° F and 29.92 in. Hg
C = concentration of Clj, mg
V , = volume of gas sampled at standard conditions, liters
Determine the number of milligrams of hydrogen chloride present in the
sample by subtracting the number of milligrams of chlorine present, as
determined -by-the -iodine-tilrationr.fcomjthe..to.tal.number_of milligrams of
chloride present as determined by the Volhard titration. Calculate the
concentration of hydrogen chloride in parts per million using the formula in
Procedure A.
Total stack gas volume must also be measured in order to determine the
emissions on a weight basis. This may be done by measuring the gas velocity
with a pitot tube.
The following equations are used to determine gas velocity and gas volume:
29.92 „ 29.0 I
Vs=172(F)(V&Pavg) [VTS.X 2|2? X
*- s
Vg = velocity in feet per minute at stack conditions
F = pitot tube-correction factor (0.85 for type S)
Mg = molecular weight of stack gas
Tg = average stack gas temperature ° R
Pg = average stack gas pressure, in. Hg
AP = pitot tube manometer reading, in. water
40 HYDROCHLORIC ACID EMISSIONS
-------
The total stack gas volume-is then:
QS
A = stack area, sq ft
Qs = gas volume, ft3/min at 70° F and 29.92 in. Hg
The emissions on a weight per hour basis may then be determined by the
following equation:
W = ppm X 10-* X Q X 60 X %2LZl
* 387
W. = emissions, Ib/hr
ppm = parts per million by volume of contaminant
Qs = stack gas flow rate, scfrn
mol wt = molecular weight of contaminant
Clj = 70.92 HC1 = 36.4A
387 = volume (ft3) occupied by 1 !b mol at 70° F and 29.92 in. Hg
Discussion of Procedure:
The estimated error for the combined sampling and analytical procedure is
±10 percent. The precision of the analytical methods is ±2 percent on standard
samples containing Nad and NaAsOj .
The usual volumetric errors are encountered with the Volhard titraUon.
Premature endpoints may occur if the NH« CNS is not added by drops near the
equivalence point and the solution shaken before the next addition.
Nitrobenzene-is used-to effectivery
the precipitate and preventing reaction with the thiocyanate.1 ° Bivalent
mercury, which forms a stable complexion with the thiocyanate, and
substances that form insoluble silver salts interfere in the analysis and must be
absent from the sample. Titration should be made at temperatures below
25° C, as is customary in other titrations with thiocyanate.1 1
The chief source of error in the iodine titration of arsenite is the failure to
use sufficient bicarbonate to neutralize all the excess acid. If insufficient
bicarbonate is added, serious errors may be incurred because of a fading
endpoint. A reducing agent such as sulfur dioxide and oxidizing agents such as
iodine, nitrogen dioxide, and ozone interfere with the iodine titration and yield
high results when present in the stack gas sample.
ACID MIST SAMPLING.
The sampling apparatus is made up of a probe, a cyclone, a filter, four
impingers, a dry gas meter, a vacuum pump, and a flow meter, as shown in
Figure A-4.12'13
Appendix A 41
-------
6-30-81
APPENDIX A - REFERENCE TEST METHODS
* * * * *
METHOD . DETERMINATION OF TOTAL PARTICULATE, FREE CHLORINE,
AND TOTAL CHLORIDES FROM SECONDARY ALUMINUM SMELTERS
1. Applicability and Principle
1.1 Applicability. This method applies to the determination of total
particulate, free chlorine, and total chloride emissions from chlorine
demagging operations in secondary aluminum production facilities and in
other sources when specified in the regulations.
1.2 Principle. Total particulate, free chlorine, and total chloride
emissions are withdrawn isokinetically from the source using a modified EPA
Method 5 procedure. The total particulate and any particulate chlorides
present are collected on the glass fiber filter maintained at a temperature
not to exceed 93°C (200°F); free chlorine and the remaining chlorides'are''
collected in a 0.1 N KOH impinger solution. The particulate mass is deter-
mined gravimetrically after removal of combined water. Free chlorine is
measured in the impinger catch using the DPD Ferrous Titrimetric Method(1).
Total chloride is measured in the impinger catch, probe wash, and filter
extract using the Mercuric Nitrate Method^).
2. Range
2.1 For Determination of Total Free Chlorine by the DPD Ferrous Titri-
metric Method. The minimum chlorine concentration detectable by this
method has not been determined. The maximum for which this method is
suitable, as described here, is 4 mg/L. Where the total available chlorine
exceeds 4 mg/L, a smaller sample, diluted to a total volume of 100 ml,
should be used.
-1-
-------
6-30-81
2.2 For Determination of Total Chloride by the Mercuric Nitrate Method.
The mercuric nitrate method can be used to measure total chloride when the
sample to be titrated contains 0.15 to 10 mg Cl. For illustrative purposes,
a 10 ml sample containing 0.15 mg Cl would be equivalent to a concentration
of 15 mg/L, and a 10 ml sample containing 10 mg Cl would be equivalent to
a concentration of 1000 mg/L. Depending upon the type of sample, the
quantity in the portion to be titrated could be adjusted to the measurable
range by dilution or by evaporation of excess liquid.
3. Interferences
3.1 For Determination of Total Chlorine by the DPP Ferrous Titrimetric
Method. The only interferences for this method are copper and oxidized
manganese. Interference by copper up to approximately 10 mg/L is overcome
by the EDTA (ethylenediamine tetraacetate dihydrate) incorporated in the
reagents. A suitable correction for interference due to oxidized manganese
is provided in Reference 1. It is not included here because oxidized
manganese is not expected to occur in samples from secondary aluminum
smelters in high enough concentrations to cause significant interference.
The interference of combined chlorine is insignificant in the determination
of free chlorine except at high temperature and long waiting times.
This method has been adapted from a standard method for the measurement
of chlorine in water. In such cases, ammonia and certain nitrogenous
compounds are often present which can combine with the free chlorine to
form chloramines. While it is unlikely that such compounds will be en-
countered in analyses of samples of aluminum smel'ter -stack•-effluents",- the ---
potential interference which could be caused by such compounds and the
procedure for compensating for this interference are discussed in
Reference 1.
-2-
-------
6-30-81
3.2 For Determination of Total Chloride by the Mercuric Nitrate Method.
Bromide and iodide are titrated with mercuric nitrate in the same manner as
chloride. Chromate, ferric, and sulfite ions interfere when present in
excess of 10 mg/L. These should not present a problem in samples from a
secondary aluminum smelter.
Where metal ion interference•is expected, its presence can be es-
tablished by passing a sample aliquot through an ion exchange column of
Rexyn 101H. This procedure may also be used to eliminate such interference.
The presence of these ions is not expected in secondary aluminum reverberatory
furnace exhaust gases except possibly where the air pollution control scrubber
uses low quality water and there is droplet carryover.
4. Precision, Accuracy, and Stability
4.1 Precision -and Accuracy.
4.1.1 For De termination-ofr'-To'tal-'-Free -Chlorine-by—the- DP-D-Ferrous-?-^
Titrimetric Method. The precision and accuracy of the method, and several
other methods for chlorine analysis, were studied under the auspices of
the American Public Health Association, the American Water Works Associa-
tion, and the Water Pollution Federation. A more detailed discussion of
the study may be found Reference 3. Briefly, samples of known concentra-
tions of free and combined chlorine were distributed to 15 to 30
participating laboratories for analysis. The results of the two studies
are summarized in Reference 1. Sample concentrations for Study No. 2 were:
Sample No. 1, 440 jig/L free chlorine;
Sample No. 2, 980 jig/L free chlorine, and
Sample No. 3, 660 Pg/L combined chlorine.
The DPD Ferrous Titrimetric Method was one of the more accurate and
precise of those tested. Data presented in Reference 1 attribute a relative
-3-
-------
6-30-81
standard deviation of 39.8% for results of 19 different laboratory analyses
for 0.8 mg/L Cl£ by the DPD Ferrous Titrimetric Method.
4.1.2 For Determination of Total Chloride by the Mercuric Nitrate
Method. A synthetic unknown sample containing 241 mg/L chloride, 108 mg/L
Ca, 82 mg/L Mg, 3.1 mg/L K, 19.9 mg/L Na, 1.1 mg/L nitrate N, 0.25 mg/L
nitrite N, 259 mg/L sulf ate , and. 42.5 mg/L total alkalinity (contributed by
NaHC03) in distilled water was analyzed in 10 laboratories by the mercuri-
metric method, with a relative standard deviation of 3.3%, and a relative
error of 2.9%.
4.2 Stability.
4.2.1 For Determination of Total Free Chlorine by the DPD Ferrous
Titrimetric Method. Samples containing free chlorine are relatively
unstable and cannot be stored. Further, exposure of samples to strong light
or ag-H-flt-inn- wH 1 -jtrcelpratia- rgiHiTp-Elrrn nf- -their •
fore, chlorine 'analyses must be completed on-site within two hours of
sample collection, avoiding excessive light and agitation.
4.2.2 For Determination of Total Chloride by the Mercuric Nitrate
Method. Samples containing total chloride are sufficiently stable to be
returned to the laboratory for analysis.
5. Apparatus
5.1 Sampling Train. The schematic diagram of the sampling train in
Figure 1 shows that the Method _ train is similar in construction to the
Method 5 train. Apparatus is the same as Method 5, Sections 2.1.1, 2.1.2,
2.1.3, 2.1.4, 2.1.5, 2.1.8,- 2.1.9,- and 2.1.10.
5.1.1 Filter Heating System. Should be capable of maintaining the
temperature around the filter up to but not exceeding 93° C.
-4-
-------
Participate, Chlorine and Total Chloride Sampling Train
THERMOCOUPLE
POINT A
6"
PHOT TUBE
PROBE
THERMOCOUPLE
PROBE
PI TOT TUBE
STACK WALL
PITOr MANOMETER
HEATED AREA
THERMOMETERS
ORIFICE
NOTE: FIRST 3 IMPINGERS SHOULD BE CHARGED
WITH O.TN KOII SOLUTION
DRY HAS METER
FILTER HOLDER
THERMOCOUPLE
IMPINGER TRAIN
1HERMOCOUPIE
CHECK VALVE
I MI-in nut
ICE HA III
VACUUM LINE
VACUUM GIIAGE
MAIN VALVE
BY-PASS VALVE
AIR-TIGHT PUMP
-------
6-30-81
5.1.2 Condenser/Impingers. Same as Method 5, Section 2.1.7, except
the first three impingers will each be charged with 100 ml of 0.1 N
potassium hydroxide (KOH) in chlorine-demand-free water. An empty impinger
will not be used.
5.2 Sample Recovery. Same as Method 5, Section 2.2, except all washing
will be done with chlorine-demand-free water and no acetone will be used.
5.3 Analysis. Same as Method 5, Section 2.3, adding the following:
5.3.1 Beakers. 1000 ml.
5.3.2 Pipettes. Volumetric type 1-, 5-, 10-, 15-, 20-, and 25-ml sizes.
5.3.3 Burettes. 10- and 25-ml sizes.
5.3.4 Graduated Cylinder. 500-ml, graduated in divisions of 2 ml,
capable of measuring volume to HH 1 ml.
5.3.5 Thermometer. 0-1008C.
5.3.6 Volumetric Flasks-.-'50--,-100-r 200-—•and^250-^n-l-sizes-.* -—••>--
5.3.7 Analytical Balance. Capable of weighing to 0.1 mg.
5.3.8 Petri Dishes. For filter storage.
5.3.9 pH Meter with Non-chloride Reference Electrode and/or pH
Indicator Tape.
6. Reagents*
Use ACS reagent-grade chemicals or equivalent, unless otherwise
specified.
6.1 Sampling and Sample Recovery. The reagents used in sample
recovery are as follows:
6.1.1 Silica Gel, Crushed Ice, and Stopcock Grease. Same as Method 5,
Sections 3.1.2, 3.1.4, and 3.1.5.
Mention of trade names or specific products does not constitute
endorsement by the U»S. Environmental Protection Agency.
-5-
-------
6-30-81
6.1.2 Water, Deionized-Distilled, Chlorine-Demand-Free (DDCDF).
Chlorine-demand-free deionized-distilled water will be used for all sampling,
sample recovery, and analysis. A minimum of 20 liters should be prepared
for a sample series. To prepare chlorine-demand-free water, add sufficient
chlorine to deionized-distilled water to destroy ammonia, other nitrogen
compounds, and organic material. • The resultant chlorine excess (residual)
can be reduced by addition of sodium thiosulfate. It is preferrable to err
on the side of excess chlorine rather than chlorine demand, as the chlorine
content can be subtracted as a blank. Using the DPD Ferrous Titrimetric
Method or other chlorine measurement methods specified in Standard Methods
for Examination of Water and Wastewater^-) , a chlorine residual of less than
0.05 mg/L should be achieved.
Each Chemical Company of Ames, Iowa, manufactures a simplified chlorine
detect ion-apparattts-^chat-detects -^Mx>cine»*fcrT50=^ant3graia&3:per; -J-iteEi.. ...By.^ -, .^
slowly adding chlorine to deionized-distilled water and stirring to assist
reaction, a slight chlorine excess can be achieved and the addition of
sodium thiosulfate to redvice chlorine concentration may not be necessary.
In either of these methods, sufficient time and mixing should be allowed
for the chlorine demand to react with added chlorine. Chlorine-demand-free
water can also be prepared by ion exchange as described in Method 409G of
Reference 1. Chlorine-demand-free water should be stored in glass containers,
and displacement air passing into the container should pass through a
charcoal filter.
6.1.3 Potassium Hydroxide, 0.1 N. Dissolve 5.6 g KOH in 800 ml
of DDCDF water in a 1-liter flask, and dilute to exactly 1000 ml with
DDCDF water.
-6-
-------
6.2 Analysis. The reagents needed for analysis are listed below:
6.2.1 For Determination of Free Chlorine by the DPD Ferrous
Titrimetric Method.
6.2.1.1 Deionized-Distilled Chlorine-Demand-Free (DDCDF) Water. Same
as in Section 6.1.2.
6.2.1.2 Phosphate Buffer Solution. Dissolve 24 g anhydrous disodium
hydrogen phosphate, Na2HP04, and 46 g anhydrous potassium dihydrogen
phosphate, KH2P04, in DDCDF water. Combine with 100 ml DDCDF water in
which 0.800 g disoidum ethylenediamine tetraacetate dihydrate, also called
(ethylenedinitrilo) tetraacetic acid sodium salt, have been dissolved.
Dilute to 1 L with DDCDF water and add 0.020 g HgCl2 to prevent mold
growth and to prevent interference in the free available chlorine test
caused by any trace amounts of iodide in the reagents.
6. 2. 1 .3 - N, N-DJPthyl-p-phenyjjrMefliami-ng-fDPP) -TntH rafnr ~Sr>1 nflrm ^ -
Dissolve 1 g DPD Oxalate*, or 1.5 g p-amino-N:N-diethylaniline sulfatet, in
DDCDF water containing 8 ml 1+3 H2S04** and 0.200 g disodium ethylenediamine
tetraacetate dihydrate, also called (ethylenedinitrilo)tetraacetate acid
sodium salt. Make up to 1 L, store in a brown glass-stoppered bottle, and
discard when discolored. (The buffer and indicator sulfate are commercially
available as a combined reagent in stable powder form.)
CAUTION: The oxalate is toxic — take care to avoid ingestion.
* Eastman chemical No. 7102, or equivalent.
t British Drug House chemical available from Gallard-Schlesinger Chemical
Mfg. Corp., 584 Mineola Avenue, Carle Place, N.Y. 11514.
** See paragraph 6.2.1.5 of this section.
-------
6.2.1.4 Standard Ferrous Ammonium Sulfate (FAS) Titrant. Dissolve
1.106 g Mohr's salt, Fe(NH4)2(S04)2'6H20 in DDCDF water containing 1 ml
of 1+3 H2S04* and make up to 1 L with freshly boiled and cooled DDCDF
water. This primary standard may be used for 1 month, and the titer
checked by potassium dichromate (see Section 8, Calibrations). The FAS
titrant is equivalent to 100 g Cl/1.00 ml.
6.2.1.5 1+3 Sulfuric Acid Solution. Cautiously add 250 ml concen-
trated H2S04 to 600 ml DDCDF water and dilute to 1 L (i.e., 1 part
concentrated H2S04 + 3 parts DDCDF water).
6.2.1.6 Potassium Iodide (KI). Crystals of KI.
6.2.1.7 Hydrogen Peroxide. 3 percent solution.
6.2.1.8 Potassium Dichromate Crystals of I^C^Oy.
6.2.1.9 Ferroin Indicator.
6. 2.2 For 'DeterminatrioTTTjf^-Total-Chloride1 "by-^the=Merca«i3, 0.1 N. Dilute 3.2 ml concentrated HN03
(69%) to 500 ml with deionized-distilled water.
6.2.2.4 Sodium Hydroxide, NaOH, 0.1 N. Dissolve 2 gm NaOH in
deionized-distilled water and dilute to 500 ml.
6.2.2.5 Reagents for Low Chloride Titrations.
-------
6.2.2.5.1 Indicator Reagent. Dissolve, in the order named, 0.250 g
s-diphenylcarbazone, 4 ml of concentrated nitric acid, and 0.030 g xylene
cyanol FF in 100 ml of 95% ethyl alcohol or isopropyl alcohol. Store in
a dark bottle in a refrigerator. This reagent is not stable indefinitely.
Deterioration causes a slow end point and high results.
6.2.2.5 Standard Mercuric Nitrate Titrant, approximately 0.0141 N.
Dissolve 2.3 g Hg(N03)2 or 2.5 g Hg(N03)2*H20 in 100 mL deionized-distilled
water. Dilute to just under 1 L (see Section 8 for calibrations). The
Hg(N03)2 titrant is equal to approximately 0.50 mg Cl~/ml.
6.2.2.6 Reagents for High-Chloride Titrations.
6.2.2.6.1 Mixed Indicator Reagent. Dissolve 5 g diphenylcarbazone
powder and 0.5 g bromphenol blue powder in 750 ml 95% ethyl or 100% isopropyl
alcohol and dilute to 1 L with ethyl or isopropyl alcohol.
6.2.? 6.-2 :.Strong Standard-Mercurlc^-Ni-.trate-.Ti^rant,.~.OuJs4L.N«^—E.-^. ^-_ .:-
Dissolve 23 g Hg(N03)2 or 25 g Hg(N03)2'H20 in 900 ml deionized-distilled
water containing 5.0 ml cone. HN03. Dilute to just under 1 L.
6.2.3 For Determination of Total Water Soluble Particulate Chloride
by the Mercuric Nitrate Method.
6.2.3.1 Water. Deionized-distilled.
6.2.3.2 Other reagents same as in sections 6.2.2.2 through 6.2.2.6
since samples are to be analyzed by the mercuric nitrate method.
-------
7. Procedure
7.1 Sampling* Because of the complexity of this method, testers
should be trained and experienced with the test procedure to ensure
reliable results.
7.1.1 Pretest Preparation. Follow the general procedure given in
Method 5, Section 4.1.1.
7.1.2 Preliminary Determinations. Follow the general procedure
given in Method 5, Section 4.1.2, except as follows: Select a nozzle
size based on the range of velocity heads in order to maintain isokinetic
rates below 28 liters/min. (1.0 cfm).
7.1.3 Preparation of Sampling Train. Follow the general procedure
given in Method 5, Section 4.1.3, except place 100 ml of 0.1 N potassium
hydroxide in each of the first three impingers, and place the preweighed
silica gel in the fourth impinger.--.-Assemble-.the~train-as_shown_in
Figure 1.
7.1.4 Leak-Check Procedures. Follow the leak-check procedures given
in Methc i 5, Sections 4.1.4.1 (Pretest Leak-Check), 4.1.4.2 (Leak-Checks
During Sample Run), and 4.1.4.3 (Post-Test Leak-Check).
7.1.5 Sampling Train Operation, follow the general procedure
given in Method 5, Section 4.1.5. For each run, record the data required
on a data sheet such as the one shown in Method 5, Figure 5-2. Maintain
isokinetic sampling rates at less than 28 liters/min. (1.0 cfm). Maintain
filter and probe temperatures at a minimum of 10°C above stack temperature,
but no higher than 93°C.
7.1.6 Calculation of Percent Isokinetic. Same as Method 5, Section
4.1.6.
-------
7.2 Sample Recovery. Begin proper cleanup procedure as soon as the
probe is removed from the stack at the end of the sampling period. Allow
the probe to cool.
When the probe can be safely handled, wipe off all external particulate
matter near the tip of the probe nozzle, and place a cap over it to
prevent losing or gaining particulate natter. Do not cap off the probe
tip tightly while sampling train is cooling down. Capping would create a
vacuum and thus draw water from the impingers into the probe.
Before moving the sampling train to the cleanup site, remove the
probe from the sample train, wipe off the silicone grease, and cap the
open outlet of the probe. Be careful not to lose any condensate that
might be present. Wipe off the silicone grease from the impinger inlet
where the probe was fastened and cap it. Remove the umbilical cord from
the 1 a g t- •tmpi ngoi-—a-TvJ--fvgp.-l-hp-J mpKj>gg.1%. TfL a ,f 1 ovj >» 1 P 11 tl" 1 .q.._llfi<\d. .~___^....
between the filter and the probe, disconnect the line at the probe and
let any condensed water or liquid drain into the impingers or condenser.
Either ground-glass stoppers, plastic caps, or serum caps may be used to
close these openings.
Transfer the probe-filter-impinger assembly to the cleanup area.
This area should be clean and protected from the wind to minimize the
chances of contaminating or losing the sample.
7.2.1 Container No. 1. Carefully remove the filter from the filter
holder and place it in its identified petri dish container. Use a pair
of tweezers and/or clean dispoable surgical gloves to handle the filter.
If it is necessary to fold the filter, do so such that the particulate
cake is inside the fold. Carefully transfer to the petri dish any
-------
particulate matter and/or filter fibers which adhere to the filter holder
gasket, by using a dry Nylon bristle brush and/or a sharp-edge blade.
Seal the container.
Container No. 2. Probe Nozzle, Probe Fitting, Probe Liner, and
Front-half of Filter Holder. Taking care to see that dust on the outside
of the probe or other exterior surfaces do not get into the sample,
quantitatively recover particulate matter or any condensate from the
probe nozzle, probe fitting, probe liner, and front-half of the filter
holder by washing these components with DDCDF water and placing the wash
in a glass container. Save a water blank for analysis. Perform the
rinses as follows:
Carefully remove the probe nozzle and clean the inside surface by
rinsing with DDCDF water from a wash bottle and brushing with a Nylon
bristle .brush-.- Knv»h iint-M rho tjafpr rinsf -ghnwg-Tin--*H-g-t>i 1 P --pa-rt if. IPSJ ------
after which make a final rinse of the inside surface with DDCDF water.
Brush and rinse the inside parts of the Swagelok fitting with
DDCDF water in a similar way until no visible particles remain.
Rinse the probe liner with DDCDF water by tilting and rotating the
probe while squirting DDCDF water into its upper end so that all inside
surfaces will be wetted with DDCDF water. Let the DDCDF water drain
from the lower end into the sample container. A funnel (glass or poly-
ethylene) may be used to aid in transferring liquid washes to the container,
Follow the DDCDF water rinse with a probe brush. Hold the probe in an
inclined position, squirt DDCDF water into the upper end as the probe
brush is being pushed with a twisting action through the probe; hold a
sample container underneath the lower end of the probe, and catch any
-------
DDCDF water and particulate matter which is brushed from the probe. Run
the brush through the probe three times or more until no visible particulate
matter is carried out with the DDCDF water or until none remains in the
probe liner on visual inspection. Rinse the brush with DDCDF water, and
quantitatively collect these washings in the sample container. After the
brushing, make a final DDCDF water rinse of the probe as described above.
It is recommended that two people be used to clean the probe to
minimize sample losses. Between sampling runs, keep brushes clean and
protected from contamination.
After ensuring that all joints have been wiped clean of silicone
grease, clean the inside of the front half of the filter holder by rubbing
the surfaces with a Nylon bristle brush and rinsing with DDCDF water.
Rinse each surface three times or more if needed to remove visible parti-
culate. Make a final-rinse of the brush-:and filter holder.—Carefully :
rinse out the glass cyclone also (if applicable). After all DDCDF water
washings and particulate matter have been collected in the sample container,
tighten the lid on the sample container so that DDCDF water will not
leak out when the container is shipped to the laboratory. Mark the
height of the fluid level to determine whether or not leakage occured
during transport. Label the container to clearly identify its contents.
Container No. 3. Silica Gel. Weigh the spent silica gel (or silica
gel plus impinger) to the nearest 0.5 g using a balance. This step may
be conducted in the field.
-------
Container No. 4 - Impinger Solution and Back-Half Wash. Treat the
first three impingers as follows: leaving each impinger intact to transfer
the liquid, cap off the inlet, and pour the liquid through the outlet
into a graduated cylinder or directly into a tared sample container.
Record the volume to within + 1.0 ml or determine the liquid weight to
within j^ 0.5 g. Transfer the liquid to the sample container. Next, rinse
surfaces of the first three impingers and the glass connecting joints
with DDCDF water. Use rinse water sparingly to avoid excessive dilution
of the sample. A Nylon bristle brush may be used to facilitate removal
of any adhering material.
After transferring the impinger solution and back-half wash to the
sample container, tighten the lid on the sample container so that water
will not leak out when it is shipped to the laboratory. Mark the height
of the fluid...leyeJL.to-determine.later whether leakage occurred during
transport. Label the container to clearly identify its contents.
"Water-Blank" Container - Fill a 200-ml container with DDCDF water
taken from the same source, as that used for washing of the sampling
train.
"KOH Collecting Solution Blank" Container - Fill a 500-ml container
with KOH collecting solution taken from the same batch as that used in
the impingers.
"Filter Blank" - Take a tared unused filter from the field supply
for transfer to the laboratory for analysis.
7.3 Analysis.
Container No. 1 - Filter. Same as in Section 4.2 except change
temperature in oven drying; option to 90°C. Also after completion of the
-------
filter gravimetric analysis, extract water soluble chloride from the
filter by mascerating in 100 ml of deionized-distilled water with
a glass stirring rod. Deteirmine total chloride by the Mercuric Nitrate
Method (Section 7.3.2). Chloride determination of the filter extract can
be conducted separately from the impinger solution and back-half wash,
which would allow the latter to be analyzed without waiting for filter
desiccation; or can be combined for analysis with the impinger solution
and wash.
Container No. 2 - Probe and Front-half Wash. Transfer to appropriately
sized volumetric flask.. Dilute to the mark with deionized-distilled
water. With a pipette, take exactly 50 ml for chloride titration, and
dry-down the remainder for total particulate determination.
Container No. 4 - Impinger Solution and Back-half Wash. Bring to
500 ml volume. .-.Pipette "A-iQQ.-Jila aliquot «fQT.. jr^eLXhlnr±ne.-ran:ai.ysls:s------.-.-•;_
This should be conducted within two hours of test completion. The
remaining 400 ml of solution will be returned to the laboratory for total
chloride analysis.
"Water Blank" Container. Measure deionized-distilled water in
this container either volumetrically or gravimetrically. Transfer the
deionized-distilled water to a tared 250-ml beaker and evaporate to
dryness at ambient temperature and pressure. Desiccate for 24 hours and
weigh to a constant weight. Report the results to the nearest 0.1 mg.
NOTE: At the option of the tester, the contents of Container No. 2
as well as the deionized-distilled water blank container may be evaporated
at temperatures higher than ambient. If evaporation is done at an elevated
temperature, the temperature must be below the boiling point; also, to
-------
prevent "bumping", the evaporation process must be closely supervised,
and the contents of the beaker must be swirled occasionally to maintain
an even temperature.
"KOH Collecting Solution Blank" Container - Methods for determination
of blank values to be used in chlorine and total chloride analysis are
described in the following Section 7.3.1 and 7.3.2.
"Filter Blank" - Determine blank gravimetric and blank chloride
anlaysis identical to that conducted on the Container No. 1 filter.
7.3.1 Determination of Total Free Chlorine by the DPD Ferrous
Titrimetric Method. This method has been adapted from a standard method
for water analysis. Thus, sample conditions often encountered in water
analyses but not likely in analyses of stack samples, are not discussed.
On the other hand, sample conditions for stack effluent analyses which are
different "from those encountered in water-analyses,- -must "te-taken'irrto-"
account.
Careful pH control is essential for accurate results with this
method. As received for analysis, the sample may be strongly alkaline.
Further, the titration must be carried out at a pH of 6.2 to 6.5 to
produce a sharp colorless end point, and accurate measurement of free
chlorine. The pH of the sample must, therefore, be checked prior to
analysis and reduced, if necessary.
From the 100 ml aliquot of the impinger solution and back-half wash
withdrawn from Container No. 3 (paragraph 7.3) for free chlorine analysis,
treat a small measured test sample with determined quantities of 1..+ 3
sulfuric acid solution (Section 6.2.1.5) until the pH has been reduced to
7.5 to 8. Test for pH with pH paper as a reference electrode. With the
-------
6-30-81
quantity of acid known to-reduce the test sample pH to 7.5 to 8, ratio this
amount to the amount required to achieve the same pH range in a separate
exact 25 ml sample. After reducing the 25 ml sample pH, add deionized-
distilled water to bring the volume of sample up to exactly 100 ml.
To determine total available chlorine (i.e., total free and combined
available chlorine), place 5 ml each of buffer reagent and DPD indicator
solution in a 250 ml titration flask and mix (or use about 500 mg of DPD
powder). Add the 100 ml sample and mix.
To liberate combined chlorine, should any be present (i.e. as
chloramine, nitrogen trichloride), add several crystals of KI (about 1/2
to 1 g) and mix to dissolve. Let stand for 2 minutes and titrate rapidly
with standard FAS titrant until the red color is discharged. Record the
volume of titrant used as reading A. If color driftback indicates slightly
incomplete reaction, let stand 2 minutes and titrate until the red color
is again discharged. Record the quantity of titrant used and add to
reading A. •
Determine the blank by titrating 100 ml of KOH collecting solution in
the same manner as above, making sure all pH adjustments are correctly
made.
Explanatory Notes: The quantities given above are suitable for
concentrations of total available chlorine up to 4 mg/L. Where the total
chlorine exceeds 4 mg/L, use a smaller sample and dilute to a total volume
of 100 ml. Mix the usual volumes of buffer reagent and DPD indicator
solution, or the usual amount of DPD powder, with deionized-di-still-ed " - '-•
water before adding sufficient sample to bring the total volume to 100 ml.
Too high a pH may cause dissolved oxygen to give a color.
-17-
-------
6-30-81
Compensation for Interferences: Should the presence of oxidized
manganese be found or suspscted, the interference can be corrected by
placing 5 ml of buffer solution, one small crystal of potassium iodide,
and 0.5 ml sodium arsenite solution (500 mg NaAs02 plus 100 ml distilled
water) in the titration flask. Add 100 ml sample and mix. Add 5 ml DPD
indicator solution, mix, and titrate with standard ferrous ammonium
sulfate titrant until any red color is discharged. Subtract the reading
from reading A. If the combined reagent in powder form is used, add the
potassium iodide and arsenite first to the sample and mix, then add the
combined buffer-indicator reagent afterwards.
7.3.2 Determination of Total Chloride by the Mercuric Nitrate
Method. When using the mercuric nitrate method to determine total chloride
in outlet samples, it is necessary to pretreat any impinger sample aliquot
with hydrogen peroxide in order to destroy any residual chlorine remaining
in the sample. Residual chlorine gradually decomposes to form chlorides,
so in order to eliminate questions concerning the percentage conversion,
all field samples will require the pretreatment . After a total chloride
has been determined, that portion contributed by the chlorine will have
to be subtracted (i.e. two times the chlorine concentration). Thus, in order
to convert any free chlorine or hypochlorite to chloride, add to the
collected sample 2 ml of 3 percent hydrogen peroxide solution per 100 ml
of sample and mix.
The hydrogen peroxide reacts with hypochlorite in a basic solution
according to the equation.: ..... ......... ,.
H202 _ 02 + Cl" + H20
-18-
-------
6-30-81
The use of excess hydrogen peroxide has the added advantage of
destroying any reducing ions which may be present in the sample and which
interfere in the mercuric nitrate titration procedure.
Because pH control is critical in this method, adjust the pH of
each sample to 2.5 + 0.1 with 0.1 N HN03 or NaOH before titrating. Use
a pH meter with nonchloride type 'of reference electrode for the pH
adjustment. If only the usual chloride-type reference electrode is
available for pH adjustment, determine the amount of acid or alkali
required to achieve a pH of 2.5 + 0.1 and discard this sample portion.
Treat a separate sample portion with the determined amount of acid or
alkali and continue the analysis to its prescribed end. If no pH meter
is available, make the necessary adjustments using pH indicator paper.
7.3.2.1 Titration of Low Chloride Concentrations. Use a 100 ml
sample or—smaHer-'poTtiuu-so* that -t4ie cM-orlde- con tent- -l-y-les-s -thatt^iQ-mgs
Add 1.0 ml of indicator reagent to the sample. (The color of the
solution should be green-blue at this point. A light green indicates a
pH of less than 2.0; a pure blue indicates a pH of more than 3.8.)
Titrate the treated sample with 0.0141 N mercuric nitrate titrant to a
definite purple end point. The solution will turn from green-blue to
blue a few drops from the end point.
Determine the blank by titrating 100 ml of KOH collecting solution.
Explanatory Note: Chloride can be titrated with mercuric nitrate
because of the formation of soluble, slightly dissociated mercuric
chloride. In the pH range 2.3 to 2.8, diphenylcarbazone indicates the
end point of this titration by formation of a purple complex with the
excess mercuric ions. The error in titration is-about 1% of the volume
-19-
-------
6-30-81
of titrant used per change of 0.1 pH unit in the pH range 2.1 to 2.8. If
exact pH adjustment is not possible because no pH meter is available, it
is felt that keeping within a range of + 0.1 pH unit using pH paper is
sufficient for most analyses. In the low chloride procedure, the alcohol
solution of diphenylcarbazone also contains xylene cyanol FF which is
used as a pH indicator and as a background color to facilitate end-point
detection. Increasing the strength of the titrant and modifying the
indicator mixture enable determination of higher chloride concentrations.
7.3.2.2 Titration of High Chloride Concentrations. Place 50.0 ml
sample in a 150-ml beaker (5.00 ml sample may be used when more than 5 ml
titrant is needed). Add approximately 0.5 ml mixed indicator reagent and
\
mix well. The color should be purple. Add 0.1 N HNC-3 dropwise until the
color just turns yellow. Titrate with 0.141 N mercuric nitrate titrant to
the first permanent "dark purple. Titrate'a-50-ml-volume of-"K0H correct ing ~
solution as a blank using the same procedure.
8. Calibrations
8.1 Sampling Train. Sampling train calibrations are the same as EPA
Method 5, Section 5.
8.2 Standard Ferrous Ammonium Sulfate (FAS) Titrant (For DPP Ferrous
Titrimetric Method). Prepare 0.025 N standard potassium dichromate^)
solution by dissolving 1.226 g I^C^Oy, ACS reagent grade or equivalent,
previously dried at 103°C for two hours, in distilled water and dilute to
1000 ml. For standardizing the FAS titrant (Section 6.2.1.4) dilute 1.0 ml
standard I^C^Oy solution to about 25 ml. Add 3 ml concentration H2S04
and cool. Titrate with the ferrous ammonium sulfate titrant, using 2 to
3 drops (0.10 to 0.15 ml) ferroin indicator.
-20-
-------
6-30-81
Normality of FAS - nl K?Cr?07 x 0.025
ml Fe(NH4)2(S04)2
Prepare ferroin indicator solution by dissolving 1.485 g 1,10 -
phenanthroline monohydrate, together with 695 mg FeS04*7H20 in water and
dilute to 100 ml. This indicator solution may be purchased already
prepared.
8.3 Standard Mercuric Nitrate Titrant (For Mercuric Nitrate Method)
Prepare 0.0141 N standard sodium chloride solution by dissolving
8.243 g NaCl dried at 105°C in distilled water and dilute to 500 ml.
Dilute 50 ml of the standard sodium chloride solution to 1000 ml; 1.00 ml
equals 0.500 mg chloride. Make a preliminary standardization of the
mercuric nitrate titrant by following the titration procedure described
in Section 7.3.2.1. Use replicates containing 5.00 ml standard NaCl
solution diluted to 100 ml with deionized-distilled water. Store the
Hg(N03)2 titrant away from the light in a dark bottle.
8.4 Strong Standard Mercuric Nitrate Titrant (For Mercuric Nitrate
Method). Using 0.0141 standard sodium chloride solution from 8.3, perform
a preliminary standardization by following the titration procedure
described in 7.3.2.2. Use replicates containing 25 ml standard NaCl
solution and 25 ml deionized-distilled water. The chloride equivalence
of the titrant is approximately 5 mg/1 ml. Store the H(N03)2 titrant
away from the light in a dark bottle.
9. Calculations
9.1 Calculations for Total Particulate are the Same as Method 5, _
Section 6.
-21-
-------
6-30-81
9.2 For Determination of Total Free Chlorine by the DPP Ferrous
Titrimetric Method. Use the following equation to calculate the total
mass in the sample volumes.
J 35.45 2
Where:
MS
Ms
tb
Vs
N
35.45
Va
2
Va
= Total mass chlorine in the sample (mg)
= Volume of titer for aliquot (ml)
» Volume of titer for blank (ml)
• Volume of sample
™ Normality of ferrous ammonium sulfate titer
(as described in this method is
N 1.106 gm Fe(NH&)? - 0.00282)
392.15 gm/molar equivalent
- Molecular weight of chlorine
• Volume of aliquot (ml)
- Only one of the two chlorine atoms is available for
titration
9.3 For Determination of Total Chloride by the Mercuric Nitrate
Method. Cal ulation of total chloride in collected samples is accomplished
as follows:
Where:
- (Vta - Vha)V« N 35.45
Va
Ms » Mass chloride in sample (mg)
vta ™ Volume of titer for aliquot (ml)
vba " Volume of titer for blank aliquot (ml)
Vs ** Volume of sample in ml
N - Normality of Mercuric Nitrate (by the method is 0.1409)
35.45 » Molecular weight of chlorine
Va « Volume of aliquot (ml)
-22-
-------
6-30-81
10. Bibliography
1. Standard Methods for Examination of Water and Wastewater, 14th
Edition, 1975, APHA, AWWA and WPCF, Standard Method 409E. DPD Ferrous
Titrimetric Method, pp. 329-332.
2. Ibid. Standard Method 408B. Mercuric Nitrate Method, pp. 304-306.
3. Ibid. pp. 310-313.
4. Ibid. Oxygen Demand (Chemical), pp. 552.
-23-
-------
APPENDIX B
CHLORIDE PROCEDURES
-------
^ •"
Or
1. Principle and Applicability
l.i Princlpla. A gas sarnp'ia is exzractac: fron: the stack and the
r.ydrcchloric acid content is measured using the mercuric nitrate titra-
«"on ".einod.
1.2 Applicability. This method is applicable for the determination
of hydrochloric acid emissions from stationary sources. The upper limit
of the method has not been established though samples containing 4000 parts
per .Trillion hydrochloric acid have been efficiently collected and determined
using two midget impingers each containing 15 milliliters of O.iN sodium
hyarcxide.
Possible interferences are ~.etal ions (usually Cu, Fe, Ni , Mg, and V)
and the sulfite ion. These tend to mask the analysis endpoint. A strong
acid type cationic exchange 'res i;rv t,aS:'te'en'''cemcr,strate-d~to~:rBmovff''iTiter- —
ferir.g metal ions satisfactorily. When gas stream containing SJ^ in excess
of 300 parts per million are sampled, the S02 can be effectively removed by
treating the sample with 3 percent hydrogen perbxi.de ^:
2. Apparatus
2.1 Sampling. The sampling train is shown in Figure 1 and the compon-
ent parts are discussed below.
2.1.1 Probe. A five-foot, borosilicate. glass or Teflon-lined probe is
used. The inlet is packed with glass wool to remove particulate matter.
The probe must have a heating system to prevent water condensation while
*
sampling:
2.1.2 Midget Impingers. Four midget impingers are connected
-------
2
•in series with leak-free glass connectors. Silicone grease may be used
to insure a tight seal.
2.1.3 Glass Wool. Borosilicata or quartz.
2.1.4 Temperature Gauge. Dial thermometer, or equivalent, to
maasura temperature of gas leaving impinger train to within 1°C(2°F). .
2.1.5 Drying Tube. Tube packed with 6-to 15-mesh indicating type
silica gal or equivalent to dry the gas sample and to protect the .dry
gas meter and pump.
2.1.5 Valve. Needle valve, to regulate sample gas flow rate.
2.1.7 Pump. Leak-free diaphragm pump, or equivalent, to pull gas
through the train. Install a small surge tank between the pump and the
rate meter to eliminate the pulsation effect of the diaphragm pump on
the rotamater.
2.1.3 Rate Meter. Rotameter, or equivalent, capable of measuring
flow rate to within 4 percent of the selected flow rate of 2 1/min.
2.1.9 Volume Meter. Dry gas meter, « ifficiently accurate, to-'mea-
sure the sample volume to three decimal places in cubic feet, calibrated
at the selected flow rate and conditions actually encountered during
sampling, and equipped with a temperature gauge (dial thermometer, or
equivalent) capable of measuring temperature to within 3°C (5.4*F).
2.1.10 Barometer. Mercury, aneroid,, or other barometer capable
of measuring atmospheric pressure to within 2.5 mm Hg (0.1 in Hg).
."2.1.11 Vacuum Gauge. At least 760 mm Hg (30 in. Kg) gauge, to be
used for leak check of the sampling train.
2.2 Sample Recovery. The following equipment is needed for sample
recovery.
-------
3
2.2.1 Storage Containers. Leak-free ICO ml polyethylene or
glass bottles.
2.2.2 Wash Bottle. Polyethylene or glass.
2.3 Analysis. Volumetric glassware should be class A only.
2.3.1 lr"i enmeyer Fl ask.. 125 ml.
2.3.2 Graduated Burette. 25ml.
2.3.3 Volumetric Pipet. 1 ml.
2.3.4 Volumetric Flasks. ICC ml, one for each sample.
2.3.5 Small Funnel. For transferring sample.
2.3.5 Magnetic Stirrer.
2.3.7 Magnetic Stirring Bar.
2.3.8 Dropping Pipet or Dropper.
2.3.9 Filter Paper. Whatman No. 40.
3. Reagents
Unless otherwise indicated, all reagents must conform to the
specifications established by the Committee on Analytical Reagents
of the American Chemical Society. Where such specifications are not
available, use the best available grade.
3.1 Sampling.
3,1.1 Water. Triple deionized, distilled to conform to AS7M
Specification D 1193-74, Type 3.
3.1.2 Sodium Hydroxide, 0.1 N. Dissolve 4g NaOH in deionized,
distilled water and dilute to 1 liter.
3.1.3 Hydrogen Peroxide, 3 Percent. Dilute 30 percent hydro-
gen peroxide 1:9 (V/V) with deionized, distilled water (15 ml is
needed per sample). Prepare fresh daily.
-------
3.2 Sampla Recovery.
3.2.1 Sodium Hydroxide, 0. IN. As in 3.1.2.
3.2.2 Sodium Hydroxide, 0.05 N. Dissolve 2g NaOH in deionizad,
distilled water and dilute to 1 liter.
3.3 Analysis.
3.3.", Mercuric Nitrate Solution, 0.01 N. Dissolve 1.63 g
mercuric nitrate in deionized, distilled water and dilute to 1 liter.
Shake well. Transfer to an amber colored reagent bottle. Standardize
this solution according to the procedure in Section 5.5.
3.3.2 Bromophenol-Diphenylcarbazone Indicator. Dissolve 0.005 g
bromophenol blue and 0.5 g diphenyl carbazone in 75 ml pure
ethanol, then dilute to 100 ml therewith.
3.3.3 0.05 N HN03. Dilute 1.6 ml concentrated HNOj (69 percent)
to 500 ml with distilled, deionized water.
3.3.4 0.0100 N Nad Solution. Dissolve 0.584 g NaCl in
deionized, distilled water and dilute to 1 liter.
3.3.5 Cation Exchange Resin. Rexyn 101 H (Fisher Scientific)
is one type that has been used successfully. The resin column is
prepared as follows:
A small amount of the fresh resin is poured into a 250 ml
beaker containing 100 ml distilled water and allowed to soak for
30 minutes. A 25 ml buret is packed at the bottom end with a small
portion of glass wool to prevent plugging the tip with resin par-
ticles. The buret is-then filled approximately ha If-full, .with - -.._..
distilled water, and the wetted resin is poured in until it reaches
the 11 ml graduation mark. Care must be taken to maintain
-------
the we tar level at least one 'inch above the resin level at all t1-r.es to
prevent res-in channelling. Tha buret is filled to the 0 ml graduation mark
with distilled water, and the stopcock is opened to allow the column to drain
until the final water level is at the 10 rnl gracuation mark. This rinsing
process is repeated several ti~;as and the rinses discarded. A column blank
is determined by passing a 10 ml aliquot of distilled water through the column
followed by a 10 ml rinse. The blank volume is then adjusted to 100 ml with
distilled water and titrated with standard mercuric nitrate. One drop of
titrant should be sufficient to reach the endpoint.
4.. Procedure
4.1 Sampling
; 4.1.1 Preparation of Collection Train. Measure 15 ml of 0.1 N sodium
hydroxide into.-each-of-the -.first-two, midget impiT.gers^ancL^li mUof-...3..percent,-=
hydrogen peroxide into the third one. Leave the final impinger dry. Assemble
the train as in Figure 1. Adjust probe heater to a temperature sufficient to
prevent water condensation. Place crushed ice and water around the impinger .
4.1.2 Leak-Check Procedure. A leak-check prior to the sampling run is
optional; however, a leak-check after the sampling run is mandatory. The
leak check procedure is as follows:
Temporarily attach a suitable (e.g., 0-40 cc/min) rotameter to the out-
let of the dry gas meter and place a vacuum gauge at or near the probe inlet.
Plug the probe inlet, pull a vacuum of at least 250 mm Hg (10 in. Hg), and
note the*-flow rate as imiicated--by the rotameter.- - A"l-e-akage rate -not"tn
excess of 2 percent of the average sampling rate is acceptable.
-------
0
Note: Carefully release the probe inlet plug before turning
077 u.12 pump.
It is suggested (not mandatory) that the pump be leak-checked
Separately, either prior to or after the sampling run. If done
prior to the sampling run. the pump leak-check shall precede the
leak-check of the sampling train described immediately above; if
done after tha sampling run, the pump leak-check shall follow the
train leak-check. To leak-check the pump, proceed as follows:
Disconnect the drying tube from the prcbe-impinger assembly. Place
a vacuum gauge at the inlet to either the drying tube or the pump,
pull a vacuum of 250 mm (10 in.) Kg, plug or pinch off the outlet
of the flow meter and then turn off the pump. The vacuum should
remain stable for at least 30 seconds.
Other leak-check procedures-may-be usedr"subject"to^the-- —
approval of the Administrator, U. S. Environmental Protection Agency.
4.1.3 Sample Collection. Record the initial dry gas meter
reading and barometric pressure. To begin sampling, position the
tip of the probe at the sampling point, connect the probe to the
bubbler, and start the pump. Adjust the sample flow to a constant
rate of approximately 2.0 liters/min as indicated by the rotameter.
Maintain this constant rate (+JO percent) during: the entire sample
run. Take readings (dry gas meter, temperatures at dry gas meter
ar.d at impingar outlet and rate meter) at least every 5 minutes.
Add more ice during the run to keep the temperature of the gases
leaving the last impinger at 20°C (SS°F) or less. At the conclusion
of each run, turn off the pump, remove probe from the stack, and
-------
record the final readings. Conduct a leak-check as in Section 4.1.2.
(This leak-check is mandatory). If a leak is found, void the tast run or
use procedures acceptable to the Administrator to adjust the sample volume
for the leakage. Drain the ice bath, and purge the remaining part of the
train by drawing clean ambient air through the system for 15 minutes at
the sampling rate.
C3 4.2 Sample Recovery. Disconnect the impingers after purging. Dis- .
card the contents of the hydrogen peroxide SC^ scrubber impinger. Pour
the contents of the other impingars containing sodium hydroxide into a
leak-free polyethylene or glass bottle for shipment. Rinse the impingers
and the connecting tubes with deionized, distilled water, and add the
washings to the same storage container. The glass wool probe plug is
recovered and placed in a second storage container. The probe is rinse
with 25 ml distilled water into a second storage container. The probe is
rinsed 25 ml distilled water into a third storage container and marked
accordingly. Seal, mark the fluid level, and identify all containers.
4.3 Sample Analysis. Note level of liquid in each container, and
confirm whether any sample was lost during shipment; note this on analytical
data sheet. If a noticeable amount of leakage has occurred, either void
the sample or use methods, subject to the approval of the Administrator,
to correct the final results.
Using a small funnel, the contents of each sample container are
quantitatively transferred to separate 100 ml volumetric flasks. The
probe wash is filtered through Whatman Number 40 paper into a flask. The
probe plug is rinsed three times with water and likewise filtered into
another flask. Each flask is diluted to the mark with deionized, dis-
tilled water.
-------
3
An aliquot of one out of every five sa-.ples Is passed through a
catiom'c exchange resin (prepared as In Section 3.3.5) and compared to
what of untreated samples to see If interfering metal ions are present.
If values differ by more than 1 percent, an aliquot of all samples should
be passed through the resin prior to the analysis. To oxidize any S0«
present, add and mix 2 ml 3 percent hydrogen peroxide solution per 100 ml
sample. ••
Before analysis, the pH of the sample must be adjusted to 3.2 - 3.4.
Pour 25 ml of the 100 ml sample into a 125 ml Erlenmeyer flask and add 5
drops of bromophenol/diphenyl carbazone while stirring the sample contin-
uously using a magnetic stirring bar. If the sample is initially alkaline,
it will turn blue or red in color after addition of the indicator. In this
case, add_Q.,l~N_nitrie acid dropwise:-until Jtne-sampJe4us£-J;ueas~yeUow iru.~_
color. Then add an additional 1.0 ml of 0.05 N nitric acid to bring the
pH to 3.2 - 3.4. If the sample is initially acidic, the solution will turn
yellow after addition of the indicator. In this case, add 0.1 N sodium
hydroxide dropwise until the sample turns blue in color. Then add 0.05 N
nitric acid dropwise until the sample just turns yellow. Add an additional
1.0 ml of 0.05 N nitric acid to bring the sample to the proper pH of 3.2 -
3.4. After the pH is adjusted, titrate the sample with standard 0.01 N
mercuric nitrate until the first appearance of a violet color which
persists for several minutes with continuous stirring. Repeat and average the
titration volumes. Run a blank with each series of samples..---
5. CaTibration
5.1 Metering System. Before its initial use and after each field
test series, leak check the metering system as described in Method 6,
-------
Section 5.1.
5.2 Thermometers. Calibrate against mercury-in-glass thermometer.
5.3 Rotameter. The rotameter need not be calibrated but should be
kept clean according to manufacturer's instructions.
5.4 iarc,matar. Cali:/ra';a c^iinst mercury barometer.
5.5 Mercuric Nitrata Solution. Standardize the mercuric nitrate
solution by titrating against triplicate aliquots of 0.01CO N standard NaCl
solution.
5. Calculations
Carry out calculations, retaining at least one extra decimal figure
beyond that of the acquired; data. Round off figures after the final calcu-
lation. Other forms of the equations may be used as long as they give
equivalent results.
6.1 Nomenclature
C.HC1 = Concentration of hydrogen chloride, dry basis cor-
rected to standard conditions, mg/dscm (Ib/dscf).
N = Normality of mercuric nitrate titrant, milliequiva-
1ents/ml.
P. = Barometric pressure at the exit orifice of the dry
gas meter, mm Hg (in. Kg).
P „ , = Standard absolute pressure, 760 mm Hg (29.S2 in. Kg).
S wfl
T = Average dry gas meter absolute temperature, °K (°R).
T . , = Standard absolute temperature, 293°K,(528°R)._^ . ...
S uU
V = Volume of sample aliquot titrated, ml.
a
V = Dry gas volume as measured by the dry gas meter,
tH
con (dcf}.
-------
10
V ff<.-* = Dry gas volume measured by the dry gas mater,
in \ S iG ;
corrected to standard conditions, dscni (dscf).
V_- = Total volume of solution in which the hydrochloric
Sw i it
acid sample is contained, 100 ml.
V^ = Volume of mercuric nitrate titrant used for the
t,
sample, ml (average of replicate titrations).
V^_ = Volume of mercuric nitrate titrant used for the
ull
blank, ml .
Y = Dry gas meter calibration factor.
70.90 = Equivalent weight of hydrochloric acid.
6.2 Dry sample gas volume, corrected to standard conditions.
Equation 5-1
Where:
K, = 0.3358 °K/mm Kg for metric units.
= 17.54 °R/in. Hg for English units.
6.3 Hydrochloric acid concentration.
c .„ (vt •
HC1
Equation 6-2
*
"Where:
;<2 = 70.90 mg/meg. for metric units
-4
= 1.563 x 10 Ib/meg. for English units.
-------
T>
t I
7. Sibl •iocrasi'.y
1. Cheney, J., C. Fortune, and R. Rollins. Report on the
Development and Evaluation of Stationary Source Hydrochloric Acid
Emissions Measurement Methodology. U. S. Environmental Protection
Agency, Stationary Source Emissions Research Branch. Research
Triangle Park, N. C. December, 1975.
2. Rcm, J. o. Maintenance, Calibration, and Operation of
Isokinetic Source-Sampling Equipment. Office of Air Programs,
Environmental Protection Agency. Research Triangle Park, N. C.
APTD-G57S. March, 1972.
3. Annual Book of ASTM Standards. Part 31; Water, Atmospheric
Analysis. American Society for Testing and Materials. Philadelphia,
Pa. 1974. pp. 40-42.
1 Mention of trade names or specific products does not constitute
endorsement by the Environmental Protection Agency.
-------
408 A. Argentometric Method
I. General Discussion
a. Principle: In a neutral or slightly
alkaline solution, potassium chromate
can indicate the end point of the silver
nitrate titration of chloride. Silver chlo-
ride is precipitated quantitatively before
red silver chromate is formed.
b. Interference: Substances in
amounts normally found in potable wa-
ters will not interfere. Bromide, iodide,
and cyanide register as equivalent chlo-
ride concentrations. Sulfide, irhiosulfate,
and sulfite ions interfere but can be re-
moved by treatment with hydrogen per-
oxide. Onhophosphate in excess of 25
mg/1 interferes by precipitation as silver
phosphate. Iron in excess of 10 mg/1 in-
terferes by masking the end point.
2. Reagents
a. Cblaride-frcc water: If necessary,
use redistilled" or deionized' distilled-wa--
ter.
b. Potassium cbromate indicator solu-
tion: Dissolve 50 g KiCrO* in a little
distilled water. Add silver nitrate solu-
tion until a definite red precipitate is
formed. Let stand 12 hr, filter, and di-
lute to 1 1 with distilled water.
c. Standard silver nitrate titrant,
0.014IN.- dissolve 2.395 g AgNCb in
distilled water and dilute to 1,000 ml.
Standardize against 0.0141N Nad by
the procedure described in ^3b below.
Store in a brown bottle. Standard silver
nitrate solution 0.0141 N— 500 jig Cl/
1.00ml.
d. Standard sodium chloride,
0.014 hV; Dissolve 824.1 mg NaCl
(dried at 140 C) in chloride-free water
and dilute to 1,000 ml; 1.00 ml=500
MgCI.
e. Special reagents for removal of in-
terference:
1) Aluminum hydroxide suspension:
Dissolve 125 g aluminum potassium
sulfate or aluminum ammonium sulfate,
AlKCSOOz- 12H2O or AlNHXSCUh--
12H:O, in I 1 distilled water. Warm to
60 C and add 55 ml cone NH^OH
slowly with stirring. Let stand about 1
hr. transfer the mixture to a large bottle,
and wash the precipitate by successive
additions, with thorough mixing and de-
cantations of distilled water, until free
from chloride. When freshly prepared,
the suspension occupies a volume of
approximately 1 I.
2) Pbenolphtbalein indicator solu-
tion.
3) Sodium hydroxide, NaOH. IN.
4) Sulfuricadd, H:SO4, \N.
5) Hydrogen peroxide, HzOz, 30%.
3. Procedure
a. Sample preparation: Use a 100-ml
sample or a suitable portion diluted to
100 ml.
If the sample is highly colored, add 3
ml AKOHb suspension, mix, let settle,
filter, wash, and combine filtrate and
washing.
If sulfide, sulfite. or thiosulfate is
present, add 1 ml H:O: and stir for 1
min.
b. Titration: Titrate samples in the
pH range 7 to 10 directly. Adjust sam-
ples not in this range with H:SO* or
NaOH solution. Add 1.0 ml KjCrCh
indicator solution. Titrate with standard
silver nitrate titrant to a pinkish yellow
-------
end point. Be consistent in end-point
recognition.
Standardize the silver nitrate titrant
and establish the reagent blank value by
the titration method outlined above. A
blank of 0.2 to 0.3 ml is usual for the
method.
4. Calculation
mg/l Q
ml sample
where A = ml titration for sample, B —
ml titration for blank, and JV= normal-
ity of AgNOs.
mg/INaCl=mg/IClXI.65
5. Precision and Accuracy
A synthetic unknown sample contain-
ing 241 mg/I chloride, 108 mg/l Ca,
82 mg/I Mg, J.I mg/l K. 19.9 mg/l
Na, l.l mg/l nitrate N, 0.25 mg/l ni-
trite N, 259 mg/l sulfate, and 42.J
mg/l total alkalinity (contributed by
NaHCOj) in distilled water was ana-
lyzed in 41 laboratories by the argento-
metric method, with a relative standard
deviation of 4.2% and a relative error of
1.7%.
408 B. Mercuric Nitrate Method*
1. General Discussion
a. Principle: Chloride can be titrated
with mercuric nirrate because of the for-
mation of soluble, slightly, dissociated—
mercuric chloride. In the pH range 2.3
to 2.8, diphenylcarba/one indicates the
end point of this titration by formation
of a purple complex with the excess
mercuric ions. The error in titration is
about 1 % of the volume of titrant used
per change of 0.1 pH unit in the pH
range 2.1 to 2.8. Because exact pH ad-
justment is not feasible except by use of a
pH meter, it is felt that keeping within a
range of ±0.1 pH unit is sufficient for
• United Sam Patent No. 2.784.064 has been is-
sued to F.E. Clarke, relative to the mercurimetric titra-
tion of chloride. Nothing contained in this manual is to
be construed as granting any right, by implication or
otherwise, for manufacture, sale, or use in connection
with any method, apparatus or product covered by pat-
ent, nor as insuring anyone against liability for innSnge-
ment of patent.
most water analyses. Therefore, in this
method, a specific mixture of nitric acid
and diphenylcarbazone is added to a
water sample, adjusting the pH of most
parable "waters-to pH-ijlipJ—Avthird-
substance in this alcoholic mixture, xy-
lene cyanol FF, is used as a pH indicator
and as a background color- to facilitate
end-^oint detection. The introduction of
10 r g sodium bicarbonate to both the
blan; and the standard titration pro-
vides a pH of 2.5±0.1 when 1.0 ml in-
dicator-acidifier reagent (1f2^l)l is
added. Increasing the strength of the ti-
trant and modifying the indicator mix-
ture enable determination of the higher
chloride concentrations common in
wastewater.
b. Interference: Bromide and iodide
are titrated with mercuric nitrate in the
same manner as chloride. Chromate.
ferric, and sulfite ions interfere when
present in excess of 10 mg/l.
-------
2. Reagents
a. Standard sodium chloride 0.0141
;V. See Method A. ^2dabove.
b. MmcddJ.HNCb.O.hV.
c. Sodium hydroxide, NaO H. 0.1N.
d. Reagents for lo-^-chloridt titra-
tinns:
1) Indicator-addi/ier reagent: The
nitric acid concentration of this reagent
is an important factor in the success of
the determination and can be varied as
indicated in a) or b) to suit the alkalinity
range of the sample being titrated. Re-
agent a) contains sufficient nitric acid to
neutralize a total alkalinity of 150 mg/l
as CaCOj to the proper pH in a 100-ml
sample.
a) Dissolve, in the order named, 250
mg s-diphenylcarbazone. 4.0 ml cone
HNOs, and JO mg xylene cyanol FF in
100 ml of 95% ethyl alcohol or isopro-
pyl alcohol. Store in a dark bottle in a
refrigerator. This reagent is not stable
indefinitely: Deterioration causes a slow
end point and high results.
b) Because pH control is critical in
this method, adjust the pH of highly al-
kaline or acid samples to 2.5±0.1 with
OAN HNCh or NaOH. not with
NazCOi. Use a pH meter with a non-
chloride type of reference electrode for
the pH adjustment. If only the usual
chloride-type reference electrode is
available for pH adjustment, determine
the amount of acid or alkali required to
achieve a pH of 2.5±0.1 and discard
this sample portion. Treat a separate
sample portion with the- determined
amount of acid or alkali and continue
the analysis to its prescribed end. Under
these circumstances, omit the nitric acid
from the indicator reagent to maintain
the proper sample pH. Alternatively.
vary the nitric acid concentration of the
indicator-acidifier reagent to accommo-
date conditions wherein water samples
of very high or very low alkalinity are
being analyzed.
2) Standard mercuric nitrate titrant,
0.0141/V: Dissolve 2.5 g Hg(NCbh or
2.5 g Hg(NCh):-H:O in 100 ml dis-
tilled water containing 0.25 ml cone
HNO3. Dilute to just under 1 1. Make a
preliminary standardization by follow-
ing the procedure described in fi Ja. Use
replicates containing 5.00 ml standard
NaCl solution and 10 mg NaHCCh di-
luted to 100 ml with distilled water. Ad-
just the mercuric nitrate titrant to ex-
actly 0.014IN and make a final
standardization. Store away from the
light in a dark bottle. Standard mercuric
nitrate titrant, exactly 0.014 hV, is
equivalent to 500 ^g Cl/1.00 ml.
e. Reagents for hiyb-cbloriiit? tirra-
tims:
D-Mixed-indtcator reagent :~D\ssolve-
5 g diphenylcarbazone powder and 0.5
g bromphcnol blue powder in 750 ml
95% ethyl or isopropyl alcohol and di-
lute to 1 I with ethyl or isopropyl alco-
hol.
2) Strong standard mercuric nitrate
titrant, 0.141N: Dissolve 25 g
Hg(NO3)2-H:O in 900 ml distilled wa-
ter containing 5.0 ml cone HNOj. Di-
lute to just under 1 1, and perform a pre-
liminary standardization by following
the procedure described in fl \b. Use rep-
licates containing 25.00 ml standard
NaCl solution and 25 ml distilled water.
Adjust the titrant to 0.141 /V and make a
final standardization. The chloride
equivalence of the titrant is 5.00 mg/
1.00ml.
-------
3. Procedure
a. Titration of lov chloride concen-
tratwns.- Use a 100-ml sample or
smaller portion so that the chloride con-
tent is less than 1 0 mg.
Add 1 .0 ml of indicator-acidifier re-
agent to the sample. (The color of the
solution should be green-blue at this
point. A light green indicates a pH.of
less than 2.0; a pure blue indicates a pH
of more than 3.8. For most potable wa-
ters. the pH after this addition will be
2.5±O.I. For highly alkaline or acid
waters. adjust pH to about 8 before add-
ing the indicator-acidifier reacent.)
Titrate the treated sample with
0.01 41 N mercuric nitrate titrant to a
definite purple end point. The solution
will turn from green-blue to blue a few
drops from the end point.
Determine the blank by titrating 100
ml distilled water containing 10 mg
NaHCOi.
b. Titration of high chloride concert-
trations: Place^^.OVmfcsample;in^ir;h5G-5r
ml beaker (5.00 ml sample may be used
when more than 5 ml titrant are
needed). Add approximately 0.5 ml
mixed indicator reagent and mix well,
The color should be purple. Add O.lN
HNCh dropwise until the color just
turns yellow- Titrate with ?• ! 4J N mer*
cunc mt™f tltrant 1° the fi"< Per'
maner" da,rk PurPle- Turate a distllled
water blank usmg che same Procedure.
4 Ca,culation
m /I Cl = (
ml sample
where /4 = m, tjtration for sample% B =
m, titration for b|ank% and ^normal-
jty Of He(NCb):.
mg/lNaCl=mg/lClXl.6;
5 precision and Accuracy
A synthetic unknown sample contain-
ing 241 mg/1 chloride. 108 mg/1 Ca,
82 mg/1 Mg. 3.1 mg/1 K, 19.9 mg/1
Na. 1.1 mg/I nitrate N. 0.25 mg/l ni-
trite N, 259 mg/1 sulfate, and 42.5 mg/
NaHCO3> in distilled water was ana-
lyzed in 1 0 laboratories by the mercuri-
metric method, with a relative standard
deviation of 3.3% and a relative error of
2.9%.
-------
APPENDIX C
METHOD 5A
PARTICULATE AND CONDENSABLE HYDROCARBONS PROCEDURE
-------
APPENDIX A - REFERENCE TEST METHODS
METHOD 5A. DETERMINATION OF PARTICULATE AND
CONDENSIBLE ORGANIC MATTER FROM STATIONARY SOURCES
1. Applicability and Principle
1.1 Applicability. This method applies to the determination of
particulate and condensible organic matter within the context of the
following definition: "Particulate matter" means any finely divided solid
or liquid material, other than uncombined water, that condenses at or above
the filtration temperature, and is collected by the front-half of the
sampling train. "Condensible organic matter" means any material which
remains after extraction, filtration, and ambient evaporation of the ether-
chloroform extract of the impinger portion of the train.
1.2 Principle. Particulate and condensible organic matter is withdrawn
isokinetically from the source. Particulate matter is collected on a glass
fiber filter maintained at temperatures of 120 + 14°C, or such other
temperature as specified by an applicable subpart of the standards or by
the Administrator for a particular application." The part'iculate mass is
determined gravimetrically after removal of uncombined water. Condensed
organics are collected in the water-filled Greenburg-Smith impingers at
temperatures less than 20 + 5°C. The condensed organics are analyzed
gravimetrically after extraction from the water by ether and chloroform*.
2. Apparatus.
2.1 Sampling Train. Same as Method 5, Section 2.1, except the Greenburg-
Smith system shall be used to determine stack gas moisture content and
condensible organic matter, and the sampling train must be assembled without
the use of stopcock grease. Also, in the laboratory and before the first
test, clean the impingers and filter assembly with chromic acid cleaning
solution followed by deionized distilled water and methylene chloride.
*Warning: Chloroform is a suspected carcinogen. The vapors of ethyl ether
are an explosion hazard. All work should be done under an explosion-proof
hood with the analyst protected from exposure to these chemicals. Ether
should not be retained beyond the container date; opened containers should
not be stored for more than six months.
-1-
-------
2.2 Sample Recovery.* Same as Method 5, Section 2.2, except in the
laboratory before the first test all glassware used to handle the sample
should be cleaned with chromic acid cleaning solution as described in
Section 2.1. Also, WYTEX fiber should be used in bristle cleaning brushes
rather than nylon, as tests have shown nylon to react with methylene chloride
to produce a residue.
2.3 Analysis. Same as Method 5, Section 2.3, with the following
additional items:
2.3;1 Separatory Funnels. Two, 100-ml capacity.
2.3.2 Buchner Funnel. 47-mm diameter.
2.3.3 Erlenmeyer Flask. 500-ml.
3. Reagents
3.1 Sampling. Same as Method 5, Sections 3.1.1, 3.1.2, 3.1.3, and
3.1.4.
3.1.1 Water. Deionized distilled to conform to ASTM specification
Dll93-74, Type 3. Analyze blanks prior to field use to eliminate a high
blank on test samples.
3.1.2 Chromic Acid Cleaning Solution. Prepare by slowly adding 1
liter concentrated sulfuric acid, with stirring, to 35 ml saturated sodium
dichromate solution. HANDLE WITH CAUTION.
3.2 Sample Recovery. Methylene Chloride. ACS reagent grade <0.001
percent residue, in glass bottles. Methylene chloride blanks shall be run
prior to field use and only methylene chloride with low blank values «0.001
percent) shall be used.
3.3 Analysis.
3.3.1 Methylene Chloride. Same as 3.2.
3.3.2 Ethyl Ether.- .Reagent.grader. C 0^001 .percent xesidue -ini-.suitable -
containers to retain low residue blank but still meet safety requirements.
*Mention of trade names on specific products does not constitute endorsement
by the U.S. EPA.
-2-
-------
3.3.3 Chloroform. Reagent grade, <_ 0.001 percent residue, in glass
bottles.
3.3.4 Water. Deionized distilled as described in Section 3.1.
3.3.5 Filters. 47-mm diameter glass fiber filters as described in
Method 5, Section 3.1.1.
3.3.6 Hydrochloric Acid. Approximately 1 N. Bring 83 ml concentrated
HCL to 1 liter. HANDLE WITH CAUTION.
3.3.7 Sodium Hydroxide, 10 N. Dissolve 40 g NaOH in deionized distilled
water and dilute to 100 ml. HANDLE WITH CAUTION.
3.3.8 Dessicant. Anhydrous calcium sulfate, indicating type. Alternately,
other types of desiccants may be used, subject to approval of the administrator.
4. Procedure
4.1 Sampling. Same as Method 5, Section 4.1, except use deionized
distilled water in the impingers, and do not use silicone grease on joints
as the silicone grease is soluble in methylene chloride.
4.2 Sample Recovery. Same as Method 5, Section" 4.2, except substitute
methylene chloride for acetone. Also, add the following:
4.2.1 Container No. 4 (Impinger Water). Treat the first three impingers
as follows: Leaving each impinger intact to transfer the liquid, cap off
the inlet, and pour the liquid through the outlet into a graduated cylinder
or directly into a tared sample container. Record the volume to within jf
1.0 ml or determine the liquid weight to within +0.5 g.
After transferring the water and condensible organic matter to the
sample container, tighten the lid on the sample container so that water
will not leak out when it is shipped to the laboratory. Mark the height of
the fluid level to determine later whether leakage occurred during transport.
Label the container to clearly identify its contents.
4.2.2 Container No. 5 (Methylene Chloride Wash). Rinse the inside
surfaces of the first three impingers and the glass connecting joints with
methylene chloride. Rinse all surfaces three times or more if necessary to
remove visible particulate. A WTCEX bristle brush may be used to facilitate
removal of any adhering material.
-------
After all methylene chloride washings and particulate matter are
collected in the sample container, tighten the lid on the sample container
so that methylene chloride will not leak out when it is shipped to the
laboratory. Mark the height of the fluid level to determine later whether
leakage occurred during transport. Label the container to clearly identify
its contents.
4.2.3 Water Blank. Save a portion of the deionized distilled water
used in the impingers. Take 200 ml of this water and place it in a glass
sample container labeled "water blank".
4.2.4 Methylene Chloride Blank. Place 200 ml of unused methylene
•chloride in a glass sample container labeled "methylene chloride blank".
4.3 Analysis. Same as Method 5, Section 4.3. In addition, handle
Containers No. 4 and 5 as follows (an example data sheet is shown in Figure
5A-1):
4.3.1 Container No. 4. Note level of liquid in container and confirm
on analysis sheet whether leakage occurred during transport. Adjust the
sample to a pH of 2 with hydrochloric acid using short range pH paper.
Transfer the contents to a separatory funnel. Extract the water with
three 25-ml portions of ether, followed by three 25-ml portions of chloroform.
(Note: Due to the flammability of ether and the health related effects of
chloroform, the use of these reagents should be carefully supervised.)
Combine the organic phases in another separatory funnel and back extract
with 100 ml of deionized distilled water. The aqueous phase is not retained
for analysis.
Filter the organic phase through a 47-mm glass fiber filter to remove
any particulates entrained in the solvent. Then, evaporate the filtrate at
ambient temperature and pressure until the solvent appears evaporated.
Desiccate the remaining residue for 24 hours, and weigh to a constant
weight. Report the results to the nearest 0.1 mg.
4.3.2 Container No. 5. Note level of liquid in container and confirm
on analysis sheet whether leakage occurred during transport. Transfer the
contents to a tared 250-ml beaker, and evaporate to dryness at ambient
temperature and pressure. Desiccate for 24 hours and weigh to a constant
weight. If more than 10 mg of residue is present, dissolve in 25 ml of
ether followed by 25 ml chloroform. Combine the extracts, and filter through
-------
1 Plant.
2. Data
tun No..
.fethylene Chloride
5. wash volume, ml
Methylene chloride
-7 wash blank, mg
9. Water volume in Impingers, ml.
1C E-C wash blank, mg
Methylene chloride
blank volume, ml
Methylene Chloride
ft
°- blank concentration, mg/mg
8. E-C water blank volume, ml -
11. Filter No..
CONTAINER
NUMBER
1
2
TOTAL
WEIGHT OF PARTICIPATE COLLECTED,
mg
FINAL WEIGHT
^^xd
TARE WEIGHT
. •
^xd
Less methylene chlorid
blank
Weight of paniculate matter
WEIGHT GAIN
k
mn-
CONTAINER
NUMBER
4
5«
. TOTAL
WEIGHT OF CONDENSIBLE ..
ORGANIC MATTER COLLECTED, mg
FINAL WEIGHT"
^XI^
TARE WEIGHT
HxC
Less E-C wale r blank
Weight of condensible organic matter
WEIGHT GAIN
mc-
mb*
mo"
•Data corrected for methylene chloride blank
FINAL
INITIAL
LIQUID COLLECTED
TOTAL VOLUME COLLECTED
VOLUME OF LIQUID
, WATER COLLECTED
IMPINGER
VOLUME,
ml
SILICA GEL
WEIGHT,
9
g» ml
CONVERT WEIGHTOFWATER TO VOLUME BY
TOTAL WEIGHT INCREASE BY DENSITY OF WATER (1g/ml).
INCREASE, g
Ig/ml
VOLUME WATER, ml
Figure 5A-1. Analytical data.
-------
a 47-mm glass fiber filter into a tared 250-ml beaker. Evaporate to dryness
at ambient temperature and pressure. Desiccate for 24 hours and weigh to a
constant weight. Combine with the weight obtained for Container No. 4 in
Section 4,3.1.
4.3.3 Water-Ether-Chloroform Blank. Transfer 100 ml of water taken in
the field as a water blank into a separatory funnel. Extract the water with
three.25-ml portions of ether followed by three 25-ml portions of chloroform.
Combine the organic extract in a tared container and evaporate the contents
at ambient temperature and pressure. Desiccate for 24 hours and weigh to a
constant weight. Report the results to the nearest 0.1 mg.
4.3.4 Methylene Chloride Blank Container. Measure methylene chloride
in this container either volume trie ally or gravimetric ally. Transfer the
methylene chloride to a tared 250-ml beaker and evaporate to dryness at
ambient temperatures and pressure. Desiccate for 24 hours and weigh to a
constant weight. Report the results to the nearest 0.1 mg.
5. Calibration
The calibTat-ion-^pTo-cedures-are—the—samB"aa'S^Met:ho*-5i^-Sect±OTI-^.—--r-— «• -
6. Calculations
All calculations and nomenclature are the same as in Method 5, Section
6, with the fc .lowing additions.
6.1 Nomenclature. Add the following:
Ce » Ether-chloroform-water blank residue concentration, mg/ml.
m^j = Residue mass from methylene chloride rinse of impingers after
ether-chloroform extraction and evaporation, mg.
mn* =» Total amount of particulate matter collected (probe wash and
filter), mg.
mc = Residue mass from EC extraction of impinger water after
evaporations, mg.
me = Extract residue -mass—of- water "blank,- mg.- -
mQ = Total amount of condensed organic matter collected, mg.
Ve = Volume of water blank used in the EC extraction, ml.
*Repeat from Method 5.
-------
vww a Volume of liquid charged to the impingers (see Figure 5A-1),
ml.
We = Total extract residue mass of water blank, mg.
^m(std)* = Volt™6 °f Sas measured by the dry gas meter, corrected to
standard conditions dsm.
Ct » Total particulate and condensed organic matter concentration,
gm/dsm
6 . 2 Ether-Chloroform-Water Blank Concentration.
Ce » m^ Eq. 5A-1
Ve
We - CgVww Eq. 5A-2
n^ = m^ + m^j - We Eq. 5A-3
6 . 3 Total Particulate and Condensed Organic Matter Concentration.
Ct • (0.001 g/mg) (mn + mo) / Vm(std) Eq. 5A-4
7. Bibliography
7.1 Same as Method 5, Citations 1 to 9.
7.2 McGaughey, J.F., and D.E. Wagoner. Special Analyses of Samples from
Sinter Plants in the Iron and Steel Industry. Research Triangle Institute.
Research Triangle Park, N.C. July 1978.
7.3 Evaluation of Sampling Techniques for Sintering in the Iron and
Steel Industry. A Draft Report by PEDCo Environmental, Inc., 11499
Chester Road, Cincinnati, Ohio 45246.
*Repeat from Method 5.
-------
APPENDIX D
CHLORINE TEST FIELD DATA SHEETS
-------
DATA SHEET
CHLORINE COMPOUND TESTS
PLANT
DATE
SAMPLE NUMBER
CHLORINE
y
,
I
Impinger No.
2.
Dilution
Required
Absorbance
| . 3 5"
'
1,2.1
mg/1 Cl?
C
ml ^02
Added
* ml 30% H202 required
0.003
(sample -volume, ml) (Cl2 concentration, mg/1) x
CHLORIDE
Impinger No.
Cl~ mg/1 HC1 mg/1 Cl~ mg/1 HC1 mg/1.
By HqCl? By HqCl? By AgCl By AgCl
-------
PLANT
DATE
DATA SHEET
CHLORINE COMPOUND TESTS
6//W, l2J* •
SAMPLE NUMBER
O.S
IS]
CHLORINE
Impiager No.
3
*
mL 30%
0.003
Dilution
Required Absorbance mg/1 Cl?>
ml H202*
Added
.
2. <*>,/»*/& "4
"
required * (sample volum, ml) (C12 concentration, mg/1) x
Ne»
Cl" mg/1 921 mg/1 Cl" mg/1 HC1 mg/1
By HqCl? By HqCl? By AqCl By AqCl
-------
DATA SHEET
CHLORINE COMPOUND TESTS
PLANT
DATS
a--ia-3o
SAMPLE NUMBER
CHLORINE
Impinger No.
z
z
3
3A
Dilution
Req*"iTed
0,
Absorbance
mq/1 C1-?
ml 1202*
Added
ml 30%
required » (sample volume, ml) (Cl2 coacentration, mg/1) x
Cfl-
i
CHLORIDE
Intpinger No.
Cl" mg/1
By HqCl?
HC1 mg/1
By HqCl?
d" mg/1
By AqCl
HC1 mg/1
By AqCl
-------
DATA SHEET
CHLORINE COMPOUND TESTS
PLANT NASCQ IVi'lmi'ufrJBn .~L>
DATE I2--I ft -
SAMPLE NUMBER 3 fi
af-fcr- Oi
^ **nu_ 0,0/0
«-' ' >-•
' - 6S
/
CHLORINE '"^i "5svr-4r><^i'C ctfS&'M-te. */6^>clt 2- >».) ^ 5"£) *»-f
Dilution *- "^ H2°2*
|- I • Imgincrer No. Required Absorbanee mg/1 Cl? Added
-Janu. ^i^ a- •
z- -
Z
* ml 30% H202 required » (sample volume, ml) (C12 concentration, mg/1) x r
0.003 f I -L
C^,C
CHTiORIDB
Cl" mg/1 331 mg/1 Cl~ mg/1 HC1 mg/1
Impiaggr No. By HqCl? By HgCl? By AoCl By AqCl
-------
DATA SHKJS7
CHLORINE COMPOUND TESTS
PLANT AJASCQ Ut1*1nnQ4on
DATE "PgCSmr 19 . (930
^f A
SAMPLE NUMBER »_^_____________________________
/3
-------
DATA SHEET
CHLORINE COMPOUND TESTS
PLANT
A/A SCO
DATE ""Pe.c'e.m\9g.^ 19 I960
SAMPLE NUMBER
\*/<\™\*a4ori , LXg.ldVQd.rg.
CHLORINE
Imoinaer No
j.
/
3
3
Dilation
Required
Absorbance
ml H2O2*
mg/1 Cl? Added
0.05^/50 4
Q.O\5
ml 30%
0.003
required » (sample •volume/ ml) CC12 concentra-fcion, mg/1) x
CHLORIDE
latpinqer No.
Cl' mg/1
By HqCl?
El mg/1
By HqCl?
Cl- mg/1
By AqCl
HC1 mg/1
By AoCl
-------
'MoteSbok No.I
5OJECT.
Continued From Page-
Read and Understood By
Signed
Data
Signed
Date
-------
Notebook No..
UJ\v
Continued From Pag«
**««•
•>•*
™«M
1
1
]
1
1
j
„' 1.
^ !
r -
F
_
z
"L^"
~ • •>—
•i^*'
j-
~~
—
r
U
— '
—
r
*••• nil . . _
_.,_
__
••»•••-
""
l
1
1
]
1
P
LJ
rP
J
P
_;.
L,
J :'
r
r
W">^H
r
Ltt /ggr
z
L
~~
P
t
L
_
t
L
r
r
r
r
P
b
t
r
n
_
q
q
q
3
4 .
,..(
— i —
ii
•
^^^ - -T
^— — |L_
»
P
__
z
^
•f.i
q
p
_
p
__
z
t
p
r
p
p
p
h
P
u
P
P
_
q
z
q
q
3
^^
"~~
~
r
z
w
c
~
c
p
b
p
p
P
p
p
p
p
P
p
i
z
3
__ ^
"~
-
"^•"•^
q
zl
c
p
b
p
~
p
z
2Jl
r
z
— ^
t
p
b
P
P
P
p
q
P
z
— -
2u
Z
3
q
q
P
i
z
z
p
b
p
b
p
p
_
c
t
_
>^_
-
p
b
p
b
_l
q
q
3
_j
^^^•.-*
3^
q
_
*"^««»
_
q
q
__
q
3
3
i
3
P
b
p
_
b
b
b
~
fe
p
r
a'
P
b
b
t
b
z
q
b
••.^•^
_j
a
q
3
•^••••B
q
q
P
_
•" i ••
c
I
t
z
p
p
p
^
p
b
p
•c
~
P
__
t
P
r
P
P
P
P
n
3
q
z
z
_^
q
q
q
q
3
3
r:
__
•'••• M
*• • 1.
~
P
t
11
__
LZ_
p
t
L
^
Z
r
P
P
_
. A
r
"' i..
;-,J/
LZ~
P
Z
t
t
Z
P
t
b
b
JI
3
3
—
3
__
3
3
3
_
P
3
p
t
_2
±
P
q
P
p
P
P
P
•ss
p
s.
z
z
__
q
__
3
q
j
r
p
t
z
z
z
H
z
t
z
z
/i
z
c
*
r
i
E
p
z
p
P
E
<3
3
»
3
3
J
_
3
q
z
3
3
3
P
P
h-
P
Z
Z
.I
j<
P
_
?
£±
c
s
~4
P
P
d
t ^f
q
&<
P
_
z
_
P
_
P
p
P
3
•MM*
~
Z
t
^
£k
~
*
P
£
P
M
P
P
if
P
3
P
LH
J
Z
_
3
_
,
Zl
3
P
Z
Z
r
^
~~
p
L
Z
*£
Z
_,
t
1-!,
*t-
<.T
r
M
P
P
n
q
P
4^^*
q
3
q
~
~-
p
_i
_
,
q
q
3
p
z
t
^
p
_
p
p
r~
• • "i s«
i«
p
o
-k
q
P
n
p
p
i V
p
H
5
q
_
__
_
•^™^^«
__
-H
P
_
••••••
r
p
Z
r
z
p
_
a
z
r
P
p
r
p
p
to
~
•^•MPB
P
^J
r
P
Z
^~
z
r
h~
P
r
te
z
K
p
p
z
__
p
— _
* .
3
M
H
P
U
P
q
q
q
z
3
Z
3
^•^••^
i
3
n
n
q
q
z
_
q
,
3
3
3
c
t
Z
r~
P
t
t
^
z
p
^
z
fc
_t
G
P
r
Q
h
3
P
r~
~ i
^
— .
S
^^
z
z
q
q
P
_
3
Vi
z
_J
P
z
r
p
Z
h
t
Z
p
b
z
»•*— -i— p
z
z
p
z
z
p
p
p
p
p
^k
••«••••»•
^^i^»«
~
r-
~
p
p
r
z
z
u
p
z
p
p
p
p
p
p
s
q
u
^3
p
p
__
1
q
q
P
3
3
'^•^'•^
•
P
_
Z
_,
3
q
_>
P
q
3
r
p
z
^
c
p
p
p
z
p
__
p
z
p
_
eTVj
p
P
^.
S3
~
.
/ft
~
3
Vf,
Z
z
q
P
3
p
r~
p
t
Z
P
~
Z
r
_
_
t
P
fe
u
Z
t
p
"•"i •
JV\
^ • II
1
5
~i
.2
z
__
_
-,
_
P
q
q
q
3
p
p
t
p
z
1
t
t
h
t
z
c
p
z
H
B
p
3
s
p
Q
I!
••• .
P
Z
Z
r
p
Z
Z
_
_
t
p
g
H
ra
P
P
q
q
3
p
a
_j
_^_
1 yn^
ZT
,
Z
_J
_j
n
3
P
3
q
,
q
_j
q
q
q
_j
P
_
Z
Z
Z
_
P
~
P
z
p
t
L
Z
t
P
r
d
q
6.,
3
3
3
UX
I
P
z
_
3
__
3
3
J
C
p
p
r
p
z
z
t
z
z
z
q
q
q
*
P
P
pi
P
_,
,
q
'*>
q
i
3
Zl
z
q
q
3
z
L
z
p
h~
~
p
p
p
z
p
p
p
r
P
b
23
3
3
Z^
q
q
q
P
P
q
3
3
q
q
p
3
C
L
i
c
P
^~
p
p
b
z
t
z
b
P
r
P
A
1 —
~
~
3
q
3
q
3
,
3
q
3
3
3
*!•! •!,
P
P
P
b
p
z
z
t
p
b
p
p
_
t
t
b
t
•> i •••
E
S4<)
"•^ • ^l
__
1M
P
z
q
^« ••*
•^••MH
•^••^W
q
z
_i
i
•• •
i
3
E
L
r
r
r
h
t
r
r
r
r
r
_
_
r
r
ic»(.
—
z
\^<
— '•
• •. ».
•l ...
•• 1-
^••. 1
^•^•fc
Continued on Page
. — _
Read and Understood By
Signed
Oat*
Signed
Date
-------
•KOJECT.
Signed
Notebook No..
Continued From Ptgt
*mmd »"<* Understood By
Oat*
Signed
Oete
-------
"Notebook Not
•ROJECT
Continued From Page,
Read and Understood By
Signed
Date
Signed
-------
Notebook No..
u
ROJECT.
Continued From Page.
3A-
-------
OJ pmr
Continued From Page-
<5>L
00
T
to
b.zi
7*
Continued on Page
Read and Understood By
Signed
Data
Signed
Date
-------
•JECT_
1MOT6OOOK IMO._
Continued From Page
D
xti.
-
n
>
fr
\J
0
a
(TO
•«•
r'4
t
i.kJ
9?
o
.». :
Continued on Page
Read and Understood By
Signed
Data
Signed
°»««
-------
<.'ROJECT_
Continued From Page.
*M
LL
1
^
D
i '.
; l/
o o
0, 2Jo t
^
/ p
Pi
412-
U
,3 '
/C
^
-
-=- 12.
^ e?7
51
.'/< O
HC£7
b,«to
Continued on Page
Read and Understood By
Signed
Data
Signed
Data
-------
FIEI.I. ,ITA
riAiu__
IIAIE
lU'EIIAIOII
AMIHUII IIUrCNAlllllE 7^^
IIAIIOUEIIIIC riiEiiiiiiE _2^£^_Z__
l>A|ir. rillUIIIIF (P I ~3r\ ' 1J
,. • -n « L\
iiUEiiiimmEfi(i).
1
nor/i cm..
SAMI'I E IKIX IIIIUIIEM
iiEicnnnxmnuiER
HEIEII Ally
crAcimi C
flinilEIIEAIEItSETTIIlO.
IIEAIIIMIDXJEIIIIIC
rrr
nifiMw:i.At__n°
o.OO
' '''
jciiHiAiicoFiiiAVEnJE.romuAYoiir
HEAO AIIO iiEconn ALL DATA EVEIIY.
MIHIIICI
in AVI /HE
rnuir
IIIIMIII n
A
cine* nut
O
CAS UEIEII REAnillG
VEincuir
HEAD
OIIIFICE niEJIIUIE
pirrEncnikAi.
(A!!!. !-. !!0>
nEJIIIEI)
ACTUAL
•iti b
HACK
UUrUlAIIIIIE
nnwi MEIEII
HMI'EnAIIUlE
IKIEI
OIIILET
VACIIIIfl.
JAMJ'I E IIOX
lEurniAiiiiiE,
luniir.Eil
lEIU'EHAIIIIlE.
VZ
So
X.
?o
bo
2.64
2.16
D
n
&<>
2(
80
lu
Q.-sr
10
//o
10
/o
_!ii
2
rj
90
JttU,
33-
l.H
2.U
So
/o
• 7
,-i
-------
inAvmiE
rniiir
HiiMiitn
0
3
C
FIELf. .ITA
riAiir
I»A1 E /a-13-SO
jAurimr, incAiioii
JAMH E ivrE
IIIIHIIIIMIIEII.
on: n A i on _J
AMI!IIIII HHrEIIAIIIIlE
DAIIOUEIIIIC rilEJJIHIE . 3
-------
I |LU» .11A
n.Aiii
IIATE 13-tE-y (
SAuri.uiGiocAiioii
WE
1'IIOIIE 11lit! III MIR TYPE
linn titiMiiEii .3 fl
on I! A Kill .
IILIEIUIIIUIIEIUO
"f.3
IIEAIEO IIDX SEIIII1C / OO
IIEn.HEIICE At %'°
JCIIIMAIICOF TIIAVEniE.rOlllf IAY1UII
HEAD AMD IIECOim ALL DATA EVEIIY_JL__ MIHIIIEI
0.'JOC^~» *•*' 51
inAVIIIlE
rotnr
IIIIMUlli
CLOCK HUE
r.AiUEiEiuiEAniiir,
°
,
^4-3.700
VELOCIIY
HEAD
1.0
oairicE riiESJiinE
uirrcnifHiiAL
nCJIHEO ACTUAL
; afr
STACK
UUrCllATUIlE
IHIYCAIUEUII
UMI'EnAtllME
HUH fllllLET
I'lIMP
VAcuini.
In. I'l
JAMI1 E IIOX
lEUrfllAIUIIE,
IMI-IIICEII
inimiAiiiiiE.
'f
/.O
.2-/S-
i
/. y
_43
IbOT,
/. 0
ZJ
15"
/g
fc,3oo
/.
7V
0^o_
/.o
ZD
21
M 50 lf 0
1. 0
Z.O
^i&
U_
l.o
/.o
•7-f
_W
•to
l.o
1,0
21
-;-i
TT
(.0
1-0
_to
to
v/
12
U
J
\c\
V
a
-------
TIEIJ. ,UA
-go
SAuri.iiio i OCA IIOH
$«iri.f
/ r/ £1 c-
rnuiiE i muni Aim WEjPVjktfs
i»._
1
HUH rilNllllUI J.
(in ir A i on
A
ASSIIMEnilQISIIIflE.t.
SAMri.EIIOXIIIIUIlEri_
MEIER mix NuunEn_
uricii AH.,
A Will III lllirCIIAIIIIlE
IMIIOUEIIIIC rilESIIHIE 7?P- °3
IICAIfli!
IIEAILIMIOXSEIimC.
1CIIIMAIIC OF THAVEHIE rOIIII I AYOIir
t
HEAP AJ(0 HECOHD ALL DAIA EVEIIY MI/HUH
in AVI ME
rnmr
NiiMiitn
CLOCK IWL
tad
fiA(UEICIIREAf)llia
VELOCIIY
HEAD
oniricE wEisunE
inrf rntnriAi.
(AMI. I
DEWED
ACIUAL
STACK
uurntATuiiE
(Ijl/f
IIIIVr.AtVEUII
IIIIEI
nurLET
nwr
VACiiiin.
In. Hf
E nnx
iiurniAiiMiE,
luniir.Eii
lEIU'EIIAIIIIlE
•f
1-0
.1ST
\*
-------
\.
riAiu_
jAur HIM OCA noil
I AMI-IE IVI'I
HUH IHM1IIEII
on ii A inn!
i nifiin Aim TYPE.
twin E in ;. / z f
AJSIJMI;I» HIHSIWIE. x
JAMI-I E (IOX IIIIMIIEII
.JMEIEIImixmiunitt
HEIIII A»Jj£2
.11
AMIIIIIII mirniAIIIME J&V
IIAIIOWC line rnnuiiiE J?o.
mm: riirnmif jr^ " 4--
I (LIEN IIIIHIIEII (i)
rilOliE HFAIEll SETTIlin /
iiEAiin unit iEiij(iB /'oo
)$
-%
sciiEUAiieoF TiuvEnJE ronu LAYuur
DEAD Alii) IIECOlin ALL DATA EVEIIY.
rnAVfiiiE
rnmr
IIIIMIIIfl
o
/•/,&-/
18
33
fiASUEICJI HEADING
100
5-5/^00
SS5.
SV7
VEincnr
IIEAO
/. 0
/. 0
.
o
f. 0
I.O
l.o
t.o
i. o
ho
t.o
l.o
1,0
oinncE
(All), la. ll,n>
. 2VS
ACIUAL
STACK
HIIPUIAIUIIE
Jf,!/F
i.i.
7-V
IS"
iinyr.AJwntii
iiwrniAiiiiiE
nut i
niiiiEi
7
VAcniiri.
SAMITE nnx
IfUriHAIIIIlE.
luniir.Eii
lEIII'rilAlllllE.
•f
_2
ll
Zl
O
5T
sv
\.o
SY
1
3L
.1:0
J5i
sr
r. o
• i
/-.
i
=!£.
111
t-
6
-------
FIEU>.,ATA
mm.
-J1AI
jAiir i mo i nr A lion
IAMI-I E ivre
IMIII tiinmrn
on HA ion
riiuiiEidiraiiAiinivrE.
IIOHIE 10
AMIIIMII mirtiiAiuiiE
IIAIIOUf HIIC f IIIJIIHIE.
AUIJMIinilOIIHIIlC. X.
J AMI-IE IIOX IIIIMHEII_
iiLiiiinoxiiiiuntn_
III 11II All,
'tf-
riinilEIIEAIEIISETIIIIO.
IIEAIIIMIOX 1EIIIIIG
ttf.fi.ntncf.Af
HEAD AIIO IIECOni) ALL DATA EVEHY
LAYOUT
MIIIIIIEJ
inAVIIIlE
mini
r.AtUEIEHIIEAniHG
VELOCIIY
HEAD
omricE rnEisunE
inrrfntiuiXL
ncsmtn ACTUAL
HACK
TIUi'EIIATUIIE
DIIVr.AlMCHU
IIMI'rilAllIHE
nun
OIIILET
lf.rtl.-F
I'llMI'
VAciiiin.
!-.. Ui
JAMPI E IIOX
HUrtllAIIIIIE.
luriiifiin
TEIU'EIIAIIIIIE.
"f
c
-6
6;
s-7
7*5
"-&
r
6 c?
Slo.J
75"
•7
510.1
.tfr
TV
<<> 6
/.o
7
/.o
n
(.0
7 '
iSL
te_
JMLl
r.o
(.0
io
A*. 1
576. 'i
_L
-------
FIELD DATA
pi/wr
PROBE LENGIH AND TYPE
NOZZLE 1.0. _-../?*!_ ........
SAMPLING LOCATION
SAMPLE TYPE .HtjUrvM>»
RUN NUMBER £
OPERATOR _
AMOIENT TEMPERATURE Jfofjl
BAROMETRIC PRESSURE
S1ATIC PRESSURE. (Pj)
FILTER NUMBER irv)b
«'«"" F,M*c.^fe
READ AND RECORD ALt DATA EVERY
COMMENTS:
-------
PHOUE LENGTH AND TYPE
NOZ2LE I.D. _, L?i
ASSUMED MOIS1URE.
SAMPLE BOX HIIMOEn
METEH UOX NUMUER
MEIER All
COMMENfS.
-------
ANALYTICAL DATA
PLANT.
DATE_
SAMPLING LOCATION
SAMPLE TYPE
RUN NUMBER
SAMPLE BOX NUMBER.
CLEAN-UP MAN
FRONT HALF
ACETONE WASH OF NOZZLE, PROBE, CYCLONE (BYPASS),
FLASK, FRONT HALF OF FILTER HOLDER
FILTER NUMBER
3(A
7
BACK HALF
IMPINGER CONTENTS AND WATER WASH OF/
IMPINGERS, CONNECTORS, AND BACH
HALF OF FILTER HOLDER
ACETONE WASH OF IMPINGERS, CONNECTORS,
AND BACK HALF OF FILTER HOLDER
,O
M01STURE-
COMMENTS:
"AN
- 0>
^"^ {\
CONTAINER
LABORATORY RESULTS
CONTAINER
.mg-
FRONT HALF'SUBTOTAL"
CONTAINER
ETHER-CHLOROFORM
EXTRACTION
BACK HALF SUBTOTAL
.mg
.mg
mg
TOTAL WEIGHT
me
IMPINGERS
FINAL VOLUME
INITIAL VOLUME
NET VOLUME
SILICA GEL
FINAL WEIGHT *•
INITIAL WFlRHT •
NFT WFICHT
EPA (Dur) m
4/72
ml
ml
ml
^Lfi^.t-f- , ' 3.3.Q, 3 a
S I
e
e
.B'
TOTAL MOISTURE
-------
ANALYTICAL DATA
PLANT.
DATE_
COMMENTS:
SAMPLING LOCATION
SAMPLE TYPE
RUN NUMBER
£> v*
SAMPLE BOX NUMBER
CLEAN-UP MAN
FRONT HALF
ACETONE WASH OF NOZZLE, PROBE, CYCLONE (BYPASS),
FLASK, FRONT HALF OF FILTER HOLDER
LABORATORY RESULTS
FILTER NUMBER
BACK HALF
L
IMPINGES CONTENTS AND WATER WASH OF
IMPINGERS, CONNECTORS, AND/BACK
HALF OF FILTER HOLDER/y
ACETONE WASH OF IMPINGERS, CONNECTORS,
AND BACK HALF OF FILTER HOLDER
3ri l^i^fitJ^&f^ — 5i''i
MOISTURE
fr
IMPINGERS
FINAL
INITIAL VOLUME.
NET VOLUME .
SILICA GEL
FINAL WEIGHT
.ml
.ml
.ml
.( o
CT
NET WE
EPA (Dur) 221
4/72
CONTAINER
CONTAINER
mg
Ac
FRONT HALF SUBTOTAL
.mg
£ ' I
7 no-1
CONTAINER
ETHER-CHLOROFORM
EXTRACTION
BACK HALF SUBTOTAL
_rog
TOTAL
mg
?§&
g
g
.£
TOTAL MOISTURE
-------
ANALYTICAL DATA
PLANT.
DATE_
COMMENTS:
SAMPLING LOCATION
SAMPLE TYPE
RUN NUMBER
SAMPLE BOX NUMBER.
CLEAN-UP MAN
FRONT HALF a\\ ot
ACETONE WASH OF NOZZLE, PROBE, CYCLONE (BYPASS),
FLASK, FRONT HALF OF FILTER HOLDER
FILTER NUMBER
BACK HALF
IMPINGER CONTENTS AND WATER WASH OF
IMPINGERS, CONNECTORS, AND BACK
HALF OF FILTER HOLDER
ACETONE WASH OF IMPINGERS, CONNECTORS,
AND BACK HALF OF FILTER HOLDER
MOISTURE-
IMPINGERS
FINAL VOLUME
INITIAL VOLUME
NET VOLUME
SILICA GEL
FINAL WEIGHT
INITIAL WEIGHT
NET WEIGHT
EPA (Dni) 231
4/72
LABORATORY RESULTS
CONTAINER.
CONTAINER
.mg
FRONT HALT-SUBTOTAL":
.mg
CONTAINER
ETHER-CHLOROFORM
EXTRACTION
BACK HALF SUBTOTAL
TOTAL WEIGHT
.mg
.mg
.mg
_mg
5^3.3
TOTAL MOISTURE
-------
ANALYTICAL DATA
PLANT
DATF
SAMPLING LOCATION CCf-
SAMPLE TYPE lM°
RUN NUMBER _JJl_
SAMPLE BOX NUMBER
CLEAN-UP MAN
FRONT HALF
i(
ACETONE WASH OF NOZZLE, PROBE, CYCLONE (BYPASS),
FLASK, FRONT HALF OF FILTER HOLDER
FILTER NUMBER
I
TC
BACK HALF
IMPINGER CONTENTS AND W TER WASH OF
IMPINGERS, CONNECTOR-, AND BACK
HALF OF FILTER HOLDER
ACETONE WASH OF IMPINGERS, CONNECTORS,
AND BACK HALF OF FILTER HOLDER
MOISTURE
IMPINGERS
FINAL VOLUME .
INITIAL VOLUME.
NET VOLUME .
f
.ml
SILICA i
FINAL WEIGHT
INITIAL WEIGHT
NET WEIGHT
EPA (Dur) 231
4/72
.ml
COMMENTS:
Scr ucS*
U
LABORATORY RESULTS
CONTAINER.
CONTAINER
.nig-
FRONT HALF-SUBTOTOri
.mg
CONTAINER
ETHER-CHLOROFORM
BACK HALF SUBTOTAL
.m|
.m|
.ml
WEIGHT
TOTAL MOISTURE
-------
ANALYTICAL DATA
PLANT.
DATE_
COMMENTS:
SAMPLING LOCATION.
SAMPLE TYPE ,
RUN NUMBER
SAMPLE BOX NUMBER
CLEAN-UP MAN
FRONT HALF
ACETONE WASH OF NOZZLE, PROBE, CYCLONE (BYPASS),
FLASK, FRONT HALF OF FILTER HOLDER
LABORATORY RESULTS
CONTAINER.
.rag
FILTER NUMBER
/M/
V-
Cci
u
v
BACK HALF
IMPINGER CONTENTS AND WATER WASH OF
IMPINGERS, CONNECTORS, AND BACK
HALF OF FILTER HOLDER
ACETONE WASH OF IMPINGERS, CONNECTORS,
AND BACK HALF OF FILTER HOLDER
MOISTURE
CONTAINER
.nig'
FRONT H"ADJ~SUBTOTAT
CONTAINER
ETHER-CHLOROFORM
BACK HALF SUBTOTAL
.mg
.rag
_mg
TOTAL WEIGHT
me
IMPINGERS
FINAI VOLUMF
INITIA1 V01UMF
NFTvni IIMF
SILICA GEL .c,-, -,
FINAI WFIGHT 1 l '' '
INfTl41 WFIRHT l-^"1- ^
NFT WFIf^HT
EPA (Our) 231
4/72
0 V x, i o ^ *
ml f-y t f "^~ C L ^
__| J £i
mi w
ml
S 2
» 1 !£ • ~i j
! I
J
1
g
g
S'
FOTAL MOISTURE
-------
ANALYTICAL DATA
PLANT.
DATE_
\-Uvr
SAMPLING LOCATION
SAMPLE TYPE _
RUN NUMBER_
(-\ 0
SAMPLE BOX NUMBER.
CLEAN-UP MAN
FRONT HALF
ACETONE WASH OF NOZZLE, PROBE, CYCLONE (BYPASS),
FLASK, FRONT HALF OF FILTER HOLDER
FILTER NUMBER
COMMENTS:
dUA)S£
CONTAINER.
CONTAINER
LABORATORY RESULTS
.mg
.mg
- . (Ov£
"^FR(
RONT'tTAlTF-SUBTOTAL-
.mg
BACK HALF
1MPINGER CONTENTS AND WATER WASH OF
IMPINGERS, CONNECTORS, AND BACK
HALF OF FILTER HOLDER
ACETONE WASH OF IMPINGERS, CONNECTORS,
AND BACK HALF OF FILTER HOLDER
0
CONTAINER
ETHER-CHLOROFORM
BACK HALF SUBTOTAL
—mg
.mg
.mg
IMPINGERS
FINAL VOLUME .
INITIAL VOLUME.
NET VOLUME .
SILICA GEL
FINAL WEIGHT .
INITIAL WEIGHT .
EPA (Dur) 221
4/72
.ml
, mi
, ml
g
g
I
TOTAL MOISTURE
-------
APPENDIX E
CHLORIDES TITRATION LAB SHEETS
-------
fc 5
5 \
VJo.
/k
SI
TT
I
.'50
c.ooq-is
s
c.»o
if..no
o. is
1 1i FI
SI
TI
3 V^T
ST.
ST
Tl
TI
3*2 TX
ST.
ST.
TT. ..
TX
TX
^ -x
ST.
TX
TT
SX
TT:
TX
S Ji ' 0.00
...I
^S v
SJL .'o.0<
. 1 ...
•ZS
. . 5
"2.5. .v
SJ. o.cc
Jr
'Z'S
. 5..
10
T,S
Svi Q.-OC
to
I a
£* v >
^ o'c
1 C?
50
q-l? f"3».35 4-335"
0.3? 0.^0
4.1
^1S ...'."% SO. ^.50
O.SO c.^5
"2.-I - -
o.M-o a<^o .
' - '-T0 —
«nS -Z.^O 1.T5
0.10 C"^0
c. yo ~~
<:.os o.io
0^5 -o. i o •
CT.S
>^-i5 if.HO ^^^o
^.Sc 7.sc
C.to C.-LO
r CHS —
^ns ^T ^^7
I o.*S o.'\o
i c.*>o —
-------
Cl
SJl
coo^TS
S.30
C.70
S.30
~^\'~
S nA . C. 0-0
O
"pvj IS
k ^ u
V
C--LK^ i
V
*T V 1.^5 2.3^
i _ O.^O |.OO
?»Jl C-.Ooqn5 I.SO' I. SO
•o "? oo 3 -o S
' ' '
*pw -2,5 V o.io Qio
CiJ-s^ . . 5" »JL 0. . oo^lS .. *S.T9 . 3.15
V.
i
^w 7'5 * - - <2>5 0<"LS
v C^J-fc— i <
3 o oooqns q.\5 «=\.\5
'pW aS V O.S5 O.SO
srtl_
3 ..JL O.ooqis vs.'^o is.o~o
TV "Z'S . . v .. . Q-fS O.SS*
re^- *ni s
-n\
733
-no
TTt
TTZ- >
,JL o. CC^TZ, 7-3 c *7.*3O
-
•• -
5
ITS 5"
T7"S . . 1
"TTO \
S
o {^c^-^N
l.TO ».feS
"o Tol
\.10 \.-?S
a ^o a»o
0. 1 S o. » S
o. i o c. i o
0.15 O.XS
3.0.0 3. ? o
-------
APPENDIX F
HYDROCARBONS FIELD DATA SHEETS
AND STRIP CHARTS
-------
nan. ,»TA
rum
IIAIE <£-!
i-niint iriinniAnn me 5'
mi?/iEin .
j AMD mo i OCA i inn
HIEivrE
AHIIMIill IKIIS HIRE". X .2-
oriiiiAioii
C-r r
JAMI'lE HOXmiMllfll^JjJU
HE i EII nnx imunEn J£_L£
HE an an,, t 87
IIAIIOUE line riinitiiiE 3o/o<
SIAIIC riiEitiuiE. (rti_±iijL_
IIUEIIHIIIIIIEIIM
IIEAHHJETIHIQ
IIEAILII IIOX SEIIinC /OO
scitruAiicoF uiAvcniE twin i Avniir
flEAO AM) IIECOIID ALL DAI A EVEIIY_J_ MIHIIIES
i.
inAvriu.E
mm i
iiiiMinn
A 12-
A- IO
CHICK IIMIi
O
J.
.4-
101 1
cAtiiEiciinEAnuin
VEinciir
HEAD
,•3
omricEriuJsi/nE
DirrcnEiuiAL
IflllLln.HO}
HESinED
3-10
ACIIIAL
ITzT
1
HACK
UUI'EIIATUIIE
71
imyr.Ai VETCH
Illll'EIIAIIIHE
nun
DIIILET
VACIIIltl.
(x. !'£
O
JAMCI E nnx
IfUrHIAIIIIIE.
mirniAimiE.
•f
A-rl
JOO.-OOO
3.30
S"
28
A- 1
3.30
7s-
iOlu
/OU-//00
3.1s-
.30
VJ
102.5
/ 07. 2.00
M/.ooo
'73
•J
'73
3
/n.ooo
1035"
/20
'.13
S
3
2
3
IZi/ioo
1.10
7S'
IZt.'lfrO
r- C
f
/zc . r/o
7Z '
21
A- /I
tf- r/
M-.ffO
o
li
6-'"f
iii
III2
(tfO-lOO
.-56
6-70
'H
II
7.(0
1/10
'7S"
Co
'7
-------
TRAVERSE
POINT '
NUMBER
t?~<.r
0-ll
f$~3
fU
G-i
0>-i
\CLOCK TIME
"—-
1S'1 . IIT-I
GO l|2«f
/,,3 /I"Z~7
fli3o
I13J
77, | !$(,
GAS METER READING
lVnl. II1
1*3.100
/62./00
lie t>.^00
fiiM o o
/'I C». l1/"2-
*'-YV
VELOCITY
HEAD
ldpsl. in. HjO
I/ f "\
LJ-Cl
'^
.Lf«
-^/S
•^>6
ORIFICE PRESSURE
DIFFERENTIAL
(AH). In. HjOl '
DESIRED
?. 3^7
67
/"> /
f i>jt)
''3. 0
DRY GAS MEIER
TEMPERATURE
INLET
6"
OUTLET
" ' W Qy|
»
6,f-
C?li2
(,
7
s'y
7
[>
PUMP
VACUUM.
In H|
-7
7
7
F
^
SAMPLE OOX
TEMPERATURE.
°F
'ZtfX-
2$V
1^1
2^0
wo
IMPINGER
TEMPERATURE.
°F
41/
6 z_
67^
fcfi
bV
-------
PROUE LENG1II AND
NOZZLE 1.0.
SAMPLING LOCATION
SAMPLETYPE
//,, ASSUMED MOISIURE. t _r_.._.
AMPLE BOX NUMBER i>~
ETER BOX NUMBER
METER
C FACTOR
RUM NUMBER N b" - 3-
OPERATOR
AMDIENF TEMPERA! URE
RHODE HEATER SETTING_L^O,*1?
IIEATEH COX SEUIHC_/
-------
IVC
TNAVCntE
POINT
NUMOCR
E
CLOCK HUE
%
CAS DEI EH HEAPING
(V.,1. II
1
•22-2$
VELOCIIV
HEAD
U»ti. in. HjO
ORIFICE PHEUUIIE
DlffEHENTIAL
(All), in HjOt
DESIRED
5-1
ACTUAL
STACK
1EUPERATUH
•T.I.'F
OnVCASyEIER
TEUPERATURE
INLET OUTLET
PUMP
VACUUM.
la ll|
10
SAMPLE BOX
TEMPERATURE.
IWINCER
TEMPERATURE.
•f
4
-2 +C
tf,
ife?
-nil
I'll
%
•Z2M
•rz
lPA{llui)?3C
-------
ANALYTICAL DATA
PLANT.
DATE_
SAMPLING LOCATION.
SAMPLE TYPE _J^_;
RUN NUMBER
SAMPLE BOX NUMBER
CLEAN-UP MAN A^fe-"
COMMENTS:
FRONT HALF
ACETONE WASH OF NOZZLE, PROBE, CYCLONE (BYPASS),
FLASK, FRONT HALF OF FILTER HOLDER
-^r—. ko"-liv'
FILTER NUMBER
LABORATORY RESULTS
CONTAINER.
CONTAINER
FRONT HALF SUBTOTAL
.rag
BACK HALF
IMPINGER CONTENTS AND WATER WASH OF
IMPINGERS, CONNECTORS, AND BACK
HALF OF FILTER HOLDER
ACETONE WASH OF IMPINGERS, CONNECTORS,
AND BACK HALF OF FILTER HOLDER
MOISTURE
IMPINGERS
FINAL'
INITIAL VOLUME.
NET VOLUME .
.ml
.ml
.ml
5^0. (a
-H.1
-0
V'
SILICA GEL
FINAL WEIGHT
INITIAL WEIGHT
NET WEIGHT - "S--5
5 ."
EPA (Our) 231
4/72
CONTAINER
ETHER-CHLOROFORM
EXTRACTION
BACK HALF SUBTOTAL
ms
.1
-7
3,-
- «r.
TOTAL MOISTURE
-------
ANALYTICAL DATA
PLANT.
DATE_
u/n
SAMPLING LOCATION.
NAO
SAMPLE TYPE _L_L:L
RUN NUMBER L
SAMPLE BOX NUMBER
CLEAN-UP MAN & L
J
COMMENTS:
FRONT HALF
CONTAINER.
LABORATORY RESULTS
_mg
FLASK, FRONT HALF OF FILTER HOLDER
FILTER NUMBER *7 ~7 2x-
(
***•-.£>*«.•*• -frJ-T
D APtf Ufll F £ \t**^ " J * M. € W&. If CD ?cf
DMufv nnUr i L. »
IMPINGER CONTENTS AND WATER WASH OF
IMPINGERS, CONNECTORS/AND BACK
HAI F OF FILTFR HOLDER
APFTDNP WASH Or IMPINKFRS CONNECTORS
AND BACK HALF OF FILTER HOLDER
MOISTURE
IMPINGERS ?«rs*0 4d "
FINAl vpl I1NIF, . , ml ^?' i ^'
INIT1AI VniUMF ml >^°' * ^b1^,'
NET VOl UMF. ., . ml ^"™TP~T ^ (- f
SILICA GEL ^ ife-^
FINAL WFinHT L^^ g ^w^.w ,
INITIAL WF'^HT /J-^>'2-0 « ' 2^^ £
NFTWFIRHT — \.T g 1.H.-1 p
CONTAINER me
( (7
FRONT HAI-F SIIRTnTAlr— - - mp
CONTAINER m?
ETHER-CHLOROFORM
EXTRACTION mE
BACK HALF SUBTOTAL mg
TOTA1 WPI^HT ™
B "/ 'e>' \ ^, ^
0 ^ cj
g
2 U? C •'
j- TOTAL MOISTURE "*-/.*? :
EPA (Dur) 231
4/72
•A2.S-
-------
-10-
-30-
-iO-
-6O-
-70-
-90-
-90-
-100
\l
too-
-90-
-0 —
• ~
-30-
-90-
-100 —
-------
-40-
-50-
-90;
-70-
-40-
il-
3-
too-
-70-
-50-
-60-
-9O-
-100
-------
a .
-10-
-20-
-40-
L£
100-
-90-
-30-
-30-
-ao-
-------
'' 0-
m
-30-
-80-
• 80
i
-100
toc-
-90;
-30-
-TOu.
-90
-30-
^207
-10-
-------
10O;
9O
-40-
-70-
-BO-
-90-
-100
Si
=3;
=30=
dfc
-------
!00_-
-ao-
-ao-
—70 —
-30-
-40-
100- —90- 30-
-10-
-------
-10-
-20-
-30-
-50-
-(iO-
-70-
-90-
-90-
-130
3S-,
3? •
HSi ~
100 • 90 -90-
-20 t. MJ>-
-10-
-JO-
-30-
-70-
-100
. STT n *-JfI
-------
-------
APPENDIX 6
HYDROCARBONS LAB DATA
-------
ENGINEERING- S CIENC2
PLANT/LOCATION
DAIZ
?ARTICULATE WEIGHT, FILTER
Run No. /lilt ar No. i 1 1 i
Filter, Gross Wt. ,
z 1
1
I
llaltial Wt., g
We., z
RESIDUE WEIGHT
Acetone Back Half
Run No. ' l&* I/G 5T?
Acetone Blank Vol. . ml
Acetone Blank Residue, z
Acetone Blank Cone., g/1
Wash Volume, ml
Gross Residue We., z
Blank Wt. , z
tfS.I
/) • OOC ^>
o . oc^ 3
^cfc. c)
A.0/5i»
0.50/3
•Net Residue We., * IO-O/V3
Ru.J/w55?
6*.l.
D < CCO (e
C -CCb3
3W' 0
O.OC3I
O 'tCI?
0-00-73
I
RESIDUE WEIGHT
Acetone Probe Wash
ERun No.
[Acetone Blank Vol. , ail
[Acetone Blank Residue, z
{Acetone Blank Cone. , 2/1
[Wash Volume, ml
'Gross Residue We. , z
*.»//6S*t> !$,•,* IQSV
7
Wt
D -CO Ofe
O.COCoZ
\I=>-1
b.oJQ-r
O • OO 7 /
O • C23 (o
1
1
-------
ESCI.NEERINC-SCIEUCE
PLANT/LOCATION
DATE
?ARTICULATE WEIGHT, FILTER
No./?ilgar No.
l7ilear. Grsss Wt., g
ilalsial We. , g
Nee VS.. g
RESIDUE WEIGHT, T£
JRua No . /Beakar No .
ITCE Blank Vol. . al
JTCE 31ank Residua, e
It" Cg s Blank Cone., ^/l
.Wash Voluae, al
!Gross Residua We., 2
[Blank We. , 2
iNac Residue We. . g
1 Pu « / /tf 5^
I *3)?.'5
lO • QGV~I
lO- ,sy«?O
1 'Ce&7
i o . o/^-£
1/CffJ. 3
io • o/«^ B
10 .&&/ 9
i o • o/ oy
>
"Run M Q .'
Uol,
\lcluA-4.
,
£c*\c. /Jl
sw;
o .
0 .
*
O . 0:0-31
a. ooo g
0 . oo S4)*
O,
0 . O/3 I
O .
' OOOO
o . o oo .5
O.
O.OOI/
-------
ANALYSIS
SOLVCNT
jlJANALYSIS STARTED
V t
PROJECT NUHOER
PROJECT NAME
PROJECT HANAGEIl L-
FINISHED
ITEM WE
TARE WE.1GIIT
TEltP
Std^lt,
AUALVSL
PATE
a.ih (t\
Itp'f
**
*•(*
3*
llUMQEfl
FIELD SAMPLE
wt.
3rd w,t»
Uli.Wt,
6th
6th wt,
FINAL TARE
WL. (y)
a
557
rl^V-
ArAt-l
'^fa J^yj'^IrA
5^ctl|]
go£
rcg"
Tcfi
TCls
/7/o
95?.
96-
FINAL VIC!GUT
DATE - -
tElJPJ'F)
Std.Mt,'
AfMLVSt
IlllllUER
a
r a
70
C
Ut Wt.i,
%- 17 V*
IS ,
i v> -i f • /n
_, f Ot
tl
FINAL SAH'L^
TARE WL (g)
QA.HOO
DIFFERENCE
IQ)
o.ool/
0.00 5T4/
-------
ANALYSIS . bOLVENI £
PROJECT NUMBER . 9//£ PROJECT NAME
c
ST/miuD
PROJECT MANAGER
ITU1 UlilGIIEI)
TARE WEIGHT
DATE ,
R.ll.l*
I-IP «f
TEi-IP
Std.VIL
ANALYST
/Of):
/CO •
UUMQER
FIELD SAMPLE
NUHQEI1. •/:•;: MV
Wt.
"4fc**l 1*11* .'
JI u nu«
Ath.Wt,
5th wtri.
6th
G
a
a
Tii><
ry-
G_
/cro,
L4SI
u 9V.9V/0
FIHM. TARE
Wt.
9V.
c/ /,
FINAL
I
DATE • -
.11.11,(«)'
tEl|P(«F)
Ud.Utr
AHALYSt
ilUHQER
70
/QO
"^.0-7 1 8
C15.0130
10
IS ST373
FINAL SAM
U
TARE Ubi (g)
OlFrEilENCE
(g)
0.0007
o.oo/
-------
ANALYSIS
PROJECT
SOLVENT
PROJECT NAME
Ah
. ANALYSIS STARTED
JPROJECT MANAGER
.. G
TARE HEIGHT
FIHISIIEU
ITEM W:i
DATE
l.!l,(*(
7£HP(°F)
Std.Wt,
ANALYST
ff
FINAL TARE
HI. («j)
NUMQER
FIELD SAMPLE
NUHQEfl, ;•;;.)
Wt.
3rd
6th
6th wtr'
Old.?
7 ¥12
32..
«77°
tf-
30
Q77*-
Tun A.
3 - 9
3 3 .0 7^9
FINAL WEIGHT
DATE
TElip(aF)
Std.Mt,
ANALYSt
10
,Jo 00*0
c <-
>=
tfi
"Jo, O£>Ot>
liUIIOER
Ut
FINAL SAM
L^
TAftE Wt, (g)
DiFrERENCE
(g)
-no
ID
A ?72,
o 77^
31.
I?
0,
07
-------
ANALYSIS
PROJECT
SOLVENT
I'HOJECT NAME
ANALYSIS STARTED
MANAGER
G
FINISHED
ITEM U
TARE WEIGHT
{SfH7
TEIIPPF)
Stii.Ut.
.AllALVST
NUMOER
FIELD SAMPLE
'77-
:/.'. 2nd -Wt.
3rd wtii
Uh.wt,
&tl» wt4:
6 til wU
FINAL TARE
Wt. (g)
one
. 57*7^
O 7?*
lurt Mrt
TO.
•a
Q -77?
Q -7**
la
A A
t? IfY
739 1
30.7
FINAL ME I GUT
DATE •
n.ii,(t)'
f£llP(flF)
Std.Wt,-
AllALVSt
llUIIOER
I
Til
A
-a,.
7o
C.U
3 /.
33
.2-9.^0
/?*/
1 .4th:wtj'
.,"..•] i -1?
7-"?.
6$
t». *U OO
Z-J.XZ.l4
FINAL SAM
LE,
TARE UL (g)
33.TFPO
•30, -73
OiFFEItENCE
(g).
o.
(3, /
o.
-------
SHGXSZZZZSG-SCI2JC2
DAE
?ASTXCULAZS EZXGST, 7ILIZ2.
iSua Ho./Fiicer No.
„:?/? /<3 7*3!
Pun
Filrar, Grass Wt. , 2
-77.
Nec Wtt. ,
Ron So./?ilcer So.
Run SL /O 77?
Filrarv Gross Wt.
31, 3/02.
?/.
Jaldal Wt.
1393
.??. mo
. 9/4.2,
0,1
<7.*j99 Q
-------
APPENDIX H
NYLON BRUSH/METHYLENE CHLORIDE EVALUATION LETTER
-------
ENGINEERING-SCIENCE
7903 WESTPARK DRIVE • McLEAN, VIRGINIA 22102 • 703/790-9300
CABLE ADDRESS: ENGINSCI
TELEX: 89-9401
May 27, 1981
9105.00/1
Mr. Gene Riley
U. S. Environmental Protection Agency
Emission Measurement Branch, (MD-13)
Research Triangle Park, N.C. 27711
Re: EPA Contract 68-02-3541, W.A. 1
Dear Mr. Riley,
The purpose of this letter is to report the results of the tests
conducted by ES-revalua.-ting •.ther:suitabili.ty-..afJ nylon, probe., .brushes .for- I
with methylene chloride (MeCl2) solvent. The evaluation was conducted
during May, 1981 in conjunction with the standards development testing
program for secondary aluminum smelters. Provided below is a brief
summary of the objectives of the tests, the procedures used, the results,
and the conciusior ;.
Objectives
The purpose of the evaluation was to establish whether nylon brush
material contaminates MeCl2 as a result of extraction of materials from
the nylon into the MeCl2 solvent or partial solubility of the nylon.
Procedures
Three new nylon brushes identical to the ones used by ES for wash-
ing sampling probes (Fisher Catalog No. 03-578) were tested. Each
brush was washed in soapy water, rinsed in distilled dionized water, and
then air dried.
Twelve beakers were desiccated and tare weighed (a set of four beakers
for each brush) and 100 ml of MeCl2 was added to each beaker. Each brush
was then soaked successively three times, once in each of three beakers.
The fourth beaker in each set was used as a blank. The soaking times in
the MeCl2 were varied from 1 to 15 minutes as shown in Table 1.
After soaking the brushes, the MeCl2 in each beaker was evaporated
and the beakers were desiccated, and the residue weighed.
OFFICES IN PRINCIPAL CITIES
-------
ENGINEERING-SCIENCE
Letter to Mr. Riley
May 27, 1981 V
Page Two
Results
Table 2 shows the residue weights in the beakers corrected for the
MeCl2 blank. Each brush was then soaked successively two additional
times, once for 60 minutes each in tare weighed beakers containing 100
ml of MeCl2 and once for 5 minutes in tare weighed beakers containing
100 ml of MeCl2« The 60 minute soaking was performed in order to
establish whether further conditioning of the brushes in MeCl2 would
reduce the contamination to a negligable level during a subsequent 5
minute soaking.
Table 3 shows the residue wieghts for the 60 minute soakings and
the 5 minute soakings for each brush, corrected for blanks.
»
Conclusions
Based on the results of the tests, it is concluded that nylon brushes
are unsatisfactory for use with methylene chloride for probe washing. The
nylon is apparently slightly soluble in the MeCl2« The results show that
the largest amount of residue remains after the initial soaking (all resi-
due weights were grerater than 3mg) and also after the one hour soaking.
The nylon material appears to become less resistant to attack from MeCl2
following a prolonged 'soaking. All residue weights were greater than 2 mg
following the final 5 minutes soakings.
It is believed that friction between the brush and probe during wash-
ing would increase the amount of residue, and that the residue contribution
from the probe brush would be greater than 2 mg. For this reason it is
recommended that washing of the probe be conducted by first rinsing the
probe with MeCl2 three times without a brush, followed by brushing with
acetone three times, and a final rinse with MeCl2 without a brush.
Sincerely,
ENGINEERING-SCIENCE
Jonathan S. Greenberg
Air Engineering & Monitoring
JSG/rl
cc: George Weant
Don Holtz
------- |