United States
             Environmental Protection
             Agency
            Environmental Monitoring and Support
            Laboratory
            Research Triangle Park NC 27711
EPA-600 4-79-058
September 1979
             Research and Development
&EPA
Test  Methods to
Determine  the
Mercury Emissions
from Sludge
incineration Plants

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination  of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental  Health Effects Research
      2.  Environmental  Protection Technology
      3.  Ecological Research
      4.  Environmental  Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical  Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special"  Reports
      9.  Miscellaneous Reports

This report has been assigned to the ENVIRONMENTAL MONITORING series.
This series describes research conducted to develop new or improved methods
and  instrumentation for the identification and  quantification of environmental
pollutants at the lowest conceivably significant concentrations. It also includes
studies to determine the ambient concentrations of pollutants in the environment
and/or the variance of pollutants as a function of time or meteorological factors.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
TEST METHODS TO DETERMINE THE MERCURY EMISSIONS
        FROM SLUDGE INCINERATION PLANTS

                       by

 W.  J.  Mitchell, M. R.  Midgett, and J. C. Suggs
Environmental Monitoring and Systems Laboratory
     U.S.  Environmental Protectfon Agency
 Research Triangle Park, North Carolina 27711

                       and

                   D.  Albrinck
             PEDCo Environmental,  Inc.
              Cincinnati, Ohio 45246
   ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
        OFFICE OF RESEARCH AND DEVELOPMENT
      U.S.  ENVIRONMENTAL PROTECTION AGENCY
  RESEARCH  TRIANGLE PARK,  NORTH CAROLINA 27711

-------
                               DISCLAIMER
    This report has been reviewed by the Environmental  Monitoring Systems
Laboratory, U.S.  Environmental  Protection Agency, and approved for pub-
lication.   Mention of trade names or commercial products does not con-
stitute endorsement or recommendation for use.
                                     11

-------
                                FOREWORD

     Measurement and monitoring research efforts are designed to
anticipate potential environmental problems, to support regulatory
actions by developing an in-depth understanding of the nature and
processes that impact health and the ecology, to provide innovative
means of monitoring compliance with regulations and to evaluate the
effectiveness of health and environmental protection efforts through
the monitoring of long-term trends.   The Environmental Monitoring
Systems Laboratory, Research Triangle Park, North Carolina, has
responsibility for: assessment of environmental monitoring technology
and systems; implementation of agency-wide quality assurance programs
for air pollution measurement systems; and supplying technical support
to other groups in the Agency including the Office of Air, Noise and
Radiation, the Office of Toxic Substances and the Office of Enforcement.

     This study was conducted to develop improved test methodology for
measuring the mercury emissions from sludge treatment plants.   Two
methods were developed and evaluated.   One method deals with measuring
the mercury content of the stack gases while the other deals with
measuring the mercury content of the sludge.
                                   m

-------
                               PREFACE

     U.S.  Environmental Protection Agency (EPA) regulations limit the
total atmospheric emissions of mercury from sewage sludge drying and
incineration plants to 3200 grams per 24-hour day (1).

     These emissions can be measured either directly by means of a
stack test or, as is usually done, indirectly by means of a sludge
analysis [EPA Method 105] (1). If sludge analysis is selected, three
sludge samples are collected over a 24-hour period such that each
sample represents a composite of eight consecutive 200- to 400-ml grab
samples collected at one-hour intervals.  The total weight of sludge
processed in that 24-hour interval is then multiplied by the average
concentration of mercury found in the samples to obtain an estimate
of the daily mercury emission from the plant.

     The sludge analysis procedure assumes that all the mercury in
the sludge enters the atmosphere, which is unlikely.  Thus, if a
sludge incinerator potentially exceeds the emission standard based
on this sludge analysis, a stack test is generally conducted using
EPA Reference Method 101 (1,2); a test method in which the mercury
emissions are collected in acidic iodine monochloride (IC1) and
analyzed by cold vapor (flameless) atomic absorption spectrometry
(FAA).

     After promulgation of the mercury standard, an unpublished study
by EPA indicated that both methods were inadequate for characterizing
the mercury emissions from these plants.  The studies described below
were done to find suitable methods for determining the mercury content
of sewage sludge and the mercury emissions from sewage sludge incin-
erators.  For simplicity, the procedure for sludge analysis is de-
scribed separately from the procedure for measuring the stack emissions.
The actual test methods are described in Appendix A (Stack Test Method)
and Appendix B (Sludge Test Method).
                                     iv

-------
                               ABSTRACT

     Two test methods for mercury are described along with the labora-
tory and field studies done for their development and validation.

     One method describes how to effectively homogenize large volumes
of sewage sludge using a mortar-mixer prior to mercury analysis.   It
was discovered that manual mixing of sludge was inadequate for ensuring
an even distribution of mercury throughout the bulk sample prior to
analysis.  Furthermore, mercury content of sludge varied in a non-
predictable manner with time.   The second method described an improved
procedure for measuring the mercury emissions from stacks on sewage
sludge incinerators.   Stack samples were collected in a solution of
4 percent potassium permangate in 10 percent sulfuric acid and analyzed
for mercury using fTameless atomic absorption spectrometry.

-------
                               CONTENTS
Foreword	iii
Preface	iv
Abstract	   v
     1.   Stack Sampling Method  	
               Introduction  	  1
               Conclusions and Recommendations 	  2
               Experimental	3
               Results and Discussion  	  6
     2.   Sludge Analysis Method 	
               Introduction  	 10
               Summary and Conclusions 	 11
               Experimental  	 11
               Results and Discussion  	 15
References	16
Appendicies
     A.  Method to determine particulate and gaseous mercury
         emissions from sewage sludge incinerator stacks  	 19
     B.  Method *o determine the mercury content of sewage sludge.  . 40
                                      vi

-------
                                  SECTION 1


                            STACK SAMPLING METHOD


                                INTRODUCTION




     Method 101 was originally promulgated as the Reference Method for


measuring mercury emissions from mercury cell chlor-alkali plants (2).


These plants differ significantly from sludge incineration plants in the


following ways:  1)   chlor-alkali plant stack emissions are essentially

                                                             3
elemental mercury and often range from 1000 to 30,000 ug Hg/m , whereas


the sludge incineration plant stack emissions may contain mercury in compound


form and can range from 50- to 300-ug Hg/m ; and 2) chlor-alkali stackgas is


quite clean (containing a mixture of air, chlorine, and elemental mercury).


In contrast, the incineration plant stack gas can contain sooty particulate,


15 to 30% water, mercury, other metals and significant amounts of oxidizable


organic material.


     When Method 101 was subjected to a collaborative test in 1974 using


samples-collected from a mercury cell chlor-alkali plant, it yielded


imprecise and inaccurate results (3).  Based on these results, the method


was later revised to simplify the analytical procedure and to improve the


mercury aeration system (4).  A collaborative test showed that the revised


method was highly precise and quite accurate (4).


     One restriction in the revised procedure limited the IC1 concentration


in the aeration system to 0.0001M.  At higher IC1 concentrations, potassium


iodide (KI), which is used to prepare the IC1 reagent, inhibits the


reduction of mercury by tin(II)chloride (4,5).   This restriction, coupled


with the low mercury levels present in the stack gas, caused the revised


Method 101 to be unsuitable for sludge incinerator stack samples (6).

-------
     The study described below was done to find a suitable method for
determining mercury emissions from sludge incinerator plant stacks.  Two
methods were investigated in-depth.  One method employed IC1 as the
collecting liquid and mercury analysis by neutron activation analysis
(NAA).  The other method used KMnO. as the collecting liquid and mercury
analysis by flame!ess atomic absorption (FAA).  To check for method bias
some KMnO. samples were also analyzed by NAA.

CONCLUSIONS AND RECOMMENDATIONS
     Saopling results from sludge incinerators show that the final mercury
concentration in a sampling train operated at 20 £/min for two hours will
range from 0.2- to 0.6 pg tig/ml.  Based on sample stability considerations,
we have made the following recommendations:
     1)  Use 0.3M IC1 or 4% KMn04/10% H2S04 as the collecting liquid in
         place of a more dilute solution to increase the oxidative
         capacity of the sampling train and do not exceed the oxidation
         capacity of the solution during sampling.
     2)  If KMnO^/FLSO^ is used as the collecting solution, prepare the
         reagent immediately before use.
     3)  Place 50 ml of the collecting reagent in the first impinger and
         at least 200 ml in the second and third impingers.
     4)  Place the contents of the entire sampling train in one jar and
         rinse all glassware with the collecting reagent.
     5)  Store the samples only in acid-washed glass jars with Teflon
         cap liners and process the samples as soon as possible after

-------
         collection.



     6)  At least one sample from each plant should be analyzed using the



         Method of Standard Additions regardless of the procedure used.







EXPERIMENTAL



     Prior to conducting field tests, attempts were made to remove the



interference of KI on the SnCK, but no practical way was found.  One



simple idea, decreasing the iodine monochloride concentration from 0.1M



to 0.05M or less, was impractical because of the high oxidizable organic



content of some of the stack gases that might be sampled.   Attempting to



collect more mercury by increasing the sampling time from two hours to



six  or eight hours was also impractical  because of the high organic



content and high moisture content of the stack gas.   Consideration was



also given to changing the reducing agent, but as shown by Melcher and



Welz (5) an extensive evaluation would be required on numerous samples



from many plants to ensure that another constituent of the stack gas or



in the collecting liquid did not interfere with the analysis.   For these



reasons, it was decided to evaluate two procedures previously found to



yield reliable results.  In one procedure, the sample is collected in IC1



and analyzed by NAA.  In the other procedure the sample is collected in a



KMn04/H2S04 solution and analyzed by FAA (7).



     Relative collection efficiency and sample stability were evaluated



in four field tests.  Collection efficiency was primarily evaluated by



employing single-point, isokinetic sampling with two or more trains



sampling simultaneously at essentially the same point in the stack.



Stability of the mercury-containing samples was evaluated by:



1) analyzing the stack samples immediately after collection and also



                                   3

-------
at intervals spanning several months; 2) spiking stack samples with
mercury and measuring its concentration as a function of storage time;
3) placing aliquots from some samples in 15% nitric acid and comparing
their long-term stability to the original sample.  During this testing
program, the usefulness of placing a filter between the probe and first
impinger and between the third and fourth impinger was also studied.
     The basic sampling train consisted of four Greenburg-Smith impingers
connected in series.  The first, second and fourth field tests were con-
ducted at a municipal sludge incinerator located in Ohio and the third
test was conducted at one located in Maryland.  The Ohio incinerator
processed sludge  from residential sources, while the Maryland incinerator
processed sludge  primarily from industrial sources.  Both incinerators
were equipped with wet scrubbers and had stack temperatures of about 65°C.
     The first field test was conducted in April, 1978, the second in
July 2978; the third in September 1978; and the fourth in May 1979.
During the first  test, the plant suffered a major process upset, and it
was not possible  to obtain a complete sampling run before the large
amount of organic material in the stack completely exhausted the 0.1M
IC1 and 2% KMn04/10%H2S04 mercury collecting solutions.  Therefore, the
test was repeated three months later at a time when plant operation was
normal and the stack's oxidizable-organic material content was con-
siderably lower.  In this second test, four trains sampled simultaneously
during each two-hour sampling run.  Two of the trains contained 0.1M IC1;
the others contained 2% KMn04/10%H2S04.  Only the color in the first
impinger of all trains was bleached during sampling.
     Four-train-sampling was also employed in the third test, but each
train contained a different collecting solution, i.e., 0.1M IC1; 0.3M
                                   4

-------
IC1; 2% KMn04/10%H2S04; and 4% KMn04/10%H2S04-  Four, 3-hour sampling



runs were conducted.  The oxidizable content of the stack gas was so



minute that not even the first impinger solution was completely bleached



during sampling.



     These field tests indicated that the 4% KMn04/10%H2S04 was the best



collecting solution and fTameless atomic absorption spectrometry the best



analysis procedure.  Thus, the fourth test was conducted only to determine



the reserve capacity of this solution.   In this study the paired-train



testing was used.   One train contained 0.5% KMn04/10%H2S04 and the other



contained 4% KMn04/10%H2S04.  Sampling was stopped just before the



oxidative capacity of the train that contained 0.5% KMn04/10%H2S04 was



exhausted, giving a sampling time of approximately two hours.
     Table 1.   SAMPLING RESULTS AT SLUDGE INCINERATION PLANTS.

IC1 Results (pg
Test
Date
7/78



9/78



4/79

Sampling
Run
1
2
3
4
1
2
3
4
1
2
0.1M
NAA
58
125
65
76
74
128
149
90
-
-
0.1M
NAA
42
94
55
91
-
-
-
-
-
-
Hg/m3)
0.3M
NAA
_
-
-
-
110
116
111
107
-
-
KMnO,
0.5% !
FAA FAA
35
85
50
71
83
- 134
- 106
- 115
49
64
, Results
a
NAA
_
-
60
80
87
121
110
107
-
-
2%
FAA
35
87
48
71
-
-
-
-
-
-
(ug Hg/m3)
4%
NAA FAA
_ _
-
67
110 . -
- 80
- 129
- 136
- 141
- 51
- 74


NAA
_
-
-
-
97
121
139
142
-
-

-------
RESULTS AND DISCUSSION

     The results from the three successful field tests are presented
in Table 1.  These investigations combined with sample stability and
filter position studies conducted on stack samples yielded the following
conclusions:
     1)   On the average the  IC1/NAA results for the July 1978 test are
slightly higher than the KNn04/NAA results.  On the average, the IC1/NAA
results for the September 1978 test are  identical to the KMn04/FAA results.
These observations indicate that the two collecting liquids have an equal
collecting efficiency for nercury.
     2)   within individual sampling runs, the agreement between two KMnO.-
containing trains is superior to that between two IC1-containing trains.
This larger variation in the  IC1 results is likely due to the imprecision
in the NAA procedure and is not due to sample instability or sample
recovery errors.  Thus, repeat analyses  on the same sample showed that the
precision of the KNnO./FAA procedure was much better than that for the
IC1/NAA and KMnO./NAA procedure.  For example, compare the following pairs
of results in ug Hg/m  obtained on three samples analyzed by NAA (164, 134;
97, 83; 93, 81) to pairs of results obtained on three samples analyzed by
FAA (80, 77; 83, 80; 129, 139).
     3)   The oxidizing capacity of the  collecting solution should not be
exceeded during testing or collection efficiency or sample stability may
suffer.  The samples themselves seem to  be stable for at least several
weeks as long as an oxidizing medium is  maintained in the solution.

-------
     4)   The suitability of locating a filter upstream of the impingers


to remove some of the oxidizable material was initially studied in the


first field test where it was found that the filter and probe temperature


affected the amount of particulate collected on the filter.  As the

filter/probe temperature increased, the amount of collected material de-


creased.  Due to the poor plant operation conditions during this period

the stack gas contained large quantities of sooty particulate matter.

When this test was repeated in May 1979, the plant was operating normally


and filter temperature did not affect the particulate catch for the single,

paired-train, sampling run conducted.   The Method 5 train operated at 65°C

                                                 o
measured a particulate concentration of 29.2 mg/m  and the Method 5 train
                                                                     o
operated at a 140°C measured a particulate concentration of 28.3 mg/m .   The
                      o
difference of 0.9 mg/m  is within the statistical variation expected for this

type of test (8).


     These results indicate that a filter located upstream of the impingers

can assist in removing particulate matter but may not remove all  of it if the


plant is emitting sooty particulate.  Furthermore, it would be unwise to

attempt to simultaneously measure both particulate and mercury content using

the same sampling train, because it is presently unknown whether any mercury


collected on the filter would be lost during the filter-conditioning steps

necessary to determine the particulate catch.


     5)   In the May 1979 test, glass-fiber filters followed the third

impinger in each train to discover if mercury was passing through the train.

At the conclusion of the sampling run, each filter was placed in a jar that

contained 30 ml  of 4% KMnO.AU^^SO. solution and returned to the laboratory


for mercury analysis.   In all four trains a greyish material  was found on

-------
the filter, but less than 1% of the mercury collected in a train was found
on the filter.
     6)   The brown Mn02 precipitate that formed in the KMnO.-containing
trains during sampling never contained more than one percent of the mercury
collected in the train.
     7)   The standards and samples must contain the same reagents for
accurate FAA results and at least one sample should be subjected to the
Method of Standard Additions (9).
     Overall, the test results show that collection in IC1 followed by  NAA
analysis or collection in KMn04/H2S04 followed by FAA or NAA analysis will
yield acceptable results.  The FAA procedure is preferred from the view of
cost of analysis (NAA can cost as muc as $100 per sample and the quartz vials
must be sealed with an oxy-hydrogen torch), precision and methodology
presently there is no standardized method for mercury analysis by NAA).  On
the other hand, the IC1/NAA procedure is preferred from the view of sample
stability, (once placed in the vial, sample stability does not affect the
results) reagent stability, (the KMnO./HpSO. solution must be prepared within
12 hours of use) and sample recovery (KMn04 decomposes during sampling and
storage to give a brown precipitate and a brown ring on the impinger wall
that must be removed with 8N HC1.).  Before NAA can be used routinely, it
must be standardized in the following areas: sample volume, quartz vial
purity, irradiation time, flux strength, number of analyses per sample,
time interval between irradiation and counting, counting time, wavelength(s)
to be used for counting purposes, and number and concentration of standard
samples to be processed with source samples.
     Regardless of the collecting solution used, the oxidizing capacity of
the collecting reagent must not become exhausted during sampling.  Thus,
                                   8

-------
even though samples appear to be stable for at least several weeks, they



should be processed within several days of collection.   For NAA processing



involves placing the sample in a quartz vial.  For FAA processing means



the complete analysis.   Samples should be processed as soon after collection



as possible, because dilute mercury solutions are notoriously unstable.



Find!ay (7) reported that a 10 ug Hg/ml (in 1% HNO-) standard solution was



stable for about one week, but mercury solutions containing 10 ng/ml in a



completely reduced permanganate solution were stable for less than one



hour.  Feldman (10) reported that solutions containing below 0.025 |jg/ml



were stable for less than 24-hours even in 5% nitric acid.   He also noted



that the stability of dilute solutions was affected by suspended parti-



culate matter and by storage container material,  i.e.,  solutions stored in



polyethylene bottles were much less stable than those stored in glass



bottles.

-------
                                 SECTION 2
                          SLUDGE ANALYSIS METHOD

INTRODUCTION

     In March 1977, the Quality Assurance Branch (QAB) of the Environmental
Monitoring and Support Laboratory, conducted a collaborative test of EPA
Method 105 (1).  In the collaborative test ten laboratories analyzed
portions of grab samples of dewatered sludge taken from three sludge treat-
ment plants.  (Grab rather than conposite samples were used to minimize the
effect of sample non-homogeneity on the results, since Method 105 was con-
cerned only with sample analysis and not sample collection.)'  Each 150-g
grab sample was manually mixed for 15 minutes and divided into 10-g portions.
Each laboratory received a 10-g portion from each plant, dried each sample
and then digested and analyzed 0.2 g portions in triplicate as described in
EPA Method 105 (1).  To provide an alternate determination of Hg content, an
EPA laboratory analyzed a portion of each grab sample  in duplicate  using
neutron activation analysis (NAA).
     The results of this collaborative test clearly indicate that for a
given sample between-laboratory and, at times, with-laboratory, agreement
was poor (Table 2).  A statistical analysis of the results indicated that
non-homogeneity in mercury concentration was a significant cause of the
within-sample variation in at least two of the three samples - an indication
that hand-mixing was an ineffective way to homogenize sludge samples for
mercury analysis.  This finding was important because Method 105 normally
would be used on a composite sludge sample composed of eight grab samples,
and the regulations did not specify any definitive sludge mixing procedure
                                   10

-------
for this composite sample.



     The following study was done to develop and validate a method to



ensure that sludge samples were properly homogenized before chemical



analysis.  This study used sludge samples collected at three sewage sludge



incinerators.   The sample weight and sampling interval were two major para-



meters investigated in detail.







SUMMARY AND CONCLUSIONS



     Dewatered  sewage sludge samples cannot be reliably homogenized,  at



least for mercury (Hg) analysis by manual mixing alone.   However, effective



homogenization can be obtaied using a commercially available mortar mixer.



One unexpected result from the study was the observation that the mercury



content of dewatered sludge entering the conveyor line varied in an un-



predictable way with time.  For example, it was found that a 27 kg (60 Ib)



bulk sewage sludge sample composed of grab samples taken at 5 minute



intervals could have a mercury content statistically different from that



in a similar weight sample collected in the same period composed of grab



samples collected at 10 minute intervals.  Individuals collecting sludge



samples for trace element analysis should consider this factor when



sampling the sludge conveyor line.







EXPERIMENTAL



     The collaborative test results indicated that even within a small grab



sample of sludge there could be a non-homogeneous distribution of mercury.



Therefore, it seemed likely, that a more extreme non-homogeneous distribution



would exist on the sludge conveyor line itself.   Thus, the larger the  bulk



sample taken from the sludge conveyor line, the more representative the



                                   11

-------
                            TABLE  2.   COLLABORATOR RESULTS ON SEWAGE SLUDGE SAMPLES
                                       (ug  Hg/g DRIED  SLUDGE).

NAA
Sample Results
1 1.8
3.9

2 3.0
2.2

3 2.3
2.4


Lab 1
3.8
3.3
0.87
4.1
2.7
1.0
3.8
3.9
0.79

lab 2
2.8
2.7
2.8
2.5
2.3
2.3
3.0
2.9
3.0

Lab 3
2.8
2.8
3.7
2.7
2.9
3.6
2.6
2.8
3.5

Lab 4
6.66
0.82
0.91
0.57
0.48
0.45
0.71
1.1
1.6
Method
Lab 5
1.0
1.5
1.0
1.1
1.6
5.1
2.0
3.8
2.2
105 Results
Lab 6
3.1
3.1
3.5
1.8
2.0
2.2
3.3
2.2
3.8
Lab 7
1.4
1.2
1.2
10.
12.
13.
2.3
1.9
2.7
Lab 8
1.2
1.6
1.4
1.5
1.4
1.3
2.7
2.7

Lab 9 Lab 10
0.33
0.45
0.63
0.18
a
0.94
0.49
0.82

• — 	 	 	 	 	 	
"No Hg detected.

-------
mercury analysis should be, provided of course that this bulk sample could


be thoroughly blended and homogenized before analysis.

     For this reason, it was decided to develop and evaluate a procedure


that could quickly but effectively homogenize large amounts of sewage
                                              *
sludge.  When various sludge mixing procedures were evaluated by adding

zinc oxide to dewatered sludge and analyzing for mercury and zinc as a
                                                           3
function of mixing time, it was found that a 57 liter (2 ft ) capacity

wheel barrow-type commercial mortar mixer rotating at 30 rpm for at least


30 minutes would effectively mix sludge samples.   Mixing times less than

30 minutes were found to be inadequate.




Sludge Sampling

     The effectiveness of this new mixing technique was verified using

time-composite dewatered sludge samples from three sludge treatment plants.


Simultaneously a study was conducted to determine the optimum sampling

interval for collecting samples from the conveyor line for the time-

composite (bulk) sample.  To accomplish this, the sludge conveyor line was

sampled so that over an eight hour interval, 27-kg bulk samples comprised

of grab samples taken at different sampling intervals were collected.   For


example, one bulk sample would be composed of 0.3-kg grab samples collected

at 5-minute intervals and another would be composed of 1.0-kg samples


collected at 15-minute intervals.  Sampling intervals of 5,- 10,- 15,- 20,-

25,- 30,- and 60-minutes were used in this study, although not all  of  these

intervals were used at every plant.
                                   13

-------
Sludge Analysis
     Each 27-kg sample was returned to the laboratory where the entire
sample was placed in the mortar mixer and mixed for at  least 30 minutes
at 30 rpm.  Two to four, 500-ml samples were then withdrawn from the mixer
and each 500-ml sample was blended in a 2-liter Waring-type blender for
two minutes.  Three to five, 20-ml sludge samples were  then withdrawn from
the blender, put into separate tared 125-ml Erlenmeyer  flasks and the flasks
reweighed to obtain the exact weight of sludge added.   Twenty-five ml of
aqua regia was added to each sample and the sample was  digested at low
temperature for 30 minutes on a hot plate, cooled, filtered through S and S
No. 588 filter paper, diluted to 100 ml and analyzed for Hg using the
apparatus described by Mitchell and Midgett (4) and the analytical pro-
cedure summarized below.
     A 5- to 25-ml aliquot of the digested sample was added to 25-ml of dis-
tilled, deionized water in the aeration bottle.  Three  drops of Baker Anti-
foam B Silicon Emulsion were added, followed in sequence by 5-ml of 15% (V/V)
nitric acid and 5-ml of 5% (by weight) potassium permaganate.  The solution
was then mixed well using a magnetic stirring bar.  A solution, prepared by
dissolving 12 g of sodium chloride and 12 g of hydroxylamine sulfate in 100 ml
of distilled, deionized water, was then added in 5-ml aliquots until a color-
less solution was obtained.  Five ml of 10% (by weight) stannous chloride (4)
was added to the aeration cell and the mercury aerated  into the atomic
absorption cell using nitrogen as the purge gas.  Selected samples were also
spiked and reanalyzed by the Method of Standard Additions  (9) to confirm the
accuracy of the original analysis.
                                    14

-------
RESULTS AND DISCUSSION



     Table 3 summarizes the results obtained at each sludge treatment plant



in relation to: frequency of sludge removal from the conveyor belt over an



eight-hour interval (bulk sample); number of 500-ml samples withdrawn from the



mortar mixer; number of aliquots analyzed fronreach blended 500-ml sample;



and the grand mean (ug Hg/g dried sludge) for each bulk sample.   During this



study, a total of 48, 500-ml samples were taken from 18 bulk (27-kg) samples



and at least three analyses were performed on each 500-ml sample after



blending.



     A comparative statistical  analysis on all  results showed that:  1) all



20-ml aliquots from a 500-ml sample were statistically identical; 2) all



500-ml aliquots from a bulk sample were statistically identical  with the



exception of one sample from Plant 1;  and, 3) bulk samples from the  same



plant were not always statistically identical.   For example, the 10- and



20 minute interval samples from Plant 3 (July 1978) are statistically



different from the other four bulk samples.   Analogously, the 5-, 15-,



and 30-minute interval samples  from Plant 1 are statistically the same,



but the 60-minute interval sample is statistically different from the



5- and 15-minute interval samples.



     These results show that the bulk sludge samples were effectively



homogenized but that the actual Hg content of the bulk samples was



affected in an unpredictable manner by the sampling frequency.   Although



the cause of this time variation remains unidentified,  it might originate



from each plant having at least two processes contributing sludge to the



conveyor belt at the same time  but on an intermittent basis, i.e., sludge



would be deposited at 1- to 5-second intervals.   Further, the volume of



sludge deposited on the conveyor line at any one time by each process also



                                   15

-------
varied from minute-to-minute.  If this is the origin, then perhaps a
similar variation will occur when a sludge line is sampled to determine
other trace elements present.  Since this variation would result from the
process itself, the simplest way to minimize its effect on analytical
results would be to collect a  large bulk sample composed of many grab
samples, e.g., take grab samples no further than 15 minutes apart.
     Although undesirable, this rapid variation in mercury content of the
sludge line can probably be tolerated, since the objective of the sludge
analysis is only to determine  if a stack test is necessary.
                                  SECTION  3
                                 REFERENCES

 1.    U.S.  Environmental  Protection  Agency.   National  Emission  Standards  for
      Hazardous Air  Pollutants.   Federal Register, 40:48302-48311.  October
      14,  1975.
 2.    U.S.  Environmental  Protection  Agency.   National  Emissions Standards for
      Hazardous Air  Pollutants.   Federal Register, 38:8835-8845.  April 6,
      1973.
 3.    Mitchell, W. J.,  and M.  R.  Midgett.  Evaluation  of the  EPA Method for
      the  Determination of Mercury Emissions  from Stationary  Sources.
      Quality Assurance Branch,  Environmental  Monitoring and  Support  Labora-
      tory,  U.S.  Environmental  Protection  Agency, Research  Triangle Park,
      North Carolina 27711.   October 1974.  Available  upon  request.
                                    16

-------
4.   Mitchell, W. J. , and M. R. Midgett.   Improved  Procedure  for Determining



     Mercury Emissions from Mercury Cell Chlor-Alkali  Plants.   APCA Journal,



     26:674-677 (1976).



5.   Melcher, M., and B. Welz.  The Use of Sodium Borohydride  as a



     Reductant in Mercury Determinations.  Paper presented  at  the 29th



     Pittsburgh Conference on Analytical Chemistry  and Applied Spectro-



     scopy, Cleveland, Ohio (1978).



6.   Mitchell, W. J.  unpublished results, 1976.



7.   Findley, W. J., K. Li and J. C. Hilborn.  The  Collection  and Analysis



     of Mercury from Mercury Cell Chlor-Alkali Plant Gaseous Process



     Streams.  Paper No. 77-61.1, presented at the  APCA 70th Annual Meet-



     ing, Toronto, Canada (June 1977).



8.   Mitchell, W. J., and M. R. Midgett.  Means to  Evaluate the  Per-



     formance of Stationary Source Test Methods.  Environ. Sci.  and



     Technol., 10:85-88 (1976).



9.   Klein, R., and C. Hach.  Standard Additions, Uses and Limitations in



     Spectrophotometric Analysis.  Amer. Lab. 9:21-27 (1977).



10.  Feldman, C.  Preservation of Dilute Mercury Solutions.  Anal.  Chem.



     46:99-102 (1974).
                                   17

-------
                              TABLE 3.   EFFECT OF SAMPLING FREQUENCY ON MERCURY RESULTS
Sampling
Date
1/78
2/78
4/78
7/78
Plant
No.
1
2
3
3
No. of 500-ml No. of 20-ml
Samples from Aliquots from
Each Bulk Each 500-ml
Sample Sample
2
2
2
4
5
5
5
3
Mercury Content (yg Hg/g dried sludge)
in Each Bulk Sample
5-mina
1.28
2.88
2.65
4.78
10-mina
*
*
*
4.01
15-mina 20-ml a 25-mla
1.19 * *
2.95 * *
2.12 * *
4.82 5.40 4.46
30-mla
1.57
2.61
2.92
4.50
60-mla
1.87
3.33
3.25
*

00

          a
           Sampling interval used in collecting the 8-hour bulk sample.

          *
           Not done at this plant.

-------
                             APPENDIX A

            METHOD TO DETERMINE PARTICULATE AND GASEOUS
        MERCURY EMISSIONS FROM SEWAGE SLUDGE INCINERATOR STACKS
1.   Principle and Applicability

     1.1   Principle.  Participate and gaseous mercury emissions are
withdrawn isokinetically from the source and collected in acidic
potassium permanganate (KMnO.) solution.   The mercury collected (in the
mercuric form) is reduced to elemental mercury, which is then aerated
from the solution and passed into an optical cell in an atomic
absorption spectrometer, where the maximum absorbance at 253.7 nm
is measured.

     1.2   Applicability.  This method is applicable for the determi-
nation of particulate and gaseous mercury emissions when the carrier
gas stream is principally air, flowing in a duct or stack.

2.   Range and Sensitivity

     2.1   Range.  As written, the analytical method is applicable to
source test samples that contain (after initial dilution) between 20
and 800 ng Hg/ml  Samples containing in excess of 800 ng Hg/ml can be
analyzed by this method if diluted.   Samples containing less than 20 ng
Hg/ml per milliliter should not be analyzed by this method because they
are stable for less than 24 hours.

     2.2   Sensitivity.  This depends on the recorder/spectrometer
combination selected.

3.   Interfering Agents

     3.1   Sampling.   Excessive oxidizable-organic matter in the stack
gas will cause premature depletion of the KMnO. solution, thereby
preventing further collection of mercury.

     3.2   Analysis.   Condensation of water vapor on the optical cell
windows produces a positive interference.

4.   Precision

     4.1   Within-Laboratory Standard Deviation.   The estimated within-
laboratory standard deviation of the~procedure in the concentration range
of 50 to 130 ug Hg/m  is 4.8 ug Hg/m .  This estimate is based on eight,
paired-train tests conducted during the study to develop and validate this
method.

5.   Apparatus

     5.1   Sampling Train.  A schematic diagram of the sampling train
is shown in Figure 1.  Complete construction details are given in APTD-0581
(Citation 4, Section 11); commercial models of this train are also available.
For changes from APTD-0581 and for allowable modifications of the train
                                19

-------
                                                                THERMOMETER
       TEMPERATURE SENSOR
             PROBE
             REVERSE-TYPE
              PITOTTUBE
ro
o
                                     THERMOMETER
                                                                                                             CHECK
                                                                                                             VALVE
FILTER HOLDER
  (OPTIONAL)
                                                                                                              VACUUM
                                                                                                                LINE
                                                Figure 1. Mercury sampling train.

-------
(Figure 1), see the following subsections.  Note that use of a flexible
line between the probe and the first impinger is unacceptable.

     The operating and maintenance procedures for the sampling train are
described in APTD-0576 (Citation 5, Section 11).  Since correct usage is
important in obtaining valid results, all users should read APTD-0576 and
adopt its operating and maintenance procedures, unless otherwise specified.
The sampling train consists of the following components:

     5.1.1   Probe Nozzle.  Stainless steel (316) or glass with sharp,
tapered leading edge.   The angle of taper shall be <30° and the taper
shall be on the exterior to preserve a constant internal diameter.   The
probe nozzle shall have a button-hook or elbow design, unless other-
wise specified by the Administrator.   If made of stainless steel,  the
nozzle shall be constructed from seamless tubing; other materials  of
construction may be used, subject to approval of the Administrator.

     A range of nozzle sizes suitable for isokinetic sampling should
be available, e.g., 0.32 to 1.27 cm (1/8 to 1/2 in.) inner diameter
(ID) nozzles in increments of 0.16 cm (1/16 in.).   Each nozzle shall
be properly calibrated.

     5.1.2   Probe Liner.  Borosilicate or quartz glass tubing.   A
heating system capable of maintaining a gas temperature at the probe
exit during sampling of 120 + 14°C (248 + 25°F) may be used to prevent
water condensation.  If a filter is used ahead of the impingers,  use
of the probe heating system is mandatory. Since the actual  temperature
at the outlet of the probe is not usually monitored during sampling,
probes constructed according to APTD-058 and utilizing the calibration
curves of APTD-0576 (or calibrated according to the procedure outlined
in APTD-0576) will be considered acceptable.  NOTE:   Metal  probe liners
are unacceptable.

     5.1.3   Pitot Tube.   Type-S, (as described in Section 2.1 of  Method 2,
Citations 1,2, Section 11) or other devices approved by the Administrator.
The pitot tube shall be attached to the probe (as shown in Figure  1) to
allow constant monitoring of the stack gas velocity.  The impact (high
pressure) opening plane of the pitot tube shall be even with or above
the nozzle entry plane (see Method 2, Figure 2-6b) during sampling.   The
Type S pitot tube assembly shall have a known coefficient,  determined
as described in Section 4 of Method 2.

     5.1.4   Differential Pressure Gauge.  Inclined manometer or
equivalent device (two),  as described in Section 2.2 of Method 2.   One
manometer shall be used for velocity head (Ap) readings, and the other,
for orifice differential  pressure readings.

     5.1.5   Impingers.   Four Greenburg-Smith impingers connected  in
series with leak-free ground glass fittings or any similar leak-free
noncontaminating fittings.  The first,  third and fourth impingers  may
be modified by replacing the tip with a 13 mm (0.5 in.) ID glass tube
extending to 13 mm (0.5 in.) from the bottom of the flask.
                                21

-------
     5.1.6   Acid Trap.  Mine Safety Appliances*  Air Line Filter,
Catalogue Number 81857, with acid absorbing cartridge and suitable
connections, or equivalent.

     5.1.7   Metering System.  Vacuum gauge, leak-free pump, thermo-
meters capable of measuring temperature to within 3°C (5.4°F), dry
gas meter capable of measuring volume to within 2 percent, and re-
lated equipment, as shown in (Figure 1) other metering systems capable
of maintaining sampling rates within 10 percent of isokinetic and of
determining sample volumes to within 2 percent may be used, subject
to the approval of the Administrator.  When the metering system is
used in conjunction with a pitot tube, the system shall enable checks
of isokinetic rates.

     5.1.8   Filter.  Glass fiber, Gelman Spectrograde or equivalent
filter.

     5.1.9   Barometer.  Mercury, aneroid, or other barometer capable
of measuring atmospheric pressure to within 2.5 mm Hg (0.1 in. Hg).
In many cases, the barometric reading may be obtained from a nearby
national weather service station, in which case the station value
(which is the absolute barometric pressure) shall be requested and
an adjustment for elevation differences between the weather station
and sampling point shall be applied at a rate of minus 2.5 mm Hg
(0.1 in. Hg) per 30 m (100 ft) elevation increase or vice-versa for
elevation decrease.

     5.1.10  Gas Density Determination Equipment.  Temperature sensor
and pressure gauge, as described in Sections 2.3 and 2.4 of Method 2,
(See Citation 1, Section 11).  The temperature sensor shall, preferably,
be permanently attached to the pitot tube or sampling probe in a fixed
configuration such that the tip of the sensor does not touch any metal.
Alternatively, the sensor may be attached just prior to use in the field.
However, if the temperature sensor is attached in the field, it must be
placed in an interference-free arrangement with respect to the pitot tube
opendings (see Method 2, Figure 2-7).  As a second alternative, provided
that a difference of not more than 1 percent in the average velocity
measurement is introduced, the temperature gauge need not be attached to
the probe or pitot tube.  (This alternative is subject to the approval
of the Administrator.)

     5.2     Sample Recovery.

     5.2.1   Leakless Glass Sample Bottles.  1000 ml and 100 ml with
Teflon-lined caps.

     5.2.2   Graduated Cylinder.  250 ml.

     5.2.3   Plastic Container.  Approximately 350 ml.


^Mention of trade names or specific products does not constitute
endorsement by the U.S. Environmental Protection Agency.
                                22

-------
     5.2.4   Balance.  Most laboratory balances are capable of weighing
to the nearest 0.5 g or less.  Any of these balances are suitable for
use here.

     5.2.5   Funnel and Rubber Policeman.  To aid in transfer of silica
gel to container; not necessary if silica gel is weighed in the field.

     5.2.6   Funnel.  Glass, to aid in sample .recovery.

     5.3     Analysis.

     5.3.1   Atomic Absorption Spectrometer.  Perkin-Elmer 303, or
equivalent, containing a hollow cathode mercury lamp and capable of con-
taining the optical cell described in Section 5.3.2.

     5.3.2   Optical Cell.   Cylindrial-shape with quartz end windows
and having the dimensions shown in Figure 2.  The cell is wire-wound
with approximately 2 meters of 24 gauge nichrome heating wire and
wrapped with asbestos insulation tape or equivalent; the wires must
not touch one another.

     5.3.3   Aeration Cell.  Constructed according to the specifi-
cations in Figure 3.  A glass frit is not an acceptable substitute
for the blown glass bubbler tip shown in Figure 3.

     5.3.4   Recorder.  Matched to the output of the spectrometer
described in Section 5.3.1.

     5.3.5   Variac.  To vary the voltage on the optical  cell  from
0 to 40 volts.

     5.3.6   Hood.  For venting optical  cell exhaust.

     5.3.7   Flow Metering Valve.

     5.3.8   Flow Meter.   Rotameter or equivalent,  capable of measuring
a gas flow of 1.5 liters/minute.

     5.3.9   Nitrogen Cylinder.   Equipped with a single stage regulator.

     5.3.10  Connecting Tubing.   All  connecting tubing between the
solution cell and the optical  cell shall  be made of glass (ungreased
ball and socket connections are recommended).   Tygon tubing,  other
types of flexible tubing, and metal tubing are not acceptable sub-
stitutes.  Teflon, steel, or copper tubing should be used between
the nitrogen tank and the flow meter valve (5.3.7).   Tygon, gum or
rubber tubing may be used between the flow metering valve and the
solution cell.

     5.3.11  Flow Rate Calibration Equipment.   Bubble flowmeter or
wet test meter for measuring a gas flow rate of 1.5 liter/minute
to within +0.1 liter/minute.
                                23

-------
     5.3.12  Volumetric Flasks.  Class A with penny head standard
taper stoppers; 100 and 500 ml.

     5.3.13  Volumetric Pipets.  Class A; 1, 2, 3, 4 and 5, 10, 20 ml.

     5.3.14  Graduated Cylinder.  25 ml.

     5.2.15  Magnetic Stirrer.  General purpose laboratory type.

     5.3.16  Magnetic Stirring Bar.  Teflon coated.

     5.3.17  Filter.  Whatman No. 40 or equivalent.

6.   Reagents

     6.1     Stock Reagents.

     6.1.1   Water.  Distilled, meeting ASTM specifications for Type I
Reagent Water - ASTM Test Method D-1193-72.  The KMnO. test for oxidizable
organic matter may be eliminated when high concentrations of organic matter
are not expected to be present.  This water must be used in all dilution
and solution preparations.

     6.1.2   Nitric Acid, Concentrated.  American Chemical Society (ACS)
reagent grade, or equivalent.

     6.1.3   Sulfuric Acid, Concentrated.  ACS reagent grade, or
equivalent.

     6.1.4   Hydrochloric Acid, Concentrated.  ACS reagent grade,
or equivalent.

     6.1.5   Potassium Permanganate.  ACS reagent grade, or equivalent.

     6.1.6   Sodium Chloride.  ACS  reagent grade, or equivalent.

     6.1.7   Hydroxylamine Sulfate.  ACS reagent grade, or equivalent.

     6.1.8   Tin(II) Chloride  (Crystal).  Baker Analyzed Reagent Grade
or any other brand of analyzed reagent grade tin (II) chloride that
will give a clear solution when dissolved as described in Section
6.3.3.  Tin (II) sulfate  (crystal)  of the same purity may be sub-
stituted for the Tin (II) chloride, provided it gives a clear
solution.

     6.1.9   Mercury(II)Chloride.   ACS reagent grade, or equivalent.

     6.1.10  Mercury Stock Solution (1 mg Hg/ml).  Completely dis-
solve 0.1354 g of mercury (II) chloride  in 75 ml of distilled water,
add 10 ml of concentrated nitric acid and adjust the volume to 100 ml
with distilled water.  Mix thoroughly.  This solution should remain
stable for at least one month.
                                24

-------
                                    18/9 FEMALE BALL SOCKET
                 LENGTH NECESSARY TO FIT SOLUTION CELL
                     TO SPECTHOPHOTOMETER
                                                   TO VARIABLE TRANSFORMER
  VENT TO HOOD
                                                   331 cm DIAMETER QUARTZ
                                                   WINDOWS AT EACH END
NOTES:
CELL WOUND WITH 24-GAUGE NICHROME WIRE
TOLERANCES ±5 PERCENT
         Figure 2.  Optical cell.
                           25

-------
                                     FROM TANK


                               |     /   *   \ 18/9 MALE BALL
                                     L  ^^^J-	
                                                               4mm BORE TEFLON STOPCOCK
 19/22 GROUND GLASS JOINT
•          WITH STOPPER
              19/22 GROUND GLASS JOINT
                                                                                TO OPTICAL CELL
                                                                      18/9 MALE BALL JOINT
L
v=2
s.
M
!




/a
/



i
t
Vi





1 1 i
•"
±XJ
1




1
•\
JT
2
NL
1

'







5




                                                             ALL DIMENSIONS IN cm
                                                            UNLESS OTHERWISE NOTED
                       BLOWN GLASS BUBBLER       BOTTLE PORTION
                       APPROX. 0.6 cm X 1.0 em    4.0 cm O.D. X 35 cm t.D.
                            Figure 3.  Aeration cell.
                                         26

-------
     6.1.11  Sulfuric Acid Solution  (10 Percent V/V).  Dilute  100 ml
of concentrated sulfuric acid to 1000 ml using distilled water.

     6.1.12  Nitric Acid Solution (15 Percent V/V).  Dilute 15 ml of
concentrated HN03 to 100 ml using distilled water.

     6.1.13  Hydrochloric Acid (8N).  Dilute 67 ml of concen-
trated HC1 to 100 ml using distilled water.

     6.2     Sampling Reagents.

     6.2.1   Water.  Same as 6.1.1.

     6.2.2   Absorbing Solution, 4% KMnO..  Prepare fresh daily by
dissolving 40 g of KMnO. in sufficient 10% (V/V) H2SO. to make one
liter.  The solution should be prepared only in glass bottles to
prevent degradation.

     6.2.3   Wash Acid.  1:1 V/V Concentrated Nitric Acid and Water.

     6.2.4   Silica Gel.  Indicating type, 6- to 16-mesh.  If
previously used, dry at 175°C (350°F) for 2 hours.  New silica
gel may be used as received.

     6.3     Analysis Reagents.

     6.3.1   Intermediate Mercury Standard Solution, 10 ug Hg/ml.
Prepare fresh weekly.  Pipet a 5.0 ml aliquot of the mercury stock
solution (6.1.11) into a 500-ml volumetric flask and add 20 ml of
15% (V/V) HNO. solution.  Adjust the volume to 500 ml using distilled
water.  Thoroughly mix the solution.

     6.3.2   Working Mercury Standard Solution, 200 ng Hg/ml.
Prepare fresh daily.  Pipet a 5.0 ml aliquot from the "Intermediate
Mercury Standard Solution" into a 250-ml volumetric flask.   Add 5 ml
of the absorbing solution and 5 ml  of 15% (V/V) HNOo.   Adjust the volume
to 250 ml with distilled water.  Mix thoroughly.

     6.3.3   Sodium Chloride-Hydroxylamine Sulfate Solution.   Dissolve
12 g of sodium chloride and 12 g of hydroxylamine sulfate in distilled
water and dilute to 100 ml.   Hydroxylamine hydrochloride may replace
hydroxylamine sulfate.

     6.3.4   Tin(II) Chloride Solution.   Prepare daily and keep sealed
when not being used.  Completely dissolve 20 g of tin(II) chloride (or
25 g of tin(II) sulfate) in 25 ml of concentrated hydrochloric acid.
Bring the total volume to 250 ml using distilled water.   Other strong
acids such as nitric acid and sulfuric acid are not to be substituted
for the hydrochloric acid.
                                27

-------
7.    Procedure

     7.1     Sampling.  The complexity of this method is such that,
in order to obtain reliable results, field personnel should be
trained and experienced with the test procedures.  Sampling guide-
lines are given in the following sections.  These guidelines are
generally applicable; however, source alterations such as stack
extensions or expansions often are required to ensure the best
possible sample site.  Note also that since mercury is a hazardous
pollutant, care should be taken to minimize exposure to it.  Finally,
since the total quantity of mercury to be collected is generally small,
the method must be carefully applied, to prevent contamination or loss
of sample.

     7.1.1   Pretest  Preparation.  All the components shall be main-
tained and calibrated according to the procedure described in APTD-0576,
unless otherwise specified herein.

     Weigh to the nearest 0.5 g several 200 to 300 g portions of silica
gel in air-tight containers.  Record the total weight of the silica gel
plus contained, on each container.  Alternatively, the silica gel may be
weighed directly into its impinger just prior to train assembly.

     7.1.2   Preliminary Determinations.  Select the sampling site
and the minimum number of sampling points according to Method 1
(Citation 1, Section  11), or as specified by the Administrator.  Deter-
mine the stack pressure, temperature, and the range of velocity
heads using Method 2, (Citation 1, Section 11).  It is recommended that
a  leak-check of the pitot lines (see Method 2, Section 3.1) be per-
formed.  Determine the moisture content using Reference Method 4,
(Citation 1, Section  11), or its alternatives to decide isokinetic
sampling rate settings.  Determine the stack gas dry molecular weight,
(See Method 2, Section 3.6).

     Select a nozzle  size based on the range of velocity heads, such
that it is not necessary to change the nozzle size in order to
isokinetically maintain a sampling rate of between 14 and 28 1pm
(0.5 and 1.0 cfm).  During the run, do not change the nozzle size.
Ensure that the proper differential pressure gauge is chosen for
the range of velocity heads encountered (see Section 2.2 of Method 2,
Citation 1,2, Section 11).

     Select a suitable probe liner and probe length such that all
traverse points can be sampled.  For large stacks, consider sampling
from opposite sides of the stack to reduce the length of probes.

     Samples shall be taken over periods of time as are necessary to
accurately determine  the maximum emissions occurring in a 24-hr period.
In the case of cyclic operations, sufficient tests shall be made to
allow accurate determination of the emissions that occur over the
duration of the cycle.  A minimum sample time of 2 hours is recommended.
In some instances, high oxidizable-organic content of the stack gas may
make it impossible to sample for the desired minimum time.  This is
                                 28

-------
indicated by complete bleaching of the purple color of the permanganate
solution.  In these cases, a sample run may be divided into two or more
subruns to ensure that the absorbing solution is not depleted.

     7.1.3   Preparation of Sampling Train.  Prior to assembly, clean
all glassware (probe, filter holder [if used], impingers, and
connectors) by rinsing with wash acid, tap water, 8N HC1, tap water,
and finally distilled water.  Place 50 ml of absorbing solution in
the first impinger; place 100 ml of absorbing solution in the second
and third impingers and place approximately 200 g of preweighed
silica gel in the fourth impinger.  Care should be taken to prevent
the absorbing solution from contacting any greased surfaces.

     Using a tweezer, place a filter in the filter holder.   Be sure
that the filter is properly centered and the gasket properly placed
to prevent the sample gas stream from circumventing the filter.
(Note:   The use of a filter is optional and if one is used, it must be
stored in contact with the absorbing solution after use.  It may not be
desiccated before analysis.)

     Install the selected probe liner using a Viton A 0-ring when
stack temperatures are less than 260° (500°F) and an asbestos string
gasket for higher temperatures.   (See APTD-0576 for details.)  Other
connecting systems using either 316 stainless steel or Teflon ferrules
may be used.  Mark the probe with heat resistant tape or another method
to denote the proper distance into the stack or duct for individual
sampling points.   Assemble the train as shown in Figure 1,  using (if
necessary) a very light coat of silicone grease on all  ground glass
joints, greasing only the outer portion (see APTD-0576) to  avoid
possibility of contamination by the silicone grease.

     After the sampling train has been assembled, set the probe and
if applicable, the filter heating system at the desired operating
temperature.  Allow time for the temperature to stabilize.   Place
crushed ice around the impingers.

     7.1.4   Leak-Check Procedures.   Follow the leak-check  procedures
outlined in Method 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).   (See
Citations 1,2, Section 11.)

     7.1.5   Mercury Train Operation.   During the sampling  run, maintain
an isokinetic sampling rate (within 10 percent of true isokinetic un-
less otherwise specified and a temperature around the filter (if
applicable) of 120 + 14°C (248 + 25°F).

     For each run, record the data required on a data sheet such as
the one shown in Figure 4.   Be sure to record the initial  dry gas
meter reading.  Record the dry gas meter readings at the beginning
and end of each sampling time increment, when changes in flow rates
are made, before and after each leak-check, and when sampling is
halted.  Take other readings required by Figure 4 at least  once at
each sample point during each time increment and additional readings
                                29

-------
     PLANT
     LOCATION	
     OPERATOR	
     DATE	
     RUN NO	
     SAMPLE BOX NO..
     FILTER BOX N0._
     METERAH@	
     CFACTOR	
AMBIENT TEMPERATURE.
BAROMETRIC PRESSURE.
ASSUMED MOISTURE. % _
PROBE LENGTH, m (ft)	
NOZZLE IDENTIFICATION NO..
AVERAGE CALIBRATED NOZZLE DIAMETER, cm (in.!
PROBE HEATER SETTING*	
LEAK RATE, m3/min (cfm)	
PROBE LINER MATERIAL	
     PITOT TUBE COEFFICIENT, Cp.
                                              SCHEMATIC OF STACK CROSS SECTION
STATIC PRESSURE, mm HgJm.Hg).
FILTER NO.*	
TRAVERSE POINT
NUMBER












TOTAL
SAMPLING
TIME
(0), min.













AVERAGE
VACUUM
mm Hg
(in.Hg)














STACK
TEMPERATURE

-------
when significant changes (20 percent variation in velocity head)
necessitate additional adjustments in flow rate.  Level and zero
the manometer.  Because the manometer level and zero may drift due
to vibrations and temperature changes, make periodic checks during
the traverse.

     Clean the portholes prior to the test run to minimize the chance
of sampling deposited material.  To begin sampling, remove the nozzle
cap and verify that the probe and, if applicable, filter are at the
proper temperature and that the pitot tube and probe are properly
positioned.  Position the nozzle at the first traverse point with the
tip pointing directly into the gas stream.  Immediately start the
pump and adjust the flow to isokinetic conditions.  Nomographs are
available, which aid in the rapid adjustment of the isokinetic
sampling rate without excessive computations.  These nomographs are
designed for use when the Type S pitot tube coefficient is 0.85 +
0.02, and the stack gas equivalent density (dry molecular weight}
is equal to 29 + 4.  APTD-0576 details the procedure for using the
nomographs if C  and M. are outside the above stated ranges.

     When the stack is under significant negative pressure (height
of impinger stem), take care to close the coarse adjust valve before
inserting the probe into the stack to prevent impinger liquid from
being forced backward.  If necessary, the pump may be turned on with
the coarse adjust valve closed.

     When the probe is in position, block off the openings around
the probe and porthole to prevent unrepresentative dilution of the
gas stream.

     Traverse the stack cross-section, as required by Method 1
(Citation 1,2, Section 11) or as specified by the Administrator.
To minimize the chance of extracting deposited material be careful
not to bump the probe nozzle into the stack walls when sampling
near the walls or when removing or inserting the probe through
the portholes.

     During the test run, add ice and, if necessary, salt to main-
tain a temperature of less than 20°C (68°F) at the silica gel  out-
let.   Also, periodically check the level  and zero of the manometer.

     A single train shall be used for the entire sample run,  except
in cases where simultaneous sampling is required in two or more
separate ducts or at two or more different locations within the
same duct, or, when equipment failure necessitates a change of trains.
In all other situations, the use of two or more trains will be subject
to the approval of the Administrator.

     At the end of the sample run, turn off the coarse adjust valve,
remove the probe and nozzle from the stack, turn off the pump, record
the final dry gas meter reading, and conduct a post-test leak-check (see
Section 7.1.4).  Also, leak-check the pitot lines as described in
Section 3.2 of Method 2, (Citations 1,2,  Section 11).   The lines must
pass this leak-check in order to validate the velocity head data.
                                31

-------
     7.1.6   Calculation of Percent Isokinetic.  Calculate per-
cent isokinetic (Section 6.11 of Method 5, Citations 1,2, Section 11)
to determine whether the run was valid or another test run should be
made.  If there was difficulty in maintaining  isokinetic rates due to
source conditions, consult with the Administrator for possible variance
on the isokinetic rates.

     7.2     Sample Recovery.  Proper cleanup  procedure begins 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 any external parti -
culate matter near the tip of the probe nozzle and place a cap over it.
Do not cap off the probe tip tightly while the sampling train is cooling
since this would create a vacuum and draw liquid out from the impingers.

     Before moving the train to the cleanup site, remove the probe, wipe
off the silicone grease, and cap the open outlet of the probe.  Be careful
not to lose any condensate present.  Wipe off  silicone grease from the
impinger (or filter holder) inlet where the probe was fastened and cap it.
Remove the umbilical cord from the  last impinger and cap the impinger.
Either ground-glass stoppers, plastic caps, or serum caps may be used to
close these openings.

     Transfer the entire sampling train to the cleanup area.  This area
should be clean, protected from the wind, and  free of mercury contamination.
(Ambient air in laboratories located in the immediate vicinity of mercury-
using facilities is normally not free of mercury contamination.)

     Inspect the train prior to and during disassembly and note any
abnormal conditions.  Note:  All glass storage bottles and the graduated
cylinder must be precleaned as described in Section 7.1.3.   Treat the
samples as follows:

     Container No. 1.  Measure the  liquid in the first three impingers
to within + 1 ml by using a graduated cylinder.  Record the volume of
liquid present (e.g., see Figure 5-3 of Reference Method 5, (Citation
1, Section 11).  This information is required  to calculate the moisture
content of the effluent gas.  Place the contents of the first three
impingers into a 1000 ml glass sample bottle.

     Taking care to see that dust on the probe or other exterior
surfaces does not contaminate the sample, quantitatively recover the
mercury (and any condensate) from the probe nozzle, probe fitting, probe
liner and impingers by  rinsing these components with a total of 250 to
400 ml of fresh absorbing solution.  Add all washings to the 1000 ml
sample bottle.  Remove  any residual brown deposits on the glassware using
the minimum amount of 8 N HC1 required and add this HC1 rinse to the
1000 ml sample bottle.
     After all washings have been collected in the sample container,
tighten the lid on the  container so that leakage will not ocur when
                                 32

-------
it is shipped to the laboratory.  Mark the height of the fluid  level
to determine whether or not  leakage occurred during transport.  Label
the container to clearly identify its contents.

          Container No. 2.   If a filter was used in the sampling train
carefully remove it from the holder and place it in a 100 ml jar
equipped with a teflon-lined cap.  Add 20-40 ml of 4% KMNO./10% H?SO.
to the bottle and cap.

     Container No. 3.  Place an unused filter in a 100 ml bottle
equipped with a teflon-lined cap.  Add 20-40 ml of 4% KMNO.710%
HgSO^ to the bottle and cap.  This will serve as a filter Blank.

     Container No. 4.  For a blank, place 500 ml of 4% KMnO.710% H2SO.
in a 1000 ml sample bottle and seal the container.

     Container No. 5.  Note the color of the indicating silica gel to
determine if it has been completely spent and make a notation of its
condition.  Transfer the silica gel from its impinger to its original
container and seal.  A funnel may make it easier to pour the silica
gel without spilling.  A rubber policeman may be used as an aid in
removing the silica gel from the impinger.  It is not necessary to
remove the small amount of dust particles that may adhere to the
impinger wall and are difficult to remove.  Since the gain in weight
is to be used for moisture calculations, do not use any water or
other liquids to transfer the silica gel.   If a balance is available
in the field, weigh the spent silica gel (or silica gel plus impinger)
to the nearest 0.5 g; record this weight.


8.   Calibration and Standards.

     8.1     Aeration System Assembly.  Assemble the aeration
system as shown in Figure 5.  Flow Calibration.   Set the outlet pressure
on the nitrogen cylinder regulator to a minimum pressure of 500 mm Hg
(10 psi) and use the flow meter valve and a bubble flow meter or wet test
meter to obtain a flow rate through the solution cell  of 1.5 + 0.1
liters/min.   After flow calibration is complete, remove the bubble flow
meter from the system.

     8.2     Optical Cell Heating System Calibration.   Using a 25 ml
graduated cylinder, add 25 ml of distilled water to the bottle section
of the aeration cell and attach the bottle section ta-the bubbler
section of the cell.  Attach the solution cell  to "the optical cell, and
while aerating at 1.5 liters/minute, determine the minimum Variac setting
necessary to prevent condensation in the optical cell  and in the connecting
tubing.   (This setting should not exceed 20 volts.)

     8.3     Preparation of Glassware For Use in Analysis.   Before use, all
glassware, both new and used, should be brushed with soap and water,
liberally rinsed with tap water, soaked for one hour in a 1:1 nitric
acid/water solution and then rinsed with distilled water.
                                33

-------
     The rate at which elemental mercury is released from a solution
is affected by the solution temperature. ; Consequently, the shape of the
absorption curve (area) and the point of maximum absorbance (peak height)
are affected by the solution temperature.  Therefore, to obtain reproducible
results, it is important to bring all solutions to room temperature before
use.

     8.4     Analytical System Calibration for Analysis.

     8.4.1   Spectrophotometer and Recorder Calibration.  Set the spectro-
photometer wavelength dial to 253.7 nm  and make certain the optical cell is
at the minimum temperature that will prevent water condensation from
occurring.  Then set the recorder scale as follows:  Using a 25 ml
graduated cylinder, add 25 ml of distilled water to the aeration cell bottle
and pipet a 5.0 ml aliquot of the working mercury standard solution into the
aeration cell.  (NOTE:  The mercury-containing solution should always be
added to the aeration cell after the 25 ml of distilled water.)  Place a
Teflon-coated stirring bar in the bottle.  Add 5 ml of 4% KMn04/10% H^SO.
followed by 5 ml of 15% nitric acid and 5 ml of 5% potassium permanganate
to the aeration bottle and mix well.  Now, attach the bottle section to
the bubbler section of the aeration cell and make certain that:  (1)  the
aeration cell exit arm stop-cock (Figure 3) is closed (so that mercury will
not prematurely enter the optical cell  when the reducing agent is being
added); and (2) there is no flow through the bubbler.  Add 5 ml of sodium
chloride-hydroxylamine sulfate solution to the aeration bottle through the
side arm and mix.  If the solution does not clear, add additional sodium
chloride hydroxylamine sulfate in 1-ml  increments until the solution is
colorless.  Now add 5 ml of 10% stannous chloride solution to the aeration
bottle through the side arm and immediately stopper the sidearm.  Stir the
solution for 15 seconds, turn on the recorder, open the aeration cell exit
arm stopcock and then immediately initiate aeration with continued stirring.
Determine the maximum absorbance of the standard and set this value to read
90 percent of the recorder full scale.

     8.4.2     Calibration Curve.  Repeat the procedure in Section 8.4.1
using 0.0, 1.0, 2.0, 3.0, 4.0 and 5.0-ml aliquots of the working standard
solution.  Repeat this procedure on each aliquot size until two con-
secutive peaks agree within 3 percent of their average value.  The final
amount of mercury in the aeration bottle for the 0.0, 1.0, 2.0, 3.0, 4.0
and 5.0-ml aliquots will be 0, 200, 400, 600, 800 and 1000 ng, respectively.
NOTE:  To prevent mercury carry-over from one sample to another, it is
important that the nitrogen tank valve  remains open and that the aeration
cell remains connected to the optical cell until the recorder pen has
returned to the baseline to ensure that all mercury has been purged from the
system.  After separating the bottle and bubbler sections of the aeration
bottle, place the bubbler section into  a 600-ml beaker containing approxi-
mately 400 ml of distilled water to remove all traces of the reducing agent.
In addition, to prevent the loss of mercury prior to aeration, all traces
of the reducing agent (tin(II)chloride) must be removed between samples.
This is accomplished by washing with distilled water, however, washing the
aeration cell parts with concentrated hydrochloric acid is necessary if any
of the following conditions occur:  (1) a white film appears on any inside
surface of the solution cell; (2) a sudden change is observed in the cali-
                                 34

-------
                                                 NEEOLt VALVE FOR

                                                  FLOW CONTROL
u>
en
                                     N2 CYLINDER
                                                                                                   TO VARIABLE TRANIPORMER
                                                    Figure 5.  Schematic of the aeration system.

-------
bration curve; (3) reproducible results are not obtained on replicate
samples; or, (4) after a sample with an abnormally high mercury content is
analyzed.

               Subtract the averaged peak height of the solution blank
(0.0-ml aliquot) - which should be less than 2 percent of the recorder
full scale -  from the averaged peak heights of the 1.0-ml, 2.0-ml, 3.0-ml,
4.0-ml and 5.0-ml aliquot standards.  (If the blank absorbance is greater
than 2 percent of full-scale, the probable cause is mercury contamination
of a reagent or carry-over of mercury from a previous sample.)  Plot the
corrected peak height of each standard solution versus the corresponding
total weight of mercury in the aeration cell and draw the best fit straight
line.  This plot should pass either through the origin, or through a point
no further from the origin than + 2 percent of the recorder full-scale.
If a discovery occurs, check for nonlinearity of the curve and for in-
correctly prepared standards.

     8.5     Analysis of Stack Samples.

     8.5.1   Container No. 2.  (Filter).  Place contents, including filter
into a 250 ml beaker and heat this on a steam bath until most of the liquid
has evaporated.  Do not take to dryness!  Add 20 ml concentrated nitric
acid to the beaker, cover it with a watch glass and heat it on the steam bath
for 2-hrs.  Remove and filter the solution through Whatman No. 40 filter
paper.  Save the filtrate for analysis for mercury.  Discard the filter.

     8.5.2     Container NO. 3.  (Filter Blank).  Treat as described in
Section 8.5.1.

     8.5.3     Container No. 1.  (Probe/Impinger Samples) Filter the impinger
contents through Whatman No. 40 rapid filter paper to remove the brown
Mn02 precipitate.  Wash the filter with 50 ml of 4% KMnO.710% H2SO. solution
and add this wash to the filtrate.  Save the filtrate for analysis for mercury.
Discard the filter.

     8.5.4     Container No. 4.  (Absorbing Solution Blank).  Treat as de-
scribed  in Section 8.5.3.  Combine this filtrate with the filtrate from
Container No. 3 (Section 8.5.2) to serve as the sample blank and treat as
described in Section 8.5.5.

     8.5.5     Analysis.  Combine the filtrates from Containers No. 1 and
No. 2, mix and dilute to a known volume with distilled water.  Repeat the
procedure used to establish the calibration curve (Sections 8.4.1, 8.4.2)
using an appropriate sized aliquot (1-10 ml) from the stack sample.  (NOTE:
If the 10-ml sample is below the detectable limit, a larger aliquot should
be used  (up to 20-ml), but the volume of water added to the aeration cell
should be decreased accordingly to prevent the solution volume from exceeding
the capacity of the aeration bottle.  If the peak maximum of a 1.0-ml aliquot
of sample is off scale further dilution of the original sample must be done
to bring the mercury concentration into the calibration range of the spectro-
photometer.
                                 36

-------
               Analyze successive aliquots of each sample and the field sample
blank until two consecutive peak heights agree within + 3 percent of their
average value.  Check the spectrophotometer calibration frequently by running
a blank and standard at least after every five samples; recalibrate as
necessary.

               It is also recommended that at least one sample from each
stack test be checked by the Method of Standard.Additions (See Citation 9,
Section 11) to confirm that matrix effects have not interfered in the
analysis.

9.   Calculations

     9.1     Abbreviations and Symbols
                                 2
             A   =  Stack area, m .

           D.F.  =  Total dilution factor for the mercury containing
                    solution before addition to the aeration cell.

          [Hg]y  =  Total ug of mercury in the source sample after
                    initial dilution.

       [Hg]      =  Total ng of mercury in the aeration cell (from
            '  '     the calibration curve constructed as described
                    in Section 8.4.2), corrected for solution blank
                    value.

       P.        =   Barometric pressure at the sampling site, mm Hg.

       P        =   Absolute stack gas pressure, mm Hg.

       R        =   Rate of mercury emission, g/day.

       S        =   Aliquot size in ml added to the aeration cell.

       T        =   Absolute stack temperature, °K.

       v        =   Average stack gas velocity, m/sec.

       V        =   Total volume of gas sample as measured by the dry
        m           gas meter, corrected for leakage,  m .
       v        =   Total gas sample volume (stack conditions),  m3
       vtotal
       Y        =   Dry gas meter calibration factor.

       AH       =   Average pressure differential across the orifice
                    meter (see Figure 4), mm H^O.

     9.2       Calculate V. the total volume of dry gas metered
(corrected for leakage, if%ecessary, as outlined in Section 6.3 of
Method 5, Citation 1,2 in Section 11).
                                37

-------
     9.3       Calculate the volume of water vapor and the moisture content
of the stack gas, according to Equations 5-2 and 5-3 of Method 5, (Citation
1,2, Section 11).

     9.4       Calculate v , the average stack gas velocity, using
Equation 2-9 of Method 2 (Citation 1,2, Section 11).

     9.5       Calculate the total gas sample volume at stack conditions,
using the following equation:
         Vtotal  =  Vm M  "L  /I        PC           J      Equation 1


     9.6       For each source  sample, correct the average maximum
absorbance of the two consecutive samples whose peak  heights agree within
+ 3 percent of their average for the contribution of  the field blank.

     Calculate the total mercury content, [Hg]T> in each sample correcting
for any dilutions made to bring the sample into the working range of the
spectrophotometer.

     9.7       Total Mercury Emission.  Calculate the total amount of
mercury emitted from each stack per day by Equation 2.  This equation
is applicable for continuous operations.  For cyclic  operations, use only
the time per day each stack is  in operation.  The total mercury emissions
from a source will be the summation of results from all stacks.

               (Hg)n
                                                   86,400 seconds/day
                        Vtotal              )   \    106  ug/g

                                                             Equation 2

     9.8       Isokinetic Variation.  Calculate percent isokinetic for each
sampling run, as outlined in Section 6.11 of Method 5, 40 CFR 60, Appendix A.

10.  Evaluation of Results

     10.1       Acceptable Isokinetic Results.

     10.1.1     The following range sets the limit on acceptable isokinetic
sampling results:

     If 90 percent < I < 110 percent, the results are acceptable.  If the
results are low in comparison to the emission standard and I is beyond the
acceptable range, or, if I is less than 90 percent, the Administrator may
opt to accept the results.  Otherwise, reject the results and repeat the
test.
                                38

-------
11.   Bibliography

     1.   U.S. Environmental Protection Agency.  Standards of  Per-
          formance for New Stationary Sources  (40 CFR 60).  Federal
          Register, 42(160):41754-41789, August 18, 1977.

     2.   U.S. Environmental Protection Agency.  Amendments to
          Reference Methods 1-8 (40 CFR 60).   Federal Register 43(57):
          11984-11986.  March 23, 1978.

     3.   Hatch, W. R., and W. I. Ott.  Determination of Sub-Microgram
          Quantities of Mercury by Atomic Absorption Spectrophotoaetry.
          Anal. Chem. 40:2085-87.  1968.

     4.   Martin, R. M.  Construction Details  of Isokinetic Source
          Sampling Equipment.  U.S. Environmental Protection Agency,
          Research Triangle Park, North Carolina.  APTD-0581,  April,
          1971.

     5.   Rom, J. J.  Maintenance, Calibration and Operation of
          Isokinetic Source Sampling Equipment.  U.S. Environmental
          Protection Agency, Research Triangle Park, North Carolina.
          APTD-0576.  March, 1972.

     6.   Mitchell, W. J., and M. R. Midgett.  Method for the  Analysis
          of Mercury Emissions from Mercury Cell Chior-Alkali  Plants.
          U.S. Environmental Protection Agency, Research Triangle Park,
          North Carolina  May, 1975.

     7.   Shigehara, R. T.  Adjustments in the EPA Nomograph for
          Different Pitot Tube Coefficients and Dry Molecular Weights.
          Stack Sampling News 2:4-11.  October, 1974.

     8.   Vollaro, R. F.  Recommended Procedure for Sample Traverses
          in Ducts Smaller Than 12 Inches in Diameter.  U.S. Environ-
          mental Protection Agency, Research Triangle Park, North Carolina.
          November, 1976.

     9.   Klein, R., and C. Hack.  Standard Additions, Uses and Limitations
          in Spectrophotometric Analysis.   Amer. Lab. 9:21-27  (1977).
                                39

-------
                             APPENDIX B
       METHOD TO DETERMINE THE MERCURY CONTENT OF SEWAGE SLUDGE


1.    Principle and Applicability

     1.1       Time-composite sludge samples are withdrawn from the conveyor
belt after dewatering and before incineration or drying.  A weighed portion
of the sludge is digested in aqua regia followed by oxidation by potassium
permanganate.  Mercury in the digested sample is subsequently measured by
the conventional cold vapor technique.

     1.2       Applicability.  This method is applicable for the determination
of total organic and inorganic mercury content in sewage sludges.  The normal
range of this method is 0.2 to 5 pg/g of sludge.  This range may be extended
above or below the normal range by increasing or decreasing sample size and
through instrument and recorder control.

2.    Apparatus

     2.1       Sampling

     2.1.1     Container, 50-liter, plastic

     2.1.2     Scoop for removing sludge sample from conveyor line.

     2.2       Mixing of Composite (8-hour) Sludge Sample

     2.2.1     Mortar mixer, wheelbarrow-type, 57-liter (or equivalent) with
electrically driven motor.

     2.2.2     Blender, Waring-type, 2-liter.

     2.2.3     Scoop for removing 100-ml and 20-ml samples of blended sludge.

     2.3       Analysis

     2.3.1     Atomic Absorption Spectrometer.  Perkin-Elmer 303, or equiv-
alent, containing a hollow cathode mercury lamp and capable of containing
the optical cell described in Section 2.3.2.

     2.3.2     Optical Cell.  Cylindrical-shape with quartz end windows and
having the dimensions shown  in Figure 1.  The cell is wire-wound with approxi-
mately 2 meters of 24 gauge  nichrome heating wire and wrapped with asbestos
insulation tape or equivalent; the wires must not touch each other.

     2.3.3     Aeration Cell.  Constructed according to the specifications in
Figure 2.  A glass frit is not an acceptable substitute for the blown glass
bubbler tip shown in Figure  2.

     2.3.4     Recorder.  Matched to the output of the spectrometer
described in Section 2.3.1.
                                40

-------
     2.3.5     Varlac.  To vary the voltage on the optical cell from 0 to 40
volts.

     2.3.6     Hood.  For venting optical cell exhaust.

     2.3.7     Flow Metering Valve.

     2.3.8     Flow Meter.  Rotameter or equivalent, capable of measuring a
gas flow of 1.5 liters/rain.

     2.3.9     Nitrogen Cylinder.  Equipped with a single stage regulator.

     2.3.10    Connecting Tubing.  All connective tubing between the solution
cell and the optical cell shall be made of glass (ungreased ball and socket
connections are recommended).  Tygon tubing, other types of flexible tubing,
and metal tubing are unacceptable substitutes.  Teflon, steel, or copper
tubing should be used between the nitrogen tank and the flow meter valve
(See Section 2.3.7).  Tygon, gum, or rubber tubing may be used between the
flow metering valve and the solution cell.

     2.3.11    Flow Rate Calibration Equipment.  Bubble flowneter or wet test
meter for measuring a gas flow rate of 1.5 liter/rain, to within + 0.1
liter/min.

     2.3.12    Volumetric Flask.  Class A with penny head standard taper
stoppers; 100-ml.

     2.3.13    Volumetric Pipets.  Class A; 1, 2, 3, 4, and 5 ml.

     2.3.14    Graduated Cylinder.  50 ml.

     2.3.15    Magnetic Stirrer.  General purpose laboratory type.

     2.3.16    Magnetic Stirring Bar.   Teflon coated.

     2.3.17    Filter Paper.  S & S No.  588 (or equivalent).

3.   Reagents

     3.1       Water.  Distilled, meeting ASTM specifications for Type I
Reagent Water-ASTM Test Method D-l193-72.   At the option of the analyst,
the KMNO. test for oxidizable organic matter may be eliminated when high
concentrations of organic matter are not expected to be present.   This
water must be used in all dilutions and solution preparations.

     3.2       Aqua Regia.  Prepare immediately before use by carefully
adding one volume of concentrated HNO, to three volumes of concentrated HC1.

     3.3       Antifoam B Silicon Emulsion.   J. T.  Baker Co., (or equivalent).

     3.4       Nitric Acid (15% V/V).   Prepared by diluting 15 ml  concen-
trated HNO~ to 100 ml using distilled, deionized water.
                                41

-------

                                    18/9 FEMALE BALL SOCKET
                 LENGTH NECESSARY TO FIT SOLUTION CELL
                     TO SPECTROPHOTOMETER
                                                   TO VARIABLE TRANSFORMER
 VENT TO HOOD
                                                   181 cm DIAMETER QUARTZ
                                                   WINDOWS AT EACH END
NOTES.
CELL WOUND WITH 24-GAUGE NICHROME WIRE
TOLERANCES  ' 5 PERCENT
         Figure  1.  Optical cell.
                            42

-------
19/22 GROUND GLASS JOINT
         WITH STOPPER
             19/22 GROUND GLASS JOINT
                                                                            TO OPTICAL CELL
c
h
!


fa

Ittl
fc».
r

41
./ i
y
•


                                                         UNLESS OTHERBISE NOTED
                    BLOHN GLASS BUBBLER      BOTTLE PORTION
                    AimOX.a6cnX1.Oon    U en O St. X IS OB U>.
                      Figure 2. Aeration cell.
                                        43

-------
     3.5       Sodium Chloride-Hydroxylanrine Sulfate Solution.  Dissolve
12 grams of sodium chloride and 12 grams of hydroxylamine sulfate in
distilled water and dilute to 100 ml.  Hydroxylamine hydrochloride may
be used in place of the hydroxylamine sulfate.

     3.6       Potassium Permanganate - 5% solution, w/v.  Dissolve 5
grams of potassium permanganate in sufficient distilled water to make
100 ml.

     3.7       Tin(II)Ch1oride Solution.  Prepare fresh daily by
dissolving 20 grams of tin(II)chloride in 25 ml concentrated HC1 and
dilute to 250 ml using distilled water.  For an accurate and reproducible
mercury analysis to be obtained, it  is necessary to have a clear solution
of tin(II)chloride.  If the solution is cloudy, the tin(II) ion has been
oxidized to tin(IV), and fresh reagent grade tin(II)chloride should be
obtained.  Alternately, a clear, tin(II)chloride solution is available
from Perkin Elmer Corporation, Norwalk, Connecticut.  Tin(II)sulfate may
be substituted for the tin(II)chloride.

3.8       Mercury (II) Stock Solution (1 mg/ml).  Completely dissolve
0.1354 g of ACS Reagent Grade HgCl?  in 75 ml of distilled water, add 10 ml
of concentrated HNO-, and adjust tne volume to 100.0 ml with distilled water.
Mix thoroughly.  (Tnis solution should be stable for at least 1 month.)

     3.9       Intermediate Mercury  Standard Solution, 10 ug Hg/ml.  Prepare
fresh weekly.  Pipet a 5.0 ml aliquot of the mercury stock solution into a
500-ml volumetric flask and add 10 ml of the 15 percent nitric acid solution.
Adjust the volume to 500 ml using distilled water.  Thoroughly mix the
solution.

     3.10      Working Mercury Standard Solution, 200 ng Hg/ml.  Prepare
fresh  daily.  Pipet a 5.0 ml aliquot from the "Intermediate Mercury
Standard Solution" into a 250-ml volumetric flask.  Add 5 ml of 1:1 nitric
acid and adjust the volume to 250 ml with distilled water.  Mix thoroughly.

4.   Procedure

     4.1       Sludge Sampling.  Sludge samples should be withdrawn at a
maximum  of 30 minute intervals over  an eight  hour period and placed in a
rigid  plastic containers.  A minimum of 25  kg of sludge  should be obtained
in this  period.

     4.2       Sludge Mixing.  Transfer the entire  25 kg sample to a
57-liter capacity  (2 ft  ) mortar mixer equipped with an  electrically
driven motor.  (This operation  is best done  in a hood or  outside).  Allow
the  samples  to mix for a minimum of  30 minutes at 30 rpm.  Take six,
100-ml aliquots of sludge and  combine  in  a  2-liter  blender.   Blend sludge
for  five minutes adding  small  portions of distilled deionized water as
necessary to give  a fluid consistency.  Immediately after stopping the
blender, use a small beaker to withdraw four, 20-ml aliquots  of blended
sludge and place them  in separate, tared  125-ml Erlenmeyer flasks.
Reweigh  each flask to  determine the  amount  of sludge added.   (Three of the
samples  will be used to  determine  the mercury content in the  sludge and
the  fourth will be used  to measure the  solids content of the  blended sludge.)
                                 44

-------
     4.3       Solids Content of Blended Sludge.  Dry a 20-ml blended sample
in an oven at 105°C until a constant weight is obtained.  Allow to cool in a
desiccator and record the dry weight of the sample.

     4.4       Aqua Regia Digestion of Blended Samples.  To each of the
three remaining 20-ml samples from Section 4.2 add 25 ml of aqua regia and
digest the samples on a hot plate at low heat for 30 minutes.  Remove fron
the hot plate and allow to cool.  (Samples should be a pale yellow-brown
color and void of the dark brown color characteristic of organic matter.)

               Filter each digested sample separately through S and S No.
588 filter paper and rinse the filter contents with 50 ml of distilled
deionized water.  Transfer the filtrate and filter washing to a 100-ml
volumetric flask and carefully dilute to volume.

     4.5       Solids Content of Sludge Before Blending.  Remove two, 100-ml
portions of mixed sludge from the mortar mixer and place in separate, tared
400-Bl beakers.  Dry in an oven at 105°C to a constant weight and cool in a
desiccator.

     4.6       Analysis for Mercury

     4.6.1     Equipment Assembly.  Assemble the aeration system as de-
scribed in Figure 3.

     4.6.2     Flow Calibration.  Set the outlet pressure on the nitrogen
cylinder regulator to a minimum pressure of 500 mm Hg (10 psi) and use the
flow meter valve and a bubble flow meter or wet test meter to obtain a flow
rate through the solution cell of 1.5 + 0.1 liters/min.  After flow cali-
bration is complete, remove the bubble flow meter from the system.

     4.6.3     Optical Cell Heating System Calibration.  Using a 25-ml
graduated cylinder, add 25 ml of distilled water to the bottle section of
the aeration cell and attach it to the bubbler section. Attach the
solution cell to the optical cell, and while aerating at 1.5 liters/min,
determine the minimum Variac setting necessary to prevent condensation of
moisture in the optical cell and in the connecting tubing.  (This setting
should not exceed 20 volts).

     4.6.4     Spectrometer and Recorder Calibration.  Set the spectro-
meter wavelength dial to 253.7 nm and make certain the optical cell is at
the minimum temperature that will prevent water condensation from occur-
ring.  Then set the recorder to 90 percent of full scale as follows:
Using a 25-ml graduated cylinder, add 25 ml of distilled water to the
aeration cell bottle.  Add three drops of Antifoam B to the bottle and
then pipet a 5.0 ml aliquot of the working mercury standard solution into
the aeration cell.   (The mercury-containing solution should always be
added to the aeration cell after the 25 ml of distilled water.)  Place a
Teflon-coated stirring bar in the bottle.  Add 5 ml of 15 percent nitric
acid and 5 ml of 5% potassium permanganate to the aeration bottle and mix
well.  Now, attach the bottle section to the bubbler section of the
aeration cell, and make certain that:  (1) the aeration cell exit arm
stopcock (Figure 3) is closed (to avoid mercury prematurely entering the
optical cell when the reducing agent is added); and (2) there is no flow
through the bubbler.

                                45

-------
             NEEDLE VALVE FOR
              FLOW CONTROL
Nj CYLINDER
                                                             TO HOOD


                                                                 TO VARIABLE TRANSFORMER
                                      MAGNETIC STIRRER
              Figure 3. Schematic of the aeration system.

-------
Add 5 ml of 12% sodium chloride hydroxylanrine sulfate solution to the
aeration bottle through the side arm and mix.  If the solution remains
un- clear, add additional sodium chloride hydroxylamine sulfate in 1-ml
increments until the solution is colorless.  Now add 5 ml of tin(II)
chloride solution to the aeration bottle through the side arm and
immediately stopper the sidearm.  Stir the solution for 15 seconds, turn
on the recorder, open the aeration cell exit arm stopcock and then
immediately initiate aeration with continued stirring.  Determine the
maximum absorbance of the standard and set this value to read 90 percent
of the recorder full scale.

     4.6.5     Calibration Curve.  Repeat the procedure in Section 4.6.4
using 0.0, 1.0, 2.0, 3.0, 4.0 and 5.0-ml aliquots of the working standard
solution.  Repeat this procedure on each aliquot size until two consecutive
peaks agree within 3 percent of their average value.  The final amount of
mercury in the aeration bottle for the 0.0, 1.0, 2.0, 3.0, 4.0 and 5.0 ml
aliquots will be 0, 200, 400, 600, 800 and 1000 ng.   Note:  To prevent
mercury carry-over from one sample to another, it is important that the
nitrogen tank valve is open and that the aeration cell remains connected
to the optical cell until the recorder pen has returned to the baseline
and the sample is purged from the system.   After separating the bottle and
bubbler sections of the aeration bottle, place the bubbler section into a
600-ml beaker containing approximately 400 ml of distilled water and rinse
the bottle section of the aeration cell with a stream of distilled water
to remove all traces of the reducing agent.   In order to prevent the loss
of mercury prior to aeration, it is important to remove all traces of the
reducing agent (tin(II)chloride) between samples.   This is accomplished
by washing with distilled water, however,  washing the aeration cell  parts
with concentrated hydrochloric acid is necessary if any of the following
conditions occur:   (1) a white film appears on any inside surface of the
solution cell; (2) a sudden change is observed in the calibration curve;
(3) reproducible results are not obtained on replicate samples; or
(4) following an abnormally high (in mercury content) sample.

               Subtract the averaged peak height of the solution blank
(0.0-ral aliquot) — which should be less than 2 percent of the recorder full
scale — from the averaged peak heights of the 1.0-ml, 2.0-ml, 3.0-ml, 4.0-ml,
and 5.0-ml aliquot standards.  (If the blank absorbance is greater than
2 percent of full-scale, the probable cause is mercury contamination of a
reagent or carry-over of mercury from a previous sample.)  Plot the
corrected peak height of each standard solution versus the corresponding
final mercury concentration in the aeration cell  (in ng) and draw the best
fit straight line.  This plot should either pass through the origin, or
pass through a point no further from the origin than + 2 percent of the
recorder full scale.  If a discrepancy occurs check for nonlinearity of
the curve and for incorrectly prepared standards.

     4.6.6     Analysis of Digested Samples.   Repeat the procedure used
to establish the calibration curve (Section 4.6.5) using an appropriate
sized aliquot (1*10 ml) from the digested sample.   (If the 10-ml sample
is below the detectable limit, a larger aliquot should be used (up to
25-ml) but the volume of water added to the aeration cell should be
decreased accordingly to prevent the solution volume from exceeding the
capacity of the aeration bottle.  If the peak maximum of a 1.0-ml aliquot
of sample is off scale further dilution of the original sample must be

                                47

-------
done to bring the mercury concentration into the calibration range of the
spectrometer.

               Analyze successive aliquots of each digested sample until
two consecutive peak heights agree within + 3 percent of their average
value.  Check the spectrometer calibration frequently by running a blank
and standard at least after every five samples; recalibrate as necessary.

               It is also recommended that at least one digested sample
from each 8-hour composite sample be checked by the Method of Standard
Additions to confirm that matrix effects have not interfered in the
analysis.

5.   Calculations

     5.1       Nomenclature.

               C    =  Concentration of mercury in the digested sample, ug/g.

               Fgb  =  Weight fraction of solids in the blended sludge.

               F    =  Weight fraction of solids in the collected sludge
                oir)      _^_i_    •  *
                       after mixing.

               M    =  Mercury content of the sewage sludge on a dry basis,
                       M9/9-

               m    =  Mass of mercury in the digested sample, ug.

               V    =  Volume of digested sampled analyzed, ml.
                a

               V    =  Volume of digested sample, ml.
                5
               Wf   =  Weight of empty sample flask, g.

               Wf   =  Weight of sample flask and sample, g.

               Wf.  =  Weight of sample flask and sample after drying, g.

               Wb   =  Weight of empty sample beaker, g.

               W.    =  Weight of sample beaker and sample, g.

               W..  =  Weight of sample beaker and sample after drying, g.

     5.2       Mercury Content of Digested Sample (Wet Basis).  For each
sample, correct the average maximum absorbance of the two consecutive
samples whose peak heights agree within + 3 percent of their average for
the contribution of the blank.  Use the calibration curve and these
corrected averages to determine the final mercury concentration in the
solution cell for each sludge sample.

               Calculate the total mercury content in each gram of digested
sample correcting for any dilutions made to bring the sample into the working
range of the spectrometer and for the weight of the sludge portion digested.

                                48

-------
                                  m V

                    Cm  =
                              Va (Wfs " Wf)           Equation 1
     5.3       Solids Content of Blended Sludge.  Determine the solids
content of the 20-ml aliquot dried in the oven at 105°C (See Section
4.3).
                     Fsb =
                                 f c ~  ~ff\
                                 TS    ra             Equation 2
                                wfs - wf
     5.4      Solids Content of Bulk Sample (after mixing in mortar mixer).
Determine the solids content of each 100-ml aliquot (See Section 4.5) as in
Section 5.2 and average the results.
                               u   - u
                               wbs   wbd
                  F
                   sm
                                   - Wb             Equation 3
     5.5      Mercury Content of Bulk Sample (Dry Basis).  Average the
results from the three samples from each 8-hour composite sample and
calculate the mercury concentration of the composite sample on a dry basis.



                                    Cm
                      M  =           m                Equation 4

                               Fsb      Fsm
                                49

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 . REPORT NO.
  EPA 600/4-79-058
                                                           3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
                                                           5. REPORT DATE
 TEST METHODS TO DETERMINE THE MERCURY  EMISSIONS FROM
 SLUDGE INCINERATION  PLANTS
                                                                       1979
                                                           6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
           W.J. Mitchell,  M.R. Midgett and  J.C.  Suggs,
  EPA/EMSL/RTPNC and  D.  Albrinck, PEDCo  Environmental,Inc

   ERFORMING OBC \NI2,
                                                           8. PERFORMING ORGANIZATION REPORT NO.
~P
CT\N'l2ATION NAME AND ADDRESS
  Quality Assurance  Division (MD-77)
  Environmental Monitoring Systems Laboratory
  U.S.  Environmental  Protection Agency   ,
  Research Triangle  Park,  North Carolina 27711
                                                           10. PROGRAM ELEMENT NO.

                                                             1AD800
                                                           11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
                                                            13. TYPE OF REPORT AND PERIOD COVERED
  Office of Research  and Development
  Environmental Monitoring Systems Laboratory
  U.S. Environmental  Protection Agency
  Research Triangle  Park, North Carolina  27711
                                                           14. SPONSORING AGENCY CODE
                                                             EPA 600/08
 15. SUPPLEMENTARY NOTES
  To be published  as an Environmental  Monitoring Series report
 16. ABSTRACT
       Two test  methods for mercury  are described along with  the  laboratory and field
  studies done in  developing and validating them.  One method describes how to homo-
  genize and analyze large quantities  of sewage sludge.  The  other test method
  describes how  to measure the mercury emissions from the stacks  of sewage sludge
  incinerators.   In this latter method, the samples are collected in a potassium
  permanganate/sulfuric acid solution  and analyzed for mercury using fTameless atomic
  absorption.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                               b.lDENTIFIERS/OPEN ENDED TERMS
                                                                         c.  COSATI Field/Group
 Mercury, Sewage Sludge  Incinerators,
 Stack testing, Sludge analysis
                                               Testing for mercury
                                                              43F
                                                              68A
13. DISTRIBUTION STATEMENT
   Release to Public
                                              19. SECURITY CLASS (ThisReport)

                                               Unclassified	
                                                            21. NO. OF PAGES
                                                                56
                                              20. SECURITY CLASS (Thispage)
                                               Unclassified
                                                                         22. PRICE
EPA Form 2220-1 (9-73)
                                           50

-------