EPA-60U/4-82-024
April 1982
DETERMINATION OF NITROAROMATIC COMPOUNDS AND
ISOPHORONE IN INDUSTRIAL AND MUNICIPAL WASTEWATERS
by
Kenneth H. Shafer
Battelle Columbus Laboratories
Columbus, Ohio 43201
Contract No. 68-03-2624
Project Officer
James E. Longbottom
Physical and Chemical Methods Branch
Environmental Monitoring and Support Laboratory
Cincinnati, Ohio 45268
ENVIRONMENTAL MONITORING AND SUPPORT LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
FOREWORD
Environmental measurements are required to determine the quality of
ambient waters and the character of waste effluents. The Environmental
Monitoring and Support Laboratory - Cincinnati conducts research to:
• Develop and evaluate techniques to measure the presence and
concentration of physical, chemical, and radiological pollutants 1n
water, wastewater, bottom sediments, and solid waste.
• Investigate methods for the concentration, recovery, and
Identification of viruses, bacteria, and other microbiological
organisms In water. Conduct studies to determine the responses of
aquatic organisms to water quality.
• Conduct an Agency-wide quality assurance program to assure
standardization and quality control of systems for monitoring water
and wastewater.
Under provisions of the Clean Water Act, the Environmental Protection
Agency 1s required to promulgate guidelines establishing test procedures for
the analysis of pollutants. The Clean Water Act Amendments of 1977
emphasize the control of toxic pollutants and declare the 65 priority
pollutants and classes of pollutants to be toxic under Section 307(a) of the
Act. This report 1s one of a series that Investigates the analytical
behavior of selected priority pollutants and suggests a suitable test
procedure for their measurement.
Robert L. Booth, Acting Director
Environmental Monitoring and Support
Laboratory - Cincinnati
111
-------
ABSTRACT
A method was developed for the determination of nitrobenzene, 2,4- and
2,6-d1n1trotoluene (DNT), and isophorone 1n wastewaters. The methods devel-
opment program consisted of: a literature review, determination of the
stability of the compounds 1n organic solutions, determination of extraction
efficiency for each compound from water using two organic solvents, deter-
mination of storage stability of each compound 1n water, and evaluation of
various clean up techniques.
The final method was applied to several representative wastewaters
spiked with the compounds at appropriate levels, as well as to surface
water, 1n order to determine precision and accuracy of the method. For a
wastewater sample spiked with 96 ppb of nitrobenzene, 560 ppb of Isophorone
and 8 ppb of 2,4-ONT, and 7 ppb of 2,6-DNT, recoveries were 70 t 6% for
nitrobenzene, 71 t 5< for Isophorone, 70 ± 2t for 2,4-ONT, and 78 ± 1* for
2,6-ONT. Minimum detectable levels 1n this wastewater are estimated to be 5
ppb for nitrobenzene, 5 ppb for Isophorone, and 0.05 ppb for 2,4- and
2,6-DNT. However, caution must be used in presuming these MOLs are accurate
for other wastewaters since several wastewaters were encountered which
exhibited high levels of Interfering compounds. For some complex waste-
waters it may be necessary to apply additional clean up procedures or more
selective analytical detection systems 1n order to achieve these levels of
sensitivity.
This report was submitted in fulfillment of Contract No. 68-03-2624 by
Battelle Columbus Laboratories under the sponsorship of the U.S. Environ-
mental Protection Agency. This report covers the period from November 1,
1977, to March 1, 1979, and work was completed as of March 1, 1979.
i v
-------
CONTENTS
Disclaimer 11
Foreword 111
Abstract 1v
Figures v1
Tables v111
1. Introduction 1
2. Literature Review 2
3. Experimental Procedures 3
Evaluation of Measurement Techniques 3
Gas Chromatography (GC) 3
High Performance Liquid Chromatography (HPLC) 4
Solvent Stability Studies 4
Extraction Studies 5
Preservation Studies 6
Clean up Methods Evaluation 6
Method Validation Studies 7
4. Results and Discussion 11
Evaluation of Measurement Techniques 11
Gas Chromatography (GC) 11
High Performance Liquid Chromatography (HPLC) 20
Solvent Stability Studies 22
Extraction Studies 22
Preservation Studies 29
Clean up Method Evaluation 35
Wastewater Analysis 35
5. Sumnary and Recommendations 53
References 54
Appendix
N1troaromat1cs and Isophorone, Method 609 55
v
-------
FIGURES
Number
1 Chromatographic separation of the category 4 compounds at
100° on OV-1 at the 400 ng level 14
2 Chromatographic separation of the category 4 compounds at
1200 on 1.95% qp 1/1.5X OY-17 at the 400 ng level 15
3 Chromatographic separation of the category 4 compounds at the
400 ng level on 1.95X QF-1/1.5% OV-17 using a temperature
program from 80° to 170° at I00/m1n 16
4 Response curve for nltroaromatlcs using GC/ECD 17
5 Response curve for Isophorone using ECO 18
6 Response curve for nitrobenzene and Isophorone using GC/FID 19
7 Chromatograph for HPLC separation of the nltroaromatlc
compounds pounds using a Pye-Unlcan electron capture detector
at the 200 ng level 21
8 Chromatograms for Sample 1, unsplked, using ECO at 85°C,
Isothermal (a), and ECD at 120°C Isothermal (b) 38
9 Chromatograms for Sample 1, spiked with 13 ppm of nitrobenzene,
270 ppb of Isophorone, and 0.88 ppb each of 2,4- and 2,6-DNT,
using ECD at 85<>C (a), and 120°C (b) 39
10 Chromatograms for Sample 2, unsplked using ECO at 85°C (a),
and 170°C (b) 41
11 Chromatograms for Sample 2, spiked with 13 ppb nitrobenzene,
270 ppb Isophorone, and 0.88 ppb of 2,4-and 2,6-DNT, using
ECD at 85°C (a) and 120°C (b) 42
12 Chromatograms for Sample 3, unsplked, using ECD at 85°C (a)
and 145°C (b) 43
13 Chromatograms for Sample 3, spiked with 96 ppb of nitrobenzene,
560 ppb of isophorone, 70 ppb of 2,6-DNT, and 8 ppb of 2,4-DNT,
using ECD at 85°C (a) and 145°C (b) 44
vi
-------
Figures (continued)
Number Page
14 Chromatograms for Sample 4, unsplked using FID at 85°C (a),
and ECD at 145°C (b) 45
15 Chromatograms for Sample 4, spiked with 96 ppb nitrobenzene,
560 ppb Isophorone, 7 ppb of 2,6-DNT, and 8 ppb of 2,4-DNT,
using FID at 85°C (a), and ECO at 145°C (b) 46
16 Chromatograms for Sample 5, unsplked, using FID at 85°C (a)
and ECD at 145°C (b) 48
17 Chromatograms for Sample 5, spiked with 90 ppb of nitrobenzene,
50 ppb of Isophorone, and 5 ppb of 2,4- and 2,6-DNT, using
FID at 85°C (a), and ECD at 145°C (b) 49
18 Chromatograms for Sample 6 unsplked using FID at 85°C (a)
and ECD at 145°C (b) 50
19 Chromatograms for Sample 6 spiked with 100 ppb of nitrobenzene,
Isophorone, and 2,6-DNT and 50 ppb of 2,4-DNT using FID at
85°C (a) and ECO at 145°C (b) 51
20 Chromatograms for Sample 6 spiked with 100 ppb of nitrobenzene,
Isophorone, and 2,6-DNT and 50 ppb of 2,4-DNT after Flor1s1l
clean up using FID at 85°C and ECD at 145°C (b) 52
v11
-------
TABLES
Number Paqi
1 Column chromatographic scheme 8
2 Modified clean up method 9
3 Retention times of Category 4 compounds on various stationary
phases 12
4 Detection limits for Category 4 compounds 13
5 Detection limits for nltroaromatlcs and isophorone using various
detection systems 20
6 Solvent stability of Category 4 compounds In acetone 23
7 Solvent stability of Category 4 compounds 1n methanol 24
8 ANOVA analysis of solvent stability data 25
9 Data from triplication extraction of water dosed at 40 ppb
level 26
10 ANOVA analysis of extraction data 27
11 Recovery data for effect of drying agent and extraction solvent
volume 28
12 Comparison of K-D and vortex concentration 30
13 Percent recoveries of Category 4 compounds stored 1n
duplicate at the 40 ppb level and roan temperature 31
14 Percent recoveries of Category 4 compounds stored 1n duplicate
at the 40 ppb level and 4°c 32
15 Table of F values from ANOVA analysis of preservation data 33
16 Column chromatography data 36
17 Analytical results from various wastewater samples 37
viii
-------
SECTION 1
INTRODUCTION
The U.S. Environmental Protection Agency (EPA) has established a 11st of
129 toxic substances which are either known or likely to be present 1n
selected Industrial aqueous effluents. The 114 organic substances have been
categorized on a chemical basis Into 12 categories. The EPA has established
a monitoring program for the purpose of controlling the concentration of
such substances 1n the effluents at an environmentally acceptable level. In
support of this program, analytical methodologies for these toxic substances
have been developed and evaluated by application to some selected effluents.
This 1s a report of the development and evaluation of methodology for
the determination of Isophorone, nitrobenzene, 2,4-d1n1trotoluene, and
2,6-d1n1trotoluene which represent the Category 4 compounds 1n Industrial
aqueous discharges. The major thrust of the work 1s 1n gas chromatography
(GC) using flame Ionization (FID) and electron capture detection, (ECD) but
high performance liquid chromatography (HPLC) has been evaluated as a pos-
sible method using both UV and ECD.
The successful completion of this program Involved the fulfillment of
certain directives set forth 1n the contract by EPA. An extensive litera-
ture review was first conducted to evaluate the previous work 1n the area.
Subsequent work was directed toward determination and then full evaluation
of an appropriate measurement technique, which best satisfied the require-
ments for sensitivity and selectivity, as well as the considerations of
sample cost, that 1s, equipment, time, and training which would be needed
for the method. The stability of the nltroaromatlc compounds and Isophorone
1n water mlsdble solvents and their stability 1n chlorinated and unchlorl-
nated buffered water at different pHs and storage temperatures were studied
over the prescribed time periods. Extraction efficiency of two organic
solvents was also studied for the standard compounds. The remainder of the
program Involved the study of the sample preparation and clean up steps
which would be necessary to eliminate sample Interferences. The complete
method was then applied to several representative wastewater samples and an
assessment was made of the precision and accuracy of the complete procedure.
1
-------
SECTION 2
LITERATURE REVIEW
An automated computer search was made of Chemical Abstracts, Biological
Abstracts, Analytical Abstracts, and National Technical Information Services.
Only a few articles were found pertaining to the analysis of nltroaromatlc
compounds 1n water. Two of these articles described analytical methods
based on the direct colorlmetrlc determination of nitrate' or a
Me1senhe1mer complex' formed from the parent nltroaromatlc compound.
Since the detection limits of these methods are 1n the 10-20 ppm region they
are not useful for wastewater analysis where ppb detection limits are sought.
A third article described a method for determining TNT 1n sea water^.
This method Involves extraction with benzene followed by analysis with an
electron capture gas chromatograph using a Dexsll 300 stationary phase.
Approximately 70S extraction efficiency was obtained at the 1 ppb level and
the detection limit was reported to be 20 ppt.
Several other articles were located which dealt with the gas chromato-
graphic determination of nltroaromatlcs In media other than water. Several
of these used flame ionization detection^5.6 and lacked the sensitivity
for most wastewater applications.« One article' reported the use of thin
layer chromatography for the determination of 1,3,5-trlnitrobenzene,
l,2-d1n1trobenzene, and several trinitrotoluene Isomers as well as other
nltro compounds. Absolute detection limits were 1n the plcogram region. No
articles were found dealing with the HPLC analysis of the Category 4
compounds 1n water nor were any articles found dealing with the analysis of
Isophorone 1n aqueous media.
2
-------
SECTION 3
EXPERIMENTAL PROCEDURES
EVALUATION OF MEASUREMENT TECHNIQUES
Gas Chromatography (GC)
Isothermal, linear, and step-temperature programs were run on 1.95%
QF-1/1.5% OV-17 (Gas Chrom Q 80/100 mesh), 3.0* OV-17 (Chromosorb W 80/100
mesh), 3.OX OV-1 (Chromosorb W 80/100 mesh), and 3.OX 0V-101 (Chromosorb W
80/100 mesh). Glass columns of 1/4" O.D. x 4' were used for all stationary
phases with the exception of 0V-101 which was packed Into a 1/4" 0.0. x 10'
column. A Hewlett-Packard 5830A chromatograph was used with microprocessor
controlled temperature programming and peak Integration for quantitation.
The mobile phase for the FIO was nitrogen and for the ECO, 90X
Argon/methane, which was passed through a heated filter and oxygen
scrubber. The FID was kept at 250°C, the ECD (63N1 source) was set at
330°C, and the Injector temperature remained at 250°C. Manual Injec-
tions were made with a 10 pL Hamilton syringe.
All standard solutions were prepared with Burdlck and Jackson solvents.
The analytical grade standards were purchased from Aldrlch Chemical Company.
Nitrobenzene was used as received. Isophorone was distilled under vacuum
with a water aspirator connected to the distillation apparatus. A colorless
fraction (b.p. 213°-214°C at S.T.P.) was collected and stored at 4°C.
The 2,6- and 2,4-d1n1trotoluenes were recrystalUzed from benzene by addi-
tion of hexane. Melting points were 71°C for 2,4-d1n1trotoluene and
66°C for 2,6-d1n1trotoluene.
As discussed later 1n this report, the 1.95* QF-1/1.5X OV-17 was found
to be the most suitable stationary phase for this analysis. Using this
column the linearity and sensitivity for each compound was Investigated
using both FIO and ECD and the chromatographic conditions described above.
The column temperature was 85°C for Isophorone and nitrobenzene and
145°C for 2,4-DNT and 2-6-DNT. The conditions were chosen since 1t
appeared Impractical to determine all four compounds 1n a single Isothermal
run due to the wide range of retention times. To determine the linear range
for each compound, four microliter volumes of standard solutions of the
compounds at various concentrations, In acetone, were Injected. The
quantity of material necessary to produce a peak with a signal to noise
ratio of about 5:1 was defined as the minimum detectable 11m1t (MOL).
Solution concentrations were chosen to cover the range from the MDL to the
3
-------
upper limit of linearity. The results of this study are presented 1n a
separate section of this report.
High Performance Liquid Chromatography (HPLC)
A preliminary Investigation was conducted 1n order to determine whether
or not HPLC would be an appropriate technique for the determination of these
compounds. Previous work at Battelle® showed that reverse phase chroma-
tography was not selective enough for the separation of 2,4- and 2,6-DNTs.
Furthermore, the use of electron capture detection restricted studies to
normal phase because of the electron affinity of reverse phase solvents.
Thus, only normal phase chromatography was examined 1n this study.
The following HPLC conditions were used:
Solvent delivery system - Varlan Model 8500
Stationary phase - Dupont silica gel, 4.6 mm 1.0. x 250 mm long,
10 micron particle diameter
Mobile phase - 3% ethyl acetate 1n pentane
Flow rate - 2.3 mL/m1n - -
Injection volume - 10 yL
Detection system - UV - Dupont Model 837. ECD - Pye Unlcam model LC-EC
Detector unit.
A comparison was made of the sensitivity of the two detectors for the
Category 4 compounds by Injecting a series of standard solutions at appro-
priate concentrations 1n hexane.
SOLVENT STABILITY STUDIES
Solutions containing 200 ppm of all four compounds 1n acetone and
methanol were prepared. The solutions were flame sealed 1n 10 mL glass
ampules. Immediately after sealing and after storage 1n the dark and at
room temperature for 30, 60, and 90 days, three ampules of each solution
were opened and assayed by triplicate Injection of four microliter volumes
onto the GC/FID system.
Quantitative analysis of the samples was done by comparison of the peak
areas which were obtained from the microprocessor on the HP5830 GC to those
of a 200 ppm solution freshly prepared 1n the same solvent as the sample to
be analyzed. The chromatographic conditions used are as follows:
GC - HP5830A
Column - 1.95X QF-1/1.5* OV-17 on Gas Chrom Q (80/100 mesh)
1/4" O.D. x 6'.
Flow rate - 44 mL/m1n, Helium
Injector temperature - 250°C
Detector temperature - 250°C
Column temperature - 80°-180°C at 8°C/m1n
Statistical evaluation of this data was made with the computer program,
SPSS-Statistical Package for the Social Sciences, Northwestern University
4
-------
Version 6.5. The subprogram "oneway" provided fundamental statistics of the
data besides oneway analysis of variance (ANOVA) for each compound due to
changes 1n the concentration as a function of time.
EXTRACTION STUDIES
The solvents methylene chloride and ethyl acetate were examined for the
extraction of Category 4 compounds from aqueous solutions. Extraction at pH
2,7, and 10 was made with methylene chloride and ethyl acetate 1n triplicate
for all pHs. A 200 tiL volume of a 100 ppb solution was added to a 500 mL
buffered aqueous phase and extracted with three 100 mL quantities of organic
solvent. After drying the solvent by filtration through anhydrous MgS04,
the extract was evaporated to about 10 mL on a rotating evaporator and then
further concentrated to about 200 yL on a vortex evaporator at 30°C. The
dosage volume and 200 yL final volume was measured 1n a 1-mL syringe cali-
brated 1n 10 uL Increments (Hamilton No. 1001). At no time during the
procedure did the samples reach a temperature above 30°C. Buffers were
purchased from VWR Scientific to maintain the three pH levels. Their com-
positions are as follows: _
Buffer Systems
Ionic
pH Composition Strength
2.0 Tartaric and Phosphoric Acids .05M
and Potassium Blphthalate
7.0 Sodium and Potassium Phosphate .05M
10.0 Sodium Borate and Sodlun Carbonate .05M
Mlcrocentral Laboratories, the manufacturer of the capsules, refused to
furnish the absolute quantities of each component 1n the buffer. Concentra-
tion levels were detemlned with GC-FID by comparison to an external standard
using the chromatographic conditions described 1n the previous section.
Several other experiments were conducted to validate the use of alter-
nate extraction and work up schemes. In order to determine 1f smaller
volumes of solvent and/or alternate drying agents gave comparable results
the following experiments was conducted.
A 500-mL volume of water (buffered at pH 7 as above) was spiked with
200 yL of a 100 ppm acetone solution of the four compounds. Extraction was
done with three 30 mL volumes of methylene chloride followed by drying the
extract by addition of two grams of Na2S04. The extract was then
decanted and concentrated to 10 mL on a rotary evaporator and then reduced
1n volume further to 200 yL by vortex evaporation. GC-FID was used for
quantitation. This experiment was done 1n triplicate and repeated using
5
-------
MgS04 as the drying agent. The effect of solvent volume was determined by
comparison of these recoveries to those 1n the previous study, where 100 uL
volumes were used.
In order to determine 1f K-D concentration was as effective as vortex
concentration, a 2 ml volume of an acetone solution containing 200 ppm of
each compound was diluted to 10 ml with methylene chloride. The solution
was then concentrated to 2 mL by vortex evaporation at 30°C or by K-D
concentration using a Kontes tube heater at about 100°C. Each concentra-
tion procedure was done 1n triplicate and the final concentrate assayed by
GC-FID as described earlier.
A third experiment was conducted to determine 1f any loss of these
compounds resulted during solvent displacement of the methylene chloride
with toluene, since this 1s necessary when 6C-ECD 1s used. A 500 iiL volume
of a 200 ppm acetone solution of the compounds was diluted to 10 ml with
methylene chloride. The solution was then concentrated to 1 mL by K-D using
a tube heater at 100°C. Five-hundred microliters of toluene were then
added and the sample further concentrated to 500 y. This concentrate was
then analyzed by GC-FID as before. This experiment was done 1n triplicate.
PRESERVATION STUDIES
In order to examine the stability of these compounds under various
storage conditions the following experiment was conducted. Samples were
prepared by dosing 500 mL buffered water with 200 yL of an acetone solution
containing 100 ppm of each compound. All samples were stored 1n the dark.
Two temperatures, room and 4°C, three pH levels, 2, 7, and 10, and two
chlorine levels, 0 and 2 ppm were Investigated. This resulted In a 3 x 2 x
2 storage condition matrix, 1n which each point 1n the matrix was run 1n
duplicate, giving a total of 24 samples which were studied. The three pH
levels were achieved using the buffers described 1n the Extraction Section.
The two ppm chlorine level was achieved by spiking the sample with 2 mL of a
solution containing 1500 ppm of calcium hypochlorite.
After storage for seven days, the samples were extracted using the same
extraction and work up procedures as were used for the extraction studies
described earlier, except that 30 mL quantities of methylene chloride were
used instead of 100 mL quantities. Quantitation was done by GC-FID using an
external standard as described previously. ANOVA analysis was done on the
data using the SPSS computer package previously described.
CLEAN UP METHOOS EVALUATION
Both Flor1s1l and silica gel were evaluated for the elutlon of
Category 4 compounds. F1or1s11 (Sigma Chemical Company, 60-100/PF) and
silica gel (Davidson Chemical Company, 100-200 mesh, Grade H) were prepared
by wet packing 10 gram quantities of these stationary phases (baked at
120°C for 24 hours) as a slurry 1n MeCl2« The columns were then eluted
with 50 mL of pentane. A 500 uL volume of a 5X MeCl2/pentane solution,
containing 2 mg/mL of each compound was added to the top of the column.
Both columns were then eluted with a series of solvents as shown in Table 1
6
-------
(Conditions 1 and 2) and the fractions analyzed by GC-FID as described
previously.
After examination of the separation profiles of each column, Flor1s1l
was selected as the most appropriate for this application. From the point
of view of sample concentration acetone was considered to be preferable to
MeOH, since 1t 1s more volatile. Therefore, the elutlon profile of the
Category 4 compounds was determined using a series of acetone/MeCl2
mixtures as shown 1n Table 1, Condition 3. This elutlon scheme was used for
the analysis of the first three wastewater samples described 1n the next
section.
METHOD VALIDATION STUDIES
Six water samples were received and used to validate the proposed test
procedure for Category 4 compounds. These samples were as follows:
1. Sewer pump sample receiving wastes from phenolic resins, vinyl
acetate, and polyvinyl chloride process areas.
2. Brine sample from holding tank receiving washings from ships
delivering various commodities.
3. Secondary sewage effluent (Columbus, Ohio).
4. Final effluent from UNOX treatment system receiving wastes from
plants producing plastlclzers, butyl rubber, and olefins.
5. Surface water (Olentangy River).
6. Final effluent from organic chemical plants producing nitrobenzene,
o-d1chlorobenzene, o-n1trophenol, aniline, and oil additives.
For the validation of the protocol the experimental design was as
follows. Three replicates were run undosed Initially. Based on these
results the sample was then dosed with levels of the compounds sufficient to
give a response at least five times the background level and rerun 1n
triplicate. After dosing at the same level, three allquots of the sample
were placed 1n the refrigerator at 4°C for seven days. The samples were
then reassayed.
The methodology used 1n the analysis of samples 3-6 1s presented 1n the
Appendix and 1s the optimized protocol resulting from this work. The
methodology used for samples 1 and 2 differed from the final protocol 1n two
respects. The clean up method used for samples 1 and 2 was Condition 3 1n
Table 1, presented 1n the Clean up Methods evaluation section. This method
was replaced with clean up Condition 4, Table 2, which was used for samples
3-6 and 1s Included 1n the final method. This method Involved activation of
the Flor1s1l at 200°C, rather than 120°C (as 1n Condition 3). The
elutlon solvent was also changed, as shown In Table 2, to eiute the com-
pounds with 30 mL of 10% Acetone/MeCl2. This procedure gave a more
reliable elutlon profile than the previous condition.
7
-------
TABLE 1. COLUMN CHROMATOGRAPHIC SCHEME
Stationary
Fraction
Mobile
Condition
Phase
Number
Phase
10 mL volumes
1
Silica Gel
1.
Pentane
2.
5* MeClp/Pentane
3.
20* MeCi2/Pentane
A.
50* MeCl?/Pentane
5.
100* MeC12
6.
20% Me0H/MeCl2
7.
50* MeOH/MeCl?
8.
50* Me0H/MeCl2
9.
50* MeOH/MeCl2
10 mL volumes
2
Flor1s11
1.
Pentane
2.
5* MeClj/Pentane
3.
20* MeCi2/Pentane
4.
50* MeCl2/Pentane
5.
100* MeCi2
6.
20* Me0H/MeCl2
7.
50* Me0H/MeCl2
8.
50* Me0H/MeCl2
9.
50* MeOH/MeCl2
10 mL volumes
3
F1or1s1l
1.
30* Acetone/MeCl2
2.
40* Acetone/MeCl2
3.
50* Acetone/MeC12
4.
60* Acetone/MeCl2
5.
70* Acetone/MeC12
6.
80* Acetone/MeCl2
8
-------
TABLE 2. MODIFIED CLEAN UP METHOD
Stationary
Mob1le
Condition
Phase
Phase
30 mL each
4
F1or1s1l
1. MeCl2
Activated at
200°C
2. 10X Acetone/MeCl2
3. 10X Acetone/MeCl2
9
-------
For samples 1-3 nitrobenzene and Isophorone were determined by
Isothermal elutlon at 85°C with ECD, whereas for samples 4-6 they were
determined by FID, as described 1n the Appendix. A detailed description of
the treatment and analytical consideration for each sample 1s given 1n the
results section.
The basis for selection of these particular samples was as follows.
Samples 1 and 2 were obtained from a source which had previously been found
to contain Isophorone. Sample 3 was chosen at the request of the EPA
project officer, who asked that the wastewater application phase for all
categories Include a secondary municipal sewage effluent. Sample 5 was
chosen 1n order to validate the method for surface waters as well. Sample 4
was designated by the EPA project officer since 1t represents an effluent
from an advanced waste treatment system serving a wide variety of chemical
processes. Sample 6 was a common effluent from several organic chemical
manufacturing plants and was believed to contain nitrobenzene.
10
-------
SECTION 4
RESULTS AND DISCUSSION
EVALUATION OF MEASUREMENT TECHNIQUES
6as Chromatography (GC)
The retention times of the nltroaromatlc compounds and Isophorone for
the GC columns tested are shown 1n Table 3. The OV-1 column packing
exhibits the least affinity for Category 4 compounds which 1s to be expected
since 1t 1s the least polar of the packing materials studied. Separation of
the four compounds was accomplished with 3.OX OV-1,' 3X 0V-101, and 1.95%
QF-1/1.5X OV-17. However, 3% OV-17 did not separate Isophorone and nitro-
benzene. The chromatograms shown 1n Figures 1-3, demonstrate the type of
separation that was achieved with these stationary phases. The 1.95%
QF-1/1.5% OV-17 stationary phase was chosen for use 1n the methodology over
the others because 1t showed less solvent peak tailing under the conditions
used. For confirmation purposes, either 0V-101 or OV-17 1s suitable,
however, 1t must be kept 1n mind that Isophorone and nitrobenzene are not
resolved on OV-17. OV-17 gives better peak slopes with less tailing for
2,4-DNT and 2,6-DNT than does OV-1.
The response curves for the four compounds using FID and ECD (using
1.95% QF-1/1.5% OV-17) are shown 1n Figures 4-6. The limits of detection
for these compounds on both detectors are shown 1n Table 4. From these data
1t 1s readily apparent that ECD 1s the preferred detector for 2,4-DNT and
2,6-DNT due to Its extremely high sensitivity for these compounds. The
sensitivity of the two detectors for Isophorone 1s comparable. Initially,
1t was felt that ECD would be more selective for Isophorone and thus 1t was
chosen as the preferred detector. However, 1t was later determined for
certain wastewater samples that FID was more appropriate for Isophorone
determination since 1t exhibited a lower background level. The ECD 1s more
sensitive for nitrobenzene than FID and 1s thus the preferred detector.
However, the use of FID for nitrobenzene has the advantage that, since both
Isophorone and nitrobenzene can be determined at the same Isothermal elutlon
temperature, only one chromatographic run needs to be made to determine both
compounds.
The linearity of the chromatographic column was established down to 9.0
nanograms Injected for nitrobenzene, 0.12 for 2,6-DNT, and 0.12 for 2,4-DNT
using the ECD, and 4 ng Injected for Isophorone and nitrobenzene using FID.
Due to the wide difference 1n retention times for the four compounds, 1t
11
-------
TABLE 3. RETENTION TIMES OF CATEGORY 4 COMPOUNDS
ON VARIOUS STATIONARY
PHASES
Temperature Flow
Retention Times
Column
(°C) (mL/m1n)
(m1n)
Compound
1.95% QF-1/
85 44
3.31
Nitrobenzene
1.5% OV-17
4.49
Isophorone
145 44
3.52
2,6-DNT
*
5.35
2,4-DNT
3% 0V-101*
100 44
4.31
Nitrobenzene
5.72
Isophorone
150 44
4.75
2,6-DNT
6.54
2,4-DNT
3% OV-1
100 30
1.27
Nitrobenzene
1.61
Isophorone
8.14
2,6-DNT
12.4
2,4-DNT
OV-17
120 30
5.50
Nitrobenzene
150
5.50
Isophorone
16.7
2,4-DNT
22.8
2,6-DNT
*1/4" x 10' glass column, all others were 1/4" x 4' glass column.
12
-------
TABLE 4. DETECTION LIMITS FOR CATEGORY 4 COMPOUNDS
Minimum detectable limit*
Compound Detector (ng Injected)
Nitrobenzene FID 2.5
Isophorone 3.0
2,6-DNT 2.9
2.4-DNT 3.0
Nitrobenzene ECD 1.2
Isophorone 1.1
2,6-DNT 0.06
2,4-DNT 0.04
~Calculated using five to one signal to noise ratio.
13
-------
z
nO 00
Figure 1. Chromatographic se
coapounds at 100®
paration of the category 4
on 0V-1 at the 400 ng level.
14
-------
Figure 2. Chromatographic separation of the category 4
conpounds at 120* on 1.95X QF 1/1.51 OV-17
at th* 400 ng level.
15
-------
Chromatographic separation of tha category 4
conpounds at 120* on 1.95X QF 1/1.5Z OV-17
at the 400 ng laval.
15
-------
o
%r\ r*
• ~
-------
ta SO Aft jo *0 70 ii> *6 illll
iMMt IiJmM (a|)
Figure 4. Response curve for nltroaromatlcs using 6C/EC0.
-------
Aaount Iojcctcd (us)
Figure 5. Response curve for lsophoron* using CCD.
18
-------
I ~ I ' r 1"; IJ; r Iv j .: " " • i • - iaophoroM
JlMwt Iaj*cc«4 (t|)
Figure 6. Response curve for nitrobenzene and
Isophorone using 6C/FID.
-------
was necessary either to select a temperature program condition or two
different Isothermal conditions for the determination of all four com-
pounds. Figure 3 shows a temperature programmed run (80°-170° at
I00/m1n) which separates the four compounds quite nicely in a relatively
short period of time (about 10 min). However, temperature programming
cannot be done using many ECDs since a large Increase 1n background level
results, leading to decreased sensitivity. For this reason, two Isothermal
conditions were chosen (85°C for nitrobenzene and Isophorone and 145°C
for 2,4-DNT and 2,6-DNT) so that each of the components eluted within three
to five minutes.
High Performance Liquid Chromatography (HPLC)
An evaluation was made of normal phase HPLC using ECD and UV detection,
as described 1n the experimental section.
The noise level of the HPLC-ECD changed appreciably from day to day,
possibly due to impurities 1n the solvents such as water or dissolved
oxygen. Because of the difficulties Involved with achieving a consistent,
reproducible noise level from day to day, the detector was considered not to
be desirable for this analysis. A chromatogram showing the separation and
detection of the Category 4 compounds using HPLC-ECD 1s shown in Figure 7.
The conditions used were those described 1n the Experimental Section.
A comparison of the UV and ECD sensitivities for the four compounds, as
well as the GC-FID and GC-ECD detection limits 1s shown 1n Table 5.
TABLE 5. DETECTION LIMITS FOR NITROAROMATICS AND
ISOPHORONE USING VARIOUS DETECTION SYSTEMS
HPLC
ECD
UV
GC
FID ~
ECD
Nitrobenzene
22*
4.7
2.5
1.2
Isophorone
200
13.3
3.0
1.1
2,6-DNT
4
20.0
2.9
0.06
2,4-DNT
7
11.5
3.0
0.04
* Nanograms Injected to produce 5 to 1 signal to noise
ratio.
These data show that GC/ECD 1s far more sensitive than HPLC/ECD for all
of these compounds. It 1s also apparent that HPLC/UV is nearly as sensitive
as HPLC/ECD for these compounds. Based on this data 1t was concluded that
GC/ECD or GC/FID would be a superior determination technique for these
compounds.
20
-------
c
c
a
0
01
c
Figure 7. Chrooatograph for HPLC separation of tha nitroaroaatic
compounds using a Pye-Unicaa electron capture detector
at the 200 ng level.
21
-------
SOLVENT STABILITY STUDIES
The stabilities of the four compounds were found to be quite good 1n
both acetone and methanol over the 90-day period, as shown 1n Tables 6 and
7. ANOVA analysis of the data, shown in Table 8, Indicated that at the 95X
confidence level (F.95) that there is a significant difference between the
levels of Isophorone, 2,4-DNT, and 2,6-DNT 1n methanol and 2,6-DNT in
acetone with time. However, visual Inspection of the data Indicates that
any difference which exists 1s quite small and the statistical significance
1s a result of the high precision of the assay and random bias of the
chromatographic calibration from one time interval to the next.
The 95% confidence level was chosen since it represents a medium degree
of precision whereas the 99* confidence level 1s relatively restrictive and
the 90X confidence level 1s very non-restr1ct1ve. This choice 1s somewhat
arbitrary and could be altered depending on the purpose of the particular
study.
EXTRACTION STUDIES
The solvents methylene chloride and ethyl acetate were examined for the
extraction of Category 4 compounds from aqueous solutions. Recoveries, of
all four of these compounds were greatest for methylene chloride at pH 7 and
10, as shown 1n Table 9. Visual Inspection of the data In this table
Indicates a substantial decrease 1n recoveries at pH 2 in all cases. The
effect of pH on recovery 1s much less obvious when the extraction 1s done
with methylene chloride Instead of ethyl acetate. ANOVA analysis of the
data was conducted using a three way analysis program of the extraction
efficiencies at each pH. This approach yields a single F value which points
out whether or not there 1s a significant difference between any of the data
sets, but does not decide which direction the difference Is. The ANOVA
data, shown 1n Table 10, Indicates that there is a great dependence of
recovery on extraction solvent, and thus methylene chloride 1s a signifi-
cantly better solvent than ethyl acetate. The ANOVA also Indicates that
there 1s a significant dependence of extraction efficiency on pH for both
solvents, and this dependence 1s much greater for ethyl acetate than
methylene chloride. Visual Inspection of the data shows that only the pH 2
data 1s lower using methylene chloride and these data points are undoubtedly
the reason for the significant F value from ANOVA analysis.
In suimtary, the data shows that methylene chloride 1s a far better
extracting solvent than ethyl acetate, for these compounds and that extrac-
tion at pH 7 or 10 with methylene chloride is more effective than extraction
at pH 2.
As described in the experimental section, several related experiments
were conducted to determine whether or not alternate work up procedures
could be used effectively. The first such experiment examined whether: 1)
30 mL volumes of solvent could be used in place of 100 mL volumes and, 2)
Na2S04 could be used 1n place of MgSO^j. The results of this study are
shown in Table 11. From these data, it 1s obvious that recoveries are not
affected by using Na2S04 in place of MgS04. In comparing these data
22
-------
TABLE 6. SOLVENT STABILITY OF CATEGORY 4
COMPOUNDS IN ACETONE
Compound
Mean
Concentration
(ppm)
* Stability
Nitrobenzene
Time 0 Days
200
100 ± 3.0
Time 30
202
101 ± 1.5
Time 60
199
100 ± 4.2
Time 90
197
99 ± 5.5
Isophorone
Time 0
202
101 ± 4.0
Time 30
200
100 ± 3.5
Time 60
199
100 ± 4.7
Time 90
201
101 ± 1.7
2,6-DNT
Time 0
199
100 ± 2.3
Time 30
203
102 ± 4.0
Time 60
194
97 ± 4.2
Time 90
196
98 ± 5.3
2,4-DNT
Time 0
200
100 ± 2.0
Time 30
202
101 ± 3.5
Time 60
186
93 ± 1.7
Time 90
190
95 ±3.1
23
-------
TABLE 7. SOLVENT STABILITY OF CATEGORY 4
COMPOUNDS IN METHANOL
Compound
Mean
Concentration
(ppm)
X Stability
Nitrobenzene
Time 0 Days
196
98 ± 3.5*
Time 30
196
98 t 0.58
Time 60
197
99 ± 2.3
Time 90
201
101 ± 2.5
Isophorone
Time 0
203
102 ± 0.58
Time 30
202
101 ± 0.58
Time 60
201
101 ± 1.7
Time 90
202
101 ± 2.1
2,6-DNT
Time 0
200
100 t 4.9
Time 30
182
91 ± 0.58
Time 60
197
99 ± 2.5
Time 90
209
105 ± 2.5
2,4-DNT
Time 0
201
101 ± 6.5
Time 30
206
100 ±2.0
Time 60
184
92 t 2.6
Time 90
193
97 ± 2.3
~Standard deviation for triplicate analysis.
24
-------
TABLE 8. ANOVA ANALYSIS OF SOLVENT STABILITY DATA
Fexp
Fexp
acetone
methanol
Nitrobenzene
0.466
2.93
Isophorone
0.525
186*
2,6-DNT
6.95*
14.1*
2,4-DNT
3.79
23.1*
~F.99 ¦ 7.59
4.59
4.07
~Exceeds F.95 Test for significance.
25
-------
TABLE 9. RECOVERY FROM TRIPLICATION EXTRACTION OF
WATER DOSED AT 40 PPB LEVEL
Compound
PH
Solvents
Nitrobenzene
Isophorone
2,6-DNT
2,4-DNT
Methylene chloride
81.0 ± 7.5*
Ethyl acetate
34.9 ± 7.6
7
103.0 *
2.4
83.0 ±
4.8
10
101.0 ±
3.7
65.0 t
7.8
2
80.0 ±
8.0
39.2 t
9.2
7
107.0 t
4.9
83.0 ±
3.3
10
104.0 t
3.9
60.0 ±
2.9
2
71.0 t
o
•
o
•H
O
•
fM
o
•
CM
7
88.0 t
8.2
72.0 ±
1.8
10
89.0 t
0.39
65.1 t
3.4
2
69.2 t
16.0
27.0 t
14.0
7
90.0 ±
o
•
o
70.0 t
2.2
10
91.0 ±
1.7
75.0 ±
4.9
*% recovery t standard deviation
26
-------
TABLE 10. ANOVA ANALYSIS OF EXTRACTION DATA
I) Fexp values for the effect of change 1n pH for each extraction solvent.
Compound Methylene chloride Ethyl acetate
Nitrobenzene 23.1* 109*
Isophorone 23.0* 182*
2,6-DNT 8.75* 414*
2,4-DNT - . . .. 6.51* 207*
F.95 5.14 5.14
II) Fexp values for the effect of extraction solvent at constant pH.
Compound
PH 2
PH 7
pH 10
Nitrobenzene
143*
55.8*
97.4*
Isophorone
92*
48.6*
298*
2,6-DNT
109*
14.3*
337*
2,4-DNT
39*
13.9*
47.6*
F.95
7.71
7.71
7.71
~Significant difference at F.95.
27
-------
TABLE 11. RECOVERY DATA FOR EFFECT OF DRYING AGENT
~ "AND EXTRACTION SOLVENT VOLUME
Compound
Recovery
Extraction
Solvent
Drying Agent
100 mL HeCU*
MgS04 1
30 ml MeCl-
Na2S04 c
30 mL MeCl9
MgS04 £
Nitrobenzene
103 ± 2.4**
83 t 2.0
81 t 3.3
Isophorone
107 ± 4.9
87 ± 1.5
86 ± 2.0
2,6-DNT
88 t 8.2
94 ± 2.0
98 ± 13.0
2,4-DNT
- 90 t 10.0
85 ± 3.0
84 t 3.5
* Data from Table 9.
** Percent recovery ± standard deviation for triplicate extractions at the
40 ppb level and pH 7.
28
-------
to those 1n Table 9 for pH 7 and methylene chloride, 1t can be seen that a
slight decrease 1n recovery 1s noted for some of the compound using 30 ml
rather than 100 mL quantities of solvent. However, since the recoveries are
still quite good (> 85*) using 30 mL volumes 1t was felt that this volume
could be used 1n the extraction protocol.
A second related experiment was conducted 1n which K-D concentration was
compared to vortex concentration, since many laboratories do not have a
vortex evaporator. The data for this experiment are given 1n Table 12.
From these data 1t 1s obvious that the K-D and vortex concentration are
equally effective and thus either one could be used 1n the protocol.
A third experiment was conducted to determine whether or not significant
loss occurs when methylene chloride 1s displaced with toluene, using the K-D
concentrator. When a methylene chloride solution was concentrated and
solvent displaced with toluene recoveries were 93 t 1.2% for nitrobenzene,
94 ± 1.3* for Isophorone, 100 ± 2.2% for 2,6-DNT, and 102 ± 3.IX for
2,4-DNT. Thus, no appreciable losses occurred during solvent exchange and
this protocol can be followed when using the GC-ECD, which 1s Incompatible
with methylene chloride.
Based on these studies, the preferred extraction protocol 1s as
follows: 1) solvent extract with 3 x 60 ml/1 Iter of sample of methylene
chloride at pH 7; 2) dry extract by filtration through 2 grams of Na2S04
or MgS04; and 3) concentrate the extract to about 1 mL using K-D concen-
tration. If GC-ECD 1s to be used the extract must be solvent exchanged with
toluene by addition of 1 mL of toluene and reconcentratlon of the extract to
1 mL using K-D concentration.
PRESERVATION STUDIES .
The stability data for this study are presented 1n Tables 13 and 14.
The stabilities presented were not corrected for extraction efficiency since
1) quantitative (> 90%) extraction was observed at pH 7 and pH 10 using
methylene chloride; and 2) although the extraction efficiencies at pH 2,
from the previous study, were on the order of 80%, the extraction of the
stored water samples had apparently higher recoveries, for an as yet unknown
reason.
Stabilities of the compounds appeared to be highest at 4°C and pH 7 or
pH 2 In the absence of chlorine as shown 1n Tables 13 and 14. ANOVA
analysis, as shown 1n Table 15, Indicates that some small variations 1n the
recoveries may be attributed to chlorine content 1n the samples, however,
visual Inspection of the data shows the effect of added chlorine to be
Inconsistent and quite small.
On the basis of this data the suggested storage condition 1s pH 7 and
4°C. Storage at 4°C 1s preferred since biological degradation, which
could occur 1n actual samples, will be retarded under reduced temperature
conditions.
29
-------
TABLE 12. COMPARISON OF K-0 AND VORTEX CONCENTRATION
Compound
K-0
Recovery
Vortex
Nitrobenzene
88 t 4.7*
89 t 2.9
Isophorone
88 ± 4.9
93 ± 2.0
2,6-DNT
89 l 4.3
93 ± 2.7
2,4-DNT
89 ± 5.7
94 t 2.7
~Percent recovery ± standard deviation for triplicate analysis.
30
-------
TABLE 13. PERCENT RECOVERIES OF CATEGORY NO. 4 COMPOUNDS STORED
IN DUPLICATE AT THE 40 PPB LEVEL AND ROOM TEMPERATURE
Room temperature No chlorine 2 ppm chlorine
pH 2
Nitrobenzene
85.8
±
7.6
83.7
t
1.4
Isophorone
87.0
t
13.0
82.5
t
1.2
2,6-D1n1troto1uene
96.0
t
10.0
92.3
i
2.6
2,4-D1n1trotoluene
96.5
*
16.0
90.5
t
0.6
PH 7
Nitrobenzene
72.9
t
11.0
73.8
t
8.4
Isophorone
73.4
t
13.0
74.5
±
7.8
2,6-D1n1troto1uene
81.0
±
11.0
81.5
t
7.9
2,4-D1n1trotoluene
76.3
t
21.0
86.7
±
8.5
pH 10
Nitrobenzene
77.4
±
2.0
67.7
t
0.91
Isophorone
78.8
t
2.0
72.0
±
0.0
2,6-D1n1troto1uene
83.3
t
1.2
78.3
±
0.0
2,4-Dlnltrotoluene
82.7
t
2.5
82.9
±
0.0
31
-------
TABLE 14. PERCENT RECOVERIES OF CATEGORY 4 COMPOUNDS STORED
IN DUPLICATE AT THE 40 PPB LEVEL AND 4°C
4°C No chlorine 2 ppm chlorine
pH 2
Nitrobenzene
87.7
±
0.071*
67.9
±
4.2
Isophorone
96.2
±
0.61
74.0
±
4.7
2,6-DNT
105.0
t
4.9
91.3
±
11.0
2,4-DNT
99.8
±
8.8
85.6
±
9.8
PH 7
Nitrobenzene
86.4 t 2.4
79.1 ± 23.0
Isophorone
88.0 ± 1.9
79.5 ± 22.0
2,6-DNT
97.4 ± 3.3
90.0 t 23.0
2,4-DNT
105.0 ± 4.2
83.0 ± 9.0
pH 10
Nitrobenzene
77.9
±
12.6
51.9
±
19.0
Isophorone
81.4
±
13.2
53.6
t
17.0
2,6-DNT
91.8
t
20.0
64.4
±
12.0
2,4-DNT
83.8
±
10.4
62.2
±
8.6
* Percent recovery ± standard deviation.
32
-------
TABLE 14. PERCENT RECOVERIES OF CATEGORY 4 COMPOUNDS STORED
IN DUPLICATE AT THE 40 PPB LEVEL AND 4°C
4°C No chlorine 2 ppm chlorine
pH 2
Nitrobenzene
87.7
±
0.071*
67.9
±
4.2
Isophorone
96.2
±
0.61
74.0
±
4.7
2,6-DNT
105.0
t
4.9
91.3
±
11.0
2,4-DNT
99.8
±
8.8
85.6
±
9.8
PH 7
Nitrobenzene
86.4 t 2.4
79.1 ± 23.0
Isophorone
88.0 ± 1.9
79.5 ± 22.0
2,6-DNT
97.4 ± 3.3
90.0 t 23.0
2,4-DNT
105.0 ± 4.2
83.0 ± 9.0
pH 10
Nitrobenzene
77.9
±
12.6
51.9
±
19.0
Isophorone
81.4
±
13.2
53.6
t
17.0
2,6-DNT
91.8
t
20.0
64.4
±
12.0
2,4-DNT
83.8
±
10.4
62.2
±
8.6
* Percent recovery ± standard deviation.
32
-------
TABLE 15. TABLE OF F VALUES FROM ANOVA ANALYSIS
OF PRESERVATION DATA
ANOVA by temperature
pH 2
nitrobenzene
Isophorone
2,6-DNT
2,4-DNT
F.95
No chlorine
0.126
0.936
1.18
0.062
10.1
2 ppm chlorine
25.0*
6.30
0.017
0.500
10.1
pH 7
Nitrobenzene
Isophorone
2,6-DNT
2,4-DNT
F.95
2.69
2.48
3.98
3.48
10.1
0.095
0.092
0.256
0.181
10.1
pH
10
"RTtrobenzene
Isophorone
2,6-DNT
2,4-DNT
F.95
0.003
0.076
0.360
0.023
10.1
0.890
2.36
2.75
11.5*
10.1
ANOVA by pH
Room Temperature
No Chlorine Chlorine
4°C
No Chlorine Chlorine
Nitrobenzene
Isophorone
2,6-DNT
2,4-DNT
F.95
1.35
0.805
1.74
0.887
9.55
7.32
2.88
4.68
1.19
9.55
1.02
1.83
0.553
3.59
9.55
1.25
1.41
1.82
3.91
9.55
(continued)
33
-------
TABLE 15. (continued)
ANOVA by chlorine content
pH 2
Nitrobenzene
Isophorone
2,6-DNT
2,4-DNT
F.95
pH 7
Nitrobenzene
Isophorone
2,6-ONT
2,4-DNT
F.95
pH 10 '
Nitrobenzene
Isophorone
2,6-ONT
2,4-DNT
F.95
Room temperature 4°C
No chlorine 2 ppm chlorine No chlorine 2 ppm chlorine
0.356
6.95
2.09
3.76
10.1
0.00
0.001
0.006
0.376
10.1
0.234
7.44
4.05
6.35
10.1
0.156
0.225
0.254
0.274
10.1
0.008
0.011
0.003
0.413
10.1
81.3*
22.3*
35.3*
0.120
10.1
0.595
1.01
4.24
0.016
10.1
1.49
1.12
0.200
0.118
10.1
2.17
3.61
4.77
16.7*
10.1*
43.8*
44.6*
2.55
2.31
10.1
0.209
0.297
0.211
9.69
10.1
2.56
3.36
2.78
5.14
10.1
* Significant difference at F .95.
34
-------
CLEAN UP METHOD EVALUATION
The results of this study are presented 1n Table 16. The compounds of
Interest eluted 1n several different fractions on slUca gel, but 1n a
single fraction on Flor1s1l (conditions 1 and 2 1n Table 15 respectively).
Therefore, Flor1s1l was considered most appropriate for this application.
Since acetone bolls at a somewhat lower temperature than methanol, 1t
was considered useful to Investigate the elutlon of the Category 4 compounds
using acetone as a mobile phase modifier. This scheme 1s shown 1n Table 16
as condition 3. As shown 1n Table 16, the compounds eluted primarily 1n the
40< and 50X acetone/MeCl2 fractions. This scheme, 1n which fractions 2-4
were collected, was used for the clean up of the first three wastewater
samples described 1n the next section. For the last three water samples
condition 4, Table 16, was used as described 1n the following section.
However, 1t was discovered 1n subsequent work that 1n some circumstances the
nltrotoluenes partially elute 1n the methylene chloride fraction. For this
reason the final protocol, 1n the Appendix, has been modified to Include the
use of 10% MeCl2 1n hexane, rather than 100% MeCl2 1n the Initial eluate.
WASTEWATER ANALYSIS
The procedure presented 1n the Appendix was applied to a series of six
water samples representing surface water, municipal effluents, and several
Industrial effluent types. As described 1n the experimental section, the
clean up and determinative methods were revised after the analysis of
several of the wastewaters. All samples were subjected to clean up before
analysis. A description of the problems associated with the analysis of
each sample follows. All data obtained from the analysis of these samples
1s given 1n Table 17.
Sample 1 was a sample drawn from a pump receiving wastes from various
process areas, Including phenolic resins, vinyl acetate, and polyvinyl
chloride. The sample appeared to contain about 40 ppb of Isophorone and
about 0.2 ppb of 2,6-DNT, as shown 1n Figure 8 and Table 17. Although 1t
did not appear to contain detectable levels of nitrobenzene or 2,4-DNT,
there were two peaks, one elutlng slightly before and one elutlng slightly
after 2,4-DNT, thus Increasing Its MOL. Recoveries of the spiked compounds
were relatively good, although only about 50% recoveries were realized for
2,4- and 2,6-DNT. Upon storage for seven days at 4°C the Isophorone
appeared to have decomposed whereas the other components had not (to a great
extent). Based on visual Inspection of the unsplked chromatograms and the
recoveries for spiked compounds the estimated MDLs 1n this sample are 2 ppb
for nitrobenzene, 25 ppb for Isophorone, 0.1 for 2,6-DNT, and 0.4 for
2,4-DNT. On Inspection of the spiked sample chromatogram (Figure 9) for
2,4-DNT 1t was noted that the GC Integration system Integrated only one peak
for the spiked sample, which 1s a composite of the 2,6-DNT and the two
overlapping peaks mentioned previously (and shown 1n Figure 8). In order to
obtain a reasonably accurate area for 2,4-DNT alone, the areas for the two
peaks from the unsplked samples were subtracted from the area of the single
peak for the spiked sample. This approach Is, of course, not useable 1n
obtaining data from unsplked samples.
35
-------
Condition
Stationary
phase
Mobile
phase
composition
* recovery
Silica Gel
10 aL each
Nitrobenzene
Isophoron*
2,6-ONT
2.4-ONT
1.
Pentane
NO*
NO
NO
NO
2.
5* MeCb/Pentane
NO
NO
NO
NO
3.
20* HeClj/Pentane
N>
NO
NO
NO
4.
50* NeClj/Pentane
NO
NO
NO
NO
5.
100* HeCl?
77.7 t 4.0**
NO
35.6 t 3.7
15.4 i 2.3
6.
20* MeOH/HeCl?
NO
NO
41.4 t 4.5
61.7 t 1.8
7.
50* NeOH/MeCI?
W
68.4 t 2.S
3.0 t 1.7
6.9 * 2.8
8.
50* NeOH/MeCl?
NO
NO
NO
NO
9.
50* NeOH/MeCl;
W
NO
W
NO
TOT*
77.7 t 4.0
68.4 t 2.5 80.0 t 3.3 84.0 l 2.3
norlsll Saw as condition 1. 74.6 i 3.2
120°C Only fraction 7 con-
activation tained detectable mounts
of the compounds.
75.2 t 4.2 74.1 * 4.5 77.0 t 3.9
TOTAL
74.6 l 3.2
75.2 1 4.2 74.1 t 4.5 77.0 t 3.9
10 i
each
Florlsll
120°C
activation
1.
2.
3.
4.
5.
30* Ac/MeCI;
40* Ac/HeCl?
SOS Ac/HeC1?
60S Ac/MeClj
70* Ac/MeClj
6. 80* Ac/MeCl;
NO
18.2 t 2.S
52.4 ~ 3.3
0.94 t 2.S
NO
HO
NO
6.S t
65.0 t
3.2 i 1.7
NO
NO
3.0
2.7
W
21.2 t 2.7
52.9 i 1.9
1.7 * 2.5
NO
NO
NO
17.8 t 2.8
56.7 * 3.1
1.8 t 2.9
NO
NO
TOTAL
71.5 t 2.8
74.7 ~
2.5
75.8 i 2.4
76.3 t 2.9
30 aL each
4
F1or1s11
200°C
activation
1.
2.
3.
100* MeCI?
10* Ac/MfCl?
10* Ac/HeCl2
NO
97.0 t 2.8
NO
NO
99.0 t
NO
1.0
NO
97.0 t 4.5
NO
NO
96.0 t 4.6
NO
Detection limit corresponds to ^IX recovery.
Standard deviation for triplicate runs.
TABLE 16. COLUMN CHROMATOGRAPHY DATA
36
-------
Sample Wo.
Saaple
Sapln
background level
lEEll
Level
spiked
X Recovery
nuvi
0 diyt at 4 C
u
1
Sewer piap receiving
Nitrobenzene
< 2
13
124
t
12(0
86 t 16
wastes from various
Isophorone
40.5 t 32.0
270
67
t
30
< 5
process trees.
2.6-ONT
0.25 t 0.01
0.9
58
i
IS
53 i 19
2,4-PNT
< 0.4 t (2)
0.9
47
t
18
42 t 18
2
Brine sm?1c from
III trobetuene
< 1
13
67
t
40
130 t IS
holding tenk receiving
Isophorone
< SO
270
110
t
44
100 t 54
washings froa ships
2,6-ONT
< 0.1
0.9
12
t
•
4 » 2.S
carrying various
2.4-ONT
«1
0.9
4
t
4
S * 3
co—odUles.
3
Secondary sewage
Nltrobeniene
« 2
96
84
t
8
77 i 1
effluent (Coluabus,
Isophorone
120 t 6
560
87
»
S
7i l 2
Ohio).
2,6-ONT
< 0.1
7
87
t
8
88 * 2
•
2,4-ONT
< 0.1
•
79
t
6
72 t 4
4
Final effluent froa
Nitrobenzene
10 t 3
96
70
t
6
71 i 3
UNO! treatment systea
Isophorone
* 5
560
71
t
5
77 t 2
receiving wastes froa
2.6-ONT
< 0.1
17
78
t
1
78 t I
•
various chcaical plants.
2.4-Ml
< 0.1
18
70
t
2
70 t 2
S
Surface water
Nitrobenzene
« 5
90
70
t
6
71 * 3
(Olentangy River).
Isophorone
< 5
SO
71
*
S
77 t 2
2,6-ONTi
< o.os
5
78
t
2
78 t 1
2,4-DNT,
< o.os
5
70
t
2
70 t 1
6
Final effluent froa a
Nitrobenzene
27 t 10
100
3S
*
11
93 t 2
cheaical aanuf act tiring Isophorone
plant producing nltro- 2,6-DNT
beniene, 0-dlchloro- 2,4-ONT
benzene, O-nitrophenol,
aniline, end various oil
additives.
« 20
< 20
TO Average recovery rod standard deviation.
100
SO
100
11 t 2
8 t 2
SI
2S
TABLE 17. ANALYTICAL RESULTS FROM VARIOUS WASTEWATER SAMPLES
-------
rhh*
&
-------
s
to
u
*4
¦
*
*
«
(V
V
»
(•>
Plgura 9. Chroaatograas tor Saapla 1, tplltad «rteh
13 ppb of olerobaataaa, 270 ppb of liophoroc«,
and 0.88 ppb aach of 2,4- and 2,6-DNT, ualnf
BCD at 83*C (a), and 120'C (b).
39
-------
Sample 2 was a seawater sample from a holding tank receiving washings
from ships hauling various commodities, and this holding tank had been
determined to contain Isophorone on a previous sampling visit. Chromato-
graphs for this sample, unsplked and spiked, are shown 1n Figures 10 and 11
respectively.
This sample was difficult to analyze since substantial variations 1n
background level were observed for replicate extractions. None of the
components of Interest were detected 1n the sample, based on retention time
data. However the MDLs for the components 1n this sample were considerably
higher than that observed for distilled water blanks or other wastewater
samples.
Recoveries of the compounds spiked Into sample 2 were apparently satis-
factory for nitrobenzene and Isophorone, although precise quantitation was
not possible due to wide variations 1n background. However, recoveries of
2,4- and 2,6-DNT were very low. Based on the background levels 1n the
sample and the recoveries observed, the MDLs are estimated to be 2 ppb for
nitrobenzene, 50 ppb for Isophorone, and 1 ppb for 2,4-DNT, and 2,6-DNT.
Sample 3 was a secondary sewage effluent and was the first sample
processed using the modified clean up method (condition 4, Table 2). The
chromatograms for the unsplked and spiked samples are shown 1n Figures 12
and 13 respectively. The unsplked sample did not contain detectable levels
of 2,4- or 2,6-DNT or nitrobenzene, but did appear to contain about 100 ppb
of Isophorone, based on retention time data. The extraction efficiency and
storage stability were > 70% for all components as shown 1n Table 17. Based
on the background level observed for the unsplked sample MDLs are estimated
to be 2 ppb for nitrobenzene, 40 ppb for Isophorone, and 0.05 ppb for 2,4-
and 2,6-DNT.
The only disturbing aspect concerning the analysis of this sample 1s the
apparently high background for Isophorone. It 1s possible that due to this
rather low electron capture response, a small quantity of a highly electro-
capturing species might have been mistaken as a large quantity of Isophorone.
For this reason, FID was employed for all future isophorone determinations.
Sample 4 was a final effluent from a UNOX treatment system receiving
wastes from plants producing plastlclzers, butyl rubber, and olefins. This
was the first sample where FID, rather than ECD, was used to determine
nitrobenzene and Isophorone. This approach was taken since the ECD back-
ground from all the previous samples had limited the sensitivity for
Isophorone to about 40 ppb. Nitrobenzene was also determined by FID. To
measure it by ECD would require an additional chromatographic run. However,
1f one desired only to measure nitrobenzene, ECD probably would be most
effective, based on the background levels for the samples evaluated herein.
Chromatograms for sample 4 unspiked and spiked are shown in Figures 14 and
15, respectively. For 2,4- and 2,6-DNT as well as for Isophorone, no
detectable quantities were found, whereas a small quantity (about 10 ppb) of
nitrobenzene and Isophorone appeared to be present, based solely on reten-
tion time data. Recoveries from the spiked sample were > 70X for all
components both before and after seven days' storage at 4°C. Based on the
40
-------
rtfcrri
PI
#Vi
4
*
ft
9
(•)
Cb)
Fl«ur« 10. Chtxm*totT%mm for StapU 2, mmpikmd
uaiat CCD «c $3*C <«), «*d 170*C (b).
41
-------
N
V
-hfii
(«)
(b)
rigur* 11. Chtoatcofraaj for Swpl* 2, »ptk«4
* *lth 13 ppb sicrob«Qxaa*, 270 ppb Liophoroa«.
*ad 0.3# ppb of 2.4- ud 2,6~DN?f uatB| ECD
•t 85*C («) and 120*C (b).
42
-------
-| V_r^___/Aww"
<•>
(b)
Flgura 12. Chroaatograaa (er Snplt 3, uaaplkad,
ualag (CO at W'C (a) and ItS'C (b).
43
-------
0.
-------
{IsJ
OtnMatoirta. for Supl* 4, uaaplkM
ualo| fll> *t S5*C (*), awl EO «c
HS*C (b).
45
-------
I c
(«) Cb>
Flgura IS. Chroaacograa for S*Kpl« 4, lpllud with
9t ppb altrebaattoa, 360 ppb laephoroo*.
7 ppb of 2,6-NfT, and t ppb of 2,4-OtfT.
uais( FID «t SVC (a), *nd ECS it US'C (b).
46
-------
background level and recovery data the MDLs In this sample are estimated to
be .1 ppb for 2,4- and 2,6-DNT, and 5 ppb for nitrobenzene and Isophorone,
based on a five to one signal to noise ratio.
Sample 5 was a surface water sample and contained no detectable levels
of any of the four compounds. Chromatograms for the unsplked and spiked
samples are shown 1n Figures 16 and 17 respectively. Recoveries were > 70X
for all components both before and after storage for seven days at 4°C.
The estimated MOLs for this sample are 5 ppb for Isophorone and nitro-
benzene, and 0.05 ppb for 2,4- and 2,6-DNT.
Sample 6 was a final effluent from a chemicals plant producing a variety
of chemicals Including nitrobenzene, o-d1chlorobenzene, aniline, nltrophenol,
and various oil additives. For this sample nitrobenzene and Isophorone were
determined by FID and 2,4- and 2,6-DNT were determined by ECD. The overall
background level of this sample was rather high, and was not appreciably
reduced using the florlsll clean up procedure. A small but distinct peak
was observed 1n the unsplked extract corresponding to nitrobenzene, at
approximately the 40 ppb level. There also appeared to be about 5 ppb of
2,6-DNT present. "Although no distinct peaks were observed for isophorone or
2,4-DNT, large peaks in their vicinity limited their detection to about 20
ppb. Based on the analysis of the unsplked extract, the sample was spiked
with 100 ppb of nitrobenzene, 2,4-DNT, and Isophorone and 50 ppb of 2,6-DNT.
Results from the spiked samples before and after storage for seven days are
shown 1n Table 17 and chromatograms for spiked and unsplked samples are
shown 1n Figures 18 and 19. Recoveries were rather low and somewhat
variable, especially for the DNTs. The results were compounded somewhat by
the fact that the preserved and unpreserved (spiked) samples were taken from
different one gallon containers, which were later found to have different
levels of background Interferences. A large peak elutlng near Isophorone
prevented Its determination 1n the spiked samples. However, by standard
addition to the extract 1t was determined that the MDL for Isophorone would
be about 400 ppb. Similarly a large peak, perhaps a decomposition product
of one of the background components, Interfered with the determination
product of one of the background components, Interfered with the determina-
tion of 2,4-DNT 1n the preserved samples.
The clean up procedure did not Improve the problem with background
Interferences as 1s Illustrated by comparing Figure 19 to Figure 20. It was
noted that for some sample allquots, a yellow band was eluted from the
column with methylene chloride, the first fraction to be discarded. For
those allquots, no recovery of the DNTs was observed, whereas for sample
allquots 1n which the yellow color was not eluted, higher recoveries of the
DNTs were observed. This observation may be the result of overloading the
column, although this 1s not certain. This result does serve to illustrate
the need for each analyst to verify the elutlon profiles for the components
of Interest under his laboratory conditions, since variations can occur both
within and between various laboratories.
47
-------
(•>
(b)
rifur* 16. ChroMCOfraaa for SiapU J, uotplkcd.
u«in» fXO ac i5*C (•) tad ECO at 145*C (b).
48
-------
CM
U>
rifnr* 17. Chroaatofraaa tot S«**l« i, aplka4
with VI pi* of altroMairaa, SO ppb
of iaaftunu, aad J of 2,4-
aad Z.&-OKT. ualai no at tl'C (a),
aad tea at HJ'C (b).
49
-------
1
~f
ac
cc
¥~
»/*
i
(»)
Fisur* 18, Chrooftcogrfttts for S*opl« 6 unapiked
FID at 85*C (*> «ikI ECO at 145*C (b)
-------
~
•«
<»>
Cb)
Plgura 19. CbroMtogrw for S«.pl« 6 «plk«d with 100 ppb
of alcrobanzana, laophorona, and 2,6-DNT and
50 ppb of 2,4-DKT using riD ac 85*C (a) and
KB at 145'C (b).
51
-------
-------
SECTION 5
SUMMARY AND RECOMMENDATIONS
The results of this work are sunmarlzed below:
• The components of Interest are readily chromatographed on OV-17
using the ECD for 2,4- and 2,6-DNT and the FID Isophorone and
nitrobenzene.
0 The compounds are highly stable for at least 90 days when stored
1n either acetone or methanol.
• The compounds are readily extracted from waters at pH 7 with
methylene chloride.
• The compounds are relatively stable for at least seven days when
stored at pH 7 and 4°C, with or without chlorine present.
• Although the final method appeared to work very well for several
aqueous samples, some problems were encountered with certain
samples. Primarily, these problems were associated with high
background levels of Interfering components which limited the
sensitivity of the measurement. Low recoveries were noted for
several samples, the cause(s) of which are not known at this
time.
t Special attention must be paid to the florlsll clean up step 1n
order to assure that the compounds of Interest are elutlng 1n
the proper fraction.
On the basis of these results, 1t must be concluded that while the final
method, given 1n the Appendix, appears to be satisfactory for many sample
types, 1t must be validated for any particular application where 1t 1s to be
used. In particular, each analyst must ensure that the extraction efficiency
and clean up elutlon profiles for the sample are functioning properly. It
1s anticipated that for certain sample types, more efficient columns, for
example, glass capillary columns, and/or more specific detectors, such as
mass spectrometry, must be employed to achieve satisfactory quantitation for
these specific compounds 1n the pesence of much larger amounts of Inter-
ferences.
53
-------
REFERENCES
1. Leggett, D. C. Analytical Chemistry. 49 (6):880, 1977.
2. Hess, T. L., L. J. Guldry and S. D. Sibley. Bull. Env. Contamination
and Toxicology. 13(5):579-581, 1975.
3. Hoffsomner, 0. C. and J. M. Rosen. Bull. Env. Contamination and
Toxicology. 7(2):177-181, 1972.
4. Dalton, R. W., J. A. Kohlbeck and W. T. Bolleter. J. Chromatog.
50(3):219-227, 1970.
5. Gehrlng, 0. G. and J. E. Shirk. Analytical Chemistry. 39(11):1315-1318,
1967.
6. Parsons J. M. Analytical Chemistry. 33(13):1858-1859, 1961.
7. Hoffsommer, J. C. J. Chromatog. 51(2):243-251, 1970.
8. Stanford, T. B. The Determination of Tetryl and 2,3-, 2,4-, 2,5-, 2,6-,
3,4-, and 3,5-D1n1trotoluene Using High Performance Liquid Chromato-
graphy. Report for Contract Number DAMD-17-74-C-4123, U.S. Army Medical
Research and Development Command, Washington, O.C., 1977.
54
-------
APPENDIX
NITROAROMATICS AND ISOPHORONE
METHOD 609
!• Scope and Application
1.1 This method covers the determination of certain nltroaromatlcs and
Isophorone. The following parameters may be determined by this
method:
Parameter STORET No. CAS No.
2,4-D1n1troto1uene 34611 121-14-2
2,6-01n1troto1uene 34626 606-20-2
Isophorone 34408 78-59-1
Nitrobenzene 34447 98-95-3
1.2 This 1s a gas chromatographic (GC) method applicable to the deter-
mination of the compounds listed above 1n municipal and Industrial
discharges as provided under 40 CFR 136.1. When this method 1s
used to analyze unfamiliar samples for any or all of the compounds
above, compound Identifications should be supported by at least one
additional qualitative technique. This method describes analytical
conditions for a second gas chromatographic column that can be used
to confirm measurements made with the primary column. Method 625
provides gas chromatograph/mass spectrometer (GC/MS) conditions
appropriate for the qualitative and quantitative confirmation of
results for all of the parameters listed above, using the extract
produced by this method.
1.3 The method detection limit (MDL, defined 1n Section 14.1)0) for
each parameter 1s listed 1n Table 1. The MOL for a specific
wastewater may differ from those listed, depending upon the nature
of Interferences 1n the sample matrix.
1.4 The sample extraction and concentration steps 1n this method are
essentially the same as 1n Method 606, 608, 611 and 612. Thus, a
single sample may be extracted to measure all of the parameters
Included 1n the scope of each of these methods. When cleanup 1s
required the concentration levels must be high enough to permit
selection of allquots of the extract, as necessary, to apply ap-
propriate cleanup procedures. The analyst 1s allowed the latitude,
under Gas Chromatography (Section 12), to select chromatographic
conditions appropriate for the simultaneous measurement of combin-
ations of these parameters.
55
-------
1.5 Any modifications of this method, beyond those expressly permitted,
shall be considered as major modifications subject to application
and approval of alternate test procedures under 40 CFR 136.4 and
136.5.
1.6 This method 1s restricted to use by or under the supervision of
analysts experienced 1n the use of gas chromatography and 1n the
Interpretation of gas chromatograms. Each analyst must demonstrate
the ability to generate acceptable results with this method using
the procedure described 1n Section 8.2.
2. Suirmary of Method
2.1 A measured volume of sample, approximately 1-1 Iter, 1s extracted
with methylene chloride using separatory funnel techniques. The
extract 1s dried and exchanged to hexane during concentration to
1.0 mL by evaporation. Isophorone and nitrobenzene are measured by
flame Ionization gas chromatography (FIDGCl. The dlnltrotoluenes
are measured by electron capture GC (ECGC)").
2.2 The method provides a F1or1s11 chromatographic cleanup procedure to
aid 1n the elimination of Interferences that may be encountered.
3. Interferences
3.1 Method Interferences may be caused by contaminants in solvents,
reagents, glassware, and other sample processing hardware that lead
to discrete artifacts and/or elevated baselines 1n gas chromato-
grams. All of these materials must be routinely demonstrated to be
free from Interferences under the conditions of the analysis by
running laboratory reagent blanks as described 1n Section 8.5.
3.1.1 Glassware must be scrupulously cleaned(3). Clean all
glassware as soon as possible after use by rinsing with the
last solvent used 1n 1t. This should be followed by deter-
gent washing with hot water, and rinses with tap water and
distilled water. It should then be drained dry, and heated
In a muffle furnace at 400°C for 15 to 30 minutes. Some
thermally stable materials, such as PCBs, may not be elimin-
ated by this treatment. Solvent rinses with acetone and
pesticide quality hexane may be substituted for the muffle
furnace heating. Thorough rinsing with such solvents
usually eliminates PCB Interferences. Volumetric ware
should not be heated 1n a muffle furnace. After drying and
cooling, glassware should be sealed and stored 1n a clean
environment to prevent any accumulation of dust or other
contaminants. Store Inverted or capped with aluminum foil.
3.1.2 The use of high purity reagents and solvents helps to mini-
mize interference problems. Purification of solvents by
distillation in all-glass systems may be required.
56
-------
3.2 Matrix Interferences may be caused by contaminants that are
coextracted from the sample. The extent of matrix Interferences
will vary considerably from source to source, depending upon the
nature and diversity of the Industrial complex or municipality
being sampled. The cleanup procedures 1n Section 11 can be used to
overcome many of these Interferences, but unique samples may
require additional clean-up approaches to achieve the MDL listed 1n
Table 1.
4. Safety
The toxicity or carcinogenicity of each reagent used In this method has
not been precisely defined; however, each chemical compound should be
treated as a potential health hazard. From this viewpoint, exposure to
these chemicals must be reduced to the lowest possible level by whatever
means available. The laboratory 1s responsible for maintaining a
current awareness file of OSHA regulations regarding the safe handling
of the chemicals specified 1n this method. A reference file of material
data handling sheets should also be made available to all personnel
involved 1n the chemical analysis. Additional references to laboratory
safety are available and have been 1dent1f1ed(4"®) for the Information
of the analyst.
5. Apparatus and Materials
5.1 Sampling equipment, for discrete or composite sampling.
5.1.1 Grab sample bottle - Amber glass, 1-1 Iter or 1-quart volume,
fitted with screw caps lined tflth Teflon. Foil may be sub-
stituted for Teflon 1f the sample is not corrosive. If
amber bottles are not available, protect samples from
light. The container must be washed, rinsed with acetone or
methylene chloride, and dried before use to minimize contam-
ination.
5.1.2 Automatic sampler (optional) - Must Incorporate glass sample
containers for the collection of a minimum of 250 ml.
Sample containers must be kept refrigerated at 4°C and
protected from light during compositing. If the sampler
uses a peristaltic pump, a minimum length of compressible
silicone rubber tubing may be used. Before use, however,
the compressible tubing should be thoroughly rinsed with
methanol, followed by repeated rinsings with distilled water
to minimize the potential for contamination of the sample.
An Integrating flow meter 1s required to collect flow
proportional composites.
5.2 Glassware (All specifications are suggested. Catalog numbers are
included for Illustration only).
5.2.1 Separatory funnel - 2000-mL, with Teflon stopcock.
57
-------
5.2.2 Drying column - Chromatographic column, 400 mm long x 19 mm
ID with coarse frit.
5.2.3 Concentrator tube, Kuderna-Danlsh - 10-ml, graduated (Kontes
K-570050-1025 or equivalent). Calibration must be checked
at the volumes employed In the test. Ground glass stopper
1s used to prevent evaporation of extracts.
5.2.4 Evaporative flask, Kuderna-Danlsh - 500-ml (Kontes
K-570001-0500 or equivalent). Attach to concentrator tube
with springs.
5.2.5 Snyder colimm, Kuderna-Danlsh - Three-ball macro (Kontes
K-503000-0121 or equivalent).
5.2.6 Snyder column, Kuderna-Oanlsh - Two-ball micro (Kontes
K-569001-0219 or equivalent).
5.2.7 Vials - Amber glass, 10 to 15-ml capacity, with Teflon lined
screw-cap.
5.2.8 Chromatographic column - 100 mm long x 10 mm ID, with Teflon
stopcock.
5.3 Boiling chips - approximately 10/40 mesh. Heat to 400°C for 30
m1n. or Soxhlet extract with methylene chloride.
5.4 Water bath - Heated, with concentric ring cover, capable of
temperature control (± 2°C). The bath should be used 1n a hood.
5.5 Balance - Analytical, capable of accurately weighing 0.0001 g.
5.6 Gas chromatograph - Analytical system complete with gas chromato-
graph suitable for on-column Injection and all required accessories
Including syringes, analytical columns, gases, detector, and
strip-chart recorder. A data system 1s reconmended for measuring
peak areas.
5.6.1 Column 1 - 1.2 m (4 ft.) long x 2 mm or 4 mm ID, Pyrex
glass, packed with Gas-Chrom Q (80/100 mesh) coated with
1.95% QF-1/1.5X 0V-17. This column was used to develop the
method performance statements given 1n Section 14. Guide-
lines for the use of alternate column packings are provided
1n Section 12.1.
5.6.2 Column 2 - 3.0 m (10 ft.) x 2 or 4 mm ID, Pyrex glass,
packed with Gas-Chrom Q (80/100 mesh) coated with 3% 0V-101.
5.6.3 Detector - Flame ionization. This detector has proven
effective in the analysis of wastewaters for isophorone and
nitrobenzene, and was used to develop the method performance
58
-------
statements 1n Section 14. Guidelines for the use of alter-
nate detectors are provided 1n Section 12.1.
5.6.4 Detector - Electron capture. This detector has proven
effective 1n the analysis of wastewaters for the dlnltrotol-
uenes, and was used to develop the method performance state-
ments 1n Section 14. Guidelines for the use of alternate
detectors are provided 1n Section 12.1
6. Reagents
6.1 Reagent water - Reagent water 1s defined as a water 1n which an
interferent 1s not observed at the method detection 11m1t of each
parameter of Interest.
6.2 Sodium hydroxide solution (10 N) - (ACS) Dissolve 40g NaOH 1n
reagent water and dilute to 100 ml.
6.3 Sulfuric acid solution (1+1) - (ACS) Slowly, add 50 mL H2SO4
(sp. gr. 1.84) to 50 ml of reagent water.
6.4 Acetone, hexane, methanol, methylene chloride, - Pesticide quality
or equivalent.
6.5 Sodium sulfate - (ACS) Granular, anhydrous. Purify by heating at
400°C for 4 hrs. 1n a shallow tray.
6.6 F1or1s1l - PR grade (60/100 mesh), purchase activated at 1250°F
and store 1n g]ass containers with glass stoppers or fo11-11ned
screw caps. Before use, activate each batch overnight at 200°C
1n glass containers loosely covered with foil.
6.7 Stock standard solutions (1.00 tig/wL) - Stock standard solutions
can be prepared from pure standard materials or purchased as certi-
fied solutions.
6.7.1 Prepare stock standard solutions by accurately weighing
about 0.0100 grams of pure material. Dissolve the material
1n pesticide quality hexane, dilute to volume 1n a 10-mL
volumetric flask. Larger volumes can be used at the conven-
ience of the analyst. If compound purity 1s certified at
96% or greater, the weight can be used without correction to
calculate the concentration of the stock standard. Commer-
cially prepared stock standards can be used at any concen-
tration 1f they are certified by the manufacturer or by an
Independent source.
6.7.2 Transfer the stock standard solutions Into Teflon-sealed
screw-cap bottles. Store at 4°C and protect from light.
Stock standard solutions should be checked frequently for
signs of degradation or evaporation, especially just prior
to. preparing calibration standards from them. Quality
59
-------
control check standards that can be used to determine the
accuracy of calibration standards, will be available from
the U. S. Environmental Protection Agency, Environmental
Monitoring and Systems Laboratory, in Cincinnati, Ohio.
6.7.3 Stock standard solutions must be replaced after six months,
or sooner 1f comparison with check standards Indicate a
problem.
7. Calibration
7.1 Establish gas chromatographic operating conditions to produce
resolution of the parameters equivalent to those Indicated 1n Table
1. The gas chromatographic system may be calibrated using the
external standard technique (Section 7.2) or the Internal standard
technique (Section 7.3).
7.2 External standard calibration procedure:
7.2.1 Prepare calibration standards at a minimum of three concen-
tration levels for each parameter of Interest by adding
volumes of one or more stock standards to a volumetric flask
and diluting to volume with hexane. One of the external
standards should be at a concentration near, but above, the
method detection limit and the other concentrations should
correspond to the expected range of concentrations found 1n
real samples or should define the working range of the
detector.
7.2.2 Using injections of 2 to 5 iiL of each calibration standard,
tabulate peak height or area responses against the mass
Injected. The results can be used to prepare a calibration
curve for each compound. Alternatively, 1f the ratio of
response to amount Injected (calibration factor) 1s a
constant over the working range (< 10% relative standard
deviation, RSO), linearity through the origin can be assumed
and the average ratio or calibration factor can be used 1n
place of a calibration curve.
7.2.3 The working calibration curve or calibration factor must be
verified on each working day by the measurement of one or
more calibration standards. If the response for any para-
meter varies from the predicted response by more than ±10%,
the test must be repeated using a fresh calibration stand-
ard. Alternatively, a new calibration curve or calibration
factor must be prepared for that compound.
7.3 Internal standard calibration procedure. To use this approach, the
analyst must select one or more internal standards that are similar
1n analytical behavior to the compounds of Interest. The analyst
must further demonstrate that the measurement of the Internal
standard is not affected by method or matrix interferences.
60
-------
Because of these limitations, no Internal standard can be suggested
that 1s applicable to all samples.
7.3.1 Prepare calibration standards at a minimum of three concen-
tration levels for each parameter of Interest by adding
volumes of one or more stock standards to a volumetric
flask. To each calibration standard, add a known constant
amount of one or more Internal standards, and dilute to
volume with hexane. One of the standards should be at a
concentration near, but above, the method detection 11m1t
and the other concentrations should correspond to the
expected range of concentrations found 1n real samples or
should define the working range of the detector.
7.3.2 Using Injections of 2 to 5 ul of each calibration standard,
tabulate peak height or area responses against concentration
for each compound and Internal standard, and calculate
response factors (RF) for each compound using equation 1.
Eq. 1 RF - (AsC1s)/(A1s Cs)
where:
As - Response for the parameter to be measured.
A^s ¦ Response for the Internal standard.
Cis - Concentration of the Internal standard, (ug/L).
Cs - Concentration of the parameter to be measured,
(yg/L).
If the RF value over the working range 1s a constant (< 10*
RSD), the RF can be assumed to be Invariant and the average
RF can be used for calculations. Alternatively, the results
can be used to plot a calibration curve of response ratios,
As/Ais, vs. RF.
7.3.3 The working calibration curve or RF must be verified on each
working day by the measurement of one or more calibration
standards. If the response for any parameter varies from
the predicted response by more than ±10X, the test must be
repeated using a fresh calibration standard. Alternatively,
a new calibration curve must be prepared for that compound.
7.4 Before using any cleanup procedure, the analyst must process a
series of calibration standards through the procedure to validate
elutlon patterns and the absence of Interferences from the reagents.
8. Quality Control
8.1 Each laboratory that uses this method 1s required to operate a
formal quality control program. The minimum requirements of this
program consist of an Initial demonstration of laboratory capa-
bility and the analysis of spiked samples as a continuing check
61
-------
on performance. The laboratory 1s required to maintain performance
records to define the quality of data that 1s generated. After
January 1, 1982, ongoing performance checks must be compared with
established performance criteria to determine 1f the results of
analyses are within accuracy and precision limits expected of the
method.
8.1.1 Before performing any analyses, the analyst must demonstrate
the ability to generate acceptable accuracy and precision
with this method. This ability 1s established as described
1n Section 8.2.
8.1.2 In recognition of the rapid advances that are occurring In
chromatography, the analyst 1s permitted certain options to
Improve the separations or lower the cost of measurements.
Each time such modifications are made to the method, the
analyst 1s required to repeat the procedure 1n Section 8.2.
8.1.3 The laboratory must spike and analyze a minimum of 10X of
all samples to monitor continuing laboratory performance.
This procedure 1s described in Section 8.4.
To establish the ability to generate acceptable accuracy and pre-
cision, the analyst must perform the following operations.
8.2.1 Select a representative spike concentration for each
compound to be measured. Using stock standards, prepare a
quality control check sample concentrate 1n acetone 1000
times more concentrated than the selected concentrations.
Quality control check sample concentrates, appropriate for
use with this method, will be available from the U.S.
Environmental Protection Agency, Environmental Monitoring
and Support Laboratory, Cincinnati, Ohio 45268.
8.2.2 Using a plpet, add 1.00 ml of the check sample concentrate
to each of a minimum of four 1000-mL aliquots of reagent
water. A representative wastewater may be used 1n place of
the reagent water, but one or more additional aliquots must
be analyzed to determine background levels and the spike
level must exceed twice the background level for the test to
be valid. Analyze the aliquots according to the method
beginning 1n Section 10.
8.2.3 Calculate the average percent recovery, (R), and the stand-
ard deviation of the percent recovery (s), for the results.
Wastewater background corrections must be made before R and
s calculations are performed.
8.2.4 Using the appropriate data from Table 2, determine the
recovery and single operator precision expected for the
method, and compare these results to the values calculated
62
-------
1n Section 8.2.3. If the data are not comparable, the
analyst must review potential problem areas and repeat the
test.
8.2.5 After January 1, 1982, the values for R and s must meet
rigid method performance criteria provided by the U.S. EPA,
Environmental Monitoring and Support Laboratory, Cincinnati,
Ohio, before any samples may be analyzed.
8.3 The analyst must calculate method performance criteria and define
the performance of the laboratory for each spike concentration and
parameter being measured.
8.3.1 Calculate upper and lower control limits for method perfor-
mance:
Upper Control Limit (UCL) ¦ R + 3 s
Lower Control Limit (LCL) ¦ R - 3 s
where R and s "are calculated as in Section 8.2.3.
The UCL and LCL can be used to construct control charts(?)
that are useful 1n observing trends 1n performance. After
January 1, 1982, the control limits above must be replaced
by method performance criteria provided by the U.S.
Environmental Protection Agency.
8.3.2 The laboratory must develop and maintain separate accuracy
statements of laboratory performance for wastewater samples.
An accuracy statement for the method 1s defined as R ± s.
The accuracy statement should be developed by the analysis
of four allquots of wastewater as described In Section
8.2.2, followed by the calculation of R and s. Alternately,
the analyst may use four wastewater data points gathered
through the requirement for continuing quality control 1n
Section 8.4. The accuracy statements should be updated
regularly(').
8.4 The laboratory 1s required to collect a portion of their samples 1n
duplicate to monitor spike recoveries. The frequency of spiked
sample analysis must be at least 10X of all samples or one sample
per month, whichever 1s greater. One aliquot of the sample must be
spiked and analyzed as described 1n Section 8.2. If the recovery
for a particular parameter does not fall within the control limits
for method performance, the results reported for that parameter 1n
all samples processed as part of the same set must be qualified as
described 1n Section 13.3. The laboratory should monitor the
frequency of data so qualified to ensure that 1t remains at or
below 5X.
8.5 Before processing any samples, the analyst should demonstrate
.through the analysis of a one-liter aliquot of reagent water, that
63
-------
all glassware and reagents Interferences are under control. Each
time a set of samples Is extracted or there 1s a change 1n
reagents, a laboratory reagent blank should be processed as a safe-
guard against laboratory contamination.
8.6 It 1s reconmended that the laboratory adopt additional quality
assurance practices for use with this method. The specific prac-
tices that are most productive depend upon the needs of the labora-
tory and the nature of the samples. Field duplicates may be
analyzed to monitor the precision of the sampling technique. When
doubt exists over the Identification of a peak on the chromatogram,
confirmatory techniques such as gas chromatography with a dissimi-
lar column, specific element detector, or mass spectrometer must be
used. Whenever possible, the laboratory should perform analysis of
standard reference materials and participate In relevant perfor-
mance evaluation studies.
9. Sample Collection, Preservation, and Handling
9.1 Grab samples must beficol1ected in glass containers. Conventional
sampling practices '°) should be followed, except that the bottle
must not be prerlnsed with sample before collection. Composite
samples should be collected 1n refrigerated glass containers 1n
accordance with the requirements of the program. Automatic samp-
ling equipment must be as free as possible of Tygon tubing and
other potential sources of contamination.
9.2 The samples must be 1ced or refrigerated at 4°C from the time of
collection until extraction.
9.3 All samples must be extracted within 7 days and completely analyzed
within 40 days of extraction(2).
10. Sample Extraction
10.1 Mark the water meniscus on the side of the sample bottle for later
determination of sample volume. Pour the entire sample Into a
two-Hter separatory funnel. Check the pH of the sample with
wide-range pH paper and adjust to within the range of 5 to 9 with
diluted sodium hydroxide or sulfuric acid.
10.2 Add 60 mL of methylene chloride to the sample bottle, seal, and
shake for 30 seconds to rinse the Inner surface. Transfer the
solvent to the separatory funnel and extract the sample by shaking
the funnel for two minutes with periodic venting to release excess
pressure. Allow the organic layer to separate from the water phase
for a minimum of ten minutes. If the emulsion Interface between
layers is more than one-third the volume of the solvent layer, the
analyst must employ mechanical techniques to complete the phase
separation. The optimum technique depends upon the sample, but may
include stirring, filtration of the emulsion through glass wool,
64
-------
centrlfugatlon, or other physical methods. Collect the methylene
chloride extract 1n a 250-mL Erlenmeyer flask.
10.3 Add a second 60-mL volume of methylene chloride to the sample
bottle and repeat the extraction procedure a second time, combining
the extracts 1n the Erlenmeyer flask. Perform a third extraction
„ 1n the same manner. . . ...
10.4 Assemble a Kuderna-Danlsh (K-D) concentrator by attaching a 10-mL
concentrator tube to a 500-mL evaporative flask. Other concen-
tration devices or techniques may be used 1n place of the
Kuderna-Danlsh 1f the requirements of Section 8.2 are met.
10.5 Pour the combined extract through a drying column containing about
10 cm of anhydrous sodium sulfate, and collect the extract 1n the
K-D concentrator. Rinse the Erlenmeyer flask and column with 20 to
30 mL of methylene chloride to complete the quantitative transfer.
10.6 Sections 10.7 and 10.8 describe a procedure for exchanging the
"methylene chloride solvent to hexane while concentrating the
extract volume to 1.0 mL. When 1t 1s not necessary to achieve the
MDL 1n Table 2, the solvent exchange may be made by the addition of
50 mL of hexane and concentration to 10 mL as described 1n Method
606, Section 10.7.
10.7 Add 1 or 2 clean boiling chips to the evaporative flask and attach
a three-ball Snyder column. Prewet the Snyder column by adding
about 1 mL methylene chloride to the top. Place the K-D apparatus
on a hot water bath (60 to 65°C) so that the concentrator tube 1s
partially Imnersed 1n the hot water, and the entire lower rounded
surface of the flask 1s bathed with hot vapor. Adjust the vertical
position of the apparatus and the water temperature as required to
complete the concentration 1n 15 to 20 minutes. At the proper rate
of distillation the balls of the column will actively chatter but
the chambers will not flood with condensed solvent. When the
apparent volume of liquid reaches 1 mL, remove the K-D apparatus
and allow 1t to drain and cool for at least 10 minutes. Remove the
Snyder column and rinse the flask and Its lower joint Into the
concentrator tube with 1 to 2 mL of methylene chloride. A 5-mL
syringe 1s recoimtended for this operation.
10.8 Add 1 to 2 mL of hexane to the concentrator tube, and a clean
boiling chip. Attach a two-ball micro-Snyder column. Prewet the
micro-Snyder column by adding about 0.5 mL of hexane to the top.
Place this micro K-D apparatus on a water bath (60 to 65°C) so
that the concentrator tube 1s partially Immersed in the hot water.
Adjust the vertical position of the apparatus and water temperature
as required to complete the concentration 1n 5 to 10 minutes. At
the proper rate of distillation the balls will actively chatter but
the chambers will not flood. When the apparent volume of liquid
reaches 0.5 mL, remove the K-D apparatus and allow 1t to drain for
at least 10 minutes while cooling. Remove the mlcro-Snyder column
65
-------
and rinse Its lower joint Into the concentrator tube with a small
volume of hexane. Adjust the final volume to 1.0 mL and stopper
the concentrator tube and store refrigerated 1f further processing
will not be performed limedlately. Unless the sample 1s known to
require cleanup, proceed with gas chromatographic analysis.
10.9 Determine the original sample volume by refilling the sample bottle
to the mark with water and measuring the volume 1n a 1000-mL
graduated cylinder. Record the sample volume to the nearest 5 mL.
11. Cleanup and Separation
11.1 Cleanup procedures may not be necessary for a relatively clean
sample matrix. The cleanup procedure recommended 1n this method
has been used for the analysis of various clean waters and Indus-
trial effluents. If particular clrcunstances demand the use of an
alternative cleanup procedure, the analyst must demonstrate that
the recovery of each compound of interest 1s no less than 85%.
11.2 Prepare a slurry of lOg of activated Florisll 1n methylene chloride
1n hexane (1 + 9)(V/V). Use 1t to pack a 10-mm ID chromatography
column, gently tapping the column to settle the F1or1s11. Add 1 cm
of anhydrous sodium sulfate to the top of the F1or1s1l.
11.2.1 Just prior to exposure of the sodium sulfate layer to the
air transfer the 1 mL of sample extract onto the column
using an additional 2 mL of hexane to complete the transfer.
11.2.2 Just prior to exposure of the sodium sulfate layer to the
air, add 30 mL of methylene chloride 1n hexane (1 + 9)(V/V)
and continue the elutlon of the column. Elutlon of the
column should be at a rate of about 2 mL per minute. Dis-
card the eluate from this fraction.
11.2.3 Next elute the column with 30 mL of acetone/methylene
chloride (1 + 9)(V/V) Into a 500-mL K-D flask equipped with
a 10-mL concentrator tube. Concentrate the collected
fraction by the K-D technique prescribed 1n Sections 10.6,
10.7, and 10.8 Including the solvent exchange to 1 mL of
hexane. This fraction should contain the nltroaromatlcs and
Isophorone.
11.2.4 Analyze by gas chromatography.
12. Gas Chromatography
12.1 Isophorone and nitrobenzene are analyzed by Injection of a portion
of the extract Into an FIDGC. The dlnltrotoluenes are analyzed by
a separate Injection Into an ECGC. Table 1 summarizes some recom-
mended gas chromatographic column materials and operating condi-
tions for the Instruments. This Table includes retention times and
MDL obtained under these conditions. Other packed columns,
66
-------
chromatographic conditions, or detectors may be used 1f the
requirements of Section 8.2 are met. Capillary (open-tubular)
columns may also be used If the relative standard deviations of
responses for replicate Injections are demonstrated to be less than
6X and the requirements of Section 8.2 are met.
12.2 Calibrate the system dally as described 1n Section 7.
12.3 If the Internal standard approach 1s being used, the analyst must
not add the Internal standard to sample extracts until 1imied1ate1y
before Injection Into the Instrunent. Mix thoroughly.
12.4 Inject 2 to 5 tiL of the sample extract using the solvent-flush
technique i9'. Smaller (1.0 yL) volumes can be Injected 1f
automatic devices are employed. Record the volume Injected to the
nearest 0.05 uL, and the resulting peak size 1n area or peak height
units.
12.5 The width of the retention time window used to make Identifications
should be based upon measurements of actual retention time varia-
tions of standards over the course of a day. Three times the
standard deviation of a retention time for a compound can be used
to calculate a suggested window size; however, the experience of
the analyst should weigh heavily 1n the Interpretation of chroma-
tograms.
12.6 If the response for the peak exceeds the working range of the
system, dilute the extract and reanalyze.-
12.7 If the measurement of the peak response 1s prevented by the
presence of Interferences, further cleanup 1s required.
Calculations
13.1 Determine the concentration of Individual compounds 1n the sample.
13.1.1 If the external standard calibration procedure 1s used,
calculate the amount of material Injected from the peak
response using the calibration curve or calibration factor
1n Section 7.2.2. The concentration 1n the sample can be
calculated from equation 2:
(A)(V )
Eq. 2. Concentration, yg/L ¦ —^ j—
where:
A - Amount of material Injected, 1n nanograms.
Vi ¦ Volume of extract Injected (pL).
Vt ¦ Volume of total extract (uL).
Vs ¦ Volume of water extracted (mL).
13.1.2 If the Internal standard calibration procedure was used,
67
-------
calculate the concentration 1n the sample using the response
factor (RF) determined 1n Section 7.3.2 and equation 3.
(AJ(IJ
Eq. 3. Concentration, ug/L ¦ (A1 )(RF)(V )
where:
As - Response for the parameter to be measured.
A-fs- Response for the Internal standard.
I« - Amount of Internal standard added to each extract
(MS).
V0 > Volume of water extracted, 1n liters.
13.2 Report results 1n micrograms per liter without correction for
recovery data. When duplicate and spiked samples are analyzed,
report all data obtained with the sample results.
13.3 For samples processed as part of a set where the laboratory spiked
sample recovery falls outside of the control limits 1n Section 8.3,
data for the affected parameters must be labeled as suspect.
14. Method Performance
14.1 Method detection limits - The method detection Hm1t (MDL) 1s
defined as the minimum concentration of a substance that can be
measured and reported with 99* confidence that the value 1s above
zero"). The MDL concentrations listed 1n Table 1 were obtained
using reagent water*'0). Similar results were achieved using
representative wastewaters.
14.2 This method has been tested for linearity of recovery from reagent
water and has been demonstrated to be applicable over the concen-
tration range from 7 x MDL to 1000 x MDLllO).
14.3 In a single laboratory (Battelle, Columbus Laboratories), using
spiked wastewater samples, the average recoveries presented 1n
Table 2 were obtained"). Each spiked sample was analyzed 1n
triplicate on two separate days. The standard deviation of the
percent recovery 1s also Included 1n Table 2.
14.4 The Environmental Protection Agency 1s 1n the process of conducting
an Interlaboratory method study to fully define the performance of
this method.
References
1. "Methods for Organic Chemical Analysis of Water and Wastes by GC, HPLC,
and GC/MS," U.S. EPA, Environmental Monitoring and Support Laboratory,
Cincinnati, Ohio 45268, July 1981.
68
-------
2. "Determ1nat1onof Nltroaromatlcs and Isophorone 1n Industrial and
Municipal Wastewaters." Report for EPA Contract No. 68-03-2624 (In
preparation).
3. ASTM Annual Book of Standards, Part 31, 0 3694, "Standard Practice for
Preparation of Sample Containers and for Preservation," American Society
for Testing and Materials, Philadelphia, PA, p. 679, 1980.
4. "Carcinogens - Working With Carcinogens," Department of Health,
Education, and Welfare, Public Health Service, Center for Disease
Control, National Institute for Occupational Safety and Health,
Publication No. 77-206, Aug. 1977
5. "OSHA Safety and Health Standards, General Industry," (29CFR1910),
Occupational Safety and Health Administration, OSHA 2206, (Revised,
January 1976).
6. "Safety 1n Academic Chemistry Laboratories," American Chemical Society
Publication, Committee on Chemical Safety, 3rd Edition, 1979.
7. "Handbook of Analytical Quality Control 1n Water and Wastewater
Laboratories," EPA-600/4-79-019, U.S. Environmental Monitoring and
Support Laboratory, Cincinnati, Ohio 45268, March 1979.
8. ASTM Annual Book of Standards, Part 31, D 3370, "Standard Practice for
Sampling Water," American Society for Testing and Materials,
Philadelphia, PA. p. 76, 1980.
9. Burke, J. A., "Gas Chromatography for Pesticide Residue Analysis; Some
Practical Aspects," Journal of the Association of Official Analytical
Chemists. 48, 1037 (T355T:
10. "Determination of Method Detection Limit and Analytical Curve for EPA
Method 609 - Nltroaromatlcs and Isophorone," special letter report for
EPA Contract 68-03-2624. Environmental Monitoring and Support
Laboratory - Cincinnati, Ohio 45268.
69
-------
table 1
CHROMATOGRAPHIC CONDITIONS AND METHOO DETECTION LIMITS
Retention Time Method
(m1n.) Detection Limit (yg/L)
Parameter Column 1 Column 2 EC FID
Nitrobenzene
3.31
4.31
13.7
3.6
2,6-D1n1trotoluene
3.52
4.75
0.01
-
Isophorone
4.49
5.72
15.7
5.7
2,4-D1n1trotoluene
5.35
6.54
0.02
•
Column 1 conditions - Gas-Chrom Q (80/100 mesh) coated with 1.95% QF-1/1.5X
OV-17 packed 1n a pyrex glass column 1.2 m (4 ft) long x 2 irm or 4 imi ID.
Nitrogen carrier gas at a flow rate of 44 mL/m1n was used when determining
Isophorone and nitrobenzene by FID. The Column temperature was Isothermal
at 85°C. Methane (10%)/Argon (90*) carrier gas at flow rate of 44 mL/m1n
was used when determining the dlnltrotoluenes by ECGC. The column tempera-
ture was Isothermal at 145°C.
Column 2 conditions - Gas-Chrom Q (80/100 mesh) coated with 3% 0V-101 packed
1n a pyrex glass column 3.0 m (10 ft) long x 2 rrm or 4 mm ID. Nitrogen
carrier gas at a flow rate of 44 mL/m1n was used when determining Isophorone
and nitrobenzene by FID. The column temperature was Isothermal at 100°C.
Methane (10%)/Argon (90%) carrier gas flow rate of 44 mL/m1n was used when
determining the dlnltroltoluenes by ECGE. The column temperature was
Isothermal, 150°C.
A 2 run ID column was used with the FIDGC and a 4 mm ID column was used with
the ECGC.
70
-------
TABLE 2
SINGLE OPERATOR ACCURACY AND PRECISION
Parameter
Average
Percent
Recovery
Standard
Deviation
X
Spike
Range
(yq/L)
Number
of
Analyses
Matrix
Types
2,4-D1n1troto1uene
63
3.1
5-100
21
4
2,6-D1n1trotoluene
66
3.2
5-50
24
4
Isophorone
73
4.6
50-60
21
4
Nitrobenzene
71
5.9
90-100
24
4
71
-------
TECHNICAL RETORT DATA
(flmm r—d Imaructiam ai thi rmrtt otfort compiictogj
1. RSPORT NO. a.
EPA-600/4-82-024 Research Reoort
a. ricipisnt* accession no.
PBS2 20839 8
«. TITLi ANOSUSTITI.1
Determination of N1troaromat1c Compounds and Isophorone
.In Industrial and Municipal Wastewaters
S. RSPORT DAT!
March 1982
PSRPORMINO ORGANIZATION COOC
i.AufMdftii!
Kenneth H. Shafer
S. PSRPORMINO ORGANIZATION RSPORT NO.
i. psrpormino organization nams ano aoorsss
Battelle - Columbus Laboratories
505 King Avenue
Colunbus, Ohio 43201
10. PROGRAM ILIMSNT NO.
1BD612
68-03-2624
13. SPONSORIMO AOINev NAM! AMD AOORSSS
Environmental Monitoring and Support Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
IS. TVPC OP RSPORT ANO PSRIOO COVSRIO
Final
14. SPONSORING AOSNCY COOS
EPA/600/06
IS. SUPPLSMSNTARY NOTSS
A method was developed for the determination of nitrobenzene, 2,4- and
2,6-d1n1troto1uene (DNT), and Isophorone 1n wastewaters. The methods development
program consisted of: a literature review, determination of the stability of the
confounds 1n organic solutions, determination of extraction efficiency for each
compound from water using two organic solvents, determination of storage stability
of each compound 1n water, and evaluation of various clean up techniques.
The final method was applied to several representative wastewaters spiked
with the confounds at appropriate levels, as well as to surface water, in order to
determine precision and accuracy of the method. For a wastewater sample spiked
with 96 ppb of nitrobenzene, 560 ppb of Isophorone and 8 ppb of 2,4-ONT, and 7 ppb
of 2,6-DNT, recoveries were 70 t 6* for nitrobenzene, 71 ± 5% for Isophorone, 70 t
2% for 2,4-DNT, and 78 t 1% for 2,6-ONT. Minimum detectable levels 1n this waste-
water are estimated to be 5 ppb for nitrobenzene, 5 ppb for Isophorone, and 0.05
ppb for 2,4- and 2,6-DNT. However, caution must be used in presuming these MDLs
are accurate for other wastewaters since several wastewaters were encountered
which exhibited high levels of interfering compounds. For some complex waste-
waters 1t may be necessary to apply additional clean up procedures or more selec-
tive analytical detection systems in order to achieve these levels of sensitivity.
17. KIV WORM AND DOCVIM1NT ANALYSIS
i. oucriptors
b.lD«NTIPI«RS/OP*N (NDSO TIRMS
c. COSATi FM/Graup
IS. OlSTmSuTIO* ITATIMCNT
Release to public
It. StCURlTY CLAM mil Alport)
Unclassified
31. NO. OP PAOIS
78
». SiCURITY CLASS
Unclassified
33. PRICC
I PA Pmm 2224-1 (R... *~T7)
------- |