EPA-800/1-77-829
May 1977
Environmental Healtii Effects Research Series
EXTENSION OF MULTI-RESIDUE METHODOLOGY:
I. Determining Multiclass Pesticide
Residues in Soil by
Gas Chromatography
II. Dynamic Fluorogenic Labelling
Detector for Carbamates
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office.of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL HEALTH EFFECTS RE-
SEARCH series. This series describes projects and studies relating to the toler-
ances of man for unhealthful substances or conditions. This work is generally
assessed from a medical viewpoint, including physiological or psychological
studies. In addition to toxicology and other medical specialities, study areas in-
clude biomedical instrumentation and health research techniques utilizing ani-
mals — but always with intended application to human health measures.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/1-77-029
May 1977
EXTENSION OF MULTI-RESIDUE METHODOLOGY
I. DETERMINING MULTICLASS PESTICIDE RESIDUES
IN SOIL BY GAS CHROMATOGRAPHY
II. DYNAMIC FLUOROGENIC LABELLING DETECTOR FOR
CARBAMATES
by
H. Anson Moye, Sujit Witkonton and Gordon Cash
University of Florida, IFAS. Food Science Department
Pesticide Research Laboratory, Gainesville, FL 32611
Contract No.
68-02-1706
Project Officer
Robert F. Moseman
Environmental Toxicology Division
Health Effects Research Laboratory
Research Triangle Park, North Carolina 27711
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
-------
DISCLAIMER
This report has been reviewed by the Health Effects Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
11
-------
FOREWORD
The many benefits of our modern, developing, industrial
society are accompanied by certain hazards. Careful assess-
ment of the relative risk of existing and new man-made
environmental hazards is necessary for the establishment of
sound regulatory policy. These regulations serve to enhance
ttye quality of our environment in order to promote the
public health and welfare and the productive capacity of our
Nation's population.
The Health Effects Research Laboratory, Research Triangle
Park conducts a coordinated environmental health research
program in toxicology, epidemiology, and clinical studies
using human volunteer subjects. These studies address
problems in air pollution, non-ionizing radiation, environ-
mental carcinogenesis and the toxicology of pesticides as
well as other chemical pollutants. The Laboratory develops
and revises air quality criteria documents on pollutants for
which national ambient air quality standards exist or are
proposed, provides the data for registration of new pesticides
or proposed suspension of those already in use, conducts
research on hazardous and toxic materials, and is preparing
the health basis for non-ionizing radiation standards.
Direct support to the regulatory function of the Agency is
provided in the form of expert testimony and preparation of
affidavits as well as expert advice to the Administrator to
assure the adequacy of health care and surveillance of
persons having suffered imminent and substantial endanger-
ment of their health.
This report represents a research effort to extend and
improve analytical methodology for the determination of a
variety of pesticide residues in our environment. The
emphasis is on less persistent chemicals which are being
used in place of the more environmentally stable pesticides.
John H. Knelson, M.D.
Director,
Health Effects Research Laboratory
111
-------
ABSTRACT
Cf all the forty pesticides under investigation, twelve
(methyl parathion, parathion, malathion, phorate, azinphos methyl,
p
azinphos ethvl, Dursban , diazinon, dimethoate, phosphamidon,
p
Azodrin and demeton) were detected by the FPD-P detector, two
p
(Temik and methomyl) by the FPD-S detector, fourteen (captan,
p
chlorobenzilate, endosulfan, Avadex , folpet, methoxychlor, PCNB,
p p
Perthane , trifluralin, atrazine, bromacil, CIPC, Difolatan and
simazine) by EC detector. Twelve have to be derivatized by PFPA
R R
(bux, carbofuran, monuron, Zectran , Landrins and IPC), PFBC
(benomyl, monuron and amitrole) or DAM (DNOC, DNBP, and dicamba)
and then detected by the EC detector. The 4% SE-30/6% OV-210
column exhibited superior separation efficiency for most pest-
icides than other tested columns. The usefulness of the nitrogen-
specific Hall detector was limited by its erratic performance.
It was caused by the repetitive injection of unvented solvent and
subsequent acid contamination in the detection system.
The silica gel column was proven to be consistent and
reliable for soil sample cleanup and separation of multi-class
pesticides. Some impurities from blank silica gel columns were
present in various eluted fractions; however, this drawback was
offset by their effectiveness in cleanup and separation of var-
ious pesticides from soil co-extractants.
Good recovery results were obtained by the Soxhlet pro-
cedure for most of the. organophosphate and halogenated pesticides,
IV
-------
except for demeton, bromacil, endosulfan and trifluralin. Also,
captan, folpet and Difolatan reacted with methanol during Sox-
hlet extraction; however, they were subsequently extracted and
p
recovered by ambient tumbling or Polytron ultrasonic extraction,
Most of the heat-labile ^-containing pesticides were
extracted by the Polytron and recovered in good yield except
amitrole, Temik and methomyl. Figure 29 summarized the pro-
cedure for detection of forty multi-class pesticides in soils.
A dynamic fluorogenic labelling detector was designed
and characterized for the high-pressure liquid chromatographic
analysis of six N-methylcarbamate and two carbamoyl oxime pest-
icides. Lannate, Matacil, Temik, Baygon, carbofuran, Sevin and
Mesurol could be extracted from sandy soil at the 0.01 ppm
level with recoveries ranging from 83 to 115%. Somewhat lower
recoveries were experienced for sandy loam and silty loam soils.
Zectran could not be reproducibly chromatographed under the
liquid chromatographic conditions chosen for the separations.
No cleanup was required prior to the liquid chromatography of
the soil extracts; no significant interferences were observed
for the unclean extracts.
This report was submitted in fulfillment of Contract
No. 68-03-1706 by Pesticide Research Laboratory, Food Science
Department, IFAS, University of Florida under the sponsorship
of the U.S. Environment Protection Agency. This report covers
the period June 14, 1974, to June 26, 1976, and work was com-
pleted as of June 26, 1976.
-------
CONTENTS
Foreword iii
Abstract iv
Figures viii
Tables xiii
Extension of Multi-residue Methodology. I.
Determining Multiclass Pesticide Residues in Soil
Introduction 1
Experimental 9
Apparatus 9
Reagents 11
Pesticide Standard Solutions 13
Pesticide Derivatization 13
Preparation of Spiked Soil Samples and
Extraction Procedure 16
Polytron Ultrasonic Extraction Procedure 16
Silica Gel Column Chromatography 17
Results and Discussion 18
Screening of Pesticides for GLC responses .... 18
Derivatization of Pesticides 48
GLC Detection of S-containing Pesticides 70
Loss of Pesticides in the Concentration Steps . . 71
Silica gel Column Chromatography 74
Co-extractive Interferences from Crude Soil
Extracts and Cleanup by Silica Gel
Column Chromatography 79
Recoveries of Pesticides from Soils by
Soxhlet Extraction 89
Recoveries of Some Organophosphate and
Thermally Labile N-containing Pesticides
from soil by Polytron Ultrasonic Extraction
and Comparison of Results with Soxhlet Ex-
traction 103
Stimultaneous Analysis of Forty Multi-class
Pesticides in Sandy Soil 119
References 130
vx
-------
Extension of Multi-residue Methodology. II.
Dynamic Fluorogenic Labelling Detector for Carbamates
Introduction 132
Experimental 134
Apparatus and Reagents 134
HPLC Column Study 136
Studies of Fluram 140
Studies on o-Phthalaldehyde 142
Conclusions 162
References 167
vii
-------
FIGURES
Extension of Multi-residue Methodology. I.
Number Page
la Chromatograms of 14 EC-sensitive pesticides on
4% SE 30/6% OV-210 column 24
Ib Chromatograms of 14 EC-sensitive pesticides on
1.51 OV-17/1.95% OV-210 column 25
Ic Chromatograms of 14 EC-sensitive pesticides on
5% OV-210 column 26
2 Analytical curve of Avadex 27
3 Analytical curve of captan, folpet and Difolatan . 28
4 Analytical curves of Perthane, chlorobenzilate
and CIPC 29
5 Analytical curves of trifluralin and bromacil ... 30
6 Analytical curves of simazine and atrazine .... 31
7 Analytical curves of PCNB, endosulfan and
methoxychlor 32
8 Analytical curves of Azodrin 41
*
9 Analytical curves of Dursban, dimethoate and
phosphamidon 42.
10 Analytical curve of diazinon 43
11 Analytical curves of phorate and dimeton 44
12 Analytical curves of malathion and parathion ... 45
13 Analytical curves of azinphos ethyl, azinphos methyl
and methyl parathion 46
14 Analytical curve of PFPA-IPC 54
15 Analytical curves of carbofuran, bux, Zectran,
Landrins (PFPA) 55
16 Analytical curve of PFBC-amitrole 63
viii
-------
FIGURES
Number
17 Analytical curve of PFBC-benomyl 64
18 Analytical curve of PFBC-monuron 65
19 Analytical curves of dicamba, DNOC and
DNBP (DAM) 66
20 Analytical curves of Temik and methomyl
(4% SE 30/6% OV-210) 72
21 Analytical curves of Temik and methomyl
(5% Carbowax 20 M) 73
22 Chromatogram of standard PFPA-derivatives of six
carbamate pesticides (a), and after they were
heated at 100°C in a sealed test tube for 6 hr in
1:1 benzene/MeOH (b) 78
23a Outlines of halogenated and organophosphate pest-
icide standards (1 mg each) eluted from silica
gel column 80
23b Outline of multi-class pesticide standard (1 mg)
eluted from silica gel column '. 81
24a Gas chromatograms of crude control sandy soil
extract (10 mg soil equivalent) 83
24b Gas chromatograms of control and 0.01 ppm PCNB
from sandy soil (10 mg soil equivalent) 84
24c Gas chromatograms of 0.01 ppm in sandy soil of Avadex
(3), trifluralin (39), endosulfan (21), methoxychlor
(79), and Perthane~X35) in 60% benzene/methanol frac-
tion and the respective control (50 mg soil equiv-
alent) 85
24d Gas chromatograms of 0.01 ppm in sandy soil of CIPC
(12), captan (10), folpet (22) , chlorobenzilate (11)
and" Difolatan ~(T6) and the respective control (10 mg
soil equivalent^" 86
24e Gas chromatograms of 0.01 ppm in sandy soil of atra-
zine (2) and simazine (37) and the respective control
(50 mg soil equivalent). 87
IX
-------
FIGURES
Number Page
24f Gas chromatograms of 0.01 ppm in sandy soil of
bromacil (8) and the respective control (50 mg
soil equivalent) 88
25a Gas chromatograms, Top; control (500 mg soil
equivalent). Middle; 0.01 ppm OP pesticides
in fraction II (60% B/H). Bottom; fraction
III (5% A/B) 90
25b Chromatograms of 0.01 ppm OP pesticide in frac-
tion IV (10% A/B), fraction V (20% A/B) and
fraction VI (100% A/B) 91
25c Gas chromatograms of 6 FPD (P) sensitive pest-
icides from 4% SE 30/6% OV-210 column 92
25d Gas chromatograms of 6 FPD (P) sensitive pest-
icides from 4% SE 30/6% OV-210 column 93
26 Chromatograms of control crude sandy soil (2 mg
equivalent) and recoveries of 0.01 ppm of dicamba
(LS), DNOC (18) and DNBP (1£) by diazomethane
derivatization 95
27a Chromatograms, left; of 1 ppm Zectran (40) and
the respective control, right; 1 ppm IPC~(24)
and its control (1 mg equivalent), PFPA deriv-
atives . k 96
i
27b Chromatograms of 0.1 ppm bux (9) and its respec-
tive control (1 mg soil equivalent), PFPA der-
ivative 97
27c Gas chromatograms, of Landrins (25,26) and the
respective control (1 mg soil eqUivaTent) PFPA
derivatives 98
27d Chromatograms of 0.1 ppm carbofuran and its
control 99
28a Chromatograms of PFBC-control (5 mg soil equiva-
lent) (83%) of 0.01 ppm monuron from sandy soil . . 100
28b Chromatograms of PFBC-control (10 mg soil equiva-
lent) and recovery of 0.1 ppm benomyl from sandy
soil 101
29 Procedure for detection of forty pesticides in
soil . _ 129
-------
Extension of Multi-residue Methodology. II.
Figure
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Isocratic separation os six carbamate mix on
uC,g, 40% acetonitrile - 60% H20
Isocratic separation of early eluting carbamates
on yCN, 12.5% acetonitrile - 87.5% H20
Isocratic separation of late eluting carbamates
on CDS, 12.5% acetonitrile - 87.5% H20
Relative fluorescence of N-methylcarbamates as
a function of time
Relative fluorescence of N-methylcarbamates as
a function of time
Relative fluorescence of N-methylcarbamates as
a function of time
Relative fluorescence of N-methylcarbamates as
a function of time
Modular dynamic fluorogenic labelling liquid
chromatograph
Chromatograms , Top; of sandy soil, spiked at
0.01 ppm. Program No. 10 for 13 min. 25%
dioxane - H20 to 40% dioxane - H20, 1 ml/min.
Bottom; check sandy soil
Chromatogram of sandy loam soil
Chromatogram of silty loam soil
Chromatogram of sandy soil spiked at 1.0 ppm . . .
Chromatogram of sandy loam soil
Chromatogram of silty loam soil
Analytical curves for six N-methylcarbamate
pesticides
Page
138
139
139
144
144
145
145
147
149
149
150
150
151
151
154
16 Effect of withholding OPA (substituting pH 10
buffer) on response of celery spiked at 0.2 ppm
with 1, Lannate; 2, Temik; 3, Baygon; 4, carbo-
furan; and 5, Sevin 156
xi
-------
Figure • Page
17 Effect of withholding NaOH (substituting pH 7
H-O) on response of lettuce spiked at 0.2 ppm
with: 1, Lannate; 2, Temik; 3, Baygon; 4,
carbofuran; 5, Sevin; and 6, Mesural 156
18 Plot of 4 a (peak width at base) for contri-
butions of hydrolysis coil and of fluorometer
cell, as a function of flow through each 157
19 Aminco Aminalyzer, adapted to N-methylcarbamate
analyses 163
20 Aminalyzer chromatogram of: 1, Lannate; 2,
Temik; 3, Baygon; 4, carbofuran; 5, Sevin;
and 6, Mesurol 164
21 Aminalyzer extended range analytical curve of
Lannate 164
xii
-------
TABLES
Extension of Multi-residue Methodology. I.
Number Page
1 List of pesticides under multi-residue
methodology investigation 2
2 Retention times and response values relative to
aldrin of fourteen EC-sensitive pesticides on
10% DC-200 19
3 Retention times and response values relative to
aldrin of fourteen EC-sensitive pesticides on
10% QF-1 20
4 Retention times and response values relative to
aldrin of fourteen EC-sensitive pesticides on
4% SE 30/6% OV-210 21
5 Retention times and response values relative to
aldrin of fourteen EC-sensitive pesticides on
1.5% OV-17/1.95% OV-210 22
6 Retention times and response values relative to
aldrin of fourteen EC-sensitive pesticides on
5% OV-210 .' 23
7 Computed efficiencies of five GLC columns 36
8 Retention times and response values relative to
parathion of eight. OP pesticides 37
9 Retention times and response values relative to
parathion of twelve organophosphate pesticides on
4% SE-30/6% OV-21.0, 200°C 38
10 Retention times and response values relative to
parathion of twelve organophosphate pesticides
on 1.5% OV-17/1.95% OV-210, 200°C 39
11 Retention times and response values relative to
parathion of twelve organophosphate pesticides on
5% OV-210, 200°C 40
12 Retention times and response values relative to
atrazine, of Nitrogen-containing pesticides on 5%
SE-30 column, 180°C (Hall detector) 49
Xlll
-------
TABLES
Number Page
13 Retention times and response values relative to
aldrin of PFPA derivatized pesticides on two
columns (Packard 7820, EC) . . - 50
14 Retention times and response values relative to
aldrin of PFPA derivatized pesticides on two
columns (Tracor 222, EC) 51
15 Retention times and response values relative to
aldrin of 2,5-dichlorobenzenesulfonyl chloride
(pyridine system) derivatized pesticides on two
GLC columns 52
16 Retention times and response values relative to
aldrin of PFBC derivatized pesticides (pyridine
system) on 10% DC-200 column, 205°C 57
17 Retention times and response values relative to
aldrin of PFBC derivatized pesticides (pyridine
system) on 4% SE-30/6% OV-210 column 58
18 Retention times and response values relative to
aldrin of PFBC derivatized pesticides (pyridine
system) on 1.5% 0V-17/1.95% OV-210 column .... 59
19 Retention times and response values relative to
aldrin of PFBC derivatized pesticides (0.1 M
NaOH system) on 1.5% 0V-17/1.95% OV-210 60
20 Retention times and response values relative to
aldrin of PFBC derivatized pesticides (0.1 M
NaOH system) on 4% SE-30/6% OV-210 61
21 Retention times and response values relative to
aldrin of BPFT derivatized pesticides on two
columns 67
22 Retention times and response values relative to
aldrin of diazomethane derivatized pesticides on
4% SE-30/6% OV-210 68
23 Retention times and response values relative to
aldrin of selected derivatization techniques on
twelve multi-class pesticides 69
24 Percent recovery of 500 yg of pesticides each in
the Kuderna-Danish concentrator step concentrated
from 250 ml 1:1 benzene/methanol to 10 ml .... 75
xiv
-------
TABLES
Number Page
25 Effect of paraffin oil keeper on pesticide
retention 76
26 Recovery of 100 yg of nitrogen containing pest-
icides after heating for 6 hr (100°C) in a
capped test tube with 2 ml benzene/methanol (1:1) . 77
27 Average percent recoveries of phorate and para-
thion from sandy soil which were spiked at 0.00,
0.01, 0.10, 1.00 and 10.00 ppm 102
28 Soil characterization 104
29 Average percent recoveries of halogenated pest-
icides from silty loam soil which were spiked at
0.01, 0.10, 1.00 and 10.00 ppm 105
30 Average percent recoveries of halogenated pest-
icides from sandy soil which were spiked at
0.01, 0.10, 1.00' and 10.00 ppm 106
31 Average percent recoveries of halogenated pest-
icides from sandy loam soil which were spiked
at 0.01, 0.10, 1.00 and 10.00 ppm 107
32 Average percent recoveries of organophosphate
pesticides from sandy soil which were spiked at
0.01, 0.10, 1.00 and 10.00 ppm 109
33 Average percent recoveries of organophosphate
pesticides from sandy loam soil which were spiked
at 0.01, 0.10, 1.00 and 10.00 ppm 110
34 Average percent recoveries of organophosphate
pesticides from silty loam soil which were spiked
at 0.01, 0.10, 1.00 and 10.00 ppm Ill
35 Comparison of solvents for ultrasonic-Polytron
extraction of carbofuran from sandy soil 112
36 Effect of Polytron extraction time on recovery
of carbofuran from sandy soil 114
37 Effect of Polytron motor speed on recovery of
carbofuran from sandy soil 115
xv
-------
TABLES
Number
38 Average percent recoveries of pesticides from
sandy soil samples which were spiked at 0.01
and 0.10 ppm (Polytron extraction) 116
39 Average percent recoveries of pesticides from
sandy soil samples which were spiked at 0.01
and 0.10 ppm (Soxhlet extraction) 117
40 Average percent recoveries of pesticides from
sandy soil samples which were spiked at 0.01
and 0.10 ppm (Polytron extraction) 120
41 Average percent recoveries of pesticides from
sandy soil samples which were spiked at 0.01
and 0.10 ppm (Soxhlet extraction) 121
42 Multi-residue analysis of fourteen EC-sensitive
pesticides in duplicate sandy soil samples which
were spiked at 0.01 and 0.10 ppm of forty multi-
class pesticides 122
43 Multi-residue analysis of twelve FPD-P sensitive
pesticides in duplicate sandy soil samples which
were spiked at 0.01, and 0.10 ppm of forty multi-
class pesticides 125
44 Multi-residue analysis of twelve multi-class/
nitrogen containing pesticides in duplicate
sandy soil samples which were spiked at 0.01
and 0.10 ppm 126
Extension of Multi-residue Methodology- II.
Number Page
1 Lower limits of detection for seven carbamates
in static Fluram system 141
2 Soil recoveries for seven carbamates for silty
loam, sandy loam and sandy soils 159
xvi
-------
Extension of Multi-residue Methodology. I. Determining
Multiclass Pesticide Residues in Soils by Gas Chromatography
H. Anson Moye and Sujit Witkonton
Introduction
The exceptionally large number of pesticides, represent-
ing a diversity of compound classes, presents a severe prob-
lem to those charged with their analyses. Efforts to extend
the existing multi-residue methodology, largely compiled by
EPA and FDA, to those compounds and substrates presently not
included would certainly be in order for a number of reasons,
which will not be elaborated here.
Multi-residue methodology, due to its inception by the
FDA, historically has concerned itself with the analysis of
foods, primarily utilizing a gas chromatographic determinative
step. Introduction of newer pesticides, along with a growing
need to know of their presence in non-target systems (the en-
vironment) has limited the body of multi-residue methodology
in terms of its applicability, partly because these compounds
have not been thoroughly tested through the multi-residue
methods and partly because those that have been tested have
been difficult to detect gas chromatographically. A large
number, if not all, of those compounds listed in Table I would
fc -\
certainly fall into one or both of these categories.
-------
Table 1. List of Pesticides* under Multi-residue Methodology Investigation
Number
Pesticide
Number
Pesticide
Number
Pesticide
1 Amitrole
2 Atrazine
3 AvadexR
(diallate)
4 Azinphos ethyl
5 Azinphos methyl
6 AzodrinR
(monocrotophos)
7 Benomyl
8 Bromacil
9 BuxR
(metalkamate)
10 Captan
11 Chlorobenzilate
12 CIPCR
(chlorpropham)
13 Demeton
14 Diazinon
15 Dicamba
16 DifolatanR
(captafol)
17 Dimethoate
18 DNOC
19 DNBP
20 DursbanR
(chlorpyrifos)
21 Endosulfan
22 Folpet
23 FuradanR
(carbofuran)
24 IPCR
(propham)
25 2,3,5-LandrinR
(trimethylphenyl
methylcarbamate)
26 3,4,5-LandrinR
27 Malathion
28 Methorny1
29 Methoxychlor
30 Methyl parathion
31 Monuron
32 Parathion
33 Phosphamidon
34 PCNB
35 PerthaneR
(ethylan)
36 Phorate
37 Simazine
38 TemikR
(aldicarb)
39 Trifluralin
40 ZectranR
(mexacarbate)
*Where trade names are used, the common name appears in parentheses. Certain trade names
are probably more familiar to many readers than the designated common names. For this
reason, some pesticides will be referred to by their trade names through out the text.
-------
Although techniques other than gas chromatography have
been used in the determinative steps of pesticide methodology
schemes, they usually lack sensitivity and selectivity. Usually
the inability to perform a gas chromatographic analysis on an
organic pesticide can be attributed to thermal instability or
lack of volatility- These problems are well known for the car-
bamates, carbamoyl oximes, phosphonates and some substituted
ureas, and prevent full utilization of the commercially avail-
able detectors, many of which would otherwise provide adequate
sensitivity and selectivity. The use of chemical derivatives
obviates these problems in many cases (1-5).
The multi-residue method of FDA, later used and extended
by EPA to non-food samples, is largely based upon the extraction
and cleanup procedure reported by Mills, Onley and Gaither in
1963 (6). This procedure could be applied to those foods of 21
fat content or less; however, subsequent modifications have
covered essentially all food types (7-10). The Pesticide Ana-
lytical Manual (PAM) of the Food and Drug Administration con-
tains a fairly comprehensive list of those pesticides which have,
or have not, been tested through what has become known as the
"Mills" procedure. Most of the compounds listed in Table I
either were not completely recovered through the procedure or
were simply not tested.
Since the main responsibilities of the FDA have not in-
cluded soil, no information concerning its extraction and clean-
up is found within the PAM. Extraction methods in large variety,
-------
however, have been reported in the pesticide analytical liter-
ature utilizing the full spectrum of solvents and techniques
from simple swirling to lengthy Soxhlet extraction. Probably
the most extensive review of the field to date is that of Chiba
(11) in Residue Reviews, which presents somewhat conflicting
data on soil extraction efficiencies. It would be hoped that
since this project concerns itself with the extension of
existing methodology, those solvents used prior to the Mills
Florisil column, would perform adequately as extractants of soil.
Unfortunately, according to those works reviewed by Chiba,
neither acetonitrile nor petroleum ether (both used in the Mills
procedure) are satisfactory. Chiba found that solvent pairs,
one nonpolar and ther other polar, were more satisfactory than
any other arrangement for the extraction of air dried sandy
loam soil. A great many factors contribute to the success with
which soils may be extracted, including soil type, moisture
content, agitatidn, contact time and solvent. These must all
be considered in any efficiency study.
Success in gas chromatographing involatile or thermally
unstable compounds can often be achieved with one or both of
the following approaches: (1) injection of relatively large
amounts of pesticide, approaching one microgram, so that chem-
ically active sites on the column are momentarily masked, re-
sulting in a higher percentage of the pesticide reaching the
detector, or (2) chemical derivatization usually at an ionizable
function group, so that volatility and thermal stability are
-------
improved. The first approach is made more feasible by the
recent improvements in element specific detectors, such as the
flame photometric, the electrolytic conductivity and the micro-
coulometric detectors. These types of detectors possess such
large inherent selectivity that relatively large amounts of
sample can be injected without interferences becoming pro-
hibitive. Selectivity can also be achieved in the second
approach by making derivatives which are not only characteristic
of certain functional groups or molecular structures but also
give responses with element selective detectors or the electron
capture detector.
To meet the above objective, sequences and scope of re-
search work were designed as follows:
1. Screening of pesticides listed in Table 1 for gas
chromatographic response. The halogen containing pesticides
were checked with the tritium electron capture detector (EC),
the phosphorus and sulfur containing pesticides with the flame
photometric detector (FPD) and the nitrogen containing pesti-
cides with the electrolytic conductivity detector. Five columns
of 10% DC-200, 10% QF-1, 4% SE-30/6% OV-210, 1.5% OV-17/1.95%
OV-210, and 5% OV-210 and their operating conditions for chlori-
nated hydrocarbons which were specified by PAM were tried.
2. Pesticide derivatization. Pesticides which gave poor
or no response to gas liquid chromatography (GLC) were ini-
tially derivatized by applying the perfluoroacylation method
(12) and halogeriated-benzene-sulfonates formation (13). GLC con-
ditions that are normally used for the nonderivatized pesticides
-------
were used.
3. Soil sample preparation. Pesticides which gave accep-
table GLC response were studied for recovery in sandy, sandy loam
and siIty loam soils. Triplicate samples of 50-100 g of soil were
spiked at 0.00, 0.01, 0.10, 1 and 10 ppm, extracted after curing
for 30 minutes, then analyzed.
4. Recovery of pesticides from soil. One hundred gm of
each soil sample was extracted by Soxhlet extraction with 300 ml
of benzene-methanol (1:1).
5. Cleanup of soil extracts. Soil extracts were cleaned
up by silica-gel column chromatography. Alternate solvent
systems were examined in order to obtain maximum recovery of
pesticides. Results are presented in average percent recoveries
with specific accompanying chromatograms to show degree and ex-
tent of co-extractive interferences.
6. Special efforts were directed at finding suitable der-
ivatives for those pesticides which during the past year's
(1974-75) work have been unsuccessfully gas chromatographed
(Table 1). Nine additional pesticides were requested by EPA to
be included in the 1975-76 multi-residue investigation. For
those compounds which appeared appropriate (possess appropriate
functional groups), the following derivatization reagents were
tested: (1) N-methyl-bis-(heptafluorobutyryl) amide, (2) 9-
chloromethyl anthracene, (3) pentafluorobenzoyl chloride, (4) 0-
(2,3,4,5,6-pentafluorobenzyl) hydroxylamine hydrochloride, (5)
acetic acid-N-hydroxysuccinimide ester, (6) a-bromo-2,3,4,5,6-
pentafluorotoluene, (7) diazomethane, (8) pentafluoropropionic
6
-------
anhydride, and (9) N-methyl-bis-(trifluoroacetamide). When
reagents became available during the course of the contract
period and appeared to offer advantages over those listed above,
they were also tested.
Standard procedures were used to attempt derivatization on
a micro scale. Aliquots of the appropriately manipulated re-
action solution were injected onto three gas chromatographic
columns (4% SE-30/6% OV-210, 1.5% OV-17/1.95% OV-210, and 5% 0V-
210), specified by EPA, installed in a Tracor 222 gas chroma-
tograph equipped with dual tritium EC detectors. For those
pesticides which gave responses, analytical curves were made
and limits of detection established (three times noise). Re-
producibility of derivatization and gas chromatography was also
established.
Once it had been determined which problem compounds were
amenable to gas chromatography by derivatization, a soil ex-
traction study was undertaken. Emphasis was on determining the
feasibility of utilizing an ultrasound extraction device (29),
the Willems Polytron PT 20, and comparing its performance to
that of Soxhlet extraction. Since the three soil types in the
1974-75 work behaved similarly in terms of extraction effi-
ciencies, only the soil giving the most gas chromatographic
interference was studied, i.e. the sandy soil.
Preliminary to extracting all the problem compounds, one
compound was selected to use for optimizing the extraction
technique for the Polytron. Once the optimum extraction tech-
nique had been established, the problem compounds were applied
7
-------
as standards to the silica gel columns so as to determine in
which eluted fraction they appeared. Subsequently each compound
was added to soil, aged for 30 min, extracted with the Polytron,
derivatized, if necessary, and analyzed gas chromatographically.
Percent recoveries were compared to those already obtained by
Soxhlet extraction. However, for those compounds which could be
derivatized with one of the proposed reagents, Soxhlet extrac-
tions were made for comparison.
-------
Experimental
Apparatus
Electron Capture Detector. Packard model 7820 gas chrom-
atograph equipped with two tritium foil electron capture de-
tectors, two independent recorders and gas chromatographic oven
Packard model 802, was operated under the following conditions:
Two 1.8 m x 4 mm I.D. coil columns packed with 10% DC-200
and 10% QF-1 on Gas Chrom Q, 80-100.mesh; inlet temperature
220°C; outlet, 230°C; detector, 220PC; column, 195°C.
Electron Capture Detector. Tracer 222 gas chromatograph
equipped with two tritium foil electron capture detectors and
two independent Soltec recorders was operated under the
following conditions:
Two 1.8m x 2 mm I.D. U-shaped glass columns packed with
either pair of 10% DC-200, 10% QF-1, 4% SE-30/6% OV-210, 1.5%
OV-17/1.95% OV-210 or 5% OV-210 on Gas Chrom Q, 80-100 mesh;
inlet temperature 230°C; detector, 220°C; column, 205°C.
Flame Photometric Detector. Varian Aerograph series 1200
equipped with Tracer flame photometric detector, Soltec re-
corder 1 mv, high voltage supply and electrometer of Keithley
instruments, was operated under the following conditions:
1.8 m x 2 mm I.D. coil-shaped glass columns packed with the
same column packings as for the EC detectors. The packing was
-------
topped with 2.5 cm of 10% Carbowax 20 M on Chromosorb W (HP),
80-100 mesh; inlet temperature, 230°C (300°C for Temik and
methorny1); detector, 165°C; nitrogen carrier gas flow rate, 50
ml/min; C^, 15 ml/min; air, 50 ml/min; H2, 150 ml/min; high
voltage setting, 850 volts. Phosphorus or sulfur mode. The
FPD was modified by reversing the H~ and air/02 inlets so that
the flame would not be extinguished by solvent after each
injection.
The sulfur mode of FPD was modified in order to increase
its sensitivity by the introduction of S09 (25). The S09/N9
Lt £* £*
(1.46 ppm of S02) auxiliary gas was mixed through the H~ line
by means of a T-connector located approximately 15 cm before
the detector. The auxiliary gas pressure was adjusted to 50
psig. and the flow was controlled at 50 ml/min with a metering
valve.
Hall Electrolytic Conductivity Detector, Model 310. Tracer
222 gas chromatograph was equipped with one Hall electrolytic
\
conductivity detector and a Soltec recorder. The gas chroma-
tograph was operated under the following conditions:
1.8 m x 2 mm I.D. U-shaped glass column packed with 5% SE-
30 on Chromosorb W (HP), 80-100 mesh; inlet temperature, 220°C;
column, 180°C; transfer line, 270°C; Helium gas flow rate, 67
ml/min. Hall detector settings: Furnace temperature, 820°C;
Mode: Nitrogen; Reaction gas: Hydrogen, 33 ml/min; cell flow,
0.5 ml/min. attenuation, 1 x 10.
Chromaflex column, size 50-25, Kontes K-422250.
Kuderna-Danish evaporative concentrator, 500 ml.
10
-------
Culture tube, 5 ml, Scientific Products, catalog no. T1346-
5.
«
Microliter pipets, 10, 25, 50, 100, 200, and 300 pi. Eppen-
dorf.
Soxhlet extraction apparatus; Kimax brand (24005), flask,
500 ml. Fisher Scientific Co. catalog no. 9-551C.
Extraction thimbles, glass. Fisher Scientific Co. catalog
no. 9-653C.
Kerr wide mouth mason jars, 12 quarts.
Disposable pasteur pipets.
Unitized Extraction Assembly (Precision 65500). Fisher
Scientific Co. catalog no. 9-678.
Nitrogen evaporator with hot plate maintained at 40-50°C.
p
Fisher Isotemp dry bath, model 145.
Automatic shaker of Eberback Corporation, Ann Arbor, Mich.
Automatic tumbling assembly.
Maxi-Mix model M-16715, series 16700, Thermolyne Sybron
Corporation.
Willems Polytron, model PR-20, Brinkmann Instruments,
Westburg, New York.
Reagents
Burdick § Jackson's "distilled in glass" or redistilled
solvents: hexane, benzene, acetonitrile, methanol, acetone.
Anhydrous sodium sulfate, acetone-extracted.
Pyridine, redistilled.
Silica gel dry column, activity III, (20% H20) ICN Pharma-
11
-------
ceuticals, Inc., Life Sciences Group, Cleveland, Ohio, catalog
no. 404526.
Pentafluoropropionic anhydride, PCR, Inc., Research Chem-
icals Division, Gainesville, FL, catalog no. 13670.
2,5-dichlorobenzenesulfonyl chloride, Eastman Kodak Co.,
Rochester, NY 14650. Recrystallized three times from redis-
tilled isooctane.
N-methyl-bis-(heptafluorobutyryl) amide, N-methyl-bis-
(trifluoroacetamide). Regis Chemical Co., 8210 N. Austin Ave.,
Morton Grove, Illinois.
Pentafluorobenzoyl chloride, 98% pure, a-bromo-2,3,4,5,6-
pentafluorotoluene, 99+% pure, 9-Anthracene-methanol, Aldrich
Chemical Co., 3355 Lenox Rd., N.E., Suite 750, Atlanta, Georgia.
Thionyl chloride, Mallinckrodt Chemical Works, St. Louis,
Missouri.
Acetic acid N-hydroxysuccinimide ester, ICN Pharmaceu-
ticals, Inc., Life Science group, Cleveland, Ohio.
A
0-(2,3,4,5,6-Pentafluorobenzyl) hydroxylamine, Applied
Science Laboratories, Inc., State College, Pennsylvania.
Diazomethane in ether, prepared from "Diazald," Aldrich
Chemical Co., Inc. (Synthesis procedure obtained from Aldrich's
bulletin).
Phosphate buffer (pH 7): 3 g NaOH and 17 g KH2P04 in 200ml
water.
Keeper solution (1% paraffin oil in benzene).
12
-------
Pesticide Standard Solutions
Ten mg of each analytical grade pesticide (provided by EPA)
was dissolved in 10 ml of redistilled benzene and kept as con-
centrate stock solution (1 mg/ml). For those compounds which
could not be dissolved readily, benzene/acetone, 1:1, was used.
The stock solution was used for derivatization or diluted to
obtain on-scale GLC responses for non-derivatized pesticides.
Pesticide Derivatization
1. Perfluoroacylation method (12).
In the 15 ml culture tube, add 200 yg pesticide, 2 ml hexane,
50 yl pyridine, 50 yl pentafluoropropionic anhydride (PFPA).
Mix well and allow reaction to proceed for one hr at room temp-
erature. Add 3 ml phosphate buffer pH 7 and mix on Maxi-Mix
to stop reaction. Add 3 ml hexane, 50 yl acetonitrile, and
mix again. After separation of aqueous and organic phases,
remove the bottom layer of aqueous phase with a disposable
pipet and discard. Add 2 ml distilled water, mix and aspirate
bottom layer; repeat washing two more times. Dry the hexane
solution with sodium sulfate. Make final volume to 10 ml with
hexane and adjust concentration for on-scale GLC response.
2. Pentafluorobenzoylation (20).
2.1. Pyridine System. In 15 ml culture tube, add 200 yg
pesticide, 4 ml benzene, 10 yl of 0.1% pyridine in benzene,
2-5 yl of neat pentafluorobenzoylchloride (PFBC). Mix well
and allow reaction to proceed for 30 min by refluxing in dry
13
-------
bath (95°C). Add 4 ml of pH 7 phosphate buffer and mix. After
separation of aqueous and organic phases, remove the bottom
layer of aqueous phase. Add 4 ml distilled water, mix and as-
pirate bottom layer; repeat washing two more times. Dry the
organic solution with sodium sulfate. Make final volume to
10 ml and adjust concentration for on-scale GLC response.
2.2. Aqueous Sodium Hydroxide System (0.1 M). In 15 ml
culture tube, add 200 yg pesticide, 4 ml 0.1 M NaOH, 2 yl PFBC.
Mix well and allow reaction to proceed for 30 min by heating
in a dry bath (95°C). After the reaction, add 4 ml of benzene
to extract derivatized compound from aqueous phase. This was
done in the same way as described in the pyridine system.
3. The derivatization of pesticides with the following
compounds was similar to the pentafluorobenzoylation procedure
(pyridine system) except that 10 yl neat pyridine was used:
acetic acid N-hydroxy succinimide ester (AAHSE), N-methyl-bis-
(heptafluorobutyryl) amide (MHFBA), N-methyl-bis-(trifluoro-
acetamide) (MBTFA), and 9-chloromethyl-anthracene (CMA)
9-chloromethyl-anthracene was synthesized by refluxing 9-
anthracene-methanol with thionyl chloride in benzene for 3 hr
(21).
4. Pentafluorobenzylation (17). In 15 ml culture tube,
add 200 yg pesticide, 2 ml acetone, 1 ml pH 12 buffer, 10 yl
a-bromo-2,3,4,5,6-pentafluorotoluene (BPFT). Mix well and
allow reaction to proceed for 2 hr by refluxing in dry bath
(100°C). Add 2 ml benzene and mix by shaking. After separation
of aqueous and organic phases, remove the aqueous phase and
14
-------
wash the organic phase twice with distilled water as previously
described.
5. 0-(2,3,4,5,6-Pentafluorobenzyl) hydroxylamine (PFBHA)
derivatization (19). Make a stock solution of the reagent,
PFBHA, in pyri4ine to contain 50 mg per ml.
In 15 ml culture tube, add 200 yg pesticide in 200 yl ben-
zene. Remove benzene under a stream of N~. Add 0.2 ml of the
PFBHA stock solution. Vortex the tube to assure contact with
the pesticide, stopper and heat in a dry bath at 85°C for half
an hour. Remove the pyridine with a stream of N~. Add one ml
of hexane, mix, follow with 1 ml of distilled water. Shake the
tube and allow the phases to separate. Transfer most of the
hexane layer to another tube using a Pasteur pipet, and add
anhydrous sodium sulfate to dry the hexane before GLC analysis.
6. Diazomethane derivatization (18). Pipet a benzene
solution of an appropriately diluted pesticide standard into a
15 ml culture tube. Remove benzene under a stream of N~. Add
one ml of diazomethane reagent and let stand 10 min with occa-
sional shaking. Evaporate solvent with N2 and add 5 ml of ben-
zene for GLC analysis.
7. Halogenated benzene-sulfonate formation (13). Reflux
soil extract of pesticide with 2 mg of 2,5-dichlorobenzene-
sulfonyl chloride in 2 ml of acetone and 50 yl pyridine for 30
minutes. The reaction solution is cooled and 5 ml of hexane
and 2 ml buffer pH7 are added. Shake well and separate the
organic phase. Wash the organic phase three times with 2 ml
15
-------
distilled water each time. Dry the organic phase with sodium
sulfate and adjust volume for GLC analysis.
Preparation of Spiked Soil Samples and Extraction Procedure
Spike triplicate samples of 100 g (10% moisture) at 0,
0.01, 0.10, 1 and 10 ppm of pesticides and extract by Soxhlet
extractor with 300 ml 1:1 benzene/methanol for 12 hours.
Evaporate benzene/methanol extract by Kuderna-Danish evaporator
with repetitive additions of benzene to eliminate methanol and
water. Dilute the samples with benzene to obtain appropriate
volume for direct GLC analysis. For further silica gel column
cleanup, evaporate benzene almost to dryness and take the
residue up in 1 ml hexane.
As an alternative procedure, tumble or shake one hundred
grams of each soil sample for one hour with 300 ml of benzene,
acetone or benzene/methanol (1:1), filter, evaporate to dryness
on hot plate (50°C) under nitrogen, and take the residue up in
1 ml hexane. *
Polytron Ultrasonic Extraction Procedures
Spike air-dried sandy soil samples of 50 g at 0, 0.01,
and OilO ppm of pesticides and extract with the Polytron.
The generator which is immersed in the soil sample is equipped
with a saw-tooth cutting head. To the soil sample add 100 ml
of benzene/methanol (1:1), unless stated otherwise. With the
Polytron generator immersed in the sample, extract the soil
twice for 60 sec. at maximum power unless otherwise noted.
Filter the extracted soil and solvent through Whatman No. 42
16
-------
filter paper. Rinse the filter paper and the soil twice with
50 ml of the extracting solvent. Evaporate combined filtrate
to dryness under a stream of N~ at room temperature (for captan,
folpet, Difolatan, and methorny1), or by Kuderna-Danish evaporator,
chromatograph the residue on the silica gel column (as below)
and analyze gas chromatographically.
Silica Gel Column Chromatography
Transfer the hexane solutions of soil extracts to silica
gel chromatographic columns. Prepare the column by packing
16 g of silica gel topped with 14 g of reagent grade Na-SO^,.
Add one ml of soil extract (equivalent to 100 g of soil) to
the top of the bed with a disposable pipet. Wash the container
with 2, 3 and 4 ml of hexane and deliver to the column respec-
tively. Elute the column successively with 50 ml of the fol-
lowing solvent systems: (1) hexane, (2) 60% benzene in hexane,
(3) 5% acetonitrile in benzene, (4) 10% acetonitrile in benzene,
(5) 50% acetonitrile in benzene, and (6) acetonitrile. Evap-
orate these fractions and adjust to volume with benzene for GLC
analysis.
17
-------
Results and Discussion
Screening of pesticides for GLC responses
1. Electron capture detector. Results for fourteen halo-
genated pesticides on five GLC columns which gave relatively
good responses are presented in Tables 2, 3, 4, 5 and 6. Gas
chromatograms of these compounds on three GLC columns are illus-
trated by Figures la, b, and c. Their analytical curves for
quantitation of soil samples are presented in Figures 2-7.
Retention times and peak areas of fourteen halogenated
pesticides relative to aldrin (RRT and RPA respectively; see
footnotes Table 2) on a 10% DC-200 column are presented in
Table 2. All pesticides are presented according to their elution
orders from lowest to highest RRTs. Multiple peaks of Avadex
indicated the presence of impurities in the originally furnished
pesticide standard. A subsequently furnished standard pro-
duced only a single GLC peak (Tables 4, 5 and 6). Overlapping
peaks on 10% DC-200 (Table 2) are: simazine and atrazine;
captan and folpet; endosulfan and Perthane. Their decreasing
degree of sensitivities (from highest RPA to lowest RPA) are:
e
trifluralin, endosulfan, PCNB, Avadex, bromacil, methoxychlor,
Perthane, Difolatan, chlorobenzilate, folpet, captan, atrazine,
simazine and CIPC.
RRTs and RPAs of fourteen halogenated pesticides on 10%
i
QF-1 are presented in Table 3. Overlapping peaks are: PCNB,
18
-------
Table 2. Retention times and response values relative to aldrin
of fourteen EC-sensitive pesticides on 101 DC-200,
195°C (Att. = 3 x 10" , N7 flows = 60 ml/min. Packard
7820, EC). L
Pesticide
Avadex
CIPC
Trifluralin
Simazine
Atrazine
PCNB
Bromacil
Captan
Folpet
Endosulfan
Perthane
Chlorobenzilate
Difolatan
Methoxychlor
Aldrin
RRT1
0.23, 0.29
0.37, 0.39, 0.57
0.29
0.34
0.39
0.40
0.48
0.92
1.18
1.19
1.50, 2.03 0.
1.98
2.36
3.11
4.60
1 (12.5 min)
Computed as retention time relative
2Dt>A _ peak area
of pesticide/amount
Rp.2 Analytical Curve
(Figure No.)
0.500
0.002
2.300
0.015
0.017
0.530
0.302
0.050
0.055
570, 0.083
0.101
0.078
0.088
0.240
1.000
to aldrin.
of pesticide
4
5
—
—
7
5
—
—
7
4
4
—
7
—
peak area of aldrin/amount or aldrin
Only major peak of endosulfan was used for analytical curve
construction.
Note: Major peaks are indicated by underlining.
19
-------
Table 3. Retention times and response values relative
to aldrin of fourteen EC-sensitiveQpesticides
on 10% QF-1, 195°C (Att. = 1 x 10~ , N? flow
= 60 ml/min. Packard 7820, EC).
Pesticide
Avadex
CIPC
PCNB
Simazine
Atrazine
Trifluralin
Perthane
Endosulfan
Folpet
Bromacil
A
Captan
Chlorobenzilate
Methoxychlor
Difolatan
Aldrin
RRT1
0.50
0.63
0.77
0.79
0.79
1.00
1.69
2.31, 4.00
2.92
3.08
3.23
3.54
5.00
8.17
1 (2.5 min)
RPA2
0.017
0.001
0.039
0.002
0.013
0.394
0.065
0.390, 0.185
0.292
0.358
0.306
0.083
0.173
0.167
1.000
1^2See footnotes of Table 2.
Note: Major peaks are indicated by underlining.
20
-------
Table 4. Retention times and response values relative to aldrin
of fourteen EC-sensitive pesticides on 4% SE-30/6%
OV-210, 205°C (Att. = 2 x 10 , N9 flow = 67 ml/min.
Tracer 222, EC). L
Pesticide
CIPC
Avadex
Simazine
Atrazine
Trifluralin
PCNB
Bromacil
Endosulfan
Folpet
Captan
Perthane
Chlorobenzilate
Methoxychlor
Difolatan
Aldrin
RRT1
0.38
0.46
0.52
0.54
0.54
0.62
1.48
1.75, 2.56
1.76
1.76
1.91
2.53
4.30
4.40
1 (7.5 min)
nr>A2 Analytical Curve
KWV (Figure No.)
0.004
0.019
0.004
0.007
0.247
0.430
0.230
0.700, 0.210
0.136
0.105
0.008
0.025
0.242
0.105
1.000
—
2
6
6
—
—
—
—
3
3
—
—
—
3
—
" C/i« £ r\t*+-t\r\+- **c- nf T»^V.1» "J
Note: Major peaks are indicated by underlining,
21
-------
Table 5. Retention times and response values relative to
aldrin of fourteen EC-sensitive pesticides on
1.5% OV-17/1.95% OV-210, 205°C (Att. 2 x 10 ,
N flow = 53 ml/min. Tracer 222, EC).
Pesticide
Trifluralin
CIPC
Avadex
PCNB
Atrazine
Simazine
Chlorobenzilate
Endosulfan
Folpet
Cap tan
Perthane
Bromacil
Difolatan
Methoxychlor
Aldrin
RRT1
0.45
0.45
0.53
0.70
0.70
0.70
1.55, 3.32
1.97, 3.58
2.70
2.70
2.70
3.25
7.35
7.80
1 (2.8 min)
RPA2
0.204
0.002
0.031
0.420
0.009
0.008
0.012, 0.019
0.660, 0.130
0.070
0.071
0.013
0.061
0.026
0.139
1.000
footnotes of Table 2.
Note: Major peaks are indicated by underlining.
22
-------
Table 6. Retention times and response values relative to
aldrin of fourteen EC-sensitive pesticides on
51 OV-210, 205°C (Att. 2 x 10 , N? flow = 53 ml/
min. Tracer 222, EC).
Pesticide
Avadex
CIPC
Atrazine
Simazine
PCNB
Trifluralin
Perthane
Chlorobenzilate
Endosulfan
Folpet
Cap tan
Bromacil
Methoxychlor
Difolatan
Aldrin
RRT1
0.59
0.69
0.85
0.85
0.88
1.00
1.65
1.78, 5.25
2.22, 3.75
3.00
3.16
3.69
4.60
7.10
1 (2.3 min)
RPA2
0.014
0.002
0.008
0.008
0.680
0.230
0.008
0.011, 0.029
0.860, 0.180
0.039
0.049
0.129
0.108
0.009
1.000
* See footnotes of Table 2.
Note: Major peaks are indicated by underlining.
23
-------
to
CTil
to I
4% SE 30/6% 0V 210,
205°C, ATT. 2 x 102
0.1 ng ea.
EC
O)
0 5 10 15 20 25 30
MIN.
FIGURE 1 a
Chromatograms of 14 EC-sensitive pesticides. Pesticide
mixtures were grouped to prevent peak coincidence; num-
bers correspond to pesticides listed in Table 1.
-------
01
ro N
0.1 ng ea.
(NJ
1.5% 0V 17/1.95% OY 210
205°C, ATT. 2 x 10 , EC
IT)
rtr
O1
bO
(H
uo
P
vO
MIN.
FIGURE 1 b
TfT
Chromatograms of 14 EC-sensitive pesticides on 1.5% OV-17/
1.95% OV-210 column.
-------
5% 0V 210, 205°C
ATT. 2 x 102, EC
0
10
2
15 20
FIGURE 1 c
Chromatograms o£ 14 EC-sensitive pesticides
50 MIN.
-------
ts)
fj
W
w
80-
70-
60-
50-
40-
30-
20-1
10-
SE-30/6! OV-210, 205°C, ATT. 4 x 10
EC DETECTOR
0.1
0.2 0.3
NG AVADEX
0.4
FIGURE 2
Analytical curve of Avadex.
0.5
-------
tsj
06
160.
140.
120.
100
S 80-
w
w
60.
40
20
4% SE-30/6% OV-210, 205°C
EC DETECTOR, ATT.=4 x 10
FOLPET
DIFOLATAN
O CAPTAN
0.05 0.10 0.15 0.20 0.25
NG CAPTAN (X 2 = NG, FOLPET; X 10 = NG, DIFOLATAN)
FIGURE 3
Analytical curves of Captan, Folpet and Difolatan.
-------
to
140
120
E 100
H
S 80.
w
ac
w
60.
40
20
10% DC-200, 205°C, ATT. 4 x 10'
EC DETECTOR
CHLOROBENZILATE
PERTHANE
5 10 15 20
NG PERTHANE OR CHLOROBENZILATE (x4 = NG, CIPC)
FIGURE 4
Analytical curves of Perthane, Chlorobenzilate and CIPC
-------
-J O
HH Pi
s
60 .
50 .
40-
30
w
H ac 20J
HH <
[•?•[ m
= * 10-1
w
10% DC-2-&0, 205°C, ATT. 4 x 10'
EC DECTECTOR
TRIFLURALIN
ROMACIL
.02 .04 .06
NG TRIFLURALIN (x25 = NG BROMACIL)
FIGURE 5
Analytical curves of Trifluralin and Bromacil.
.08
-------
O4
M
140
120-1
100
80
g 60j
S 40
20
41 SE-30/6% OV-210, 205°C, ATT. 4 x 10
EC DETECTOR
SIMAZINE
TRAZINE
1234
NG SIMAZINE (x5 = NG, ATRAZINE)
FIGURE 6
Analytical curves of Simazine and Atrazine.
-------
Mi
W
ac
w
120.1 10% DC-200, 205°C, ATT. 4 x 10'
EC DETECTOR
100.
80
60
40
20
^ ,
.01 .02703704T057(h>.6? .08
NG PCNB (x5 = NG, ENDOSULFAN, xlO = NG, METHOXYCHLOR)
FIGURE 7
Analytical curves of PCNB, Endosulfan and Methoxychlor.
-------
simazine, atrazine; folpet and bromacil. Their decreasing degree
of sensitivities are: trifluralin, endosulfan, bromacil, captan,
folpet, methoxychlor, Difolatan, chlorobenzilate, Perthane
PCNB, Avadex, atrazine, simazine and CIPC. Table 4 presents RRTs
and RPAs of these compounds on the 4% SE 30/6% OV-210 column.
Overlapping peaks are: simazine (5_7) , atrazine (2) and tri-
fluralin (59); endosulfan (21), folpet (22) and captan (10);
chlorobenzilate (11) and endosulfan (21), methoxychlor (29) and
Difolatan (16) (see Figure 1 a for chromatograms). Their de-
creasing degree of sensitivities are: endosulfan, PCNB, tri-
fluralin, methoxychlor, bromacil, folpet, captan, Difolatan,
chlorobenzilate, Avadex, Perthane, atrazine, simazine and CIPC.
RRTs and RPAs of fourteen halogenated pesticides on 1.5%
OV-17/1.951 OV-210 are presented in Table 5. Overlapping peaks
are: trifluralin (39) and Avadex (3); PCNB (34), atrazine (2)
and simazine (37); folpet (22), captan (10) and Perthane (55)
(see Figure 1 b for chromatograms). Their decreasing degree of
sensitivities are: endosulfan, PCNB, trifluralin, methoxychlor,
captan, folpet, bromacil, Avadex, Difolatan, chlorobenzilate,
Perthane, atrazine, simazine and CIPC.
Table 6 presents RRTs and RPAs of these compounds on the
5% OV-210 column. Overlapping peaks are: atrazine (2), sim-
azine (5_7), PCNB (54.), and trifluralin (59) ; folpet (22) and
captan (10); endosulfan (21) and bromacil (8) (Figure 1 c).
Their decreasing degree of sensitivities are: endosulfan, PCNB,
trifluralin, bromacil, methoxychlor, captan, folpet, chloroben-
zilate, Avadex, Difolatan, atrazine, simazine, Perthane and CIPC.
55
-------
From Figure 1 a, the overlapping peaks of atrazine (2),
simazine (37) and trifluralin (2.9) on the 4% SE 30/6% OV-210
column could be separated by the 1.5% 0V-17/1.95% OV-210 column
(Figure 1 b). None of the columns under investigation could
be used to separate atrazine (2) and simazine (57). The best
column for the separation of captan (10), folpet (22) and en-
dosulfan (21) was the 5% OV-210 column (Figure 1 c). The over-
lapping peaks between chlorobenzilate (11) and endosulfan (21),
methoxychlor (29) and Difolatan on the 4% SE 30/6% OV-210
column (Figure 1 a) could be separated by 5% OV-210 (Figure Ic).
Most of the analytical curves for the halogenated pesticides
(Figure 2-7) exhibit good linearity and they intercept the point
of origin except for folpet, Difolatan and captan (Figure 3),
CIPC, chlorobenzilate and Perthane (Figure 4) and simazine
(Figure 6), suggesting minor loss of these pesticides in the GLC
system. This might be due to some degradation of these pest-
.»
icides on the GLC column under elevated temperature or loss by
absorption on GLC liquid phases. However, quantitation of these
pesticides was still possible and fairly accurate results could
be obtained because consistent and linear analytical curves
were observed.
Unsatisfactory results were obtained for most nitrogen-
containing pesticides and carbamate type compounds with the EC
detector. Except for the tailing peak of bromacil and low sen-
sitivity of CIPC, those pesticides giving moderately sensitive :
peaks were atrazine, bromacil, Difolatan and simazine (Tables
2-6). Zectran gave inconsistently unacceptable low sensitivity.
34
-------
Those pesticides producing inconsistent multiple peaks or no
responses were amitrole, benomyl, monuron, DNOC, DNBP, dicamba,
Landrins (2,3,5 § 3,4,5), Zectran, IPC, bux and carbofuran.
Theoretical plates were obtined from aldrin peaks for
five tested GLC columns (Table 7). 4% SE 30/6% OV-210 gave the
highest theoretical plates; therefore it exhibited superior
separations (Figure 1 a).
2. Flame photometric detector (phosphorous mode). Re-
tention times and peak areas of organophosphate (OP) pesticides
relative to parathion on five GLC columns are presented in
Tables 8-11. They are presented according to their elution
order (from lowest to highest RRTs). Analytical curves are
presented in Figures 8-13.
Table 8 presents RRTs and RPAs of eight OP pesticides on
10% DC-200 and 10% QF-1 columns. The overlapping peaks on 10%
DC-200 are phorate, Azodrin and demeton; on 10% QF-1 methyl
parathion and malathion do not separate. No GLC peaks were
obtained from the injection of Azodrin and azinphos methyl on
10% QF-1. Their decreasing sensitivities on 10% DC-200 are:
azinphos methyl, parathion, methyl paration, phorate, malathion,
azinphos ethyl, demeton, and Azodrin, on 10%- QF-1: parathion,
malathion, phorate, methyl parathion, azinphos ethyl, demeton,
azinphos methyl, and Azodrin.
Table 9 presents RRTs and RPAs of twelve OP pesticides on a
4% SE 30/6% OV-210 column. Overlapping occurs for dimethoate
and Azodrin; methyl parathion and malathion; parathion and phos-
phamidon. Their decreasing degree of sensitivities are: diaz-
35
-------
CM
o\
Table 7. Computed efficiencies of five GLC columns (Tracer 222, tritium foil EC
detector, column temperature = 205°C. Att. 4 x 102) .
r_, __ Efficiency
L01umn (Theoretical Plates)
4% SE-30/6% OV-210
1.5% OV-17/1.95% OV-210
10% DC-200
5% OV-210
10% QF-12
2700
2300
1936
1849
676
N, Flow
(mx/min)
67
53
67
53
60
Efficiency computed from total retention and base width of aldrin peak.
2
Result was obtained from Packard 7820, EC.
-------
C/4
-J
Table 8. Retention times and response values relative to parathion of eight organo-
phosphate pesticides on two GLC columns (Att. 8 x 10~8, Varian 1200, FPD-P)
10% DC-200: 200° C, 50 ml/min
Pesticide
»
Phorate
Azodrin
Demeton
Methyl parathion
Malathion
Parathion
Azinphos methyl
Azinphos ethyl
RRT1
0.36
0.37
0.39
0.68
0.89
1 (4.4 min)
5.45
7.25
•'•Computed as retention time relative
2RPA = Peak area
of pesticide/amount
RPA2
0.63
0.01
0.11
0.71
0.26
1.00
1.06
0.20
to parathion.
of pesticide
10% QF-1: 200
RRT1
0.19
no peak
0.34
0.78
0.76
1 (3.8 min)
no peak
6.10
0 C, 50 ml/min
RPA2
0.53
—
0.11
0.42
0.64
1.00
—
0.29
peak area of parathion/amount of parathion
responses when 1 yg of pesticides were injected
-------
Table 9. Retention times and response values relative to
parathion of twelve organophosphate pesticides
on 4% SE-30/6% OV-210, 200° C (Att. 8 x 1Q-8,
N2 flow = 50 ml/min. Varian 1200, FPD-P).
, 7 Analytical Curve
Pesticide RRT1 RPA^ (Figure ^
Demeton 0.17, 0.32 0.300, 0.500 (11)
Phorate 0.26 1.590 (11)
Diazinon 0.29 1.840 10
Dimethoate 0.48 0.414 9
Azodrin 0.50 0.100 8
Dursban 0.64 1.700 9
Methyl parathion 0.77 0.696 (13)
Malathion 0.82 0.555 (12)
Parathion 1.00 (4.9 min) 1.000 (12)
Phosphamidon 0.74, 1.045 0.047, 0.285 9
.»
Azinphos methyl 5.26 0.278 (13)
Azinphos ethyl 6.50 0.370 (13)
1^2See footnotes of Table 8.
Parentheses indicate other GLC conditons were used; see these
conditions from respective figures.
-------
Table 10. Retention times and response values relative to
parathion of twelve organophosphate pesticides fi
on 1.5% 0V-17/1.95% OV-.210, 200° C (Att. 8 x 10 ~°
N2 flow = 50 ml/min. Varian 1200, FPD-P).
Pesticide
Pemeton
Phorate
Diazinon
Dimethoate
Azodrin
Phosphamidon
Dursban
Methyl parathion
Malathion
Parathion
Azinphos methyl
Azinphos ethyl
RRT1
0.16, 0.32
0.24
0.32
0.54
0.59
0.68, 0.97
0.76
0.77
0.88
1 (2.6 min)
9.65
12.00
RPA2
0.395, 0.762 ;
.- 1
1.240
0.620
0.830
0.162
0.034, 0.149
1.270
1.490
1.030
1.000
1.070
0.780
1^2See footnotes of Table 8.
39
-------
Table 11.
Retention times and response values relative to
parathion of twelve organophosphate pesticides
on 5% OV-210, 200° C (Att. 32 x 10"9, N? flow =
50 ml/min. Varian 1200, FPD-P).
Pesticide
Demeton
Phorate
Diazinon
Dursban
Dimethoate
Azodrin
Malathion
Methyl parathion
Parathion
Phosphamidon
Azinphos methyl
Azinphos ethyl
RRT1
0.12, 0.31
0.17
0.19
0.43
0.52
0.67
0.74
0.79
1 (3.04 min)
1.32
5.10
6.20
RPA2
0.292, 0.780
0.500
1.610
1.285
0.450
0.083
0.525
0.875
1.000
0.136
0.233
0.212
152
See footnotes of Table 8.
40
-------
240
200-
4% SE-30/6% OV-210, 210°C, ATT. 8 x 10
(FPD, PHOSPHORUS MODE)
-8
w
X
160
120
80
40
20 40 60 80
NG AZODRIN
FIGURE 8
Analytical curve of Azodrin.
100
-------
ts)
W
a:
w
a,
120
100
80
60
40
20
4% SE-30/6% OV-210ao200°C
ATT = 8 x10 '
(FPD, PHOPHORUS MODE],
DURSBAN
10
HOSPHAMIDON
40
50
20 30
NG
FIGURE 9
Analytical curves of Dursban, Dimethoate and Phosphamidon
-------
•p*
tri
H
a:
w
a:
w
a,
100-
80.
60
40
20
41 SE-30/6% OV-210. 200°C, ATT,
8 x 10~8
(FPD, PHOSPHORUS MODE)
2 4_ 6
NG DIAZINON
FIGURE 10
Analytical curve of Diazinon
10
-------
140
10% DC-200, ATT. = 8 x 10
(Phosphorus Mode)
DEMETON
(column = 200°
HORATE
(column = 180°C)
2 34567
NG PHORATE (x5 = NG DEMETON)
FIGURE 11
Analytical curve of phorate and demeton.
-------
on
w
ac
100
80 .
60 .
40 .
w
20 .
10% DC-200, 200°C, ATT,
8 x ID'8
(Phosphorus Mode)
MALATHION
PARATHION
2 4 6 8 10
NG MALATHION OR PARATHION
12
FIGURE 12
Analytical curves of malathion and parathion
-------
-pa.
ON
160
120
80
W
S 40.
1.5% OV-17/1.95% OV-210, ATT. =
10
"8
(Phosphorus Mode)
AZINPHOS ETHYL*
(Col. = 240°C)
AZINPHOS METHYLO
(Col. = 220°C)
METHYL PARATHIOND
(Col. = 200°C)
1 234567
NG METHYL PARATHION (xlO = NG, AZINPHOS ETHYL OR AZINPHOS METHYL)
FIGURE 13
Analytical curves of Azinphos Ethyl, Azinphos Methyl and Methyl Parathion
-------
inon, Dursban, phorate, parathion, methyl parathion, malathion,
demeton, dimethoate, azinphos ethyl, phosphamidon, azinphos
methyl and Azodrin.
table 10 presents RRTs and RPAs of twelve OP pesticides on
1.5% 0V-17/1.951 OV-210 column. The overlapping peaks are:
demeton and diazinon; dimethoate and Azodrin; Dursban and methyl
parathion. Their decreasing degree of sensitivities are:
methyl parathion, Dursban, phorate, azinphos methyl, malathion,
parathion, dimethoate, azinphos ethyl, demeton, diazinon,
Azodrin and phosphamidon.
Table 11 presents RRTs and RPAs of twelve OP pesticides
on 5% OV-210 column. The overlapping peaks are: phorate and
diazinon; malathion and methyl parathion. Their decreasing
degree of sensitivities are: diazinon, Dursban, parathion,
methyl parathion, demeton, malathion, phorate, dimethoate, azin-
phos methyl, azinphos ethyl, phosphamidon and Azodrin.
It was uncertain whether the minor peak for phosphamidon
was an impurity or derived from degradation on the GLC column
(Tables 9-10). However, the reproducibility of the major peak
was good and only this peak was used for recovery computation.
All of the analytical curves for the organophosphate pesticides
(Figures 8-13) exhibit good linearity; and intercept the point
of origin except for phosphamidon (Figure 9), suggesting in-
creasing loss of phosphamidon when larger amounts were injected.
3. Hall electrolytic conductivity detector. Table 12
presents retention times and response values relative to atra-
zine of some nitrogen-containing pesticides. Most of the
47
-------
tested pesticides did not give acceptable GLC peaks, which could
be due to the unsuitable GLC column or thermal decomposition of
pesticides. 28% Pennwalt 223/4% KOH and 5% Poly A-135 on 80/100
mesh Gas-Chrom R, which are used primarily for the separation
of simple amines or anilines, were tested for these nitrogen-
containing pesticides. Unfortunately, no GLC peaks from any of
the tested pesticides were observed (column temperature 160°-
180°C). It might be worthwhile to note that benomyl produced
a peak which appeared immediately after the hexane solvent peak
(5% SE-30 column). This peak might not be the intact benomyl
peak because one would not expect it to have such a short re-
tention time. It is assumed that benomyl may have undergone
thermal decomposition to give butylisocyanate.
Some solvents such as ethylacetate, methylene chloride or
other halogen-containing compounds would severely contaminate
the Hall detection system and consequently create noisy base-
lines. Hexane seems to be the best solvent for this detector;
however, it was observed that a periodic injection of pyridine
(1 ul) corrected the noisy baseline. Pyridine probably neutra-
lized the excess acids in the GLC system generated by repet-
itive sample injections.
Derivatization of pesticides
Derivatization of nitrogen-containing pesticides was at-
tempted by using pentafluoropropionic anhydride (PFPA) (Tables
13-14) and 2,5-dichlorobenzenesulfonyl chloride (DBSC) (Table
15). Most of the pesticides which have a proton attached to
the nitrogen and an aromatic structure gave positive results
48
-------
Table 12.
Retention times and response values relative to atrazine, of Nitrogen-
containing pesticides on 5% SE-30 column, 180° C CHall detector, att. =
1 x 10; furnace, 820° C; Ho flow, 33 ml/min; cell flow, 0.5 ml/min; He
flow, 67 ml/min; inlet, 225° C; transfer line, 280° C).
Pesticide
Atrazine
CIPC
Carbofuran
Simazine
IPC
Zectran
RRT
1 (1.87 min)
0.76
1.00
1.05
0.333
0.28, 1.38
RPA Remarks
1 9 mm^/ng, 141 Full scale response
0.206
0.035
4.200
2.600
0.48, 0.30 two peaks
DNOC (500 ng)
Temik (1 yg)
Benomyl (1 yg)
Avadex (1 yg)
Bux (1 yg)
Amitrole (1 yg)
Azodrin (1 yg)
Monuron (1 yg)
no peak
inconsistent tailing peak
no peak except enlarged solvent peak
noisy multiple peaks
multiple tailing peaks
no peak
multiple peaks
inconsistent tailing peaks
-------
Table 13. Retention times and response values relative to aldrin of PFPA derivatized
pesticides on two columns (Packard 7820, EC).
tn
o
10% DC-200; 195° C, 60 ml/min
Pesticide
IPC
CIPC
Simazine
Atrazine
Monuron
Furadan
Bux
Zectran
Aldrin
Att. 3
RRT1
•*>
0.08
0.16
0.26
0.28
0.28
0.29
0.34, 0.41
0.38
1 (13 min)
x 10"9
RPA2
0.230
0.088
0.640
0.240
0.087
0.158
0.167, 0.0206
0.0111
1.0000
10% QF-1; 195° C, 60 ml/min
Att.
RRT1
0.31
0.38
0.62
0.62
1.00
0.62
0.54, 0.69
0.62
1 (2.5 min)
1 x 10"9
RPA2
0.159
0.0336
0.346
0.236
0.0727
0.150
0.184, 0.0455
0.018
1.000
152
See footnotes of Table 2.
The pesticides which were also derivatized and produced no responses were Avadex
Benomyl, Bromacil, Difolatan, Amitrole, Diquat, DNOC, Ferbam, Temik and Azodrin
(PFPA of Azodrin was also checked by FPD, no response was observed).
-------
Table 14. Retention times and response values relative to aldrin of PFPA derivatized
pesticides on two columns (Tracor 222, EC, Att. 4 x 102, xolumn at 195°C).
Pesticide
Aldrin
2,3, 5-Landrin
3,4,5-Landrin
2-AB3
DNBP
Dicamba
4% SE-30/6%
N0 flow = 6
RRT1
1.000
0.350
0.425
UD
UD
UD
OV-210
7 ml/min
RPA2
1.000
0.063
0.030
—
—
—
1.51
No
RRTl
1.000
0.270
0.346
—
—
—
OV-17/1.95% OV-210
flow =53 ml/min
RPA2
1.000
0.218
0.087
—
—
—
See footnotes of Table 2.
2-AB = 2-aminobenzimidazole, a degradation product of benomyl.
UD = underivatizable.
-------
Table 15. Retention times and response values relative to aldrin, of 2, 5-dichloro-
benzenesulfonyl chloride derivatized pesticides (Pyridine system) on two
GLC columns (Tracer 222, EC,-Att. 4 x
Z
en
tsJ
Pesticide
Carbofuran
Temik3
DNOC
Monuron
Alpha-naphthol
Bux
Aldrin
41 SE-30/6%
220° C, 60
RRT1
*r
0.50
0.69
0.75
1.00
4.00
5.00, 6.00 0
1 (1.25 min)
OV-210
ml/min
RPA2
0.013
0.040
0.400
0.067
0.040
.020, 0.003
1.000
1.51 OV-1
220° C
RRT1
3.75
0.75
0.75
0.50
6.00
3.00, 4.00
1 (0.63 min)
7/1.95% OV-210
, 60 ml/min
RPA2
0.014
0.051
0.700
0.030
0.034
0.01.5, 0.008
1.000
See footnotes of Table 2.
Similar results were obtained for methomyl.
When carbaryl was derivatized in buffer pH 12 system, peaks of RRT 4.0 and 0.69
appeared.
The pesticides which were also derivatized and produced no responses were Amitrole,
Avadex, Benomyl, Azodrin, Simazine, Atrazine, Bromacil, CIPC, IPC, Zectran
Seviri and Difolatan.
-------
(Table 13-14). Linear analytical curves are presented in Figures
14 and 15. The thermal instability of N-methylcarbamates and
their reaction with acylating agents were briefly discussed by
Khalifa and Mumma (5). Damico and Benson (14) postulated that
the proton attached to nitrogen is readily transferable to the
phenolic oxygen when the carbamate is subjected to elevated temp-
eratures or under mass spectroscopic conditions. When the proton
attached to the nitrogen is replaced, the products have greater
thermal stability (15).
Moye's original derivatization techniques (13) of DBSC with
carbamates were modified by substituting the pH 12 buffer with
a pyridine system (Table 15). This relatively high pH system
may split- the intact carbamates before they react with DBSC.
From comparison of the results obtained from carbaryl and alpha-
naphthol in these two reaction systems, one might be able to
postulate the reaction mechanisms of carbaryl and other similar
carbamates with DBSC. In the pH 12 buffer system, carbaryl was
split into two reactive parts, namely alpha-naphthol and methyl-
amine or methylisocyanate which subsequently reacted with DBSC
to yield l-naphthyl--2,5-dichlorobenzene sulfonamide (RRT = 4.00)
and N-methyl 2,5-dichlorobenzene sulfonamide (RRT = 0.69) re-
spectively (Table 15). In either the pH 12 or pyridine system,
l-naphthyl-2,5-dichlorobenzenesulfonate (RRT = 4.00) was ob-
tained when alpha-naphthol was substituted for carbaryl. In the
pyridine system, carbaryl is likely to stay intact because no
GLC peaks were produced. However, DBSC derivatives of Temik
CRRT = 0.69) and methomyl (RRT = 0.69) in the pyridine system
53
-------
en
H
5
HH
w
w
DH
280.
240 .
200 .
160.
120.
80
40
SE-30/6% OV-210, 160°C. EC DETECTOR
ATT. = 2 x 10
0.02 0.04 0.06
NG PFPA-IPC
0.08
FIGURE 14
Analytical curve of PFPA-IPC.
0.10
-------
tn
tn
120
f 10°
v_/
s 80 '
I—1
K 60 .
w
^ 40 .
20 •
4% SE-30/6% OV-210, 205°C,?EC DETECTOR,
ATT. = 4 x 10^
LANDRIN-3,4,5
FURADAN
BUX
LANDRIN-2,3,5
ZECTRAN
0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10
NG PFPA-DERIVATIZED PESTICIDES
FIGURE 15
Analytical curves of Furadan, Bux, Zectran, Landrins (PFPA)
-------
both gave identical RRT peaks, and they correspond to N-methyl-
2,5-dichlorobenzene sulfonamide.
Retention times and response values relative to aldrin of
pentafluorobenzoyl chloride (PFBC) derivatized pesticides in
pyridine and O.I M NaOH systems are presented in Tables 16-18
and Tables 19-20 respectively. The best results obtained in the
pyridine system were for amitrole, benomyl and monuron, although
the resulting derivatives were far from perfect. One of the
serious short-comings was the interference of excess PFBC which
could not be eliminated by simple solvent partition. The PFBC
GLC peak fused together with the solvent peak and caused serious
tailing for about 15 min. Therefore, from Table 17, the short
RRT (0.316) of amitrole was masked by excess PFBC, although it
created no serious problem for the longer RRTs of benomyl (14.3)
and monuron (2.98). A disadvantage of PFBC-amitrole was the
tailing peak, which leads one to suspect its structural stabil-
ity under normal GLC ^conditions. PFBC-benomyl seemed to be
ideal for GLC detection except for its somewhat low sensitivity
(RPA = 0.077) and long RRT (14.3). PFBC-monuron offered the
best PFBC-derivative which could be applied to routine analyses.
Attractive qualities were seen in the moderately long RRT
(2.98) which separated it from most of the soil impurities and
the good EC-GLC response (RPA = 0.73). The non-reactivity of
PFBC with most soil impurities made it even more attractive
for application to routine residue analysis. The linear ana-
lytical curves of PFBC-derivatives of amitrole, benomyl and
monurpn are presented in Figures 16-18.
56
-------
Table 16.
Retention times and response values relative to
aldrin of PFBC derivatized pesticides (pyridine
system! on 10% DC-200 column, 205°C (Att. =
4 x 10 , N2 flow - 80 ml/min. Tracer 222, EC).
Pesticide
Amitrole
Avadex
Azodrin
Benomyl
Bux
DNOC
Carbofuran
IPC
Monuron
Temik
Zectran
RRT
0.188
UD2 (2.97, 0.37, 0.56)1
1.00
10.400
0.375, 0.690
UD2
0.580
0.500
1.310
UD2
0.810
RPA
1.960
(0.033)1
0.018
0.110
0.004, 0.005
—
0.024
0.003
1.07
—
0.003
Data represents original structure of avadex, only major
peak (underlining) was computed for RPA.
UD = Underivatizable
5MBC and 2-AB PFBC-derivatives give similar RRT.
57
-------
Table 17.
Retention times and response values relative to
aldrin of PFBC derivatized pesticides (pyridine
system) on 4% SE-30/6% OV-210 column, 205° C,
(Att. = 4 x 10 , N9 flow = 80 ml/min. Tracor
222, EC). Z
Pesticide
I
Amitrole
Avadex
Azodrin
Benomyl
Bux
Dicamba
DNBP
DNOC
Carbofuran
IPC
2,3 , 5-Landrin
3,4, 5-Landrin
Monuron
Met homy 1
Phosphamidon
Temik
Zectran
RRT
0.316
UD (4.404, 0.456, 0.670)1
0.350
14.300
0.825
UD
UD
UD
0.825
UD
0.705
0.872
2.980
UD
UD
UD
1.020
RPA
1.235
(0.027)2
0.012
0.077
0.004
—
—
—
0.001
—
0.008
0.023
0.730
—
—
—
0.001
(underlining) was computed for RPA.
; maj
'MBC and 2-AB PFBC- derivatives gave similar RRT.
58
-------
Table 18.
Retention times and response values relative to
aldrin of PFBC derivatized pesticides (pyridine
system) on 1.5* OV-17/1.95% OV-210 column, 205°
C (Att. 4 x 10 , N9 flow = 80 ml/min. Tracor
222, EC). L
Pestic^e
Amitrole
Avadex
Azodrin
2
Benomyl
Bux
Dicamba
DNBP
DNOC
Carbofuran
IPC
2,3,5-Landrin
3,4,5-Landrin
Monuron
Me thorny 1
Phosphamidon
Temik
Zectran
RRT
0.522
i
UD (0.24, 0.57, 0.74)1
—
19.00
0.784
UD
UD
; UD
0.74
0.61
0.523
0.670
2.65
UD
UD
UD
0.822
RPA
0.072
i
(0.041)1
—
0.196
0.001
—
—
—
0.055
0.001
0.010
0.039
0.480
—
—
—
0.014
Data represents original -structure of avadex; only major
peak (underlining) was computed for RPA.
2MBC and 2-AB PFBC-derivatives gave similar RRT.
59
-------
Table 19.
Retention times and response values relative to
aldrin of PFBC derivatized pesticides (0.1 M
NaOH system) on 1.5% OV-17/1.95% OV-210 column,
205° C (Att. = 4 x 10 , N7 flow = 80 ml/min.
Tracer 222, EC). L
Pesticide
RRT
RPA
Amitrole
Avadex
Azodrin
Benomyl
Bux
Dicamba
DNBP
DNOC
Garbofuran
IPC
2,3,5-Landrin
3,4,5-Landrin
Monuron
Methomy1
Phosphamidon
Temik
Zectran
UD
UD
UD
UD
UD
UD
UD
UD
UD
0.792
0.580
0.700
1.57, 2.55
UD
UD
UD
13.25
0.017
0.340
0.445
0.001, 0.005
0.035
60
-------
Table 20.
Retention times and response values relative to
aldrin of PFBC derivatized pesticides (0.1 M
NaOH system) on 4% SE-30/6% OV-210 column, 205°
C (Att. = 4 x 10% N9 flow = 80 ml/min. Tracer
222, EC). L
Pesticide
RRT
RPA
Amitrole
Avadex
Azodrin
Benomyl
Bux
Dicamba
DNBP
DNOC
Garbofuran
IPC
2,3,5-Landrin
3,4,5-Landrin
Monuron
Methomyl
Phosphamidon
Temik
Zectran
UD
UD
UD
UD
UD
UD
UD
UD
UD
0.870
0.710
0.840
1.76, 5.00
UD
UD
UD
19.4
0.016
0.212
0.181
0.001, 0.004
0.031
61
-------
In the 0.1 M NaOH system, no responses were obtained from
amitrole, benomyl and most other pesticides (Table 19). Two
low response GLC peaks were obtained from monuron which might
be the result of its thermal degradation. Moderate sensitiv-
ities (RPAs 0.2-0.4) were obtained for Landrin (2,3,4 § 3,4,5)
(Tables 19-20).
Retention times and response values relative to aldrin of
a-bromo-2,3,4,5,6 -pentafluorotoluene (BPFT) derivatized pest-
icides on two columns are presented in Table 21. The best re-
sult was obtained for dicamba (RPAs = 0.7-0.9). DNBP and
monuron gave moderate sensitivities (RPAs = 0.13-0.20) and un-
acceptably low sensitivities were obtained from other BPFT
derivatized pesticides (RPAs = 0.03-0.09).
Derivatization of DNOC, DNBP, and dicamba was attempted
successfully by using diazomethane (Table 22). The relatively
simple derivatization procedure (18) and clean reagent blank
A
make it the most promising technique for the above pesticides.
In addition, methylated derivatives of these compounds gave
relatively high sensitivities, linear analytical curves, and
good separation of GLC peaks (Figure 19 and Table 22).
Retention times and responses relative to aldrin of
selected derivatization techniques for twelve multi-class
pesticides are compiled from previous results in Table 23.
Among the three selected derivatization techniques (PFPA, DAM
and PFBC), PFPA was found to be the most suitable derivatization
reagent for most of the tested carbamate pesticides (bux, car-
bofuran, IPC, Landrin -2,3,5, Landrin -3,4,5 and Zectran).
62
-------
w
100 .
80 •
60
40
20
1.5% OV-17/1.95% OV-210, 180°C,
EC DETECTOR, ATT. 2 x 102.
0.1
0.2 0.3
NG, PFBC-AMITROLE
FIGURE 16
0.4
0.5
Analytical curve of PFBC-Amitrole
-------
240.
200.
1.5% OV-17/1.95% OV-210, 230°C,
EC DETECTOR, ATT. 2 x 102
w
W
160
120'
80'
40-
1234
NG PFBC-BENOMYL
FIGURE 17
Analytical curve of PFBC-Benomyl
-------
Ol
60
50
40-
s 30
W
20
10.
4% SE-30/6% OV-210, 205°C, EC DETECTOR,
ATT. = 4 x 102
0.1
0.2 0.3
NG PFBC-MONURON
0.4
FIGURE 18
Analytical curve of PFBC-Monuron
0.5
-------
120
H
ac
w
ac
4% SE-30/6% OV-210, 205°C,
EC DETECTOR, ATT. 4 x 10
0.01 0.02 0.03 > 0.04
NG DICAMBA (X2 = NG, DNOC OR DNBP)
FIGURE 19
0.05
Analytical curves of Dicamba, DNOC and DNBP (Diazomethane Derivatives)
-------
Table 21.
Retention times and response values relative to aldrin
of BPFT derivatized pesticides on two columns (Att. =
4 x 10 , Tracer 222, column temp. 195°C, EC, pH 12
buffer system).
Pesticide
Azodrin
Dicamba
DNBP
DNOC
3,4,5-Landrin
Monuron
Methomyl
Temik
4% SE-30/6%
N2 flow = 6
RRT
2.800
1.560
6.175
5.350
3.975
0.740
4.025
4.025
OV-210
7 ml/min
RPA
0.028
0.770
0.150
0.059
0.069
0.128
0.086
0.053
1.5% OV-17/1.
N2 flow = 53
RRT
2.100
1.540
4.730
4.500
2.770
0.690
2.830
2.770
95% OV-210
ml/min
RPA
0.041
0.950
0.156
0.110
0.098
0.192
0.098
0.098
67
-------
Table 22.
Retention times and response values relative
to aldrin of Diazomethane-derivatized pest-
icides on 4% SE-30/6% OV-210 column, 205°C
(Att. = 4 x 10 , N9 flow = 67 ml/min, Tracer
222, EC). £
Pesticide
DNOC
DNBP
Dicamba
RRT1
0.900
1.300
0.300
RPA2
0.230
0.190
0.610
* See footnotes of Table 2.
Note: Diazomethane-derivatized pesticides which produced
no GC responses were 2,3,5-Landrin, carbofuran,
Zectran, and Bux.
68
-------
Table 23. Retention times and response values relative to
aldrin of selected derivatization techniques on
twelve multi-class pesticides.
Pesticide RRT1 RPA2 Analytical Curve3
(Figure No.)
PFPA Derivatization (Ref.
IPC
Landrin-2,3,5
Landrin-3,4,5
Bux
Carbofuran
Zectran
0.142
0.321
0.420
0.43, 0.54
0.462
0.530
I)4
0. 286
0.100
0.290
0.142, 0.047
0.148
0.105
15
16
16
16
16
16
DAM Derivatization (Ref. 5)4
Dicamba 0.300 0.368 14
DNOC 0.900 0.230 14
DNBP 1.300 0.190 ' 14
PFBC Derivatization, pyridine system (Ref. 5)4
Amitrole
Monuron
Benomyl
0.522
2.980
12.500
0.072
0.730
0.033
17
19
18
162
q See footnotes of Table 2.
3
See Figures for GLC operating conditons.
4
PFPA = Pentafluoropropionic anhydride; DAM = Diazomethane;
PFBC = Pentafluorobenzoyl chloride.
69
-------
PFPA-derivatives of the above carbamates gave relatively high
sensitivities with the EC detector (RPA between 0.10 and 0.29,
Table 23) and also demonstrated a high degree of reproducibility
along with relatively linear analytical curves (Figures 14 and
15). A disadvantage of PFPA-derivatized pesticides over other
methods (DAM and PFBC) was their relatively short retention
times (RRT = 0.14-0.53, Table 23), and they were eluted from GLC
columns too close together, although they were slightly dif-
ferent in RRT and could ultimately be differentiated from each
other. Problems were encountered when these carbamate pest-
icides were studied for low level (0.1 and 0.01 ppm) soil re-
covery; this will be discussed in detail later.
Negative derivatization results were obtained with the fol-
lowing reagents: 9-chloromethylanthracene (CMA), acetic acid
N-hydroxy succinimide ester (AAHSE), N-methyl-bis-(heptafluor-
obutyryl) amide (MHFBA), and N-methyl-bis-(trifluoroacetamide)
(MBTFA). «
GLC Detection of Sulfur-Containing Pesticides
None of the tested reagents produced suitable derivatives
of Temik or methomyl for either the sulfur mode of the FPD or
the EC detector. However, the non-specific methylamine deriva-
tization of Temik and methomyl by DBSC was possible as pre-
viously mentioned. Figure 20 presents analytical curves of
Temik and methomyl which were obtained by the utilization of
ir>
the multi-residue, 4% SE-30/6% OV-210 (130°C) column and de-
tected by the sulfur mode of the FPD. Lack of sensitivity on
the above column led to the selection of a_ second rather polar
70
-------
column (24), 5% Carbowax 20 M, (Figure 21), on which sensitiv-
ities of Temik and methomyl could be increased two and fifteen
fold respectively. The analytical scheme was based on the ther-
mal degradation of Temik or methomyl to nitrile derivatives at
an injection port temperature of 300°C (24, 26). Sulfur dioxide
added to the hydrogen gas of the FPD increased the sensitivity
of Temik and methomyl by a factor of 5 (Figure 21) . Zehner
and Simonaitis (25) claimed an 8-fold increase in the sensi-
tivity of Temik and malathion when S02 was introduced.
Loss of Pesticides in the Concentration Steps (Kuderna-Danish
concentrator or evaporation of solvents under room temperature
in hood)
The stability and recovery of nitrogen-containing pesti-
cides in the Kuderna-Danish concentrator step were briefly in-
vestigated (Table 24) before any soil residue studies were
undertaken. Only captan and methomyl were lost in unaccep-
table amounts (73 and 84% respectively) in the Kuderna-Danish
concentration step. The interaction between methanol and cap-
tan under elevated temperature was believed to be the major
factor contributing to the disappearance of captan. Surpris-
ingly, methomyl could be recovered quantitatively after it was
heated at 95°C in 1:1 benzene:methanol for 30 min (see footnote
of Table 24) in a sealed test tube.
Paraffin oil acted as a useful residue keeper as demon-
strated in Table 25. Thirty-five and 22 percent of Avadex could
be kept by 1 ml of 1% paraffin oil in benzene when the dry res-
idue was subjected to continuous evaporation under a hood for
15 min and 16 hr respectively. Fifty percent of Zectran could
71
-------
--J
tsl
E
O
I-H
W
160 -
140 -
120 •
100 •
80
60
40
201
100 200 300 400 500 600
NG, TEMIK, w/ and w/o S02> NG METHOMYL w/ S02 (xlO = NG, METHOMYL w/o S02)
FIGURE 20
Analytical curves of Temik and Methomyl.
SE-30/6% OV-210ifi130°C,
ATT. = 8 x 10
(FPD, SULFUR MODE)
TEMIK
w/ SO
METHOMYL
w/o SO
TEMIK
w/o S02
METHOMYL
w/ S02
-------
E-"
OS
200
160 -J
120
£ 80 J
40 •
5% Carbowax 20M, ATT. 16 x 10
(FPD, SULFUR MODE)
-9
METHOMYL
(col. 200°C,
w/S02)
TEMIK
(col. 120°C
w/S02)
TEMIK
(col. 120°C,
w/o S02)
METHOMYL
(col. 200°C, w/o S02)
50 100 150 200 250 300 350 400
NG, TEMIK OR METHOMYL
FIGURE 21
Analytical curves of Temik and Methorny1.
-------
be preserved by paraffin oil even when its dry residue was sub-
jected to continued evaporation under a hood for 48 hr.
Table 26 and Figure 22 present the degradation of seven
carbamate and substituted urea pesticides in benzene/methanol
(1:1) for 6 hr at elevated temperature (100°). The loss ranged
from 15% for Landrin - 3,4,5 to 83% for carbofuran.
Silica gel column chromatography
Pesticide standards were studied for recoveries and elution
patterns from a silica gel column prior to soil recovery studies
(Figures 23 a and 23 b). Halogenated and organophosphate pest-
icides outlined in Figure 23 a were recovered quantitatively
(79-105%) in a single fraction. Recoveries of 79 to 90% were
obtained for endosulfan, Avadex, Dursban, phorate, methyl para-
thion, parathion, CIPC, diazinon, demeton, dimethoate, phospham-
idon, and Azodrin. PCNB, trifluralin, Perthane, methoxychlor,
captan, folpet, Difolatan, chlorobenzilate, malathion, azinphos
methyl, azinphos ethyl, atrazine, simazine and bromacil were
recovered at 90 to 105%.
Elution patterns of multi-class pesticides are outlined in
Figure 23 b. Pesticides which could be recovered from the
column quantitatively (over 75%) in a single fraction were DNBP,
benomyl, DNOC, bux, IPC, Zectran, carbofuran, monuron, and
amitrole. Pesticides which could be recovered quantitatively
(over 75%) from more than one elution fraction were 2,3,5-Landrin
(46% fraction III; 38% fraction IV), and 3,4,5-Landrin (62%
fraction III; 20% fraction IV). The dicamba standard was lost
on the silica gel column; however, DAM dicamba could be eluted
74
-------
, m
Table 24. Percent recovery of 500 yg of pesticides each in the
Kuderna-Danish concentrator step concentrated from
250 ml 1:1 benzene/methanol to 10 ml.
Pesticide
Avadex
Captan
Bux
Carbofuran
Zectran
Monuron
Temik
?
Methomyl
Detection Techniques
EC,
EC,
EC,
EC,
EC,
EC,
FPD,
FPD,
non-derivatized
non-derivatized
PFPA4
PFPA4
PFPA4
PFBC
S-mode
S-mode
% Recovery
85
27
95
97
93
96
103
16
Sample of avadex standard (3 g) in 1 ml of 1:1 benzene/methanol
was heated at 95° C for 48 hrs in a sealed Teflon-capped
test tube, 86% of the heated Avadex was recovered; while
samples of captan, folpet and Difolatan underwent the same
treatment, no recoverable residues were observed.
2
Methomyl was recovered quantitatively when it was heated at 95°C
for 30 min in a sealed test tube in 1:1 benzene/methanol.
3
See Figure 21 for GLC operating conditions.
4
Samples of pesticides should be free from methanol before PFPA
derivatization, since methanol inhibited PFPA-pesticide
reaction.
75
-------
Table 25. Effect of paraffin oil keeper on pesticide reten-
tion.
Pesticide Time after complete % pesticide recovered
(5 yg) dryness of residue paraffin ^ para££ijl2
Avadex 15 min 65 100
Avadex 16 hr 22 44
Zect'ran 48 hr 50 100
Avadex was selected as a model pesticide because it is the
most volatile pesticide under investigation.
2
One ml of 1% paraffin oil in benzene was added to each sample,
No interfering GLC peaks for paraffin oil were observed.
76
-------
Table 26.
Recovery of 100 yg of nitrogen containing
pesticides after heating for 6 hr (100°C) in
a capped test tube with 2 ml benzene/methanol
(1:1).
Pesticide
IPC
Landrin, 2,3,5
Bux
Carbofuran
Landrin, 3,4,5
Zectran
Monuron
RRT
0.18
0.35
0.40, 0.45
0.41
0.42
0.51
0.55
2
% recovery
74.0
70.5
28.0
17.3
85.0
51.5
48.0
LGLC conditions: 4% SE-30/6% OV-210, 205°C, Att. = 2 x
10 , Tracer 222, EC.
'Average of duplicate samples.
77
-------
(a)
9,23,26
STD (PFPA)
0.04 ng ea,
(b)
A6 hr. 100°C
in 1:1 benzene/MeOH
) 5 10 15 20
MIN.
FIGURE 22
Chromatograms of standard PFPA-deriv-
atives of six carbamate pesticides (a)
and after they were heated at 100°C in
a sealed test tube for 6 hr in 1:1
benzene/MeOH (b). Conditions as in
Table 26.
78
-------
quantitatively (85%) in fraction II. PFBC-benomyl could be re-
covered in fraction II (801) and fraction III (15%), and PFBC-
amitrole was eluted in fraction IV (50%) and fraction V (48%).
Interference due to the PFBC reagent were removed by the silica
gel column, The major percentage (-70%) of the unreacted PFBC
reagent was observed in fraction I and some (~25%) was carried
over to fraction II. All residue keeper (paraffin oil) was
eluted in fraction I from the silica gel column. Fairly good
yield (84%) of methomyl was observed while a lower yield (66%)
was obtained for Temik.
Co-extractive interferences from crude soil extracts and clean-
up by silica gel column chromatography.
Most gas chromatograms were derived from the multiple re-
coveries of those forty pesticides (Table 1) in a single sandy
soil sample (0.01 ppm) by Soxhlet extraction. Relatively low
sensitivity pesticides which could not be presented in the same
chromatogram because of interferences from other more sensitive
pesticides are illustrated by broken line peaks.
1. Halogenated pesticides (EC). Chromatograms for a
crude control sandy soil extract and soil spiked with fourteen
EC-sensitive pesticides in five silica gel fractions and their
respective controls are presented in Figures 24a-24f.
Figure 24a presents the background interference of crude
benzene/methanol sandy soil extract for 10 mg soil equivalent.
Complete recovery of 0.01 ppm PCNB from sandy soil is seen in
Figure 24b (Fraction I, Hexane). In fraction II, Avadex,
Trifluralin, methoxychlor, and endosulfan were quantitatively
79
-------
Figure 23a. Outline of Halogenated and Orgariop'hosphate
Pesticide Standards (1 mg each) eluted from the
silica gel column (figures immediately after
pesticide indicate percent recoveries).
Fraction I Hexane
PCNB (94%)
Fraction II 60% Benzene/Hexane
trifluralin (105%) Dursban (85%)
Perthane (97%) phorate (79%)
methoxychlor (92%) methyl parathion (88%)
endosulfan (90%) parathion (87%)
Avadex (79%)
Fraction III 5% Acetonitrile/Benzene
CIPC (90%) diazinon (82%)
Silica gel captan (102%) malathion (95%)
(16 g)
folpet (98.5%) azinphos methyl (100%)
Na7SO.
flj ,4 Difolatan (100%) azinphos ethyl (94%)
chlorobenzilate (91%)
i
Fraction IV 10% Acetonitrile/Benzene
atrazine (100%) demeton (80%)
simazine (100%)
Fraction V 20% Acetonitrile/Benzene
bromacil (98%) dimethoate (81%)
phosphamidon (79%)
Fraction VI Acetonitrile
Azodrin (90%)
80
-------
Figure 23b. Outline of Multiclass pesticide standard (1 mg)
eluted from silica gel column (figures immedi-
ately after pesticide indicate percent recov-
eries) .
Fraction I Hexane
No pesticides (paraffin oil)
Fraction II 60% Benzene/Hexane
DNBP (83%) PFBC-benomyl (80%)
DAM-dicamba (85%)
Fraction III 5% Acetonitrile/Benzene
benomyl (75%) PFBC-benomyl (15%)
DNOC (83%) bux (95%)
2,5,4-Landrin (46%) IPC (92%)
Silica gel 3,4,5-Landrin (62%) Zectran (90%)
(16 g)
Fraction IV 10% Acetonitrile/Benzene
Na SO
H4 „} 2,5,5-Landrin (38%) Carbofuran (88%)
3,4,5-Landrin (20%) PFBC-amitrole (50%)
Temik (66%)
Fraction V 20% Acetonitrile/Benzene
methomyl (84%) PFBC-amitrole (48%)
monuron (95%)
Fraction VI Acetonitrile
amitrole (100%)
Unrecoverable
dicamba
Some pesticides-were eluted in more than one fraction (under-
lined) ; addition of those fractions represents total recov-
eries; recovery values were estimated by single analysis of
pesticide standards without addition of soil.
81
-------
recovered from the. sandy soil sample (91=110%) (Figure 24c) -
t
The chromatogram for Perthane was obtained from the recovery
study of the individual compound because endosulfan, when in
the same soil sample, masked the less sensitive Perthane peak. '
In fraction III captan, folpet, and Difolatan were re-
covered quantitatively (76-89%) using the Polytron extractor
with benzene/methanol (1:1) as the extracting solvent (Figure
24d). CIPC, fortified at 0.10 ppm, was partially obscured by
one sandy soil coextractive peak which prevented accurate es-
4
timation. Complete recovery (100%) was obtained for chloro-
benzilate added to a sandy soil sample which evidenced a minor
interference peak. (Figure 24d) .
Simazine and atrazine were both recovered in fraction IV
(Figure 24e). No GLC column under investigation could be used
to separate these two compounds, although they could be recovered
from soil quantitatively (83-97%). High recovery (95%) was ob-
tained for bromacil in fraction V (Figure 24f).
2. Organophosphate pesticides. No interference peaks from
the column blank, clean-up or crude control soil samples (500 mg
soil equivalent) were observed with the FPD (Figure 25a). In
fraction II (Figure 25a), recovery results were fair (66-77%)
for Dursban, methyl parathion, and parathion and poor (37%) for
phorate. In fractions III-VI, shown in Figures 25a and 25b,
poor recovery results (40-41%) were obtained for diazinon and
~4 .
demetqn. Good recoveries (76-100%) were obtained for malathion,
azinphos ethyl, and azinphos methyl (Fraction III); demethoate
and phosphamidon (Fraction V); and Azodrin (Fraction VI). Gas
82
-------
00
SE-30/6% 0V-210, 205°C
ATT. 2 x 10% EC
0
30
FIGURE 24 a
Gas chromatogram of crude control sandy soil extract (10
mg soil equivalent).
-------
FRACTION I
CONTROL
34
FRACTION I
PCNB, 0.01 ppm
0 5 10 15 20
MIN.
FIGURE 24 b
Gas chromatograms of control and 0.01
ppm PCNB from sandy soil (10 mg soil
equivalent). G.C conditions as for
Fig. 24 a.
84
-------
00
en
CONTROL
Fraction II 0.01 ppm
10
15
20
25
30
MIN.
FIGURE 24 c
Gas chromatograms of 0.01 ppm in sandy soil of Avadex (3),
trifluralin (59) , endosulfan (21) , methoxychlor (29), and
Perthane (35) and the respective control (50 mg soil equiv-
alent) . G7C. conditions as for Fig. 24 b.
-------
00
10
15
20
MIN.
25
30
FIGURE 24 d
Gas chromatograms of 0.01 ppm in sandy soil of CIPC (12)
captan (10), folpet (22) , chlorobenzilate (11) and DiTo-
latan (15J and the respective control (10 mg soil equiv-
alent) . G.C. conditions as for Fig. 24 a.
-------
00
Fraction IV, 0.01 ppm
0 5 10 15 20 25 30
MIN.
FIGURE 24 e
Gas Chromatograms of 0.01 ppm in sandy soil of atrazine
(2) and simazine (57) and the respective control (50 mg
soil equivalent). G.C. conditions as for Fig. 24 a.
-------
oo
00
CONTROL
Fraction V, 0.01 ppm
MIN.
FIGURE 24 £
Gas chromatograms of 0.01 ppm in sandy soil of bromacil
(8J and the respective control (50 mg soil equivalent).
G.C. conditions same as for Fig. 24 a.
-------
chromatograms of 12 OP pesticide standards are presented in
Figures 25c and 25d.
3. Derivatization of pesticides. Chromatograms of DAM-
control crude sandy soil sample (2 mg soil equivalent) and of
0.01 ppm of dicamba, DNOC and DNBP extracted from soil are
presented in Figure 26. Recoveries were 59%, 73% and 103% re-
spectively.
Figure 27 (a-d) presents the chromatograms of 0.1 or 1 ppm
of six carbamate pesticides recovered from sandy soil samples
by the Polytron extractor and clean up by a silica gel column.
Figures 27a-27c present the chromatograms and recoveries of
some carbamate pesticides eluted in fraction III (5% aceton-
itrile/benzene). Figure 27d represents the chromatograms for
control soil and soil fortified with 0.1 ppm and eluted in
fraction IV (20% acetonitrile/benzene). High background inter-
ferences were observed for all PFPA-derivatized carbamate pest-
icides.
Chromatograms of PFBC-derivatized monuron and benomyl re-
covered from sandy soil are presented in Figures 28a and 28b
respectively. Monuron (0.01 ppm) was recovered in relatively
low yield (50%) whereas benomyl could not be detected below
the 0.1 ppm level due to coextractive interference (Figure 28b),
although 90% recovery was obtained at the 0.1 ppm level.
Recoveries of Pesticides from Soils by Soxhlet Extraction
Soxhlet extraction for 12 hr exhibited superior efficiency
to 1 hr tumbling for recovery of phorate and parathion from soil
(Table 27). Therefore, most of the recovery studies were carried
89
-------
CONTROL (CRUDE EXTRACT)
Fraction II. 0.01 ppm
Fraction III. 0.01 ppm
0
30
35
MIN.
FIGURE 25 a
Top, control (500 mg soil equivalent)
middle, chromatogram of 0.01 ppm OP
pesticides in fraction II (60% B/H)
and, bottom, fraction III (5% A/B).
Peaks are for Phorate (36), Dursban
(20), methyl parathion (30), para-
thion (32), diazinon (14), malathion
(27), azinphos methyl (5), and azin-
phos ethyl (4)
for Fig. 24 a,
G.C. conditions as
90
-------
to
F. IV. 0.01 ppm
F. IV. 0.01 ppm
F. IV. 0.01 ppm
5 10 15 20
MIN.
FIGURE 25 b
Chromatograms of 0.01 ppm OP pesticide
in fraction IV (10% A/B), fraction V
(20% A/B) and fraction VI (100% A/B).
Peaks are demeton (13), dimethoate (17)
phosphamidon (33), and Azodrin (6).
G.C. conditions as in Fig. 24a.
91
-------
NJ
5 ng ea.
SE 30/6% 0V 210, 200°C
,-9
ATT. 15 x 10
FPD(P)
0 5 10 15 20 25 30
MIN.
FIGURE 25 c
Gas chromatograms of 6 FPD (P)-sensitive pesticide
standards. Peaks are demeton (13), phorate (36),
Azodrin (6), methyl parathion (30), parathion (32),
and azinphos ethyl (5).
-------
10
5. ng ea.
4% SE 30/61 0V 210, 200°C
,-9
ATT. 16 x 10
FPD(P)
o
f—I
»\
•^h
30
35
10 15 20
MIN.
FIGURE 25 d
Gas chromatograms of 6 FPD (P)-sensitive pesticide
standards. Peaks are diazinon (14), dimethoate (17),
Dursban (20), malathion (27), phosphamidon (33), and
azinphos ethyl (4).
-------
out by the Soxhlet extraction method using 1:1 benzene/methanol
as the extracting solvent. For those compounds unrecoverable by
the above method, alternative techniques were used (Polytron),
and their conditions were noted. Soil samples were screened,
air-dried and characterized (Table 28). Air dried soils were
adjusted to have 10% moisture content before the addition of
pesticide standards for subsequent recovery studies.
i
»1. Halogenated pesticides. Table 29 shows that recoveries
were above 70% for all compounds insiltyloam soil except bro-
macil (63% at 1 ppm), endosulfan (68% at 0.1 ppm) and triflur-
alin (50% at 0.01 ppm and 67% at 0.1 ppm).
Captan, folpet and Difolatan reacted with methanol during
Soxhlet extraction; this phenomena was proven by simply re-
fluxing these compounds in methanol for one hour. After this
treatment, no captan, folpet or Difolatan peaks could be
detected. When benzene-tumbling was substituted for Soxhlet
k «•
extraction of these compounds, the recoveries were found to
be 70 to 100%.
In sandy soil (Table 30), some low recovery results were
obtaimed from soil samples for bromacil (61.3% at 0.1 ppm and
56% at 1 ppm), endosulfan (63% at 0.1 ppm and 1 ppm), and tri-
fluralin (59.3, 50.0, 52.5% at 0.01, 0.10 and 1.0 ppm respec-
tively) . In sandy loam soil (Table 31), only trifluralin gave
low recovery results for 0.01, 0.1 and 1 ppm (47-49%). The
consistently low recovery of trifluralin from soils when the
pesticide was present below 10 ppm might be an indication that
trifluralin was tightly bound by soil particles. When the soil
94
-------
tn
y^ww—>v_^
CONTROL
4% SE-30/6%
OV-210, 205°C,
Att.=2xlO'
EC,
0.01 ppm
0.02 ng ea.
10
15
20
25
30
MIN.
FIGURE 26
Chromatograms of control crude sandy soil (2 mg soil
equivalent) and recoveries of 0.01 ppm of dicamba (15),
DNOC (18) , and DNBP (19) by diazomethane derivatization.
-------
FRACTION III
4% SE-30/6% OV-210,
,2
Att. 4x
EC.
180°C
CONTROL
FIGURE 27 a
Chromatograms, left, of 1 ppm Zectran
(4^T) , and the respective control, right,
1 ppm IPC (24) and its control (1 mg
soil equivalent), PFPA derivatives.
-------
FRACTION III
CONTROL
41 SE-30/6% OV-210,
Att. 4xl02. EC
205°C
BUX, 0.1 ppm
0 5 10 15 20 25 30
MIN.
FIGURE 27 b
Chromatograms of 0.1 ppm bux (9) and its respective con-
trol (1 mg soil equivalent), PFPA derivative.
-------
FRACTION III
CONTROL
LANDRIN, 2,3,5
1 ppm
LANDRIN, 3,4,5
1 ppm
0 5 10 15 20
MIN.
FIGURE 27 c
Gas chromatograms of Landrin (2,3,5
and 3,4,5 isomers) (25, 26) and the re-
spective control (1 mg soil equivalent),
PFPA derivatives. G.C. conditions as in
Fig. 27 b.
98
-------
CONTROL (PFPA)
FRACTION IV
CARBOFURAN
0.1 ppm
0 5 10 15 20
MIN.
FIGURE 27 d
Chromatograms of 0.1 ppm carbofuran
and its control (1 mg soil equivalent),
PFPA derivative. G.C. conditions as
in Fig. 27 b.
99
-------
FRACTION V
0.01 ppm
MONURON
5 10 15 20
MIN.
FIGURE 28 a
Chromatograms of PFBC-control (5 mg
soil equivalent) and recovery (83%)
of 0.01 ppm monuron from sandy soil
G.C. conditions as in Fig. 27 b.
100
-------
CONTROL
FRACTION III
0.1 ppm
BENOMYL
0
10
15
20
MIN.
FIGURE 28 b
25
30
Chromatograms of PFBC-control (10 mg soil equivalent)
and recovery (80%) of 0.1 ppm benomyl fronusandy soil
(1.5* OV-17/1.95 OV-210, 220°C Att. 2 x 10 .
-------
Table 27. Average percent recoveries of phorate and para-
thion from sandy soil which were spiked at 0.01,
0.10, 1.00 and 10.00 ppm. Comparison of tumbling
and soxhlet extraction methods (without cleanup,
FPD).
% recovery of pesticide from ppm spiked
Tumbling
Phorate
Parathion
Soxhlet
Phorate
Paration
0.01 ppm 0.10 ppm
86.6 46.6
81.0 71.3
96.6 94.0
93.3 99.0
1.00 ppm 10 ppm
28.0 62.7
37.6 67.3
69.0 90.3
73.6 90.6
Average percent recovery of triplicate samples.
102
-------
particles were treated wi,th trifluralin at 10 ppm, it was re-
covered from soil more readily (70.5-91.5% from all types of
soils). The low recoveries of bromacil and endosulfan from
sandy soil (Table 30) might be attributed to the higher per-
centage of organic carbon in the s^ndy soil.
2. Organophosphate pesticides. Average percent recov-
eries of seven organophosphate pesticides from crude sandy and
sandy loam soil extracts are presented in Tables 32 and 33 re-
spectively. Considerably lower recoveries were obtained for
demeton treated sandy loam soil (68-82%) and sandy soil samples
(60-73.3%). Average percent recoveries of other pesticides
ranged from 73.3% (1 ppm malathion) to 103% (0.10 ppm azinphos
ethyl) for sandy loam soil and 67% (0.01 ppm azinphos ethyl) to
102.6% (0.01 ppm methyl parathion) for sandy soil. When crude
silty loam soil extracts of organophosphate pesticides were
cleaned up by silica gel columns (Table 34), there were no
significantly different recovery results than those non-cleaned
up soil extracts (Tables 32 and 33). Therefore, it might be
concluded that silica gel retained negligible amounts of the
tested organophosphate pesticides. No major difference of per-
cent recoveries from various soil types was observed (Tables 32-
34).
Recoveries of some Organophosphate and Thermally Labile Nitrogen-
Containing Pesticides from Soil by Polytron Extraction and Com-
parison of Results with Soxhlet Extraction.
103
-------
Table 28. Soil Characterization'
Class and particle
size distribution (%)
Sand (2-0.05 mm)
very coarse (2-1 mm)
coarse (1-0.5 mm)
medium (0.5-0.2'S mm)
fine (0.25-0.10 mm)_
very fine (0.10-0.05 mm)
Silt (0.05-0.002 mm)
Clay (0.002 mm)
Sandy
84.0
0.1
3.7
31.0
42.5
6.7
12.0
4.0
Soil Type2
Sandy Loam
92.6
0.1
4.8
38.8
41.0
7.9
1.5
5.9
Silt Loam
54.2
0.0
0.1
1.5
33.0
19.6
24.3
19.9
Organic Carbon (percent)
2.64
0.23
0.33
pH
H20 (1:1)
CaCl2 (0.01 M, 1:2)
KC1 (1 N, 1:1)
4.8
3.7
3.5
5.2
4.4
4.2
4.7
4.3
3.5
Soil samples were analyzed by Soil Characterization Laboratory,
McCarty Hall, IFAS, University of Florida
'Sandy and sandy loam soil were collected from Gainesville,
Florida, silt loam soil was provided by EPA.
104
-------
Table 29. Average percent recoveries of halogenated pesticides
from silty loam soil which were spiked at 0.01, 0.10,
1.00, and 10.00 ppm [Soxhlet extraction except where
otherwiase indicated, with silica gel column clean-
up, EC detector].
Pesticide
Atrazine
Bromacil
Captan
Chlorobenzilate
CIPC
Difolatan3
Endosulfan
Folpet3
Methoxychlor
PCNB
Perthane
Simazine
Trifluralin
% recovery
0.01 ppm
87.3
79.7
100.0
ND2
84.0
82.0
94.7
100.0
108.0
73.0
85.0
95.0
50.0
of pesticide
0,10 ppm
79.0
85.8
70.3
90.3
70.3
80.3
68.0
92.7
86.3
84.7
87.0
82.3
67.0
from ppm
1.00 ppm
90.0
63.0
87.0
92.7
77.0
79.7
95.0
96.0
96.2
85.0
101.7
86.0
83.7
spiked
10 ppm
89.3
76.3
75.0
76.3
92.0
80.0
104.0
85.0
95.2
88.7
85.3
91.3
71.3
Average percent recovery of triplicate samples.
'ND = non-detectable.
These pesticides reacted with methanol when 1:1 Benzene/Methanol
was used for Soxhlet extraction; therefore, 200 ml of Benzene
was used with tumbling extraction for the spiked soil samples
for 1 hour.
105
-------
Table 30. Average percent recoveries of halogenated pesticides
from sandy soil which were spiked at 0.01, 0.10, 1.00
and 10.00 ppm [Soxhlet extraction with silica gel
column cleanup, EC detector].
% recovery of pesticide from ppm spiked
-v
Pesticide
Atrazine
Bromacil
Chlorobenzilate
CIPC
Endosulfan
Methoxychlor
PCNB
Perthane
Simazine
Trifluralin
0.01 ppm
ND
95.7
100.0
ND
78.3
82.7
69.0
77.3
91.0
.? 59.3
0.10 ppm
60.0
61.3
85.7
92.7
63.0
76.3
73.0
71.7
77.5
50.0
1.00 ppm
84.7
56.0
91.0
82.3
63.0
78.7
88.0
84.7
110.0
52.5
10 ppm
92.7
74.5
97.2
99.7
69.0
84.0
95.5
94.5
92.3
70.5
Average percent recovery of triplicate samples.
106
-------
Table 31. Average percent recoveries of halogenated pesticides
from sandy loam soil which were spiked at 0.01, 0.10,
1.00, and 10.00 ppm (Soxhlet extraction with silica
gel column cleanup, EC detector).
% recovery of pesticide from ppm spiked
Pesticide
Atrazine
Bromacil
Chlorobenzilate
CIPC
Endosulfan
Methoxychlor
PCNB
Perthane
Simazine
Trifluralin
0.01 ppm
83.7
96.3
99.7
106.0
81.3
99.0
70.0
96.0
97.0
47.0
0.10 ppm
72.7
88.0
93.3
88.0
77.0
83.3
76.0
75.7
79.0
54.0
1.00 ppm
92.3
91.0
96.0
61.0
83.0
89.7
77.0
77.0
87.2
49.7
10 ppm
95.0
93.0
80.5
83.0
86.0
91.0
92.5
87.5
97.0
91.5
Average percent recovery of triplicate samples.
107
-------
Carbofuran was selected for use in evaluating a number of
extraction parameters which included solvent type, ultrasonic
power, and extraction time. Benzene-methanol (1:1) was found to
be the best extracting solvent (Table 35). No significant dif-
ference was observed in the extraction of carbofuran from sandy
soil by varying extraction time (30 to 90 sec. Table 36). How-
ever, the best recovery of carbofuran was obtained when a max-
imum motor speed of the Polytron was applied (Table 37). Later
experience gained fr.om this investigation suggested use of
medium speed, because the Polytron generator head was being
ground away excessively by soil particles.
Average percent recoveries of sixteen multi-class pesticides
from sandy soil are presented in Table 38. Benzene-methanol
(1:1) was selected as the solvent for general extraction, al-
though methanol reacted with some pesticides during Soxhlet ex-
traction (Table 39) and subsequently yielded poor or no recov-
ery for carbamates, captan, folpet and Difolatan. Polytron ,
extraction (Table 38) apparently overcame the problem and gave
reasonably good recoveries for those pesticides at 0.01-0.1 ppm
levels (captan, 64-76%; Difolatan, 89-100%; folpet, 88-101%).
Other pesticides in Table 38 which could be detected at 0.01-
0.1 ppm levels were: Avadex (70-75%), bux (89%), dicamba (77-
86%),,DNBP (74-75%), DNOC (60-67%), carbofuran (100%) and mon-
uron (83-93%). Pesticides which could not be detected at 0.01-
0.1 ppm levels but were detectable at 1 and 10 ppm levels were
benomyl (61-69%), IPC (71-73%), Landrin -2,3,5 (80-81%), Landrin
-3,4,5 (73-82%) and Zectran (76-85%). The only pesticide which
108
-------
Table 32.
r
Average percent recoveries of organophosphate pest
icides from sandy soil which were spiked at 0.01,
0.10, 1.00, and 10.00 ppm (Soxhlet extraction with
out column cleanup, FPD detector).
Pesticide
Phorate
Parathion
Demeton
Malathion
Methyl parathion
Azinphos ethyl
2
Azinphos methyl
% recovery of pesticide from ppm spiked
0.01 ppm
96.6
93.3
60.0
70.0
102.6
67.0
.. 86.0
0 . 10 ppm
94.0
99.0
60.0
81.0
93.3
71.7
101.7
1.00 ppm
69.0
73.6
71.7
90.3
82.0
93.3
100.7
10 ppm
90.3
90.6
73.3
93.3
88.3
99.7
92.3
Average percent recovery of triplicate samples
'With silica gel column cleanup.
109
-------
Table 33. Average percent recoveries of organophosphate pest-
icides from sandy loam soil which were spiked at
0.01, 0.10, 1.00, and 10.00 ppm (Soxhlet extraction
without column cleanup, FPD detector).
Pesticide
Phorate
!
Parathion
Demeton
Malathion
Methyl parathion
Azinphos ethyl
Azinphos methyl
% recovery of pesticide from ppm spike
0.01 ppm
77.3
99.0
68.0
92.6
102.6
83.3
101.6
0.10 ppm
102.6
84.6
73.0
84.0
83.0
103.0
99.3
1.00 ppm
89.0
75.3
73.3
73.3
84.0
102.0
78.3
10 ppm
80.0
81.0
82.0
80.6
72.6
92.0
83.3
Average percent recovery of triplicate samples.
110
-------
Table 34.
Average percent recoveries of organophosphate pest-
icides from silty loam soil which were spiked at
0.01, 0.10, 1.00, and 10.00 ppm (Soxhlet extraction,
with silica gel column cleanup, FPD detector).
Pesticide
Phorate
Parathion
Demeton
Malathion
Methyl Parathion
Azinphos ethyl
Azinphos methyl
% recovery
0.01 ppm
75.0
93.0
69.3
95.7
101.0
72.0
77.7
of pesticide
0.10 ppm
81.0
101.0
72.3
96.7
91.7
90.3
91.0
from ppm
1.00 ppm
75.0
85.7
69.3
96.7
93.3
86.7
89.3
spike
10 ppm
71.7
86.0
71.3
101.0
91.0
87.3
78.3
Average percent recovery of triplicate samples.
Ill
-------
Table 35. Comparison of solvents for ultrasonic-polytron
extraction of carbofuran (1 ppm) from sandy
soil.
Solvent mean % recovery of duplicate samples
and deviation from mean
Acetone 71.5 ± 3.5
Acetonitrile 70.5 ±0.5
Methanol 83.0 ± 2.0
Hexane 24.8 ± 5.5
Benzene 74.3 ±3.8
Benzene-Methanol (1:1) 100 ± 0.0
GLC column and operating conditions: 4% SE-30/6% OV-210,
200°C, Att. = 4 x 10 (PFPA derivatization, EC).
Soil and solvent: 50 g air-dried sandy soil extracted by
100 ml x 2 of solvent and washed with 50 ml x 2.
Polytron: 1 min x 2, maximum speed.
112
-------
could not be detected even at the 10 ppm level, was amitrole.
Due to high water solubility of amitrole, most conventional
solvents used for extraction were water or an ammonium hydroxide-
glycol (5+20) mixture (26). Groves and Chough (26) suggested
that polar ppmpounds like amitrole may be bound in the clay-
organip matter ion exchange systems similar to the binding of
nutrient ions; that entrapment between atom-layers of expandable
lattice clays may occur; that insoluble compounds or complexes
may be formed; and that adsorption may occur due to electro-
static or residual valence forces. The undetectability of
benomyl at low residue levels (0.01-0.1 ppm) was attributed to
the low sensitivity of the PFBC derivative and the high back-
ground interference of the PFBC reagent as mentioned before.
IPC, Landrins, and Zectran could not be detected at low levels
(0.01-0.1 ppm) (Table 38) mainly due to the interference peaks
of the PFPA derivatives and their short retention times which
appeared in the same regions as most underivatized soil coex-
tractives.
The Soxhlet extraction (Table 39) procedure prevented the
complete recoveries of captan, folpet, Difolatan, carbamate and
substituted urea compounds due to pesticides-methanol inter-
action or thermal degradation under prolonged (12 hr) elevated
temperature. However, better recoveries were obtained with
Soxhlet extraction than with the Polytron for the following
heat stable pesticides: Avadex (73-100 vs. 70-75), dicamba (86-
97 vs. 77-86), DNBP (91-100 vs. 74-75), DNOC (99-104 vs. 60-67).
113
-------
Table 36. Effect of Polytron extraction time on recovery
of carbofuran from sandy soil.*
Extraction time, mean % recovery of duplicate
sec samples and deviation from mean
30 x 2 71.5 ± 0.5
60 x 2 77.0 ± 1.0
90 x 2 72.0 ± 2.0
*GLC, column and conditions: same as Table 34.
Soil and extraction solvent: 50 g air-dried soil extracted
by 100 x 2 ml Benzene-MeOH (1:1) and washed with 50 x
2 ml.
Polytron speed: 50% of maximum.
114
-------
Table 37. Effect of Polytron motor speed on recovery of
carbofuran from sandy soil.
Speed, % of
maximum
30
50
70
100
mean % recovery of duplicate
samples and deviation from mean
71.5 ±
71.5 ±
75.0 ±
100 ±
6.
6.
1.
0.
5
5
0
0
GLC, column and conditions: same as Table 34.
Soil and extraction solvent: same as Table 35.
Polytron extraction time: 1 min x 2.
115
-------
Table 38.
Average percent recoveries of pesticides from sandy
soil samples (triplicate) which were spiked at 0.01
and 0.10 ppm (Polytron extraction with silica gel
column cleanup, EC detector)=l
% recovery of pesticide from ppm spiked'
r co i. -H-.H4C
i T
Amitrole
Avadex
3
Benomyl
Bux
4
Cap tan
Dicamba
Difolatan4
DNBP5
DNOC5
Carbofuran
Folpet4
I PC
Landrin-2,3,5
Landrin-3,4,5
Monuron
Zectran
0.01 ppm
(undetectable)
70.0
(69.3)
undetectable
76.0
86.0
89.3
75.3
67.3
undetectable
88.3
(72.7)
(79.7)
(82.0)
83.0
(85.0)
0.10 ppm '
(undetectable)
75,0
(61.3)
89.5
64.3
77.0
100.0
74.0
60.0
100.0
101.0
(70.7)
(81.0)
(73.3)
93.0
(76.0)
See Tables 21 and 22 for RRT, RPA and GLC operating conditions
'Parentheses imply undetectable at 0.01 and 0.10 ppm levels,
subsequent attempts for 1 and 10 ppm were conducted and
recorded in parentheses respectively.
Amitrole and benomyl were derivatized before silica gel column
cleanup.
Solvents were evaporated by standing in a hood at room temp-
erature (other pesticides by Kuderna-Danish concentrator).
'Analyzed in crude extract.
116
-------
Table 39.
Average percent recoveries of pesticides from
sandy soil samples (triplicate) which were spiked
at 0.01, and 0.10 ppm (Soxhlet extraction, with
silica gel column cleanup, EC detector).
Pesticide
Amitrole
Avadex
Benomyl
Bux
Captan
Dicamba
Difolatan
DNBP
DNOC
Carbofuran
Folpet
IPC
Landrin-2,3,5
Landrin-3,4,5
Monuron
Zectran
1 recovery of pesticide
0.01 ppm
(undetectable)
100.0
undetectable
64.3
0.0
85.5
0.0
90.7
104.3
undetectable
0.0
(60.1)
(57.0)
(50.3)
50.3
30.0
from ppm spiked
0.10 ppm
(undetectable)
73.3
85.0
39.3
0.0
97.0
0.0
100.0
99.2
48.3
0.0
(65.0)
(63.3)
(64.3)
37.0
33.3
117
-------
In the Soxhlet extraction, benomyl was converted to Methyl N-(2-
benzimidazolyl) carbamate (MBC) and 2-aminobenzimidazole (2-AB)
more readily than in the Polytron procedure. The better
recoveries with the Soxhlet procedure (85% at 0.1 ppm vs. 61
69% at 1-10 ppm) reflected the detection of the degradation
products of benomyl, i.e., PFBC-MBC and PFBC-2AB.
Average percent recoveries of five organophosphate pest-
icides by Polytron and Soxhlet extraction from sandy soil are
presented in Tables 40 and 41 respectively. High percent re-
coveries were obtained for most pesticides (82-103%) by the
Polytron extraction except for methomyl and Temik (Table 40).
The inability to recover methomyl was partly due to loss during
Kuderna-Danish evaporation, whereas the disappearance of Temik
might be due to its strong binding with soil particles. There
was no evidence of any breakdown or alteration of Temik or
other pesticides by the Polytron extraction which agrees with
the findings of Johnsen and Starr (28).
The above indefinite conclusion led us to further inves-
tigate the disappearance of these two compounds. No accept-
able recovery results could be obtained for these compounds
by room temperature solvent evaporation or by mixing the con-
centrated control sandy soil sample with Temik and methomyl
standards. Pesticide standard peaks also disappeared when
they were injected simultaneously from the same syringe to-
gether with the same amount of control soil extract. All the
above evidence led us to believe that carbamoyl pesticides
reacted with soil coextractives to produce other compounds under
elevated temperature in the GLC system.
118
-------
Simultaneous analysis of forty multi-class pesticides in sandy
soil
Duplicate sandy soil samples were spiked with 0, 0.01 and
0.1 ppm of the forty multi-class pesticides listed in Table 1
and analyzed for their recoveries simultaneously. Table 42
presents the results of fourteen EC-sensitive pesticides. Ex-
cellent recoveries were obtained for most pesticides at 0.01
ppm level (90-116%) except for Perthane, chlorobenzilate,
captan, folpet and Difolatan. Perthane was masked by the more
responsive endosulfan while chlorobenzilate suffered from in-
terference by coextractives from sandy soil. Captan, folpet
and Difolatan reacted with methanol during the Soxhlet extrac-
tion procedure. At 0.1 ppm, all the halogenated pesticides
gave excellent recoveries (86-115%) except for atrazine and
simazine (72%).
Data in the last column of Tables 42, 43, and 44 are
derived from 0.1 ppm equivalent of pesticides (without soil)
and underwent the entire procedure used in the extraction of
pesticides from soil samples. The original purpose was to
determine whether low recovery results were caused by the ana-
lytical procedure itself or because of the soil binding. No
endosulfan could be recovered when it was heated in the Soxhlet
extractor without soil for 12 hours in contrast to 95% recovery
of the same amount of pesticide spiked in soil (0.1 ppm). Con-
siderably lower recoveries were obtained from non-soil samples
as compared to their corresponding soil samples, i.e., PCNB
(87 vs. 93%), trifluralin (73 vs. 86%), chlorobenzilate (80 vs.
119
-------
Table 40. Average percent recoveries of pesticides from
sandy soil samples (triplicate) which were
spiked at 0.01 and 0.10 ppm (Polytron extrac-
tion with silica gel column cleanup, FPD; P-
mode).
4
Pesticide
Azodrin
Diazinon
Dimethoate
Dursban
Met homy 1
Phosphamidon
Temik3
% recovery of pesticide
0.01 ppm
83.0
93.3
102.7
94.0
(undetectable)
88.7
(undetectable)
from ppm spiked
0.10 ppm
81.7
87-0
98.3
96.7
(0.0)
90.3
(13.0)
See Table 9 for RRT, RPA and GLC operating conditions.
7
See footnote of Table 38 for explanation.
FPD operated on S-mode, see Figure 21 for analytical curves.
Solvents were evaporated by Kuderna-Danish concentrator.
120
-------
Table 41.
Average, percent recoveries of pesticides from
sandy soil samples (triplicate) which were
spiked at 0.01 and 0.10 ppm (Soxhlet extrac-
tion with silica gel column cleanup, FPD; P-
mode).
Pesticide
Azodrin
Diazinon
Dimethoate
Dursban
Methomyl
Phosphamidon
Temik
% recovery of pesticide
0.01 ppm
91.0
61.0
91.0
66.0
0.0
76.0
0.0
from ppm spiked
0.10 ppm
98.0
99.5
102.2
104,0
0.0
100.0
0.0
121
-------
Table 42. Multi-residue analysis of fourteen EC-sensitive
pesticides in duplicate sandy soil samples which
were spiked at 0.01, and 0.1 ppm of forty multi-
class pesticides in Table 1 (Soxhlet extraction,
4% SE-30/6% OV-210, 205°C, Att. = 2 x 10Z, N2
flow = 67 ml/min, Tracor 222, EC).
Pesticides eluted in % recoveries from ppm spiked
silica *el fractions 0.01 ppm " 0.10 ppm (0.10 ppm)1
Hexane
PCNB 116 93 87
60% Benzene/Hexane
Avadex 110 115 107
Perthane undetectable 93 80
Trifluralin 91 86 73
Methoxychlor 107 88 88
Endosulfan 94 95 O2
5% Acetonitrile/benzene
CIPC " undetectable
Captan
underwent structural transformation
Folpet in the Soxhlet extracting solvents
(benzene/methanol, 1:1)
Difolatan
Chlorobenzilate undetectable 96 80
10% Acetonitrile/benzene
Atrazine
90 72 64
Simazine
122
-------
Table 42 continued.
Pesticides elutedin % recoveries from ppm spiked
silica gel fractions 0.01 ppm 0.10 ppm (0.10 ppm)
20% Acetpnitrile/benzene
Promacil 95 93 93
0.1 ppm equivalent of pesticides without soil underwent the
same extracting procedure (Soxhlet), evaporation
(Kuderna-Danish) and cleanup (silica gel).
2
Loss due to absence of Keeper.
123
-------
96%). Identical results were obtained for methoxychlor (88%)
i
and bromacil (93%).
Table 43 presents the results of multi-residue analysis of
twelve FPD-P sensitive pesticides in duplicate sandy soil samples
\
which were spiked at 0, 0.01, and 0.10 ppm of forty multi-class
pesticides. At 0.01 ppm, pesticides which were recovered above
70% wdre: methyl parathion (77%), parathion (76%), malathion
(100%) , dimethoate (91%), phosphamidon (76%) and Azodrin (91%).
Those recoveries below 70% were: phorate (37%), Dursban (66%),
diazinon (41%), and demeton (40%). At 0.1 ppm, all twelve OP
pesticides were recovered at more than 78% (79-100%). The 0.1
ppm of the non-soil samples (last column of Table 43) gave very
poor recoveries except for parathion (71%) and Azodrin (76%).
The low recoveries (5-53%) of other OP pesticides, i.e., phorate
(48%), methyl parathion (37%), diazinon (45%), malathion (32%)
demeton (5%) and dimethoate (53%), might be attributed to their
thermal degradation on the glass surfaces of the Soxhlet ex-
tractor in the absence of soil coextractives. Soil coextrac-
tives might act as protective agents in preventing pesticide
breakdown on the glass surface. It is already known in GLC that
regular injection of sample extracts or silanizing reagents will
help prevent the degradation of some relatively heat-labile
compounds.
'J'able 44 presents the results of multi-residue analysis of
twelve nitrogen containing pesticides in duplicate sandy soil
samples which were spiked at 0, 0.01 and 0.10 ppm. At 0.01 ppm,
»'
good recoveries were obtained from DNOC (73%) and DNBP (103%).
124
-------
Table 43. Multi-residue analysis of twelve FPD-P sensitive
pesticides in duplicate sandy soil samples which
were spiked at 0.01 and 0.10 ppm of forty multi-
class pesticides in Table 1 (Soxhlet extraction,
41 SE-30/6% OV-210, 200° C, Att. = 16 x 10 , N?
flow = 50 ml/min, Varian 1200, FPD-P).
Pesticides eluted in
silica gel fractions ,
60% benzene/hexane
Phorate
Dursban
Methyl parathion
Parathion
5% acetonitrile/benzene
Diazinon
Malathion
*
).01
37
66
77
76
41
100
recoveries from ppm
ppm 0.10 ppm
94
79
100
88
90
106
spiked
(0.10 ppm)1
48
72
37
71
45
32
10% acetonitrile/benzene
Demeton 40 86 2
20% acetonitrile/benzene
Dimethoate 91 98 53
100% acetonitrile/benzene
Phosphamidon 76 100 2
Azodrin 91 100 76
>2See footnotes of Table 42.
Note: Temik and methomyl were unrecoverable (detected by FPD-S).
125
-------
Table 44.
Multi-residue analysis of twelve nitrogen con-
taining pesticides in duplicate sandy soil
samples which were spiked at 0.01, and 0.10 ppm
of forty multiclass pesticides in Table 1 (Sox-
hlet extraction, 4% SE-30/6% OV-210, 205° C,
Att. = 2 x 10Z, N9 flow = 67 ml/min, Tra<
EC). L
Tracer 222,
Pesticides eluted in
silica gel fractions
% recoveries from ppm spiked
0.01 ppm
0.10 ppm (0.10 ppm)
Without column cleanup
Dicamba (DAM) 59
DNOC (DAM) 73
t
DNBP (DAM) 103
Benomyl (PFBC) undetectable
5% Acetonitrile/benzene
2,3,5-Landrin (PFPA)
3,4,5-Landrin (PFPA)
Zectran (PFPA)
IPC (PFPA)
Bux (PFPA)
20% Acetonitrile/benzene
Carbofuran (PFPA "
Monuraon (PFBC) 50
100% Acetonitrile/benzene
Amitrole (PFBC) "
103
112
117
90
undetectable
18
35
100
80
undetectable
52 38
undetectable
•See footnotes of Table 41.
126
-------
Poor recoveries were obtained for dicamba (59%) and monuron
(50%). Pesticides which were undetectable at all concentrations
were: Landrins, Zectran, IPC, bux, carbofuran, and amitrole.
Benomyl was recoverable at 0.1 ppm (90%) while monuron gave
poor results (52%) at this level. Excellent results were ob-
tained for dicamba (103%), DNOC (112%), and DNBP (117%). Poor
or no recoveries (below 38%) were observed for most of these
pesticides in the non-soil samples (last column of Table 43)
except for DNBP and benomyl.
Conclusion
Aside from a rapid deterioration of the Polytron generator
rotor and bearings, it offers a real advantage in extracting
multiclass pesticides from soil compared to either Soxhlet ex-
traction or tumbling. More specifically, of the forty com-
pounds studied, the following, which could be successfully ex-
tracted from soil with the Polytron, could not be extracted
with the Soxhlet due to degradation: CIPC, captan, folpet,
Difolatan, carbofuran, Zectran, bux and benomyl. Amitrole
could neither be extracted by Soxhlet nor Polytron. It was not
determined whether methomyl or Temik could be extracted by the
Polytron (see part II for confirmation of this) since they
could not be gas chromatographed in the presence of soil coex-
tractives. Dicamba could not be eluted from the silica gel
intact, however, it could be if it were methylated prior to
chromatography. DNOC also could not be eluted at residue levels,
127
-------
however, it could- be determined in the crude extract after
methylation. See Fig. 29 for a summary of the procedure util
izing the Polytron.
128
-------
Fraction I
50 ml hexane
EC PCNB
Fraction II
50 ml 60% B/H
EC trifluralin
Perthane
methoxychlor
endosulfan
Avadex
DNBP (DAM)3 .
dicamba (DAM)1
FPD
~Dursban
phorate
methyl parathion
parathion
Forty Multiclass Pesticides in Soil (50 g)
Polytron
100 ml benzene/methanol (I:l)(x2)
wash with 50 ml (x2)
benzene/methanol soil extract
evaporate to dryness
Silica gel chromatography: 16 g (14 g Na-SO,)
Fraction III
50 ml 5% A/B
EC
CIPC
captan
folpet
Difolatan
chlorobenzilate
benomyl (PEBC)2
DNOC (DAM)5
bux (PFPA),.
IPC (PFPA)4
Zectran (PFPA)4
2,3,5-Landrin (PFPA)4
3,4,5-Landrin (PFPA)4
FPD
diazinon
malathion
azinphos methyl
azinphos ethyl
Fraction IV
50 ml 10% A/B
atrazine
simazine
carbofuran
FPD
demeton
(PFPA)
Fraction VI
50 ml Acetonitrile
FPD
Azodrin
Fraction V
50 ml 20% A/B
EC
Erbmacil
monuron (PFBC)
FPD
phosphamidon
dimethoate
Figure 29.
Procedure for detection of forty pesticides
benzene).
in soils (A/B = Acetonitrile/
Derivative made prior to silica-gel cleanup.
^Detectable only at 1 ppm level.
^Analyzed in crude extract.
4Can only be analyzed at 1.0 ppm and higher.
5Detectable at 0.01 or 0.1 ppm unless otherwise specified.
-------
References (Part I)
1. Argauer, R. J., J. Agr. Food Chem. 17_, 888 (1969).
2. Bowman, M. C., Beroza, M. J., J.A.O.A.C. 50, 926 (1967).
3. Crosby, D. G., Bowers, J. B., J. Agr. Food Chem. 16, 839,
(1968).
4. Shafik, M.'l., Bradway, D., Biros, F- J., Enos, H. F., J.
Agr. Food Chem. 1JJ, 1174 (1970).
5. Khalifa, S., Mumma, R. 0., J. Agr. Food Chem. 20, 632,
(1972).
6. Mills, P. A., Onley, J. H., Gaither, R. A., J.A.O.A.C. 46,
186 (1963).
7. Johnson, L., J.A.O.A.C. 48, 668 (1965).
8. Porter, M. L., Young, S. J., Burke, J. A., J.A.O.A.C. 55,
1300 (1970).
9. Wessel, J. , J.A.O.A.C. 52^, 172 (1969).
10. Carr, R. L., J.A.O.A.C. S4> 525 (1971).
11. Chiba, M., Residue Reviews 50, 63 (1969).
12. Sherma, J., Shafik, T. M., Arch. Environ. Contain. Toxicol.
^, 55 (1975).
15. Moye, H. A., J. Agr. Food Chem. 23, 415 (1975).
14. Damico, J. N., Benson, W. R., J.A.O.A.C. 48, 344 (1965).
15. Fishbein, L. Zielinski, W. L., J. Chromatog. 20, 9 (1965).
16. Moye, H. A., J. Agr. Food Chem. 19, 452 (1971).
17. Kawahara, F. K., Anal. Chem. 40, 1009 (1968).
18. Yip, G., Howard, S. F., J.A.O.A.C. 51, 24 (1968).
130
-------
19. Koshy, K. T., Kaiser, D. G., Van Der Slik, A. L., J. Chrom.
Sci. 1^, 97 (1975).
20. Nash, R. G. , J.A.O.A.C. S7_, 1015 (1974).
21. Stewart, F. H. C., Aust. J. Chem. 15, 478 (1960).
22. Roth, M., Analy- Chem. 43_, 880 (1971).
23. Carey, W. F., Kenneth, H., J.A.O.A.C. 53, 1296 (1970).
24. Zehner, J. M., J. Chrom. Sci. 14, 348 (1976).
25. Knaak, J. B., Tallant, M. J., Sullivan, L. J., J. Agr. Food
Chem. 14_, 573 (1966) .
26. Groves, K., Chough, K. S., J. Agr. Food Chem. 19, 840 (1971)
27- Fung, K. K. H., J. Agr. Food Chem. 23.* 695 (1975).
28. Johnsen, R. E., Starr, R. I., J. Agr. Food Chem. 20, 48
(1972).
29. Brown, B., Goodman, J. E., "High Intensity Ultrasonic" Van
Nostrand, New York, N.Y., (1965) pp. 196-220.
131
-------
Extension of Multi-residue Methodology. II. Dynamic
Fluorogenic Labelling Detector for Carbamates.
H- Anson Moye an4 Gordon Cash
Introduction
Many pesticides now in common use are formally deriv-
atives of methylcarbamic acid, i.e. carbamate esters or
carbamoyl oximes. These pesticides are somewhat polar and
extremely heat labile, making them poorly suited for con-
ventional column cleanup techniques and separation by gas-
liquid chromatography. Several approaches have been used
to attempt to overcome these difficulties. One method is
to form chemical derivatives which improve volatility and
thermal stability (1,2). This method is often satisfactory
if the analysis is directed toward only one compound at a
time. Colorimetric procedures involving coupling with
diazonium salts also have been utilized in similar situa-
tions. Many reports of multi-residue methods for carba-
mates have described separation by thin-layer chromatography
and quantitation by various methods including chromogenisis,
enzyme inhibition and, iri situ fluorometry.
It would appear that high pressure liquid chromato-
graphy (HPLC), since it can achieve high efficiency separ-
ations at ambient temperatures could theoretically obviate
132
-------
the themal instability problem associated with the carba-
mates. However, detection at the nanogram level normally
required for sub-part-per-million analysis in foods and .en-
vironmental samples, would seem to be unlikely with commer-
cially available detectors. Preliminary experiments in this
laboratory have demonstrated this to be the case that the
carbamates as a class neither absorb U.V.-visible radiation
nor fluoresce sufficiently (with the possible exceptions of
carbaryl and benomyl) to allow for simple and straight for-
ward extraction, cleanup and detection schemes.
Recent success with post-column fluorogenic labelling
of amino acids following a HPLC separation has offered en-
couragement that this approach could be used for the anal-
ysis of the carbamates in soil (3). The fluorogen used in
p
that study was Fluram , (4-phenylspiro [furan-2(3H), 1-
phthalan]-33-dione) (4). This reagent is non fluorescent,
nor is its hydrolysis product.
Another promising fluorogenic reagent seemed to be o-
phthalaldehyde (o-phthalic-dicarboxaldehyde) reported by Roth
(5) to be extremely sensitive for primary amino acids. Like
Fluram it reacts almost instantaneously, is itself non-
fluroescent, nor are its hydrolytic products. Unlike Fluram,
however, it is stable in protic solvents. Fluram is re-
quired to be handled in aprotic solvents, such as acetone.
Additionally, two other fluorogens specific for primary
amines were investigated during the contract period, NBD-C1
133
-------
(4-chloro-7-nitrobenzo-2,l,3-oxadiazole) and dansyl chloride
(l-dimethylaminonaphthalene-5-sulfonyl chloride).
Eight commercially available, and widely used, carba-
D
mates were chosen for this study: Lannate (methomyl),
R R R R
Matacil (aminocarb), Temik (aldicarb), Baygon , Furadan
R R R
(carbofuran), Sevin (carbaryl), Mesurol and Zectran .
Lannate and Temik are actually carbamoyl oximes but can be
effictively hydrolyzed to methylamine and hence should be
detectable under the proposed system.
The three objectives for this HPLC portion of contract
work, as specified in the contract were: (1) determine the
optimum HPLC separation parameters for the above listed
pesticides that would be compatible with the dynamic fluor-
ogenic labelling detector, (2) determine the optimum para-
meters for the dynamic fluorogenic labelling detector and
(3) demonstrate the sensitivity and selectivity of the com-
bined column-detector system for the analysis of the listed
carbamates in soils.
Additionally, a brief comparison was made between the
assembled high pressure liquid chromatograph and a commer-
cially available liquid chromatograph, the American Instru-
p
ment Company Aminalyzer (Silver Springs, MD).
Experimental
Apparatus and Reagents
1. Two Waters model 6000 solvent delivery systems
(pumps) were programmed with a Waters model 660 solvent
134
-------
programmer (Waters Associates, Inc., Milford, Mass).
2. Sample injection valve Chromatronix HPSV with 25 ul
sample loop (Spectra Physics, Berkeley, CA).
3. Columns, analytical HPLC, 30 cm x 4 mm I.D.; Waters:
y Carbohydrate. 25 cm x 4 mm I.D.; pupont; Zorbax-ODS. 1 m
x 4 mm I.D.: Dupont; CDS, ETH, SAX. Reeve Angel: Pellidon.
Applied Sciences: Vydac Reverse Phase.
4. Kel-F "T"s (2), Laboratory Data Control (Riviera
Beach, FL), model CJ3031.
5. Tubing, TeflonR, 0.062 in O.D. x 0.02 in I.D., Dixon
Medical Products (Thornton, PA). Cat. No. 6-8-112.
6. Detector, U.V. absorption, 254 nm, Waters.
7. Water bath, Haake, model FK.
8. Reagent pumps (2), Milton Roy model 196.
9, Fluorometer American Instrument Co., model 125S,
with model B16-63019 flow through cell and model 416-993
mercury-xenon lamp. Excitation, 315 nm; emission, 435 nm
(for a-phthalaldehyde). Slits, 4 mm. Scale 1.
10. Recorder, Soltec model 211, 50 mv.
11. Solvents: scintillation grade dioxane, deionized
water, pH 10 0.05 M borate buffer, 0.02 M NaOH.
12. o-phthalaldehyde (OPA o-phthalicdicarboxaldehyde),
Aldrich. Stock solutions were made by dissolving 1 g. in
10 ml of ethanol adding 1 ml of mercaptoethanol and diluting
with buffer to.v.olume.
13. Aminalyzer, American Instrument Co., equipped with
135
-------
a Waters yC,« column.
14. Carbamate pesticides obtained from E.P.A., Research
Triangle Park, NC.
HPLC Column Study
Two types of columns were evaluated, reverse phase and
adsorption, at either of two particle sizes, 10 y or 37 y
diameter. Those 37 y reverse phase columns evaluated were:
CDS, ETH, SAX, Vydac and Pellidon. The 10 y reverse phase
columns were: yC,g> yCN, yNH2, yCarbohydrate and Zorbax-ODS.
Only one adsorption column was evaluated, the yPorasil. No-
tice that the SAX column is normally a strong.anion exchange
•i
column, but in this case was evaluated in the reverse phase
mode.
The adsorption mode for the separation of the carbamates
was immediately discarded after severe tailing was noted for
isooctane-methanol or isooctane-chloroform mobile phases.
This could not be overcome by water saturation of the mobile
phases or by deactivation of the silica by water impregnation.
Much effort was expended in attempting an isocratic
separation In the reverse phase mode using the 254 nm absorp-
tion detector and ug amounts of carbamates. Several columns
were immediately discarded as being unsatisfactory, either
because no reasonable separations were achieved or because
too little retention was observed: ETH, SAX, Vydac Reverse
Phase and Pellidon. The Zorbax-ODS column exhibited over
5000 psig back pressure, causing the sample injection valve to
136
-------
leak and therefore was discarded as a candidate.
Exhibiting about 5000 theoretical plates, the
column appeared to be the most satisfactory giving adequate
separation of the carbamate mix (Maticil was not included
in this early work, Fig. 1), with 40% acetonitrile--water.
At this percentage acetonitrile Baygon and carbofuran are
not fully separated, although they can be easily quantitated;
decreasing the acetonitrile percentage to 25% gave a complete
separation at the baseline, their retentions being 29 and 33
minutes respectively- A flow of one ml/min was used through-
out this work and was found to be nearly optimum while pro-
viding for rapid analyses.
The yCN column also performed well, giving about 2000
theoretical plates for a 1 foot x 4 mm I.D. column. How-
ever, at high percentages of acetonitrile (>20%) it could
not satisfactorily separate Baygon and carbofuran. By oper-
ating this column at 12ig% acetonitrile - 87%% water Lannate,
Temik, Baygon and carbofuran could be separated quite well
(Fig. 2).
By using the 12^% acetonitrile mobile phase the CDS
column separated Sevin, Zectran and Mesurol quite well. The
early eluters, while not shown (Fig. 3) elute immediately
after the injection solvent. The Mesurol response shown is
actually that of a conversion product produced by heating
at 80°C for 15-min in ethanol; unchanged, it elutes too close
to Zectran for adequate separation.
137
-------
(s)
00
W)
yC1Q 1 ft. x 4 mm I.D. COLUMN
40t8ACETONITRILE - 601 WATER
FLOW RATE I ml/min
0.16 A.U F.S.
0
16
MINUTES
24
32
FIG. 1
Isocratic separation of six carbamate mix on uC,g> 40% acetonitrile -
60% H20,
-------
yCN 1 ft x 4 mm I.D. COLUMN
12.5% ACETONITRILE - 87.5% pH 8 BUFFER
FLOW RATE 1 ml/min
0.16 A.U.F.S. t*
LO
0 11 22
MINUTES
FIG. 2
Isocratic separation of early eluting carbamates on yCN, 12.5%
acetonitrile - 87.5% H20.
ODS 1m x 4 mm I.D. COLUMN
12.5% ACETONITRILE - 87.5% pH 8
BUFFER
FLOW RATE ml/min
0.16 A.U.F.S.
0 3 6 9 12 15
MINUTES
FIG. 3
Isocratic separation of late eluting carbamates on ODS, 12.5%
acetonitrile 87.5% H~0.
139
-------
The u NH7 and y Carbohydrate columns provided poor sep-
Lf
arations for all of the carbamates and were considered un-
usable.
Studies on Fluram
Of the four fluorogenic reagents to be examined it
became obvious that NBD-C1, which requires a non-aqueous re-
action medium, would not be satisfactory, since separations
on the adsorption column (u Porasil) were unachievable.
Also, dansyl chloride was given low priority, since the
reagent itself fluorcesces considerably.
Of the remaining two reagents, Fluram was chosen for
examination first, unfortunately so.
A. Static Measurements. Since Fluram is unreactive
toward secondary amines in general, it was expected that it
would not react with the intact esters of methylcarbamic
acid. This was verified experimentally by tests with Sevin
and Lannate. Weigele ejt aJ (3), however, reported that the
secondary amino acids, proline and sarcosine, were rendered
reactive toward Fluram by preliminary treatment with aqueous
bromine or N-chlorosuccinimide under strongly acidic con-
ditions (pH 1) .
By treating Sevin with N-chlorosuccinimide (or N-
-3
bromosuccinimide, 10 M at 25°C for 2 hours) we were unable
to observe fluorescence by adding Fluram (1.5 x 10 M) after
buffering to pH9 with borate buffer. However, both Br9
t ^
and Iat 10 M could produce maximum fluorescence by incu-
140
-------
bating for 100 minutes before buffering and reacting with
Fluram. Linear analytical curves for all the carbamates
were obtined from 1 to 10 nanomoles/ml of final solution.
Table 1 gives the lower limit of detection (LLD) in nano-
moles per ml of each carbamate, assuming a signal to noise
ratio of 3:1.
Table 1
Carbamate LLD (nanomoles/ml)
Carbofuran 0.95
Sevin 0.40
Zectran 0.30
Lannate 2.7
Temik 0.75
Baygon 1.3
Mesurol 0.71
B. Dynamic Measurements. A dynamic system was assem-
bled separately from the column to simplify optimization
of parameters unrelated to carbamate separation. A Harvard
pulseless peristaltic pump was chosen for this early work
in an effort to minimize noise from the fluorometric cell
(flow through) as reagents were metered. Column effluent
was simulated by passing an additional channel through the
peristaltic pump. Zectran was used to optimize this
system because it gave the greatest response in the static
system.
141
-------
The dynamic system was assembled with a view toward
simulating the manual operations performed with the static
system. Unfortunately, the sensitivity of the dynamic
system ne\rer approached that of the static system and was
a result of a combination of thres factors: (1) since
Fluram must be pumped in the aprotic solvent, actone, no
fully elastic tubing could be found which did not rapidly
deteriorate, (2) since rapid mixing is essential for the
Fluram to provide maximum response when metered (in acetone)
into the buffered solution of primary amine, no fully satis-
factory mixing device could be designed which did not add
considerable band spreading and, (3) since complete mixing
immediately upstream of the fluorometric cell could not be
accomplished, considerable noise, due to a Schlieren effect,
could not be eliminated.
Changing the solvent to dioxane did not improve the
situation, nor did a series of elaborate in-line, mixing
devices, the description of which is irrelevent here. Con-
sequently after nine months effort with Fluram it was aban-
doned and the remaining three months given to studying o-
phthalaldehyde.
Studies on o-Phthalaldehyde
A. Static Measurements. With only three months re-
maining in the contract period allotted to the HPLC work
(no HPLC was to be done the second year of the contract
period) it was decided to go immediately to a dynamic system
142
-------
and perform a rough optimization, based on the data of Roth
(5).
It was found (see next section, Dynamic Studies on o-
phthalaldehyde) that flows for the liquid chromatograph
with the dynamic fluorogenic labelling detector, as well
as solvent an4 reagent compositions, were optimum under the
following conditions: (1) column flow 25 to 40% dioxane-
H20, 1 ml/min (2) hydrolysis reagent, 0.02N NaOH, 0.4 to
0.5 ml/min., (3) OPA reagent, 1 g./l., 1.0 to 1.3 ml/min.
By simulating these conditions in a Teflon-capped culture
tube a study was made to determine the extent of hydrolysis
occurring in the 35 foot hydrolysis coil.
To 2 ml of 25% dioxane - H-O in the culture tube the
appropriate carbamate was added to 10 yl of dioxane to
give 10 M final concentration; then 1 ml of 0.02N NaOH was
added. The tube was capped and incubated at 73°C in a
water bath, after which it was removed, cooled rapidly to
room temperature and reacted with jl ml of the OPA reagent
(1 g/1 + 1 ml mercaptoethanol). The tube was swirled and
read immediately at 334 and 454 nm in the fluorometer.
Figures 4 through 7 show the results of the various hydrol-
ysis reactions. Note that even after 30 minutes the fluor-
escent intensity is still climbing for all the carbamates,
denoting incomplete hydrolysis at the intermediate times.
When the fluorescence, for any of them, at 30 min is com-
pared with that at 2.2 min (the residence time within the
143
-------
8
7
PQ
4
2
1.
LANNATE
MATACIL
0
25 .5
10
20 30
MINUTES
FIG. 4
Relative fluorescence of N-methylcarbamates as a function of
time. See text for conditions.
CO
0
10
20 30
MINUTES
FIG. 5
Relative fluorescence of N-methylcarbamates as a function of
time. See text for conditions.
144
-------
5-
4.
HHW 3-
CARBOFURAN
=====
SEVIN
.25 .5
10
20 30
MINUTES
FIG. 6
Relative fluorescence,of N-methylcarbamates as a function of
time. See text for conditions.
MINUTES
FIG. 7
Relative fluorescence of N-methylcarbamates as a function of
time. See text for conditions.
145
-------
hydrolysis coil of the detector) it can be seen that at
most, only 20 to 50% of total hydrolysis has occurred.
Note that Lannate and Temik are hydrolyzing in a two step
fashion, as seen from an intermediate maximum in their
curves. Recommendations as to how the observed incomplete
hydrolysis ran be remedied are to be found in Section B.
B. Dynamic Measurements. Dynamic measurements were
made with the assembled dynamic fluorogenic labelling
liquid chromatograph as diagrammed in Fig. 8. Separations
were achieved with the yC,« reverse phase column. It was
anticipated that with the addition of the hydrolysis coil
(10.7 m x 0.5 mm I.D.) bandspreading might occur to the
degree that the somewhat tenuous separations achieved for
Baygon and carbofuran with 40% acetonitrile H^O might no
longer be possible. This indeed proved to be the case, not
to be demonstrated however, until acetonitrile was replaced
with dioxane, necessary due to the extremely high background
and noise experienced because of the decomposition of ace-
tonitrile to ammonia which produces fluorescence with the
a-phthalaldehyde. Methanol and isopropanol also proved to
be compatible with the fluorogenic labelling reaction,
however, neither was able to provide the column efficiencies
achievable with dioxane. Methanol produced a ten percent
reduction in theoretical plates and isopropanol a twenty
percent reduction when compared to dioxane; this is consis-
tent with their respective viscosities.
146
-------
MOBILE PHASE
1.0 ml/min
25% DIOXANE -
40% DIOXANE H0
WATERS 660
SOLVENT PROGRAMMER 35 ft.
80°C
COLUMN
SAMPLE
INJECTION
VALVE
25 ul
0.02N NaOH
0.4 ml/min
OPA
REAGENT
1.3 ml/min
FIG. 8
Modular dynamic fluorogenic labelling liquid chromatograph,
147
-------
While no careful study of mobile phase composition,
nor flow rate was made, it soon became obvious that in order
to elute Mesurol within an hour and yet maintain separations
between Lannate and Matacil, as well as between Baygon and
carbofuran, that solvent programming was required; this was
due in part to the extra peak broadening caused by the
hydrolysis coil (see section C. Detector Band Spreading
Studies) and in part to the installation of a yC,g column
of somewhat less efficiency than that used in section A.
By programming from 25% to 40% dioxane for 13 minutes (curve
#10, extremely concave) and thereafter holding isocratic,
nearly complete separations were achieved for Baygon and
carbofuran and complete separations for Lannate and Matacil.
See Figs. 9 through 14, chromatograms of spiked soil extracts
for typical separations. If 100% recoveries had been achieved
1.0 ppm represents 1.25 ug of each carbamate, 0.1 ppm rep-
«•
resents 125 ng, ancl 0.01 represents 12.5 ng, all for 25 ul
injections.
No detailed study of the fluorogenic labelling reaction
was performed; however, by making large adjustments in the
parameters involved and by bracketing techniques it was
possible to determine ranges of acceptable conditions, which,
to a first approximation, approached the optimum.
Total flow through the 5 mm square fluorometer cell
was determined by high noise on the baseline which appeared
at about 5 ml/min at the high flow end, and by the low flow
inadequacies of the reagent pumps at about 2.7 ml/min at
148
-------
o
OS
3
C/)
U4
8 12 16 20 24
MINUTES
FIG. 9
28
32 52
56
Chromatogram of sandy soil, spiked at 0.01 ppm (top). Program
No. 10 for 13 minutes, 25% dioxane - H20 to 40% dioxane - H20,
1 ml/min. Check soil (bottom).
o
OS
0 4 8 12 16 20 24 28 32 52
MINUTES
FIG. 10
Chromatogram of sandy loam soil, conditions as in Fig. 9.
56
149
-------
O
OS
w
s
0 4 8 12 16 20 24 28 32 52 56
MINUTES
FIG. 11
Chromatogram of silty loam soil, conditions as in Fig. 9.
\J V
0 4 8 12 16 20 24 28 32 52 56
MINUTES
FIG. 12
Chromatogram of sandy soil spiked at 1.0 ppm. Other conditions
as in Fig. 9.
150
-------
_J
I—I
u
<
w
E-
0
8 12 16 20 24 28 32 52 56
MINUTES
FIG. 13
Chromatogram of sandy loam soil, conditions as in Fig. 12.
A
o
4 8 12 16 20 24 28 32 52 56
MINUTES
FIG. 14
Chromatogram of silty loam soil, conditions as in Fig. 12.
151
-------
the low end. Column flow appeared near optimum at 1.0 ml/
rain. When the NaOH flow approached the OPA flow, synchro-
nous noise (blips, regularly spaced small peaks) appeared;
consequently the NaOH flow was reduced as much as possible,
consistent with reproducible pumping (0.4 ml/min). This
allowed a maximum NaOH concentration of 0.02M; higher con-
centrations began to shift the pH of the OPA buffer and
reduced the response obtainable at pH 10. Consequently,
almost 2 ml of OPA were possible before noise began to
appear on the baseline (1.7 ml/min). This relatively high
flow of OPA also diluted the dioxane in the final mix re-
ducing "roll off" in the baseline when increasing amounts
of dioxane appeared via programming. (see Fig. 20, a chrom-
atogram from the Aminco Aminalyzer for a comparison).
OPA concentrations less than 1 g/1 (with a propor-
tionate reduction in mercaptoethanol) were workable. For
example, 0.1 g/1 OPA produced less than a 50% reduction in
response to the carbamate compared to the 1.0 g/1 concen-
tration making small errors in preparation of the OPA re-
agent insignificant.
Bubble formation in the hydrolysis coil limited the
hydrolysis temperature to 80°C, sufficient for a non-optimum
but nevertheless sensitive response; the 2.2 minutes resi-
dence kept the band spreading to a minimum. Even with this
incomplete hydrolysis (see Figs. 4-7) seven serial injections
of 50 ng Lannate standards gave a relative standard deviation
of 2.7%, certainly comparable to most gas chromatographs.
152
-------
Analytical curves for the carbamates studied are shown
in Fig. 15; excellent linearity was achieved, with the
lowest practical working decade being 5 to 50 ng. About 1
ng is a realistic lower limit of detection for Lannate
(signal/noise = 3).
Recent work (antecedent to the contract period) has
shown that by placing a restrictor after the hydrolysis coil,
utilizing an integral debubbler fluorometer cell, super-
heating the hydrolysis bath (135°C) and cooling before in-
troduction of the OPA, a two to three fold increase in sen-
sitivity can be realized, approaching 100% hydrolysis of the
carbamates. An additional two fold increase in sensitivity
can be realized by increasing the NaOH concentration to 0.2N
along with an increase in borate buffer strength from 0.05
to 0.125M.
A significant feature of the OPA fluorogenic labelling
reagent approach is that check (blank) responses can be ob-
tained by either: (1) substituting the OPA reagent with pH
10 buffer, or (2) substituting the 0.02N NaOH with pH6
distilled water. This was done for an unclean celery ex-
tract (Fig. 16) showing that only the carbamate peaks dis-
appear when the OPA reagent is withheld indicating that
remaining peaks are all due to natural (underivatized)
fluorescors. Similarly, by withholding the 0.02N NaOH
hydrolytic reagent only the carbamate peaks disappear, in-
dicating that the remaining peaks are in the free form and
153
-------
200 -
160
g 120.
cx,
80.
40.
SEVIN
LANNATE
CARBOFURAN
4 MESUROL
5 BAYGON
6 TEMIK
0 20 40 60 80 100 120
ng/25 ul
FIG. 15
Analytical curves for six N-methylcarbamate pesticides; 25 ul
injections, conditions as in Fig. 9.
154
-------
do not require hydrolysis to fluoresce (Fig. 17).
C. Detector Band Spreading Studies. By connecting
the sample injection valve immediately upstream of the 5 mm
square fluorometric cell and by making 25 ul injections of
a natural fluorescor (Sevin) at various flow rates, and by
performing a similar operation after moving the sample in-
jection valve immediately upstream of the hydrolysis coil,
an analysis can be made for the bandspreading contributions
of both coil and cell individually and in tandem, applying
the relationship for chromatographic band spreading:
2 22
a coil + cell = a coil + a cell
where: a = standard deviation for a Gaussian distri-
bution, and
4a = chromatographic peak width at base line.
This was done for the OPA system; the plot of peak
width versus flow obtained is shown in Fig. 18. For typical
post column flows, a 1.4 ml/min flow through the coil pro-
duces a 4a of 27 sec; in addition a 4a of 7 sec. is realized
for 2.7 ml/min through the cell (1 ml/min, column + 0.4
ml/min, NaOH =1.3 ml/min OPA). Consequently, the total
extracolumn bandspreading can be calculated to be 28 sec,
(4a coil + cell). For a typical HSLC column of 5000 theo-
retical plates, a peak at a retention of 10 min would be 34
sec. wide at the base; if the OPA detector's 27 sec. contri-
bution is added to this quadratically, as it should, then
the peak would be 44 sec. wide at the base, equivalent to a
155
-------
CELERY
0.2 ppm
50 ng
W/0 OPA
W/ OPA
0 3 6 9 12 15 18 21 24
MINUTES
FIG. 16
Effect of withholding OPA (substituting pH 10 buffer) on
response of celery spiked at 0.2 ppm with: 1, Lannate;
2, Temik; 3, Baygon; 4, carbofuran; and^S, Sevin
1 w/0 NaOH
W/ NaOH
LETTUCE
0.2 ppm
50 ng
9 12 15
MINUTES
FIG. 17
Effect of withholding NaOH (substituting pH 7 H20) on re-
sponse of lettuce spiked at 0.2 ppm with: 1, Lannate; 2,
Temik; 3, Baygon; 4, carbofuran; 5, Sevin; and 6, Mesurol
156
-------
4.7
4.3 -
3.9 -
3.5 -
3.1 -
2.7 -
2.5 -
2.9 -
1.5 -
1.1 -
CELL + COIL
COIL
•CELL
12 16 20 24 28 32 36 40
SECONDS
FIG. 18
Plot of 4a (peak width at base) for contributions of hydrol-
ysis coil and of fluorometer cell, as a function of flow
through each.
157
-------
column of about 3000 theoretical plates, a 40% reduction in
efficiency. However, peaks eluting after the 10 min peak
will experience less broadening and peaks eluting before will
experience more.
p. So:J4 Recoveries. The three soil types studied in
Part I were spiked and analyzed to determine percent recov-
eries for the eight carbamates. Zectran exhibited highly ir-
regular behavior both in the chromatography of standards and
in the analysis of soil extracts and hence is not shown in
the chromatograms (Figs. 9-14), nor in the percent recovery
table (Table 2). Soil was extracted and analyzed in the
following manner:
1. 50 g of soil was placed in an appropriately
sized Polytron jar, spiked with the appropriate
carbamate standard mixture, and extracted with
100 ml. of beijzene - MeOH at maximum speed for
1 min.
2. The solvent was decanted into 500 ml Erlen-
meyer, 100 ml additional benzene - MeOH added to
the soil and again extracted as above, the ex-
tract combined with the previous.
3. The jar was rinsed with 20 ml of benzene -
MeOH and filtered through filter paper under
vacuum. An additional 20 ml was used to rinse
the jar.
4. The extract was evaporated to about 1 ml in
a 250 ml Kuderna-Danish on a steam bath.
158
-------
TABLE 2
SOIL RECOVERIES
Lannate
Matacil
Temik
Baygon
Carbofuran
Sevin
Mesurol
Silty Loam, PPM
.01 .02 .05
76
36
61
83
91
93
91
96
37
73
83
77
91
96
72
33
67
74
85
85
93
1.0
74
45
78
82
81
88
97
2.0
78
63
75
81
80
85
88
Lannate
Matacil
Temik
Baygon
Carbofuran
Sevin
Mesurol
Lannate
Matacil
Temik
Baygon
Carbofuran
Sevin
Mesurol
Sandy Soil, PPM
.01 .02 .05
83
84
97
87
100
115
104
.01
11
13
21
79
75
88
95
98
109
96
104
105
105
93
Sandy Loam,
.02
80
13
40
91
110
108
100
82
88
79
78
90
90
81
PPM
.05
73
19
58
76
93
87
88
1.0
70
81
81
86
96
88
91
1.0
79
35
76
81
89
84
87
2.0
66
85
62
84
86
93
92
2.0
75
52
75
93
101
91
94
159
-------
5. A small amount (3 ml) of benzene - MeOH was
used to transfer the residue to a 15 ml culture
tube, to which 2 ml acetonitrile and 4 ml 0.2N
K2HP04 was added.
6. This was partitioned twice with 3 ml portions
of methylene chloride, and the methylene chloride
dried over crystalline sodium sulfate.
7. The methylene chloride was transferred to
a clean culture tube with several rinsings and
evaporated to dryness with N--
8. The residue was dissolved in 0.2 ml of iso-
propanol and diluted with 0.8 ml of deionized
water and injected immediately on the liquid
chromatograph (25 ul).
It was discovered that the carbamate mixture degraded
rapidly in dioxane. The best solvent seemed to be isopro-
panol. In any solvent in which the higher concentrations
were made up (dioxane, isopropanol, DMF) there was a dis-
tinct autocatalytic effect causing rapidly diminishing re-
sponses. This phenomena was noticed at the 1.0 and 2.0 ppm
level (50 and 100 ng/ul) to a lesser extent at the 0.05 ppm
level (2.5 ng/ul) and not at all at 0.01 and 0.02 ppm (0.5
and 1.0 ng/ul).
From the data in Table 2 it can be seen that generally
better recoveries are realized particularly at the lower
levels, from the sandy soil as compared to the sandy loam or
silty loam. Notice the somewhat poor recoveries for the
160
-------
lower levels of Lannate, Temik and Matacil from the sandy
loam and silty loam soils. Undoubtedly these low recov-
eries would have been more pronounced if the soils had
been aged at all after spiking rather than analyzed imme-
diately. Also notice the relative lack of interference in
the three chromatograrts for the 0.01 ppm level (Figs. 9-11).
At the 1.0 level the baseline is completely smooth with no
trace of interference. Contrast these with the G.C. chrom-
atograms in Part I; also notice the extremely low recoveries
by G.C. for Lannate and Temik, even after column cleanup.
No extensive interference study was conducted; however,
it was determined that it took 10 ug of a-naphthol, a degra-
dation product of carbaryl, to produce the same sized peak
as 10 ng of carbaryl, even though a-naphthol is highly
fluorescent at very close to the wavelengths of the a-
phthalaldehyde fluorophore; it elutes about 1 minute prior
to carbaryl under the programmed mobile phase. It has not
been established whether the urea herbicides elute under
these conditions or not, nevertheless, 10 ug of monuron or
diuron produced no response whatsoever.
E. Aminco Aminalyzer. An American Instrument Co.
(Aminco) amino acid analyzer, the Aminalyzer, was converted
for the analysis of the N-methylcarbamate pesticides by
placing 25% dioxane - H-O and 60% dioxane - H90 in the
L i*
mobile phase reservoirs, placing 0.2N NaOH in the sodium
hypochlorite reservoir, and substituting the ion-exchange
161
-------
column with a 30 cm x V I.D. uC,g reverse phase column.
Flows were similar to the previously described modular
instrument, however, two major differences existed (Fig.
19). Firstly, the Aminalyzer is capable only of step
gradient operation, achieved through control of a four way
valve between four solvent reservoirs and the pump by a
digital clock. Only one step (two solvent mixtures) was
used for this work. The results of this are seen on the
chromatogram of a six carbamate mix (Fig. 20), as a step
in the baseline caused by switching over to 60% dioxane at
20 minutes into the chromatogram. Sensitivity was good,
however, as seen by the extended range analytical curve for
Lannate (Fig. 21). Secondly, hydrolysis was performed at
a "nominal" 100°C within a 15 ft. x 0.02 in. I.D. Teflon
hydrolysis coil; whether this provided complete hydrolysis
was hot determined.
Conclusions
The o-phthalaldehyde based dynamic fluorogenic labelling
high speed liquid chromatograph offers real advantages for
the analysis of N-methylcarbamate pesticides in soil, and
more than likely, in other types of agricultural and environ-
mental samples. For the previously described arrangement it
can reliably analyze seven of the eight carbamates tested
(Zectran excluded) at the 0.01 ppm level. It can be assembled
from "off the shelf" components and requires no special ex-
162
-------
MOBILE PHASE
25% DIOXANE
60% DIOXANE
DIGITAL
HCONTROL
LOGIC
DELAY
uClg COLUMN
SAMPLE
INJECTION
VALVE
OPA
n
PUMP
0.2N NaOH
FIG. 19
Aminco Aminalyze^0', adapted to N-methylcarbamate analyses.
RECORDER
FLUOROMETER
WASTE
-------
1 6 CARBAMATE MIX
50 ng
STEP-25% to 60%
DIOXANE
AMINCO AMINALYZER
5
0 8 16 24 32 40
MINUTES
FIG. 20
Aminalyzer chromatogram of: 1,
Lannate; 2, Temik; 3, Baygon; 4,
carbofuran; 5, Sevin; and 6,
Mesurol.
10
w
u
2
§ 10
w o
a 3
-3 O 4
9 H 10
CD
PP
H
10'
LANNATE
AMINCO AMINALYZER
0.1 1 10 100
NANOGRAMS
FIG. 21
Aminalyzer extended range analytical
curve of Lannate.
-------
pertise in assembly. The key reagent, a-phthalaldehyde,
can be obtained commercially in pure form (Aldrich Chem-
icals) , does not need repurification, and is reasonably
inexpensive ($14.50/25g). The reagent is stable in solu-
tion (5°C) for at least a week, possibly longer.
b
Two yC,g analytical columns were employed during the
one year contract period for this portion of work; one
would have been sufficient if it had not been inadvertently
plugged with unclean leafy vegetable extract. It was for-
tunate that the uC,R column performed adequate separations
of the carbamates because additional trials with the uCN
and yNH2 columns with the dynamic fluorogenic labelling
detector demonstrated that they continually bled a material
which reacted with the o-phthalaldehyde producing extremely
intense fluorescence. No such bleeding was observed with
the uC,g columns.
The extreme sensitivity and more importantly, selec-
tivity of this detector has made it possible to live with
the effective degrading of the column efficiency due to
band spreading within the hydrolysis coil, and to a lesser
extent, within the fluorometer cell. Analytical columns
of higher efficiency and higher hydrolysis temperatures,
allowing for shorter coil length, should moderate this
situation somewhat.
Needless to say, if a complete chromatogram of all of
the carbamates tested in this study is not necessary the
165
-------
dioxane percentage in the mobile phase can be adjusted to
provide a more rapid analysis of the compound of interest.
Dioxane percentages of up to 80% should be possible in the
dioxane - H-0 mobile phase without adversely affecting the
labelling reaction.
Although a spectrophotofluorometer, equipped with
grating monochromators, was utilized in this study, it
certainly should be possible, if not advisable, to employ
a filter fluorometer, such as the Aminco Fluoromonitor,
or others of such type.
As demonstrated in Part I of this report the Polytron
ultrasonic extraction device appears to offer a rapid,
efficient means for the extraction of soil, not-withstanding
the fact that the generator blades deteriorate rapidly due
to the abrasiveness of the soil. Minimum cleanup, an ace-
>
tonitrile water - methylene chloride partitioning, is
required for nearly interference free liquid chromatograms
of soil extract and nearly complete recovery of most of the
carbamates, even as low as 0.01 ppm.
The Aminco Aminalyzer performed adequately on car-
bamate standards, however, it appeared to offer no real
advantages over the assembled instrument, other than it is
fully cabinetted.
166
-------
References (Part II)
i
1. W. P. Cochrane, J. Chromatogr. Sci. , 13, 246 (1975).
2. H. W. Dorough and J. H. Thorstenson, Ibid, 13, 212 (1975)
3. A. M. Felix and G. Terkelsen, Arch. Biochem. Biophys.,
157, 177 (1973).
4. S. Udenfriend, S. Stein, P. Bohler, and W. Dairman,
Science, 178, 871 (1972).
5. M. Roth, Anal. Chem., 43, 880 (1971).
6. M. Weigele, S. DeBernardo and W. Leimgruber, Biochem.
Biophys. Res. Comm; 50, 352 (1973).
167
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
. REPORT NO.
EPA-600/1-77-029
2.
3. RECIPIENT'S ACCESSION>NO.
Jfc
4. TITLE AND SUBTITLE
EXTENSION OF MULTI-RESIDUE METHODOLOGY. I. Determining
Multiclass Pesticide Residues in Soil by Gas ChromatograShy^.
II. Dynamic Fluorogenic Labelling Detector for Carbamatps.
5. REPORT DATE
1Q77
FORMING ORGANIZATION CODE
7. AUTHOR(S)
H. Anson Moye, Sujit Witkonton and Gordon Cash
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Florida, I FAS
Food Science Department
Pesticide Research Laboratory
Gainesville, FL 32611
10. PROGRAM ELEMENT NO.
1EA615
11. CONTRACT/GRANT NO.
68-02-1706
12. SPONSORING AGENCY NAME AND ADDRESS
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park. N.C. 27711
13. TYPE OF REPORT AND PERIOD COVERED
RTP,NC
14. SPONSORING AGENCY CODE
EPA/600/11
15. SUPPLEMENTARY NOTES
16. ABSTRACT
A multi-residue procedure was devised for the extraction, cleanup and
determination of forty seven pesticides in fortified soil samples. Most of the
compounds were determined by gas chromatography interfaced with either the electron
capture or the flame photometric detector. Several nitrogen containing pesticides
were quantitated using a fluorescence detector in conjunction with a high performance
liquid chromatographic system. Various methods of extraction were compared for
several pesticides. A silica gel column, eluted sequentially with various solvents,
was used for cleanup and separation of compounds into groups. Several different
derivatization procedures were used to render some pesticides more amenable to gas
chromatographic detection. A dynamic fluorogenic labelling detector was designed
and characterized for the high pressure liquid chromatographic analysis of six
Y-methylcarbamate and two carbamoyl oxime pesticides. Lannate, Matacil, Temik
3aygon, carbofuran, Sevin and Mesurol could be extracted from sandy soil at the
0.01 ppm level with recoveries ranging from 83 to 115%. Somewhat lower recoveries
were experienced from sandy loam and si 1 try loam soils. Zectran could not be
ellKl* chromat°9raphed under the liquid chromatographic conditions chosen for
^uaranons. No cleanup was required prior to the liquid chromatography of the
soil extracts; no significant interference WPPP ohcpryoH f9r the unclean extracts.—
KEY WORDS AND DOCUMENT ANALYSIS
a.
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Pesticides
sxtraction
chemical analysis
gas chromatography
soil analysis
14 B
07 A
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
JL8A
22. PRICE
EPA Form 2220-1 (9-73)
168
------- |