EPA-600/2-81-
                                                   September 1981
          DEVELOPMENT OF AN IDENTIFICATION KIT
             FOR SPILLED HAZARDOUS MATERIALS
   U.S.
                    by

  A. Silvestri, M. Razulis, A. Goodman,
     A. Vasquez, and A.R. Jones, Jr.
       Chemical Systems Laboratory
Army Armament Research and Development Command
 Aberdeen Proving Ground, Maryland 21010
                     EPA-IAG-D6-0098
                   .  Project Officer

                   Joseph P. Lafornara
        Oil and Hazardous Materials Spills Branch
Municipal Environmental Research Laboratory  Cincinnati
                Edison, New Jersey 08817
       MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
           OFFICE OF RESEARCH AND DEVELOPMENT
          U.S. ENVIRONMENTAL PROTECTION AGENCY
                 CINCINNATI, OHIO 45268

-------
                                  DISCLAIMER
     This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion.  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 recom-
mendation for use.

-------
                                   FOREWORD


     The U.S. Environmental Protection Agency was created because of increas-
ing public and government concern about the dangers of pollution to the health
and welfare of the American people.  Noxious air, foul water, and spoiled land
are tragic testimonies to the deterioration of our natural environment.   The
complexity of that environment and the interplay of its components require a
concentrated and integrated attack on the problem.

     Research and development is that necessary first step in problem solu-
tion; it involves defining the problem, measuring its impact, and searching
for solutions.  The Municipal Environmental Research Laboratory develops new
and improved technology and systems to prevent, treat, and manage wastewater
and solid and hazardous waste pollutant discharges from municipal and commun-
ity sources, to preserve and treat public drinking water supplies, and to
minimize the adverse economic, social, health, and aesthetic effects of pollu-
tion.  This publication is one of the products of that research and provides a
most vital communications link between the researcher and the user community.

     This report describes the Hazardous Materials Spills Identification Kit,
which was designed as an adjunct to EPA's Hazardous Materials Detector Kit to
utilize existing instrumentation, equipment, and procedures.  The kit enables
identification of 36 different hazardous chemicals in both water and soils by
various techniques that include thin-layer chromatography, detector tubes,
test papers, and CHEMets.  The development of the kit is discussed, as well  as
the instructions for its use.

-------
                                 ABSTRACT


     The Chemical Systems Laboratory (CSL) has developed a field kit to iden-
tify spilled hazardous materials in inland waters and on the ground.   The Haz-
ardous Materials Spills Identification Kit is a two-component kit consisting
of an inverter/shortwave UV lamp unit for photochemical  reactions and a larger
package containing reagents and auxiliary equipment.   The identification kit
was designed as an adjunct to EPA's Hazardous Materials  Detector Kit to util-
ize existing instrumentation, equipment, and procedures.

     Thirty-six materials, representative of those with  the greatest hazard
potential, were selected to evaluate candidate procedures.  A wide variety of
methods from the literature and commercial sources were  screened.  Procedures
selected for the kit include:  thin-layer chromatography, detector tubes,
detector papers, CHEMets, an arsine/Gutzeit test, and a  number of color devel-
opment procedures for use with the thin-layer chromatography.  In addition,
methods were developed for recovery of contaminants from water and soil.  All
information pertinent to identification of 36 specific materials was designed
into a compact data retrieval system, which is included  in the kit.

     Two prototype kits were delivered to EPA, along  with a supply of consum-
able materials for evaluation.  In addition, manuals, engineering drawings,
and parts lists were provided.

     This report was submitted in fulfillment of Interagency Agreement No.
EPA-IAG-06-0098 and covers the period July 1976 to March 1979.  Work was com-
pleted as of March 1979.

-------
                                  CONTENTS
Foreword	iii
Abstract	   iv
Figures	   vi
Tables	  vii
Acknowledgments  	 viii

     1.  Introduction  	    1
     2.  Conclusions   	    3
     3.  Recommendations 	    4
     4.  Experimental Results  	    5
     5.  Identification Concept   	   35
     6.  Prototype Kits	   41

References	   46
List of Manufacturers and Suppliers	   48

Appendix - Operator's Manual for Hazardous Materials Identification Kit

     1.  Introduction	49
     2.  Equipment and Methodology	51
     3.  Analytical Procedures 	  63
     4.  Supplementary Identification with the Hazardous
             Materials Detector Kit	72
     5.  Consumable Materials  	  74

-------
                                  FIGURES

Number                                                                Page

   1       Elution of organic contaminants and standard dyes
            with chloroform on silica gel HL plates
            in a standard tank at room temperature	    15

   2       Elution of organic contaminants and standard dyes
            with chloroform on silica gel HL plates in a
            sandwich chamber at room temperature 	    17

   3       Elution of organic contaminants and standard dyes with
            chloroform on silica gel  HL plates in
            a sandwich chamber at 30-40°F 	    18

   4       Elution of organic contaminants and standard dyes with
            chloroform on silica gel  HP plates in a sandwich
            chamber at 90-100°F	    18

   5       Hazardous materials spills identification kit,
            inverter/UV lamp unit	    42

   6       Hazardous materials spills identification kit,
            reagents/auxiliary equipment package 	    43

   7       Hazardous materials spills identification kit,
            reagents/auxiliary equipment package 	    44

-------
                               ACKNOWLEDGMENTS

     Appreciation is given to the following CSL staff members:
     Mr.  Robert Gamson, Chief of Kits and Labs Section, who assured that the
program was adequately staffed.
     Mr.  Charles McDowell, who worked on the fabrication of the prototype
kits.
     Mr.  Lester Strauch, who conducted power consumption studies on the
inverters.
     Mr.  Henry Dutton, who prepared the engineering drawings.
     Ms.  Ella Holub and Ms.  Kathi Grant, who typed the many reports.
                                    vn

-------
                                 SECTION 1

                                INTRODUCTION


     The Oil and Hazardous Materials Spills Branch of the Environmental Pro-
tection Agency's Municipal Environmental Research Laboratory, Cincinnati (EPA)
is responsible for research and development concerning control and mitigation
of spills of oil and hazardous materials.   Very often the nature of the spill
is known; for those instances in which it is not, some method of identifica-
tion is necessary in order to permit emergency response personnel  to take
proper action.  The Chemical  Systems Laboratory (CSL), under Interagency
Agreement EPA-IAG-D6-0096, has developed a field kit to identify spilled
hazardous materials in inland waters and on the ground.  Two prototypes were
fabricated and presented to EPA with drawings, parts lists, and manuals.

     The program was concerned with about 500 materials listed in  the Federal
Register, Volume 40, Number 250, of December 30, 1975.  Identification require-
ments of the kit and representative test materials to be used in this program
were coordinated with EPA.  Among others,  EPA is concerned with identification
of the toxic metals, anions,  and classes of organic materials as follows:

        Inorganics                        Nitriles

     Antimony trioxide                  Acrylonitrile
     Arsenic trisulfide                 Benzonitrile
     Cadmium chloride
     Lead theocyanate                   Nitro Compounds
     Mercuric cyanide
     Selenium dioxide                   Nitrophenol
     Zinc fluoride
     Sodium arsenate                    Organic Acids
     Sodium arsenite
     Potassium bichromate               Acetic acid
     Potassium chromate                 2,4-0 (acid)
     Potassium permanganate             Maleic acid
                                        Benzoic acid
        Alcohols
                                Organic Chlorine Compounds
     Propyl alcohol
                                        Dieldrin
       Aldehydes                        Heptachlor
                                        DDT
     Acetaldehyde                       Chlorobenzene

-------
     Amines                         Organophosphates

     Butyl amine                         Diazinon
     Triethylamine                      Malathion
     Aniline                            Parathion

     Esters

     Amy! acetate                       Phenols
     2,4-0 (esters)
                                        Phenol
     Hydrocarbons

     Isoprene
     Naphthalene

     In a similar program completed in January 1976, CSL developed for EPA the
Hazardous Materials Detector Kit to detect (not necessarily identify) hazard-
ous materials in inland waters.'   This kit is used to trace the location of
contamination in water and also to warn water users until the contaminant can
be treated or disappears.

     EPA and CIS decided to develop an identification kit that can be used as
an adjunct to the Hazardous Materials Detector Kit.  This kit makes use of the
pH meter, spectrophotometer, conductivity meter, and other sampling and analyz-
ing procedures already available.

     Furthermore, some of the tests used in the detector kit, if used judici-
ously and in combination, can provide reliable identification, particularly
for inorganic materials.  For example, cyanide and fluoride, which are two
anions of interest, can be identified by the detector kit (Appendix, Section 4,
Supplementary Identification with  the Hazardous Materials Detector Kit).

     Since it is planned to use the identification kit to implement corrective
action in areas of high contamination, emphasis in this program was placed on
reliability of identification rather than detection.

-------
                                  SECTION 2

                                 CONCLUSIONS


     CSL has developed a kit for use in the field to identify spilled hazard-
ous materials in water.  The design is based on the findings of an extensive
investigation of literature and commercial methods in combination with labora-
tory studies.  The operation of the identification kit has been cross-
referenced and integrated with EPA's detector kit for hazardous materials.

     Information pertinent to identification has been included in a compact
data retrieval system, which is contained in the kit.  It is possible in some
cases to identify specific contaminants and, in others, to designate a class of
contaminants.

     The kit provides EPA with potential  capability for identifying hazardous
materials in the field.  The kit is not simple or routine to use.  However,
with training and familiarization, it can provide the investigator in the field
with a wide variety of information to aid him in his mission.

     The kits mark a major advance over the present capability for identifica-
tion of contamination in the field.

-------
                                  SECTION 3

                               RECOMMENDATIONS
     The program undertaken was rather ambitious.  It not only attempted to
address the field identification of some 300 materials, but further sought to
do so using some unconventional approaches.  The research resulted in a kit
that employs thin-layer chromatography, which is normally considered a rather
sophisticated laboratory process, and also uses detector tubes normally used
for airborne pollutants to analyze water and soil.  Finally, identification is
accomplished by incorporation of a data retrieval system.  While every attempt
has been made to anticipate all potential problems in use of this kit, it is
inevitable in a program of this complexity that some difficulties will be un-
covered in intensive use.

     It is recommended, therefore, that EPA subject this kit to a complete
series of laboratory familiarization studies to fully learn its response, hand-
ling, and overall capabilities.  When this objective is accomplished, the kit
should be used in a limited scope in actual field exercises.  As part of the
overall evaluation, EPA should consider the use concept for the kit, the ex-
tent of its application, interfaces with other detection and identification
equipment, relationship to cleanup equipment and methodology, and level of
skills available for use of the kit.

     If it is determined that the identification kit can play a useful role, a
meeting may be coordinated with CSL to review the findings.  The direction, at
that time, might be either of two courses.  It may be expedient to identify
all deficiencies at that time and implement corrections, or it may be more de-
sirable to screen an additional group of test materials (12, 24, or 36) through
the kit and then implement corrections in light of the larger data base.

     In the course of making modifications or corrections,  it may be desirable
to investigate other nitrile tests (one was never adopted for the kit) and
broader response aldehyde tests.  If EPA wishes to specify  a vehicle for sup-
port of this kit, it may be advantageous to set up a wiring harness under the
hood leading to a common jack in the dashboard for hookup of the inverter.

     Further, since all the identification objectives will  not be resolved
within this program, kits will be delivered to EPA at program conclusion in
July 1978.  These kits will provide EPA with an immediate "on-line" capability.
Modifications resulting from changes in scope or shortcomings found through
field experience can be made in follow-up studies.

-------
considerably slower and more complex in makeup.  Accordingly, it was decided to
pursue more conventional methods to identify inorganic contaminants.

     A variety of visualization and eluting procedures were evaluated in this
program.  The initial  effort in the TLC study was the development of visuali-
zation techniques; once a significant number of these was selected,  it was
possible to go on to the development of the elution procedures.

     The challenge materials were prepared in concentrations of 1 mg/ml  of
chloroform.  Four-microliter quantities were spotted on candidate chromato-
graphic plates; the chloroform was allowed to volatilize and then visualization
was attempted.  Primary emphasis was on the development of spray reagents,
although other methods, such as fluorescence/quench techniques and iodine stain-
ing, were also evaluated.

     Commercial spray reagents were used directly as received from the sup-
pliers.  Reagents developed in-house used Chromist sprayers (Fisher Scientific
Co., Silver Spring, Maryland 20910) for application.

     In addition to spray reagents developed in-house and from the literature,
numerous commercially available sprays were investigated.  The protocol  for
evaluating visualization procedures was as follows:

     1.   Spot 4-microliter challenge material on TLC plate.

     2.   Examine for long- or short-wave fluorescence and fluorescence  quench.

     3.   Spray with candidate reagent, examine for visible color and long-
          or short-wave fluorescence or fluorescence quench.

     4.   Heat and again examine for color or fluorescence.

In addition, some reagents were irradiated for 5- to 10-minute periods with
high-intensity short-wave ultraviolet light to develop colors.  Other plates
were exposed to iodine vapors.

     Candidate procedures were evaluated on silica gel (with and without a
fluorescent tracer), alumina, avicel, and cellulose chromatographic plates
(Analtech, Inc., Newark, Delaware 19711). The following listing represents the
major reagents that were evaluated for TLC applications.  Many'variations were
attempted that are not listed unless the outcome was successful.  Other reagents
that were evaluated in similar applications but were not promising were N-
ethylmaleimide^', 1,5-dichloro-2,4-dinitrobenzene, 2-nitrophenyl hydrazine^,
p-nitrobenzene^diazoniumtetra-fluoroborate'9, potassium dichromate'4, toluene-
sulfonic acid'0. 4-methylumbel!iferonelu, malonic acid/salicylaldehyde,  and
brilliant green'5.

     1.   Iodine Staining^.  Expose plates to iodine vapors by placement in a
          jar with a few crystals of iodine.  Tan to brown spots are found with
          triethyl amine, aniline, malathion, parathion, and phenol.

-------
2.   Fluorescence Quench.  Good quenching is observed with shortwave
     fluorescent light with diazinon, parathion, and phenol when spotted
     on silica gel GF plates.

3.   Indandione Reagent  '.  Spray with reagent; orange fluorescence is
     observed under shortwave fluorescent light with 2,4-D (esters), 2,4-D
     (acid), dieldrin, heptachlor, DDT, diazinon, and malathion on alumina
     plates.

4.   Dragendorf Reagent  (Brinkmann Instruments, Inc., Westbury, New York
     11590).  Spray with reagent; an orange color is observed with
     triethyl amine on silica gel plates.

5.   lodoplatinate (Brinkmann).  Spray with reagent; a purple color is
     observed with triethylamine on silica gel  plates.

6.   Rhodamine B (VWR Scientific, Inc., Baltimore, Maryland 21227).  Spray
     with reagent; a purple color is observed with 2,4-D (acid), maleic
     acid, and malathion on silica gel plates.

7.   4-Oimethylaminobenzaldehyde (VWR).  Spray plate and heat 2 minutes at
     100°C; yellow spots are seen with aniline on silica gel  and 2,4-D
     (acid) on cellulose and avicel plates.

8.   2',7'-Oichlorofluorescein (VWR).  Spray plate and heat 2 minutes at
     100°C; green spots are seen under longwave fluorescence  with acetic
     acid, 2,4-D (acid), maleic acid,, and benzoic acid on silica gel
     plates.

9.   Molybdatophosphoric Acid (VWR).   Spray plate; a blue color is seen
     with triethylamine on silica gel or alumina plates.

10.   Bromocresol Green (VWR).  Spray plate;  blue spots are seen with
     butylamine and triethylamine and yellow spots are seen with acetic
     acid, 2,4-0 (acid), maleic acid, and benzoic acid on silica
     gel  plates.  (See also 33 below.)

11.   Mercuric Sulfate (Brinkmann).   No response.

12.   Ninhydrin (VWR).  No response.

13.   Diphenylcarbazone (Brinkmann).  No response

14.   Aniline Phthalate (VWR).  No response.

15.   Oiphenylcarbazone (Gelman Instrument Co.,  Ann Arbor,  Michigan 48106).
     No response.

16.   lodoplatinate (Gelman).   No response.

17.   Potassium Permanganate (Gelman).  Spray plate;  yellow spots are  seen
     with aniline, maleic acid, and phenol  on silica gel plates.

-------
18.  Ninhydrin (Gelman).  No response.

19.  Benzoquinone.  Use as a spray at a concentration of 0.5 gin/100 ml
     methanol.  Aniline gives a pink spot on silica gel  plates.

20.  Fluorescein Mercuric Acetate .   Use as a spray at a concentration  of
     17 mg/100 ml of IN sodium hydroxide.  2,4-0 (esters), heptachlor,  DDT,
     malathion, and parathion show a quench on silica gel plates under
     longwave fluorescent light.
                               O K "7 Q
21.  4-(p-Nitrobenzyl) pyridine ' *  ' .  Dissolve 0.5 gm reagent, 1.0 gm
     mercuric cyanide, and 6.0 gm sodium perchlorate in  100 ml  methanol.
     Spray the plate and heat 2 minutes at 100°C.  Spray with a  solution of
     60% potassium carbonate in water.  Purple spots are found  with diaz-
     inon and malathion on silica gel plates and yellow  spots are seen  with
     parathion on silica gel plates.
                                         Q
22.  2,3-Dichloro-5,6-dicyanobenzoquinone .  Use as a spray at  a concen-
     tration of 1.0 gm/100 ml benzene.  On silica gel, triethylamine
     gives a red, aniline a purple,  maleic acid a white, and phenol a
     blue color.
              q
23.  Chloranil .   Use as a spray at  a concentration of 0.5 gm/100 ml
     benzene.  Butylamine and triethylamine give a purple color on silica
     gel plates,  while aniline gives a green color and phenol a  peach
     color.  All  reactions require 2 minutes heat at 100°C.
                                        q
24.  N,2,6-Trichloro-p-benzoquinoneimine .  Use as a spray at a  concentra-
     tion of 2 gm/100 ml of methanol.  Heat alumina plates for  2 minutes
     at 100°C.  Triethylamine produces a gray spot.  Acetic acid, 2,4-0
     (acid), and maleic acid give white spots.  Malathion gives  a red
     color and phenol a blue color.
                                             q
25.  2,6-Oibromo-N-chloro-p-benzoquinoneimine .  Use as  a spray  at a
     concentration of 1 gm/100 ml benzene.  Heat silica  gel plates for  2
     minutes at 100°C.  Triethylamine and aniline produce black  spots and
     phenol gives a purple color.

26.  Silver Nitrate '''.  Use as a spray at a concentration of 0.5
     gm/100 ml water.  Irradiate with high-intensity ultraviolet light
     for 10 minutes.  On silica gel  plates, butyl amine,  triethylamine,
     aniline, 2,4-0 (esters), acetic acid, 2,4-0 (acid), maleic  acid,
     benzoic acid, dieldrin, heptachlor, DDT, diazinon,  malathion, para-
     thion, and phenol appear as several colors, including white, gray,
     black, tan,  and dark brown.
                       12
27.  p-Phenylenediamine  .  Use as a spray at a concentration of 0.5 gm/
     100 ml methanol.  Irradiate 10  minutes with high-intensity shortwave
     ultraviolet light.  On silica gel plates, triethylamine gives an
     orange color.  2,4-D (esters),  acetic acid, 2,4-D (acid),  maleic
     acid, benzoic acid, dieldrin, heptachlor, and malathion give purple
     colors; parathion gives a white color on silica gel plates.

-------
                    12
28.   Dimethyl aniline  .   Use as a spray at a concentration of 0.5 gm/100
     ml  methanol.   Irradiate 10 minutes with high-intensity ultraviolet
     light.   On silica gel plates, 2,4-0 (esters) and 2,4-0 (acid) give
     brown spots;  dieldrin and heptachlor give purple spots, and phenol
     produces  a white spot.

                  12
29.   Diphenylamine  .  Use as a spray at a concentration of 0.1  gm/100 ml
     methanol.  Irradiate 10 minutes with high-intensity ultraviolet light.
     On  silica gel plates, 2,4-0 (esters) produces a'purple color; 2,4-D
     (acid), maleic acid, and dieldrin give blue colors.  Also,  heptachlor
     gives a gray  color, while parathion produces a white spot on silica
     gel  plates.

30.   Palladium Chloride  '  .  Use as a spray at a concentration  of 0.5 gm/
     100 ml  methanol.  Tributylamine, butylamine, aniline, 2,4-0 (acid),
     maleic acid,  and benzoic acid produce white to light brown  spots on
  •   silica gel plates.   Oiazinon, malathion, and parathion give distinct
     brown spots on silica gel plates.
                       914
31.   Tetracyanoethylene  '  .  Use as a spray at a concentration  of 0.5 gm/
     100 ml.  Phenol appears as an orange spot on alumina plates.

32.   s-Diphenylcarbazone  .  Use as a spray at a concentration of 0.1 gm/
     100 ml  methanol.  Dieldrin, DDT, and parathion appear as red sposts
     on  alumina plates.

33.   Bromocresol Green (VWR).  Spray the plate; in addition to immediate
     spots that develop, other materials may be detected on silica gel
     plates with a 10-minute irradiation with shortwave ultraviolet light.
     2,4-0 (esters), dieTdrin, heptachlor, and DOT are found as  yeHow
     spots after irradiation.

34.   Methyl  Orange  .  Use as a spray at a concentration of 0.5  gm/100 ml
     methanol.  2,4-0 (acid) and maleic acid are seen as red spots on
     silica gel plates.

35.   Methyl  Yellow  .  Use as a spray at a concentration of 0.5  gm/100 ml
     methanol.  Butylamine, triethylamine, aniline, diazinon, malathion,
     and parathion give  yellow colors on silica gel plates, while acetic
     acid, 2,4-0 (acid), and maleic acid give red spots,  also,  if the
     plate is  irradiated with shortwave ultraviolet light, blue  spots are
     seen with acetic acid, 2,4-0 (acid), and benzoic acid.
                12
36.   o-Toluidine  .  Use as a spray at a concentration of 0.5 gm/100 ml
     methanol.  Irradiate 10 minutes with shortwave ultraviolet  light.
     Aniline,  2,4-D (esters), 2,4-0 (acid), and maleic acid appear as
     brown spots on silica get plates, while parathion appears white.

37.   2,2-Oiphenyl-l-picrylhydrazyl  .  Use as a spray at a concentration
     of  0.1  gm/100 ml chloroform.  On silica gel plates at 100°C for 2
     minutes,  butylamine, 2,4-0 (acid), maleic acid, malathion,  and
     phenol  produce yellow spots, while aniline gives a blue color.


                                10

-------
were
     38.   N,N-Dimethyl-p-phenylenediamine dihydrochloride  .   Use as a spray
          at a concentration of 0.5 gm/100 ml methanol.   On silica gel plates
          with shortwave ultraviolet radiation, butylamine, triethylamine, and
          aniline appear as gray spots.

     Materials visualized are from the selected list of contaminants.   A
summary of findings of the most significant visualization procedures is given
in Table 1 and also in Section 5, Identification Concept.  Detailed use of
these methods is described in the Appendix, Section 3,  Analytical Procedures.

     From the initial  screening it was concluded that the following reagents
     the most promising for exploitation of TLC procedures:

     Iodine vapors
     Indandione
     Bromocresol green with and without UV light irradiation
     Methyl yellow with and without UV light irradiation
     2' ,7'-Dichlorofluorescein
     4-(p-Nitrobenzyl) pyridine
     Palladium chloride
     Benzoquinone
     Chloranil
     2,3-Dichloro-5,6-dicyanobenzoquinone
     2,2-Di phenyl-1-1 pi cry!hydrazyl
     Silver nitrate
     Diphenylamine
     Dimethylaniline
     o-Toluidine

     Continued testing showed some of the reagents to be unstable, unreliable,
complex, or insensitive.  Furthermore, some, which were based on fluorescence
readout, were subject to interference by the recovery methods developed for
contaminants (see Recovery and Cleanup, this section).

     Accordingly, a final selection was made for the kit that included bromo-
cresol green (with and without UV irradiation), silver nitrate,  palladium
chloride, and chloranil.  Findings with these reagents  are presented in Table
1.  These are considered to be the easiest to use and most reliable and stable
reagents.  Furthermore, they detect a good cross-section of materials  and
nalyxhibit a degree of overlap that enables them to act as backup in some
analyses.  Using these procedures, it is possible to detect 16 of the  24
organic contaminants used in this program.  The procedures detect amines,
acids, chlorinated materials, thiophosphates, and phenols.

     In the course of this investigation, numerous reagents were developed
that are ideally used in a spray form.  The method is attractive for visualiz-
ing unknowns by the TLC procedures described and also for detecting certain
inorganic contaminants (see Inorganics Spray Reagents,  this section).   Accord-
ingly, a contract was let with Case-Mason Filling, Inc., of Joppa, Maryland
21085, to package the reagents into self-contained pressurized cans.  The re-
agents, formulation, and use are given in Table 2.
                                     11

-------
     38.  N,N-Dimethyl-p-phenylenediamine dihydrochloride  .  Use as a spray
          at a concentration of 0.5 gm/100 ml methanol.  On silica gel plates
          with shortwave ultraviolet radiation, butylamine, triethyl amine, and
          aniline appear as gray spots.

     Materials visualized are from the selected list of contaminants.  A
summary of findings of the most significant visualization procedures is given
in Table 1 and also in Section 5, Identification Concept.  Detailed use of
these methods is described in the Appendix, Section 3, Analytical Procedures.

     From the initial screening it was concluded that the following reagents
were the most promising for exploitation of TLC procedures:

     Iodine vapors
     Indandione
     Bromocresol green with and without UV light irradiation
     Methyl yellow with and without UV light irradiation
     2' ,7'-Dichlorofluorescein
     4-(p-Nitrobenzyl) pyridine
     Palladium chloride
     Benzoquinone
     Chloranil
     2,3-Oichloro-5,6-dicyanobenzoquinone
     2,2-Di phenyl-1-1 pi cry!hydrazyl
     Silver nitrate
     Oiphenylamine
     Dimethyl aniline
     o-Toluidine

     Continued testing showed some of the reagents to be unstable, unreliable,
complex, or insensitive.   Furthermore, some, which were based on fluorescence
readout, were subject to interference by the recovery methods developed for
contaminants (see Recovery and Cleanup, this section).

     Accordingly, a final selection was made for the kit that included bromo-
cresol  green (with and without UV irradiation), silver nitrate, palladium
chloride, and chloranil.   Findings with these reagents are presented in Table
1.  These are considered to be the easiest to use and most reliable and stable
reagents.  Furthermore, they detect a good cross-section of materials and
nalyxhibit a degree of overlap that enables them to act as backup in some
analyses.  Using these procedures, it is possible to detect 16 of the 24
organic contaminants used in this program.  The procedures detect amines,
acids,  chlorinated materials, thiophosphates, and phenols.

     In the course of this investigation, numerous reagents were developed
that are ideally used in a spray form.  The method is attractive for visualiz-
ing unknowns by the TLC procedures described and also for detecting certain
inorganic contaminants (see Inorganics Spray Reagents, this section).  Accord-
ingly,  a contract was let with Case-Mason Filling, Inc., of Joppa, Maryland
21085,  to package the reagents into self-contained pressurized cans.   The re-
agents, formulation, and use are given in Table 2.
                                     n

-------
           TABLE 2.   CANDIDATE REAGENTS FOR PRESSURIZED SPRAY CANS
         Reagents
                '  Use
Formulation
Silver nitrate
Palladium
Chloranil
Diphenylamine
Alizarin
4-(2-Pyridy!azo)
resorcinol
Xylenol orange
s-Di phenyl carbazone
Di phenythi ocarbazone
(Dithizone)
TLC - Organ ics
TLC - Organ ics
TLC - Organ ics
TLC - Organ ics
Inorganics

Inorganics
Inorganics
Inorganics

Inorganics
5.0 gm/1 water
5.0 gm/1 methanol
5.0 gm/1 chloroform
1 .0 gm/1 methanol
0.5 gm/1 methanol

0.65 gm/1 methanol
2.5 gm/1 methanol
5.0 gm/1 chloroform

0.5 gm/1 chloroform
     The packager prepared all reagent/propel!ant mixtures in clear glass
containers, which made it possible to determine interactions when they occur-
red.  The propellent of choice was an equal mixture of Freon-11 and Freon-12.
Sometimes, the propel!ant precipitated the reagent out of solution.  When this
happened, a second propel!ant, "A46," a mixture of isobutane and propane
designed to produce pressure of 46 psi, was used.

     Alizarin red-S and xylenol orange precipitated regardless of which propel-
lant was used; silver nitrate attacked the metal can and could not be packaged.
All the other reagents were prepared in the Freon mixture and returned to the
laboratory for evaluation.  Shortly thereafter, all the reagents, with the
exception of PAR and chloranil, decomposed to -the point where they could not
be used.  Chloranil, probably due to its low solubility, tended to plug the
tip of the spray can, eventually rendering it useless.  The self-contained PAR
spray can was selected for use in the kit.  Its composition is as follows:
     Can:
6 oz Spraytainer, plain
      Button:   Seaquist 3125  Narrow  Spray,  forwardlook  style, mechanical
               breakout

      Valve:    Sea-Spray,  NS-34,  0.018-inch  stem,  dimpled mounting cap,  Buna-
               N gasket, 0.023-inch  wire  stainless  steel spring, capillary
               body orifice,  tubed 4-1/16-inch,  0.060-inch  capillary  I.D.

      Fill:     80  gm  Freon 11
               80  gm  Freon 12
               40  gm  PAR 0.63 gm/1 methanol

      Other  spray reagents  that are of  interest to  the  kit are packaged  inde-
 pendently of  the propel!ant system.  The  reagents  are  provided  in a glass jar

                                     12

-------
and a Chromist sprayer* is attached as needed.

     Once adequate visualization procedures were available, it was possible to
develop elution procedures for TLC.  Because TIC would be applicable to those
contaminants that were capable of visualization, all subsequent elution studies
were limited to these 16 materials (see Figures 1 through 4 and Table 3).
Also, because certain variables, such as plate quality, temperature, humidity,
solvent purity, etc., were expected to affect the elutions, it was decided to
include some standard reference dyes in the separations.  Out of a number  of
dyes considered, indophenol, sudan orange R, and sudan yellow 3G were deter-
mined to produce the most compact, uniform, and reproducible movements.  Sudan
yellow 3G was ultimately chosen for the kit; even with use of a reference  dye,
variations are found in the elution data.  However, the dye serves as a good
indicator of overall performance of the system in the field.

     Separations were made using prescored 10-cm alumina and silica gel plates
(Analtech, Inc.); this size was determined to be the minimum that would produce
acceptable resolution.  Screening of elution solvents was made using an eulo-
tropic series that included acetonitrile, methanol, acetone, isopropanol,
ethyl acetate, chloroform, dichloromethane, benzene, toluene, and hexane.

     It was concluded from this initial screening that silica gel plates were
sturdier than alumina for field use and resolutions were superior.  Also,
chloroform produced varied separations which would  be more useful in applying
the process to identification.  Acetone, while not  as good a separator as
chloroform, could also be useful.

     In the course of this evaluation, Analtech, Inc., started marketing of a
hardened silica gel plate.  This plate was extremely rugged and could be
written on, thus it was decided to evaluate it more extensively.   Elution  data
were derived by the following procedure:

     1.   A rectangular tank for 20- x 20-cm plates was used for elution.   The
          sides and back were lined with glass fiber paper in order to quickly
          and more evenly saturate the atmosphere.

     2.   Enough solvent was added to the tank to wet the glass fiber and  give
          a depth of 3 to 5 mm.

     3.   Four micro!iters of the materials to be chromatographed (1 mg/ml  in
          chloroform) were spotted 10 mm from the lower edge of the chromato-
          graphic plate.   The material was spotted  in several applications  in
          order to maintain a compact spot.

     4.   Marks were made on the upper side edges of the plate 70 mm from  the
          spots (80 mm from the bottom of the plate).

     5.   The plate was made to stand upright and the top was placed on the
          tank.

     6.   When the solvent front reached the upper  markings, the plate was
          removed and allowed to dry.


                                     13

-------
     7.   Visualization was accomplished by the techniques described earlier
          and the locations of contaminants were marked.

     8.   Rf values—distance traveled by contaminant divided by distance
          traveled by eluting solvent (70 mm) -- were calculated.

     9.   On some occasions, Rs values were calculated.   These were distance
          traveled by contaminants divided by distance traveled by standard
          dye.

     Separations made on silica gel  Hard Layer (Organic  Binder) TLC Uniplates
are depicted in Figure 1 in a standard laboratory tank using procedures
described above; separation and Rf and Rs data are given  in Table  3.

     In order to place TLC in a format that would be acceptable for field use,
smaller chromatographic tanks to replace the larger laboratory type were
sought.  A small tank (source unknown) measuring 21-1/2  cm by 11  cm by 5 cm
was ideally suited for the 10-cm by 20-cm chromatographic plates.   When  glass
fiber paper lining was used in the tank, results were reproducible and closely
agreed with data derived with the larger tank.  Unfortunately, an  extensive
search for the source of this tank was unsuccessful  and  had to be  abandoned.

     As an alternate choice, a number of commercial  sandwich chambers  were
investigated.  They, again, were compact, lightweight, and were claimed  to
generate reproducible data.  Three chambers that were investigated were  found
to be poorly designed, difficult to assemble, or not suited for use of glass
plates.

     A successful chamber design was prepared in-house.   (Appendix, Section 2,
Equipment and Methodology.)  Problems were first experienced with  this chamber
when the separator used between the two plates was 1905  microliters.   When
this thickness was reduced to 762 microliters, results became more reproducible.

     Figure 2 depicts separations at room temperature made with the sandwich
chamber proposed for the identification kit.  The separation and Rf and  R$
values are presented in Table 3.  Notice that some changes observed over the
standard laboratory tank are quite dramatic.

     It was long suspected in these investigations that  certain factors
particularly affected the movement of materials on chromatographic plates.
Temperature was considered to be one of the most important factors.  Accord-
ingly, the separations made in the sandwich chamber were rerun in  an environ-
mental chamber.  In Figures 3 and 4 the effects of temperature on  the  separa-
tions are seen.  Again, the separation and Rf and RS values are presented in
Table 3.

     It is evident from the above analysis, Figures 2 through 4, and Table 3,
that even the use of standard dyes is not sufficient to  produce consistent Rf
and Rs values under all conditions.   All materials are not similarly affected
by temperature.  The elution rates of contaminants may be more sensitive to
temperature than the standard dyes.   Therefore, unless separations are made at
similar temperatures, the use of standard dyes primarily serves to indicate
the general overall operability of the system in the field.

                                     14

-------
SOLVENT FRONT





















(

f



(

")




c
)






)


























L ^



C



t)


(
)






")0







)



x
L






,c
J



d


)




•)


r
L






> (
J *•




0

")





!)<







)




70 m
60
50
40
30
20
10
0
              UJ
YLAMINE

3
a

£
ETHYLAMI

s

US
UJ
Z

Z

£
0 (ESTERS

*-

31
ROPHENOL

Z

•V
CM
a
y
UJ
u

ui
Q
u
a

W

to
Q
U
u
UJ

Z

r*
(N
^
0
o
N
Z
UJ
O

39
NIUO1
UJ
5

CN
TACHLOR
a.
UJ

a


a
o

K
NONIZ

a

n
LATHION

2

•V
lATHION
u^
a.

n
O
UJ
a

a
0
iU

|
Q
Z
ORANGE f

Z
Q
irO
YELLOW 3

Z
Q
3
           Figure 1.  Elution of organic contaminants and standard
                dyes with chloroform on silica gel HL plates
                   in a standard tank at room temperature.
     In summation, it is concluded that TLC can serve a highly useful purpose
in the identification kit. Accordingly, a sandwich chamber, using hard-layer
silica gel plates and chloroform solvent, was selected.  The qualitative use
of Rf and Rs values in combination with recommended visualization techniques
provided a substantial amount of information for identification of organic
contaminants.

DETECTOR TUBES

     The procedure for applying detector tubes for airborne contaminants to
water testing has been described.!   The process is also presented in detail in
the Appendix, Section 2, Equipment and Methodology. It consisted of placing
water suspected of contamination into a small jar.  An effervescent material,
e.g., bicarbonate, was added and the jar was capped with a detector tube
affixed to the cap so that the emitted gases were forced through the tube.
The tube was noted for a color response due to the evolved contaminant.

     The method described above can be modified for soil testing by the
following procedure.   An equivalent of 40 ml of loosely packed soil  can  be
added to the effervescent jar.  Add distilled water to the 80-ml mark and stir
the mixture for several  minutes.  Add effervescent material to the mix,  cap
the jar, and complete the test as described above.
                                     15

-------
               TABLE 3.  Rf AND  RS  VALUES OF ORGANIC  CONTAMINANTS AND STANDARD DYES

                         ELUTED  WITH CHLOROFORM ON  SILICA GEL ML UNIPLATES
Standard Tank
Room
Test Material
15. Butylamine
16. Triethylamine
17. An i 1 i ne
19. 2,4-D (Esters)
24. Nitrophenol
25. Acetic Acid
26. 2,4-D Acid
21. Maleic Acid
28. Benzoic Acid
29. Oieldrin
30. Heptachlor
31 . DDT
33. Diazinon
34. Malathion
35. Parathion
36. Phenol
Indophenol
Sudan Orange R
Sudan yellow 3G
nun
Traveled
0
0
12
37
43
2
2
0
5
44
54
54
8
16
41
9
10
33
35
Rf
0.00
0.00
0.18
0.53
0.61
0.03
0.03
0.00
0.08
0.62
0.78
0.78
0.12
0.22
0.58
0.13
0.14
0.47
0.50
Rs
0.00
0.00
0.36
1.08
1.24
0.06
0.06
0.00
0.16
1.28
1.60
1.60
0.24
0.45
1.21
0.27
0.27
0.94
1.00
30-40°F
mm
Traveled
0
0
10
32
49
3
7
1
8
52
70
70
a
8
41
8
8
29
27
Rf
0
0
0.14
0.46
0.70
0.04
0.10
0.01
0.11
0.74
1.00
1.00
0.11
0.11
0.59
0.11
0.11
0.41
0.39
Rs
0
0
0.37
1.19
1.81
0.11
0.26
0.04
0.30
1.93
2.59
2.59
0.30
0.30
1.52
0.30
0.30
1.07
1.00
Sandwich Chamber
Room
inm
Traveled
0
0
23
49
58
5
6
0
9
63
70
70
19
23
53
11
21
50
50
Rf
0
0
0.33
0.70
0.83
0.07
0.09
0.00
0.13
0.90
1.00
1.00
0.27
0.33
0.76
0.16
0.30
0.71
0.71
Rs
0
0
0.46
0.98
1.16
0.10
0.12
0.00
0.18
1.26
1.40
1.40
0.38
0.46
1.06
0.22
0.42
1.00
1.00



90-100°F
mm
Traveled
0
0
14
40
51
4
5
1
8
55
70
70
10
12
43
10
9
35
34
Rf
0
0
0.20 .
0.57
0.73
0.06
0.07
0.01
0.11
0.79
1.00
1.00
0.14
0.17
0.61
0.14
0.13
0.50
0.49
Rs
0
0
0.41
1.18
1.50
0.12
0.15
0.03
0.24
1.62
2.06
2.06
0.29
0.35
1.26
0.29
0.26
1.03
1.00
•
Note:  Solvent  travel  is 70 nim.
      R  = Normalized to sudan yellow 3G.

-------
SOLVENT FRONT






(•








/ d




r


/


^

^



^

j




(\







} {







) c







5 (1

r





f)

-)







^ V,







^



f
L






-, C
J



r


)




^



r
>*





(
N
P)
S


f
k

•>)
J



> f
r






s
J •





70 mm
60
SO
40
30
20
10
0

YLAMINE
a

in
£.
ETHYLAM
ce


_i
ROPHENO
i—
Z

(N

Q
O
u
UJ
CJ

iri

Q
U
Q


s

Q
U
u
UJ
2

5
Q
O
u
0
z
UJ
a

CO
CM

z
e
Q
UJ
5

CN

TACHLOR
°- —
uj a
z a

a -
~ M

ZINON
5

n

.ATHION
S

S

ATHION
a.

ifl

-i
O
Z
UJ

n

O
Z
a.


tc
ORANGE
Z
a
V)
w
en
YELLOW
2
a
3
        Figure 2.  Elution of organic contaminants and standard dyes
                with chloroform on silica gel HL plates in a
                    sandwich chamber at room temperature.
     Soils having a high organic content will often cause excessive foaming,
which carries the water into the detector tube.  It is necessary in these
cases to provide a small ballast chamber between the cap of the jar and the
detector tube to allow the foaming to subside.

     A survey was made of the tubes of the major suppliers (Kitagawa: Matheson
Gas Products, Lyndhurst, New Jersey 07071; Bendix/Gastec: National  Environ-
mental Instruments, Inc., Warwick, Rhode Island 02888; Draeger: National  Mine
Service Company, Pittsburgh, Pennsylvania 15219; and Mine Safety Appliances,
Pittsburgh, Pennsylvania 15235).  Available data on detection reagents, de-
tected materials, probable interferences, and sensitivity were catalogued.   It
was seen from the survey that many detection systems were repetitive, not only
among companies, but also within the same company.   For example, many compan-
ies carry various tubes containing chromium salts and sulfuric acid that are
designated for detection of certain compounds.  One company alone carries 24
variations of this tube, which is used to detect 24 different organic materi-
als.  It is believed that some tubes are not as selective as may be implied.

     From the survey,  24 types of tubes were selected for evaluation as being
highly representative of those available.  In the initial screening, the tubes
were evaluated against aqueous solutions containing 1  mg/ml of the  challenge
contaminants.  Most tubes were discarded because they were not sufficiently
sensitive or else produced too many borderline responses that were  difficult to
classify.  Promising tubes were further evaluated against natural waters  and
soils.

                                     17

-------
SOLVENT FRONT







\







} '«.





j"
X.
}



(

>
*>


f
k
1





">>
J



f







I (1







) r







, r


r




")


~)




















<.







j <^



ri



;



r>



^







) <



1


V

J



J (







;



70 m
60
50
40
30
20
10
0
                           ,-,



                                                     o  a  a
  Figure  3.   Elution  of organic contaminants  and standard  dyes  with

 chloroform  on  silica gel  HL  plates  in  a  sandwich chamber  at  30-40°F
SOLVENT FRONT














* i






c




c
^

)



<
^
^
j





•s
^



^







h) f















a r


(




•)


)














^ ^-

















c


^


1

•)


r
\.











C







5 C







) '




70 m
SO
SO
40
30
10
0
         >
      >


5


i
I
   <

                                                          s
<  S
!
§
  Figure 4.   Elution of organic contaminants and standard dyes with

chloroform on silica gel HP plates in a sandwich chamber at 90-100°F
                                  18

-------
     The tubes selected for inclusion in the kit were Benzene 5/a,  Olefin
0.05%/a, Benzene No.  121, and Acetone No. 151.   Findings are summarized in
Table 1.  The initial  screening details are summarized as follows:

     1.    Arsine O.OSa (Draeger).  A light blue color forms on the  precleanse
          layer with malathion.

     2.    Formaldehyde 0.002 (Draeger}.  A brown color forms on the indicating
          layer with isoprene.   Acetaldehyde gives a reddish color  on  the
          indicating layer.

     3.    Aniline 5/a (Draeger).  A pink color is obtained on the indicating
          layer with aniline.

     4.    Benzene 5/a (Draeger).  Purple colors form on the precleanse and
          indicating layers with aniline.  Isoprene turns the precleanse layer
          brown; naphthalene colors the precleanse layer blue; 2,4-D (acid)
          turns the precleanse layer a reddish-brown; and chlorobenzene turns
          the precleanse layer pink.  Malathion turns the precleanse layer
          light brown and the indicating layer pink.  Oiazinon produces a
          reddish-violet color on the precleanse layer.

     5.    Carbon disulfide 30/a (Draeger).  Nitrophenol produces a  yellow
          color on the indicating layer.

     6.    Carbon tetrachloride 5/c (Draeger).   Isoprene turns the indicating
          layer green and causes the reagent in the ampoule to turn brown.
          Chlorobenzene produces a light brown  color with ampoule reagent.

     7.    Olefin 0.05%/a (Draeger).  Light brown colors are observed on the
          indicating layer with acetaldehyde,  isoprene, and acrylonitrile.

     8.    Polytest (Oraeger).   A brown color forms on the indicating layer
          with aniline, amyl  acetate, and isoprene.  Naphthalene produces a
          blue color and chlorobenzene produces a brownish-purple color.
          Acetaldehyde and butylamine give brownish-green colors on the
          indicating layer.

     9.    Perchloroethylene 10/a (Oraeger).   A  blue color is produced  on the
          indicating layer with chlorobenzene,  malathion, parathion, and
          phenol.

     10.  Trichloroethane 5Q/b  (Draeger).  Gives positive tests with water;
          was abandoned.

     11.  Hydrogen fluoride 1.5b (Draeger).   Negative response.

     12.  Cyanogen chloride 0.25/a (Draeger).   Negative response.

     13.  Systox 1/2 (Draeger).  Negative response.
                                     19

-------
     14.   Ammonia 25/a (Draeger).   Negative response.

     15.   Toluene diisocyanate 0.2/A (Draeger).   Negative response.

     16.   Trichlorethylene 10/a (Draeger).   A brown  color is  obtained  with
          isoprene and acrylonitrile on the precleanse layer.   Chlorobenzene
          gives an orange color on the indicating layer.

     17.   Phenol  5/a (Draeger).  A pink color is obtained with  butylamine on
          the indicating layer.

     18.   Styrene No.  124 (Bendix/Gastic).   A yellow color forms  on  the
          indicating layer with isoprene.

     19.   Ethylene No.  172 (Bendix/Gastec).   Isoprene  and malathion  produce  a
          dark blue color on the indicating layer.

     20.   Carbon  tetrachloride No. 134 (Bendix/Gastec).   Isoprene and  chloro-
          benzene give dark brown  colors on the  indicating layer.

     21.   Carbon  monoxide ILL (Bendix/Gastec).  Malathion produces a brown
          color on the indicating  layer.

     22.   Benzene No.  121 (Bendix/Gastec).   Mixtures of colors,  including
          yellows, oranges, greens,  browns,  and  grays, are obtained  with
          isoprene, naphthalene, heptachlor, DDT, chlorobenzene,  diazinon,
          malathion, parathion, and  phenol.

     23.   Acrylonitrile, No. 191 (Bendix/Gastec).  Isoprene gives  a  brown
          color on the indicating  layer; naphthalene gives a  pink color on the
          indicating layer.  Chlorobenzene  produces  a  black on  the precleanse
          layer and a  pink on the  indicating layer.   Diazinon gives  a  yellow-
          green on the precleanse  layer and a pink on  the indicating layer.

     24.   Acetone No.  151 (Bendix/Gastec).   A mixture  of colors,  including
          green,  brown, and black, is obtained with  amyl  acetate,  isoprene,
          acrylonitrile, acetic acid, chlorobenzene, and diazinon, on  the
          indicating layer.

ARSINE/GUTZEIT

     One of the most reliable methods for detection  and identification of
arsenic is the classic Gutzeit test20.  In  this  procedure, zinc metal  and
strong acid are added  to an aqueous  solution of  an arsenical  compound  (the
procedure is effective for inorganic and certain aliphatic arsenicals).
Nascent hydrogen, which is formed, reacts with he arsenic to  form arsine.
Arsine can be detected by test papers impregnated with various  mercuric compounds.

     In the procedure  developed for  the kit, Draeger Tube AsH;j  0.05/a  was
substituted for the test paper.  The assembly described above for the  detector
tube tests was used.  In the initial evaluation  of the system,  hydrogen gas
was generated by adding 1 gm (scoop  measure) 30-mesh zinc metal  and  1  gm


                                      20

-------
(scoop measure) potassium bisulfate with a crystal of cupric sulfate for
catalyst to 80 ml of test solution.  Evolved arsine gas turned the detector
tube a grayish-purple.  The procedure was effective for all the arsenical
compounds on the challenge list of materials.  It was even useful for antimony
compounds.  Some difficulties were experienced with antimony trioxide due to
its low solubility, but when more soluble antimony compounds were tested,
strong responses were obtained.

     Midwest Research Institute, Kansas City, Missouri 64110, under contract
to the Chemical Systems Laboratory for another program, studied the use of a
Gutzeit mix (zinc, potassium bisulfate, and cupric sulfate) in detection of
certain chemical agents.  In their studies, they prepared a formulation pack-
aged in compact trilaminate plastic pouches; this preparation was more sensi-
tive and easier to handle than the individually measured quantities of chemi-
cals used in this program.  Accordingly, a supply of Gutzeit pouches was
purchased and evaluated.

     Use of the Midwest Gutzeit preparation showed a marked increase in sensi-
tivity, wherein all the antimony and arsenical materials on the challenge list
were detected.  Midwest's preparation has been included in the identification
kit.

     The above procedure may be used for soil testing by adding about 40 ml  of
loosely packed soil to the effervescent jar and diluting to the 80-ml  mark
with water.  The procedure for water testing is then followed.  Details may be
found in the Appendix, Analytical Procedures.  Findings are presented in Table
1.

SPOT TESTS

     Identification of some classes of materials was not achieved with TLC or
detector tubes, namely alochols, aldehydes, esters, hydrocarbons, nitriles,
and nitro compounds.   This area is particularly difficult as there are no
commercial tests available for these materials.

     An extensive screening of spot test procedures for identification of
these compounds was carried out.  The areas investigated are described in the
following sections.

Nitro-Contaim'ng Compounds

     A very promising test for nitro-cooiaining compounds, which uses  a 2-
amino alcohol  as a colorimetric reagent  , was developed.   Reagents tested
were ethanolamine, 2-amino-2-methyl-l-propanol, 2-amino-2-methyl-l, 3-
propanediol, and 2-(2-aminoethyl-amino)-ethanol.   These reagents, used neat,
all give a bright yellow color with an aqueous solution (1 mg/ml) of nitro-
phenol.  Furthermore, these reagents appear to be highly specific for the
nitro group of a large number of aromatic nitro compounds  as well as for some
aliphatic nitro compounds, giving yellow, orange, red, brown, and blue colora-
tions.  The reagents  and nitro-containing materials were used neat except when
both compounds were solids, in which case several drops of methanol were added.

The colors obtained with ethanolamine (l-amino-2-ethanol)  and other nitro-

                                     21

-------
containing compounds from tjie .Federal Register are shown below.

     1,3-Dinitrobenzene	deep yellow
     2,4-Dinitrophenol  	 red
     Nitrobenzene 	 yellow (on standing)
     Nitrophenol 	 deep yellow
     Parathion	no color

     None of the other organic compounds in the list of challenge materials
gave a positive test with ethanolamine.

     Yagoda Confined Spot Test Papers, carried by Schleicher and Schuell,
Inc., Keene, New Hampshire 03431, were found to be an effective and inexpen-
sive method for conducting spot tests.  The papers, which measure 40 by 40 mm,
have 1-cm test spots contained by a ring of inert impervious waxlike material.
The test material is applied to the center spot followed by the reagent.   The
procedure was highly effective in detecting nitrophenol with ethanolamine.

     One-percentmethanolic solutions of ethanolamine were also studied, but
the concentrated form was more sensitive and, therefore, preferred.

     The procedure was modified for the kit as follows:  To one drop of the
water solution on a segment of silica gel  HL Uniplate, add one drop of ethanol-
amine.  A deep yellow color is produced by nitro compounds.  Compare with a
distilled water blank, if necessary.

     The test for nitro compounds using ethanolamine was also found to be
effective with water extracts of soil samples.  The findinas are presented in
Table 1.

Aldehydes

     A deep red color developed when one drop of 5% sodium nitroprusside
(Nitroferricyanide, Na2Fe(CN)5NQ.2H20) and one drop of a saturated solution of
sodium hydroxide in methanol were added to 1 ml of aqueous test solution (1 mg
acetaldehyde/ml water)22.  None of the other organic materials responded with
sodium nitroprusside.

     When the test was carried out using a chloroform solution of acetaldehyde,
the aqueous nitroprusside solution separated to the top.  However, on shaking,
a red color concentrated in the aqueous layer.

     Bases other than sodium hydroxide were investigated; organic bases are
reported to give a blue color with acetaldehyde.  If one drop of piperidine is
substituted for sodium hydroxide, a deep blue-violet color is obtained.

     Neither formaldehyde nor paraformaldehyde, which are on the comprehensive
list of organic contaminants, responded to the sodium nitroprusside test.  It
may be too narrow as a class test for aldehydes; however, it was included in
the kit because of its excellent sensitivity.  The test was modified for the
kit as follows:  To 1 ml of the water solution, one drop of 5» sodium nitro-
prusside in water and one drop of saturated sodium hydroxide in methanol  were
added.  A deep  red color was observed with acetaldehyde.  The procedure was

                                     22

-------
also effective with water extracts of soil samples.  Findings are presented in
Table 1.

Esters

     The determination of esters is based on reaction of an ester with hydroxyl-
amine in alkaline solution to form hydroxamic acid.23  Hydroxamic acids give
a violet-colored chelate with ferric ion.  Some difficulties were encountered
in repeating the tests.  There were apparent stability problems and the propor-
tions of reagents needed to be investigated.

     The procedure first developed produced a white precipitate with the
addition of hydroxylamine and sodium hydroxide to the test solution, which
tended to obscure color formation.  The precipitate was sodium chloride, which
is insoluble in the solvents used (methanol and chloroform).  This problem was
eliminated by running the test in a water solution; the sodium chloride dis-
solves and does not interfere.

     Another difficulty was found in the reproducibility of the test; it
appears that the disposable plastic tubes used for testing were a source of
interference.  When the work was repeated using glass vials, the results were
consistent.

     It was further found that when the reaction mixture is heated in a boiling
water bath, deep purple colors are obtained.  However, prolonged heating
causes decomposition of some hydroxamic acids and, therefore, should be avoided.
After heating, 5% hydrochloric acid is used to neutralize the excess sodium
hydroxide before adding the ferric ion.  The amount of acid added is critical.
If not enough is added, ferric hydroxide will precipitate.  On the other hand,
too much acid will hinder color development.  The optimum pH is about 3 to 6.
In order to provide for pH adjustment, sodium alizarin sulfonate indicator was
added to the hydrochloric acid.

     Hydroxylamine hydrochloride and sodium hydroxide reagents were prepared
in methanol and compared with similar reagents in water; the methanolic
reagents were found to be more sensitive and are recommended.  The hydrochloric
acid/sodium alizarin sulfonate reagent was modified by increasing the acid
content from 5% to 10%, which reduces the volume of reagent required for the
test.  Heat is no longer found necessary for the test.  The test, as modified,
detects both amyl acetate (deep purple color), and 2,4-0 (esters) (brownish-
amber color), although it is more sensitive to the former.

     The procedure is also effective for detecting esters in water extracts
of soils.  The procedure in the kit is as follows:

     To 1 ml of test water in a test tube, add 0.5 ml (10 drops) of
     12.5% hydroxylamine hydrochloride in methanol and 0.5 ml (10
     drops) of 12.5% sodium hydroxide in methanol.  Allow to stand
     2 to 3 minutes.  Acidify with 10% hydrochloric acid solution
     containing 0.01% sodium alizarin sulfonate (the solution turns
     purple to red to orange to yellow).  Stop acid addition when the
     solution just turns yellow.  Add two drops of ferric chloride
     solution (5 gm Fed-,  and five drops HC1 in 50 ml  water).  Esters

                                     23

-------
     may give a color ranging from brownish-amber to purple.  Xompare
     with a water blank if there is any doubt.

     Findings are presented in Table 1.

Alcohols,
                                           24
     Vanadium(V)-8-hydroxyquinoline complex   was studied as a reagent for
alcohols.  A red color was obtained with alcohols in organic solvents.  (The
test was not effective in aqueous solutions.)  It was first proposed to run
the test in chloroform since this solvent, in conjunction with XAD-2 amber!ite
cartridges, was selected for recovery of contaminants from water (see Recovery
and Cleanup).  However, chloroform was found to contain traces of alcohol  and
therefore, all test solutions produced positive blanks.  Consideration was
given to running the test in methylene chloride since this was an alternate
solvent to chloroform for recovery.  However, methylene chloride was found to
extract interfering materials from the XAD-2 column and could not be used for
water testing.  Prior washing of the column with methylene chloride failed to
remove the interferences.

     An alternate method was to extract the alcohol from water directly by
solvent.  The test solution was shaken with a toluene solution of vanadium
oxinate and acetic acid.  When the test tube was heated in a water bath, a
faint brownish-pink color was obtained; sensitivity was inadequate for kit
purposes.

     Surprisingly, when the vanadium test was applied to soil testing, alcohol
was easily detected in the methylene chloride, which was used for extraction.
This was true even in soils having a high organic content.  Accordingly, the
test was adopted for kit use, but limited to soil analysis,  the procedure
used was as follows:

     To 1 ml methylene chloride extract of the contaminant, add two drops
     of 1* vanadium(V)-8-hydroxyquinoline complex and two drops of
     glacial acetic acid.  After 1 minute, a deep red color develops in
     the presence of alcohols.

     Findings are presented in Table 1.

     Use of diammonium eerie nitrate as a reagent for alcohol was also investi-
gated.  This reaction would be ideal since it can be carried out in water;
unfortunately, the sensitivity is not adequate.  Various organic acids and
salts (HC1, ZnClgj SnCl4, and A^) were added to the reaction medium to try
to increase the sensitivity, but nothing showed promise.

Nitriles

     Tests reported in the literature for detection of the -CN group are based
on pyrolysis at temperatures of 180°C to 250°C (a technique not practical  for
field testing), or they use a method that is not specific for nitriles.

     Experimental work showed that acrylonitrile and benzonitrile form colored
compounds with certain aromatic nitrocompounds (e.g., m-dinitrobenzene,

                                      24

-------
nitrofluorene).  These compounds did not give consistent results.

     Nitriles react with hydroxylamine to produce amidoximes, which usually
give red colors with ferric ion.  Preliminary investigation of this reaction
showed that acetonitrole gives a brown-red color and benzonitrile gives a
green color.  Generally, higher heating conditions are required; a method of
adapting the test to kit use was not found.

CHEMETS

     CHEMetrics Kits (CHEMetrics, Inc., Warrenton, Virginia 22186) were screened
as tests for hazardous materials.  CHEMets are disposable glass ampoules,
about the size of a cigarette, used for the colorimetric analysis of water
samples.  Each CHEMet contains a color-forming reagent sealed under vacuum.
When the tip of the CHEMet is broken under water, water rises into the ampoule
and dissolves the reagent.  A color forms if the sought compound or ion is
present.  A color comparator composed of liquid standards is offered for
quantitative measurement.

     CHEMets were screened against aqueous solutions containing 1 mg/ml of the
challenge contaminants.  Most CHEMets were found not be be sufficiently
responsive to the test materials.  Two, however, that passed this initial and
subsequent screenings with natural waters and soils are phenols - Model P-12,
and hydrogen peroxide - Model HP-10; they are included in the identification
kit.

     Findings are presented in Table 1 and CHEMets screened in this program
are shown below.

     1.   Phenols, Model P-12.  Gives excellent response to phenol.  Nitro-
          phenol  and aniline are also detected.  Light precipitates are ob-
          tained with cadmium chloride, lead thiocyanate, zinc fluoride, and
          potassium permanganate.

     2.   Hydrogen peroxide, Model HP-10.  Gives dark brown colors with potas-
          sium bichromate, potassium chromate, and potassium permanganate.
          Weaker brown or yellow colors and light precipitates are obtained
          with cadmium chloride, lead thiocyanate, mercuric cyanide, selenium
          dioxide, sodium arsenite, propyl alcohol, butylamine, triethyl amine,
          and parathion.

     3.   Phosphate, Model PO-150.  Excellent response to inorganic phosphate,
          but does not detect phosphate esters, diazinon, malathion, or parathion.

     4.   Hydrazine, Model H-5.   Detects aniline.

     5.   Iron, Model Fe-10.  Detects only iron.

     6.   Dissolved oxygen, Model 0-12.  Detects oxygen.

     7.   Ammonia nitrogen, Model AN. Response to acetaldehyde and potassium
          permanganate.


                                      25

-------
     8.  Chlorine, Model Cl-2.  Does not detect any of the organic chlorine
         compounds.  Sensitive response to potassium permanganate.

DETECTOR PAPERS

     Most of the reagents considered for inorganic contaminants were initially
evaluated by impregnating them on test paper dip strips.  Literature procedures
were screened with the view of finding chemicals that could be adapted to a
paper similar to a pH paper, which would give a specific color response to a
particular metal or ion.  The concept is appealing because it offers a high
volume of tests in a compact, inexpensive, and easy-to-use configuration.

     A number of papers were impregnated with chemical chromogenic reagents.
The general method used was to wet filter paper (9-cm circles of Whatman 140)
with a 1% methanolic solution of the reagent, air-dry the paper, and then cut
it into strips.  Testing was accomplished with aqueous solutions of contami-
nants at concentrations of 1 mg/ml, whether or not they were soluble to that
extent.

     E.M. Quant Strips (Scientific Products, Division of American Hospital
Supply Corp., McGraw Park, Illinois 60085), test papers for water testing,
were also evaluated.  (Directions for use of the strips specify adjustment of
pH or other modification of water samples to make the tests specific.  However,
they were evaluated with untreated waters as were the in-house prepared strips.)

     The following detector paper systems were screened in this program:

     Al*o (Quant strip)            Rhodamine B
     Fe   (Quant strip)            8-Hydroxyquinoline
     Chromate4"3 (Quant strip)      Thionalide
     Ether peroxide (Quant strip)  as-Oiphenylhydrazine             „-
     Zn+2 (Quant strip)            PAR (4-(2-Pyridylazo)-resorcinol)
     Sulfite (Quant strip)         1,10-Phenanthroline
     Mn4"2 (Quant strip)            Thiourea
     Co4/? (Quant strip)            Phenylhydrazine hydrochloride
     Ni   (Quant strip)            L-Ascorbic acid
     Ag (Quant strip)   /,,-         a-Furilmonoxime
     s-Diphenylcarbazone           Indicator paper pH 6.0-8.0 and Na2S201
     Alizarin red-S                Indicator paper pH 8.0-9.5 and Na^S?0-*
     Xylenol orange                1,5-Diphenylcarbohydrazide
     Dipycrylamine                 Morin
     2-Mercaptothioazoline         Pb-Oiethyldithiocarbamate
     Dithizone                     Arsenic trisulfide
     Nal (freshly prepared)        Rubeanic acid
     Phenylhydrazine               PAN (l-(2-Pyridylazo)-2-naphthol)

     From this initial screening, a selection of the most promising detector
paper systems was made and evaluated more extensively with natural waters and
soils.  Results are presented in Table 4.

     Extensive testing showed that the papers did indeed respond to chemical
contamination.  However, a certain variance was found in the reproducibility

                                      26

-------
TABLE 4.  CANDIDATE DETECTOR PAPERS



















Challenge Material
Blank
1. Antimony trioxide
2. Arsenic trisulfide
3. Cadmium chloride
4. Lead thiocyanate
5. Mercuric cyanide
6. Selenium dioxide
7. Zinc fluoride
8. Sodium arsenate
9. Sodium arseni te
10. Potassium
bichromate
11. Potassium chr ornate
12. Potassium
permanganate
COLOR CODE: It = light
dk = dark








 a c:
o i~~i uJ co s:
HP 0 C C C


Pk R
P R
R
BR
R R
Pk


P BR GN-BL BL
GN-BL BL

BR T dkBL R BL
P = purple
W = white
T = tan
o
c
•«- 1
e x
ni o
i — CU -O
>> c t-
l/> -C O Ol «J
1 4-> C C71 O
XI CU QJ C ' —
01 E -C ID >,
OC CU Q. I- C
Xt 0 O  Xt
z: o «t^r+-> tt- d. x. — j;
WCTY YYYHP


BR-P R
dkR GR-P R P Pk

0 0 Y
Pk PR dkR Pk
P HPk
P Pk

Y Y P
Y Y P T

T Pk Pk BR R T BR
BR = brown CM = green
BL = blue C = cream
0 = orange

at
c
o

fO

l_
(O
o


>J

 O
•r- C >>r—
i — O j^ f~
to rs* . — o
C •!- >> O
0 -C C t-
•r— -*-* Q> ~O
-C -r- £ >.
(- O Q- -C
C BL C


Y
R
0
0 Y-GN HO
R
Y
Y


Y

R
= yellow
= gray


CU
c

IM
a
(.
XI
>^
JC.
r—

C.
0) 
-------
of results.  The problem is largely associated with readout, which requires
subjective evaluation of the presence and quality of certain colors.  Very
often, impregnated papers, whether prepared in-house or obtained commercially,
have an initial color and it is up to the operator to determine whether there
is a gain or a change in the hue of the original color.   All of these situations
were aggravated when testing was extended to natural  waters and soils; the
reproducibility became even more unreliable.

     As a result, with the exception of several Quant strips which did prove
to be reliable even under the most adverse circumstances, the detector paper
approach was reduced in effort.  As seen in the next section, the value of
some of these reagents was exploited further by an alternate process.
                    +2                        +2
     Quant strips Mn  , ether peroxide, and Zn   were suitable for kit use.
However,  the Mn 2 and ether peroxide strips gave identical responses.  There-
fore, Mn+2 (either could be used) was selected.  Zn*2 test strips became
unavailable and therefore Ni"1"2 was selected as a backup test for selenium.

     The findings are presented in Table 1.

INORGANICS SPRAY REAGENTS

     Many of the reagents investigated in preparation of detector papers were
examined again for use in spray form.  It was believed that spray reagents,
because they are applied fresh to the test spot and the surrounding area
(which acts as a real-time reference blank), markedly improve the reproducibil-
ity of the tests.  Because high reliability of identification is a primary
goal in this program, the tests were judged to warrant the extra weight,
volume, and cost required by the spray reagents.

     Primary effort in this area was applied to the following reagent composi-
tions:

     1.   Alizarin redS (alizarin sodium sulfonate),  0.05% in methanol
     2.   4-2(2-Pyridylazo) resorcinol (PARK), 0.125% in methanol
     3.   Xylenol orange, 0.25% in methanol
     4.   s-Diphenylcarfaazone, 0.5% in methanol
     5.   Diphenylthiocarbazone (dithizone), 0.05% in methanol

     Tests were conducted by spotting 5-microliter aqueous samples (1 mg/ml)
of contaminants on segments of silica gel HL Uniplates.   The plates were dried
in the oven or at room temperature and then sprayed with the candidate reagent.
Color judgments were easier as the surrounding area acted as a reference
blank.

     As indicated in Thin-Layer Chromatography (TLC)  earlier, these five
reagents were packaged in self-contained pressurized spray cans.  PAR (the can
formulation is given) is one of the preparations that functions and can be
stored satisfactorily.  Through continued testing with natural water and soil
samples, it was decided to include PAR, s-diphenylcarbazone, and diphenylthio-
carbazone in the kit for the detection concept.  Chromist sprayers are used in
application of s-diphenylcarbazone and diphenylthiocarbazone.  Details on use


                                     28

-------
of these reagents are given in the Appendix, Equipment and Methodology, and
Analytical Procedures.

     Findings are presented in Table 1.


RECOVERY AND CLEANUP

     Tests for contamination in water using detector papers, CHEMets, inorganic
reagent sprays, spot tests, arsine/Gutzeit, and detector tubes can be run
directly on the water sample.  Tests using TLC procedures all require that the
contaminant be separated from water.  No test was found for alcohol in water,
nor was a means for recovering it from water found.

     Primary effort for recovery of contaminants from water was placed on
evaluation of the "Drug-Skreen" system27 marketed by Brinkmann Instruments,
Westbury, New York 11590.  The system uses small prepackaged columns containing
Amberlite XAD-2 resin, which is commonly used to extract many organic materials
from aqueous solutions.  Although the system is intended for recovery of
abused drugs from urine samples, it does have a broader application.

     The use of the "Drug-Skreen" system in the identification kit is shown in
the Appendix, Equipment and Methodology.  First, 20 ml of water sample contain-
ing a dissolved or suspended contaminant was passed through the column by
gravity feed.  Then, solvent was made to pass through the column and was saved
in a collection flask.  A siliconized phase filter was used to prevent residual
water from entering the collection flask.  Chloroform and methylene chloride
were equally effective as recovery solvents.  Ethyl acetate was not found to be
a suitable solvent; it tended to form an emulsion with water, which tended to
considerably slow down or altogether stop passage through the phase filter.

     Selection of the solvent also had an influence on the subsequent TLC tests
that could be used.  For example, a number of reagents, such as indandione,
fluorescein mercuric acetate, and 2', 7'-dichlorofluorescein, were dependent on
a fluorescent readout.  Fluorescent materials extracted from the Amberlite
resin interfered with these procedures.  Also, methylene chloride, which is a
suitable solvent for alcohol-extracted material from the resin, interfered with
the vanadium(V) 8-hydroxyquincline test.

     On the basis of these findings and the fact that chloroform is used in the
kit for the TLC elutions, it was selected as the extraction solvent.  Recovery
of contaminants from water is shown in Table 5.  It is seen that only acetic
acid and maleic acid are not recovered from water using the Amberlite column
and chloroform.

     A limited investigation was made into the utility of using Conwed Sorbent
Pads (Conwed Corporation, St. Paul, Minnesota 55101) as a means of recovering
liquid contaminants.  Conwed pads are made of highly absorptive materials and
are used in cleanup of chemical contamination on the ground, other surfaces,
and in water.  The pads measure 45.7 cm x 45.7 cm x 0.64 cm and are largely
composed of cellulose-type materials bonded to a polypropylene netting for
strength.


                                      29

-------
           TABLE 5.   RECOVERY OF CONTAMINANTS FROM THE ENVIRONMENT

1.
2.
3.
4.
5.
6.
7,
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
Test Material
Antimony trioxide
Arsenic trisulfide
Cadmium chloride
Lead thiocyanate
Mercuric cyanide
Selenium dioxide
Zinc fluoride
Sodium arsenate
Sodium arsenite
Potassium chromate
Potassium bichromate
Potassium permanganate
Propyl alcohol
Acetaldehyde
Butyl ami ne
Triethylamine
An i 1 i ne
Amy! acetate
2,4-D (Esters)
Isoprene
Naphthalene
Acrylonitrile
Benzonitrile
Nitrophenol
Acetic Acid
2,4-0 (Acid)
Maleic acid
Benzoic acid
Oieldrin
Heptachlor
DDT
Chlorobenzene
Diazinon
Malathion
Parathion
Phenol
Recovery Method: - = Not appl
Water
^
-
-
-
-
-
-
-
-
-
-
-
NM
-
good
good
good
-
good
-
-
-
-
good
poor
good
poor
good
good
good
good
-
good
good
good
good
i cable














(1)
(D
(1)

(1)




(1)
(1)
(1)
(1)
(1)
(1)
(1)
(1)

(1)
(1)
(1)
(1)
, NM =
Sand
.
_
good
good
good
good
good
_
_
good
good
good
good
-
good
good
good
-
good
-
-
-
-
good
good
good
good
good
good
good
good
-
good
good
good
good
No method



(2)
(2)
(2)
(2)
(2)


(2)
(2)
(2)
(3)

(4)
(4)
(4)

(4)




(4)
(4)
(4)
(4)
(4)
(4)
(4)
(4)

(4)
(4)
(4)
(4)
available,
Organic

..
good
good
good
good
good
—
—
good
good
good
good
-
poor
poor
good
_
good
-
-
-
_
good
good
good
poor
good
good
good
good
_
good
good
good
good
Soil


(2)
(2)
(2)
(2)
(2)


(2)
(2)
(2)
(3)

(4)
(4)
(4)

(4)




(4)
(4)
(4)
(4)
(4)
(4)
(4)
(4)

(4)
(4)
(4)
(4)
(1) = XAD-2
column with chloroform, (2) - Direct extraction with water, (3)  = Direct
extraction with methylene chloride, (4) = Direct extraction with chloroform.
                                     30

-------
     Small  samples of the pads were used to collect contaminants from water and
soil .and then extracted by chloroform or methylene chloride in preparation for
analysis.  However, blank tests showed very strong response with a large
number of reagents, indicating that significant interfering materials were
being extracted from the pads.  In view of these  findings and the fact that the
pads are not intended nor quality-controlled for  analytical work, the concept
was abandoned.

     For subsequent TLC tests, chloroform was used for extraction.   For other
tests, water was used for extraction.  Methylene  chloride is used as a collec-
tion solvent for alcohol in soil.  As noted previously, chloroform cannot be
used because it contains traces of alcohol.  Unfortunately, a procedure for
recovery of alcohol from water using methylene chloride was not developed.  The
method for recovery of contaminants from soil is  given in the Appendix, Ana-
lytical Procedures.

     The proposed  procedures were evaluated using three types of soil—low
organic (beach sand), medium organic (sand/clay mix, some organic),  and high
organic (dark soil, high organic content).  Recovery of contaminants from soils
having high or low organic content was essentially the same; therefore, these
two soils were simply grouped together as organic soil.  Findings are presented
in Table 5.  It is seen that recovery from sand was good in all cases.  Recov-
ery of contaminants from soils containing organic matter was generally good;
significant exceptions were potassium permanganate, butylamine, triethyl amine,
and maleic acid.

EVALUATION IN THE NATURAL ENVIRONMENT

     Tests detailed in the manual for the identification kit (Appendix) were
evaluated on natural waters and soils.  Water samples from Winter's  Run, Otter
Point Creek, and Big Gunpowder Falls, three fresh-water streams in Harford
County, Maryland, were contaminated in the laboratory at concentrations of 1 mg
contaminant/ml of water.  Sandy, low organic, and high organic soils were
collected from areas neighboring CSL and contaminated in the laboratory at
approximate concentrations of 1 mg of contaminant per loosely packed ml of
soil.

     All contaminants were detected with the test procedures with the exception
of those in Table 5 that were marked as poorly recovered.  Findings with water-
borne contaminants were generally more reliable.  Response with TLC, detector
tubes, detector papers, arsine/Gutzeit, spot tests, and CHEMets was  good.   The
greatest variation was found with the inorganic spray reagents.  The reactions
were time-dependent and subject to a number of other factors, such as degree of
dryness of the plates, temperature, etc.  As a result, variations were found in
colors and in the minimum detection sensitivity.  However, the degree of varia-
tion was not so great as to preclude use of these reagents, which contributed
much valuable information toward identification of inorganic contaminants.

     It is evident that testing in soil is a nonroutine, highly complex job
requiring great deductive reasoning and skill on the part of the operator. Soil
presents a very complex background from which individual pieces of data must be
extracted and then presented as a meaningful assemblage of information.  The
complexity of the problem is evidenced by the chloroform extracts of the soil

                                     31

-------
samples, which can range  from colorless  for  beach sand to dark brown for some
of the soils containing a high organic content.  While the effects of back-
ground for the sand samples are minimal, soils with a high organic matter
content often show a significant loss of sensitivity.  However,  ia the case of
a very sensitive test, such as the hydrogen peroxide CHEMet, care must be taken
even with sandy soils.

     Loss of response may be attributed to interactions between  the contaminant
and/or reagent with the soil background.  This can be a particular problem with
acid/base compounds in soils of different acidity.  Another example of loss of
response is demonstrated by potassium permanganate;  it is easily identified in
sandy soils but not in those containing organic matter.  The permanganate is
reduced by organics and is no longer detected by the proposed procedures.   The
reaction of the permanganate ion with the soil background is evident from the
extracts, wherein those from sand are colored purple, and those  from organic
soils are light yellow.

     It is possible to identify some contaminants by their decomposition
products.  However, the processes are very complex and were not  within the
scope of this program.  It may be concluded that the probability of detecting
and identifying both inorganic and organic contamination in loose beach sand is
very good.  Identification in soil with high organic content is  also possible
in some cases, but one should proceed with greater caution.

     As part of this program, an interference study was conducted to determine
the proposed kit's ability to function in the natural environment.  A random
collection of water and soil samples was made from a circuitous  routs about the
upper portion of the Chesapeake Bay in Maryland extending down to the William
Preston Lane, Jr. Memorial Bridge.  Water samples were taken from fresh-water
streams that drain into the bay; soil samples were taken from locales adjacent
to the water sources.  Assuming that the samples collected were  free of chemi-
cal contamination, the number and types of interferences were noted.

     As has been indicated, interference from constituents normally found in
inland waters is negligible with the proposed procedures.  However, as has also
been indicated, extracts of soils with high organic content are  often dark and
contain numerous potential interferences.  The extracts of many  of the soils
used in this study were extremely dark, although a few were nearly colorless.

     On the basis of data presented in Table 6, one responding to a spill
incident might suspect contamination in soil samples 8 and 15.  Such informa-
tion in combination with the circumstances surrounding that particular locale
might be the basis of followup analysis at a laboratory.   On the other hand, if
there are no mitigating circumstances to support the findings, one would pro-
ceed with more caution.  Those responses noted in the table that are consist-
ently weak are probably attributable to low-level background constituents.

     Accordingly, it is recommended in the manual that sampling  should be
accomplished in the following order of preference:  (1) neat materials, (2)
water samples, (3) sand/clay-type soil samples, and (4) dark, high organic
soils (very undesirable).
                                    32

-------
                  TABLE  6.   INTERFERENCE SCREENING  OF  KIT  PROCEDURES  WITH RANDOM WATER AND SOIL  SAMPLES
                                                                                                  en   en  CD  tfl   is   ta
                                                                                                  m   

                                                             £       ill    l^«'Jr^£cgg£S'*
                                                 t   t    oc,  §   a   |s   r   e|  I  fc   -ss   s   ^  s  5   §   §g   f

                   Hater Samples                   ^  z  j^  £   o?   
-------
STORAGE
     Reagents that showed potential for use in the kit were prepared and
placed in storage to obtain an estimate of shelf life.  This evaluation
included only reagents prepared on this program, not commercially available
spray reagents, detector papers, or detector tubes.
     Except as noted in Table 7, the reagents were kept in a bin in tightly
stoppered but unsealed brown bottles.  The reagents in bottles would be simi-
larly supplied in the kit but provided with Chromist spray assemblies for
attachment when needed.  The reagents were tested at monthly intervals.  Their
useful life is noted.
     It was noted that some sediment began forming in some of the bottled
reagents after 3 to 4 months, but this did not interfere with the reagent's
ability to respond to the test materials.
     Reagents for the various spot tests were also placed in storage.  However,
since many of these were still being modified up to several  months ago, data on
them is not as extensive.  Nonetheless, they are still reactive after 2 months
and should continue to be so for many more.
                TABLE 7.  SHELF STORAGE OF CHEMICAL REAGENTS
                    Reagent
 Useful Life
     1.  PAR, 0.16 gm/250 ml methanol
     2.  s-Diphenylcarbazone, 1.25 gm/25Q ml chloroform
                  TM
     3.  Dithi Ver  , 10 capsules/250 ml chloroform
     4.  Dithizone, 0.125 gm/250 ml chloroform
     5.  Silver nitrate, 1.25 gm/250 ml water
     6.  Chloranil, 1.25 gm/250 ml chloroform
     7.  Palladium chloride, 1.25 gm/250 ml methanol
     8.  PAR, pressurized can
     9.  Chloranil, pressurized can
About 6 months
About 6 months
About 6 months
About 6 months
About 6 months
About 6 months
About 6 months
Over 6 months
Over 6 months
                                     34

-------
                                  SECTION 5

                           IDENTIFICATION CONCEPT


     It is obvious that a kit that attempts to identify 300 materials would
require a considerable number of tests.  It would not be practical, even if
the tests were available (which they were not), to include a test for each
compound.  Accordingly, it was decided to select a group of tests that had a
range of response, which, used in combination, would serve to specify certain
materials or classes.  Therefore, an attempt was made to select from the tests
studied in this program a group that was neither too narrow nor too general in
response.  These tests are specified in the manual (Appendix).  The purpose of
selecting these tests was not only to identify the 36 test materials studied in
this program, but also to provide a range of response that may be capable of
identifying other materials from the Federal list of hazardous substances.

     Documenting the response data with only the 36 compounds used in this test
program was relatively easy.  However, if additional test data were produced
with other materials, classification and retrieval would have become increasingly
difficult.  Accordingly, a simple, manually operated retrieval system was
designed based on the use of Instant Data System (IDS) cards (Professional Aids
Co., Chicago, Illinois 60606).  The identification kits each have a set of 36
cards—one for each of the test materials studied.

     The IDA cards were printed with a standard format and pertinent data,
such as material description, response to different tests, Rf and Rs values,
and graphic representations of TLC separations.  The cards were notched at
appropriate positions to correspond to the test data.  A card prepared for the
identification of malathion is shown in the Appendix, Equipment and Methodology.

     If it is decided to screen other materials through the kit procedures,
similar cards can be prepared and added to the data base.  If the collection of
cards becomes unwieldy, sorting needles may then be used to facilitate separa-
tions.

     In addition to the above procedures, the identification kit has been
cross-referenced with EPA's detector kit to obtain additional identification
capability (Appendix, Supplementary Identification with the Hazardous Materials
Detector Kit).

     Information entered on the IDS cards for the materials studied in this
program is noted below.  The Rf and Rg data is based on separations made with
chloroform in a sandwich chamber at room temperature (see Experimental, Thin-
Layer Chromatography, and also the Appendix, Equipment and Methodology).


                                     35

-------
     Consideration was given to indexing the Rf and Rs values on the IDS cards.
However, it was decided that even with use of a reference dye, values were
subject to significant variation, particularly due to temperature.   According-
ly, it was decided not to notch the values, but to include them on  the cards
along with the graphic portrayal  of TLC separations, as qualitative informa-
tion.

     Numbers associated with test descriptors refer to analytical  procedures
given in the manual (Appendix, Analytical  Procedures).

DATA ENTERED ON IDS CARDS FOR 36  HAZARDOUS MATERIALS

     1.  ANTIMONY TRIOXIDE, colorless crystalline powder, slightly  soluble in
          water.
               11.  gray-blue

     2.   ARSENIC TRISULFIDE, yellow crystals or powder,  insoluble  in water.
               11.  gray-blue

     3.  CADMIUM CHLORIDE, small  white crystals, soluble  in water.
                3.  light yellow  precipitate
                4.  light white precipitate
                5.  red-purple
                6.  red-purple
                7.  yellow

     4.  LEAD THIOCYANATE, white  or light  yellow crystals, slightly soluble in
          cold water.
                3.  light black precipitate
                4.  light white precipitate
                5.  red-purple
                6.  red-purple
                7.  orange

     5.  MERCURIC CYANIDE, colorless crystals, slightly soluble in  water.
                3.  light brown
                5.  weak purple
                6.  weak purple
                7.  yellow

     6.  SELENIUM OXIDE, colorless needles, soluble in water.
                2.  orange
                3.  light yellow
                5.  weak yellow-orange
                7.  yellow-green

     7.  ZINC FLUORIDE, white powder, slightly soluble in water.
                4.  light yellow-green precipitate
                5.  red-purple
                6.  bright red-purple
                7.  red-orange
                                     36

-------
 8.  SODIUM ARSENATE, colorless crystals, soluble in water.
           7.  small yellow spot
          11.  gray-blue

 9.  SODIUM ARSENITE, grayish-white powder,  soluble in  water.
           3.  light yellow
          11.  gray-blue

10.  POTASSIUM BICHROMATE, yellowish-red crystals,  soluble in  water.
           1.  dark green
           3.  dark brown
           5.  weak purple
           6.  purple
           7.  yellow

11.  POTASSIUM CHROMATE, yellowish-red crystals,  soluble in  water.
           1.  dark green
           3.  dark brown
           5.  weak purple
           6.  purple
           7.  yellow

12.  POTASSIUM PERMANGANATE, dark violet crystals,  soluble in  water.
           1.  dark green
           3.  dark brown
           4.  light yellow-brown precipitate
           5.  purple
           6.  red-purple
           7.  gray

13.  PROPYL ALCOHOL, colorless liquid, soluble  in water.
           3.  light brown
          21.  red; test is only applicable  to  neat samples  and soil
               testing

14.  ACETALDEHYDE,  colorless liquid,  miscible with  water.
           9.  purple
          13.  light brown
          15.  green

15.  BUTYLAMINE
           3.  light brown
          16.  blue
          18.  dark brown to black
          19.  purple
          20.  white
                 Rf = 0.0, RS  = 0.0

16.  TRIETHYLAMINE, colorless  liquid, slightly  soluble  in  water.
           3.  light brown color
          16.  blue
          18.  gray


                                37

-------
          19.  purple
          20.  light brown
                 Rf = 0.0, RS = 0.0

17.  ANILINE, colorless liquid, slightly soluble in  water.
           4.  brown
          12.  purple on precleanse layer,  purple on indicator  layer
          18.  dark brown to black
          19.  yellow-green
          20.  light brown
                 Rf = 0.33, RS = 0.46

18.  AMYL ACETATE,  colorless liquid,  very slightly soluble  in water.
          10.  purple
          15.  black

19.  2,4-D (ESTERS), colorless liquid,  very slightly soluble in water.
          10.  brown-orange
          17.  light yellow
          18.  gray
                 Rf = 0.70, RS = 0.98

20.  ISOPRENE, colorless volatile liquid, insoluble  in water.
          12.  dark brown on precleanse layer
          13.  light brown
          14.  dark yellow
          15.  dark brown

21.  NAPHTHALENE, white crystalline solid,  insoluble in water.
          12.  dark blue on precleanse  layer
          14.  dark gray

22.  ACRYLONITRILE, colorless liquid, partially  miscible  in water.
          13.  light brown
          15.  brown-green

23.  BENZONITRILE,  colorless liquid,  slightly soluble in  water.

24.  NITROPHENOL, yellow crystals, soluble  in water.
           4.  red-orange
           8.  dark yellow
                 Rf = 0.83, Rs = 1.16

25.  ACETIC ACID, colorless liquid, miscible in  water.
          16.  weak yellow
          18.  light brown-black
                 Rf = 0.07, RS = 0.10

26.  2,4-D (ACID),  white to yellow crystalline solid, slightly  soluble
      in water.
          12.  red-brown on precleanse  layer
          15.  yellow-black


                               38

-------
          16.  yellow
          18.  brown to black
          19.  light white
          20.  light white
                 Rf = 0.09, Rs » 0.12

27.  MALEIC ACID, colorless crystals, soluble in water.
          16.  yellow
          18.  white
          19.  light white
          20.  light white
                 Rf = 0.0, R$ = 0.0

28.  BENZOIC ACID, colorless to white needles or scales,  very slightly
      soluble in water.
          16.  yellow
          18.  white to light gray
          19.  light white
          20.  light white
                 Rf = 0.13, Rs = 0.18

29.  DIELDRIN, pale tan flakes, insoluble in water.
          17.  yellow
          18.  gray to white with brown ring
                 Rf = 0.90, Rs = 1.26

30.  HEPTACHLOR, white waxy solid, insoluble in water.
          17.  yellow
          18.  gray to white with brown ring
                 Rf = 1.00, Rs = 1.40

31.  DDT, colorless or white powder,  insoluble in water.
          17.  yellow
          18.  gray to white with brown ring
                 Rf = 1.00, RS = 1.40

32.  CHLORQBENZENE, colorless volatile liquid, insoluble  in water.
          12.  light red on indicating layer
          14.  gray
          15.  dark brown

33.  DIAZINON, colorless liquid, slightly soluble in  water.
          12.  red-purple on precleanse layer, red-purple on  indicating
               layer
          14.  gray-green
          15.  brown-green
          18.  white to white with gray ring
          20.  brown to yellow with brown ring
                 Rf = 0.27, Rs = 0.38

34.  MALATHION,  yellow to dark brown  liquid, very slightly soluble  in
     water.
          14.  gray-green

                                39

-------
          15.   dark green
          18.   white to white with gray ring
          20.   brown to yellow with brown ring
                 Rf = 0.33, RS = 0.46

35.  PARATHION, yellow liquid, slightly soluble in water.
           3.   light brown
          14.   yellow-green
          15.   green-yellow
          18.   white to white with gray ring
          20.   brown to yellow with brown ring
                 Rf = 0.76, Rs = 1.06

36.  PHENOL, white crystalline mass,  soluble in water.
           4.   dark purple
          18.   gray-brown
          19.   light orange
                 Rf = 0.16, RS = 0.22
                                40

-------
                                  SECTION 6

                               PROTOTYPE KITS


     This program has resulted in a two-component identification kit—an in-
verter/shortwave UV lamp unit for photochemical reactions and a larger reagent/
auxiliary equipment package.  Prototype kits fabricated according to this
design are shown in Figures 5, 6, and 7.  The manual for the kit's use is given
in the Appendix of this report.  Two prototype kits, drawings, and parts lists
were provided to EPA under separate cover.

     The inverter/UV lamp unit measures about 7 in x 8 in x 16 in (17.8 x 20.3
x 40.6 cm) and weighs 15 Ib (6.8 kg).   The reagents/equipment package measures
8 in x 10 in x 24 in (20.3 x 25.4 x 61.0 cm)  and fully packed, weighs 13.5 kg
(30 Ib).  The procedures include thin-layer chromatography, commercial detector
tubes, spot tests, an arsine/Gutzeit test, CHEMets, detector papers, a variety
of spray reagents, and cleanup and recovery equipment.

     The photochemical tests required the use of a high-intensity shortwave UV
light.  In reviewing the availability of high-intensity lamps, none were found
that are battery-powered.   Accordingly, for use in the field, it was decided to
use an inverter to drive the lamp by converting power from the transporting
vehicle's 12-V DC battery to 115-V AC.  Two reasonably compact inverters were
ordered from Trippe Manufacturing Company, Chicago, Illinois 60606.  Model  PV-
115 was found to deliver insufficient power, but Model PV-200 was adequate  to
develop the photochemical  reactions.

     Use of the inverter significantly added to the size of the kit.  It
measures 5 in x 5 in x 3 in (12.7 x 12.7 x 7.6 cm) and weighs about 6 Ib (2.7
kg).   However, the inverter was considered critical to the kit concept and was
accepted nonetheless.  Furthermore, it was found that the inverter emits a
considerable amount of heat.  Accordingly, the cooling fins were modified to
function as a small  low-temperature hotplate, which is used to develop the
chloranil test.

     The UV lamp used with the inverter is Mineralight, Model R-52, from Ultra-
violet Products, Inc., San Gabriel, California 91778.   In operation with a
standard-size vehicle, the UV lamp/inverter unit drew 7.2 amps from a 12-V  DC
battery with the engine off/ with the engine in idle,  it drew 7.5 amps.   It is
planned to run the photochemical tests and the chloranil  test together for  a
10-minute period.   There is no problem expected in using the UV lamp/inverter
unit in this mode.  The unit has a cable for hooking directly to the vehicular
battery.
                                     41

-------
ro
              Figure  5.  Hazardous materials spills identification kit, inverter/UV lamp unit.

-------
Figure 6.   Hazardous materials  spills  identification kit,  reagents/auxiliary equipment package.

-------
figure 7.  Hazardous materials  spills  identification kit,  reagents/auxiliary equipment package.

-------
     Many of the other components used in the identification kit are commerci-
ally available.   However, a wide range of reagents were developed especially
for its use.  Unfortunately, some of them have limited shelf life and,  accord-
ingly, should be prepared on a regular basis or they should be monitored.   More
information on stability has been given in Section 4, Experimental  Results.

     A loading diagram for the kit is shown in the Appendix, Equipment  and
Methodology.
                                     45

-------
                                 REFERENCES

 1.   Silvestri, A., A.  Goodman, L.M.  McCormack, M. Razulis, A.R.  Jones, Jr.,
     and M.E.P. Davis.   Development of a Kit for Detecting Hazardous Material
     Spills in Waterways.   Environmental Protection Technology Series.
     EPA-600/2-78-055.   March 1978.
 2.   Hamilton, D.J., and B.W. Simpson.  J. Chromatog.  39_.  1969.   p.  186.
 3.   Braun, R.A., W.A.  Mosher.   J.  American Chemical  Society,  80.  1958.
     p.  2749.                                                  —
 4.   Katzoff, L., A. Silvestri.  U.S. Patent 3,956,281.   1976.
 5.   Karush, F., N.R.  Klinman,  and  R. Marks.  Analytical  Biochemistry,  9.
     1964.   p. 100.
 6.   Grant, L., C.R. Sherwood,  K.A. McCully.  J. Chromatog. 44_.   1969.   p.  67.
 7.   Watts, R.R.  J. of the A.0.A.C.   48.  1965.  p. 1161.
 8.   Getz,  J.E., and R.R.  Watts.  J.  of the A.O.A.C.  47_.  1964.  p.  1094.
 9.   Fishbein, L., and  J.J. Fawkes.  J. Chromatog.  22_.  1966.   p.  323.
10.   Stanley, C.W.  J.  Chromatog.  16.. 1964.  p. 467.
11.   Krzeminsky, L.F.,  and W.A. Landmann.   J.  Chromatog.   10.  1963.   p. 515.
12.   Adamovic, V.M.  J. Chromatog.   23_. 1966.   p.  274.
13.   Blinn, R.C.  J. of the A.O.A.C.   47_.  1964.  p. 641.
14.   Fishbein, L., and  J.  Fawkes.  J. Chromatog.  20_.  1965.  p.  521.
15.   Eastman TLC Visulization Reagents and Chromatographic Solvents.  Eastman
     Kodak Company.  Rochester, New York.   1973.
16.   Dutt,  M.C., and P.H.  Seow.  Agricultural  and Food  Chemistry.   VL  1963.
     p.  467.
17.   Benesch, R., R.E.  Benesch, M.  Gutcho, and L.  Lanfer.   Science.   123.
     1956.   p. 98.
18.   Munson, J.W.  J.  of Pharmaceutical Sciences.   63_.  1974.   p.  252.
                                     46

-------
19.   Benson, W.R., and J.M. Finocchiaro.  J. of the A.O.A.C.  48. 1965.
     p. 676.                                                  ~~

20.   Feigl, F.  Chemistry of Specific, Selective, and Sensitive Reactions.
     Academic Press, Inc.  New York, New York. 1949.  p. 647.

21.   Feuer, H.  The Chemistry of the Nitro and Nitroso Groups, Part I.
     Interscience Publishers.  New York, New York. 1949.  p. 647.

22.   Pesez, M., and M. Bartos.  Colorimetric and Fluorimetric Analysis of
     Organic Compounds and Drugs.  Marcel Dekker, Inc.  New York, New York.
     1974.   p. 276.

23.   Ibid,  p. 316.

24.   Ibid,  p. 42.

25.   Flashka, H.A., and A.J. Barnard, Jr.  Chelates in Analytical Chemistry,
     Vol.  4.  Marcell Dekker, Inc.  New York, New York.  1972.  p.  116.

26.   Feigl, F., and V. Anger.  Spot Tests in Inorganic Analysis, 6th Eng. Ed.
     Elsevier Publishing Co.  New York, New York.  1972.  p. 307.

27.   Orug-Skreen.  Drug screening system for the analysis of commonly abused
     drugs  in urine.  Brinkmann Instruments, Inc.  Westbury, New York.
                                     47

-------
                     LIST OF MANUFACTURERS AND SUPPLIERS

 1.  Analtech, Inc., Newark, Delaware 19711
 2.  Bendix/Gastec: National Environmental Instruments,  Inc.,  Warwick,  Rhode
     Island 02888
 3.  Brinkmann Instruments, Inc., Westbury, New York 11590
 4.  Case-Mason Filling, Inc., Joppa, Maryland 21085
 5.  CHEMetrics,  Inc., Warrenton, Virginia 22186
 6.  Conwed Corporation, St. Paul, Minnesota 55101
 7.  Draeger: National Mine Service Co., Pittsburgh, Pennsylvania  15219
 8.  Fisher Scientific Co., Silver Spring, Maryland 20910
 9.  Gelman Instrument Co., Ann Arbor, Michigan 48106
10.  Kitagawa: Matheson Gas Products, Lyndhurst, New Jersey 07071
11.  Midwest Research Institute, Kansas City, Missouri  64110
12.  Mine Safety  Appliances, Pittsburgh, Pennsylvania 15235
13.  Professional Aids Co., Chicago, Illinois 60606
14.  Schleicher and Schuell, Inc., Keene,  New Hampshire  03431
15.  Scientific Products, Division of American Hospital  Supply Corp., McGraw,
     Illinois 60085
16.  Trippe Manufacturing Co., Chicago, Illinois 60606
17.  VWR Scientific, Inc., Baltimore, Maryland 21227
                                     48

-------
                                   APPENDIX

         OPERATOR'S MANUAL FOR HAZARDOUS MATERIALS IDENTIFICATION KIT


                                   SECTION 1

                                 INTRODUCTION


     The Hazardous Materials Spills Identification Kit (ID Kit) is intended to
aid in the identification of hazardous materials spills in water and soils.
The kit consists of two units -- an inverter/shortwave UV lamp unit for photo-
chemical reactions, and a larger package with reagents and accessory equipment.

     The ID Kit uses a variety of commercial and laboratory tests, which are
adapted for field use.  Because many tests are not "shelf-available," it is •
necessary for the operator to maintain the reagents in a ready-state condition.
This may require monitoring and preparation of fresh batches on a regular
basis.

     Included in the ID Kit are test papers, detector tubes, spot tests, spray
reagents, and thin-layer chromatography (TLC).  In addition, there is equipment
to facilitate the recovery of contaminants from water and soil.

     The skills required to operate the kit are easily learned; however, the
task of interpreting the derived data is more challenging.  It is evident that
the operator's level of training and background experience will play a signifi-
cant role in combining and interpreting certain pieces of information in order
to come to reasonable conclusions.

     Identification may be accomplished by comparing the data derived from
testing with that contained in the data system included with the kit.  By com-
paring vital pieces of information, it is possible to identify some materials
specifically, or in other cases, narrow them to particular classes.

     It will be found that a number of factors affect the performance of the
kit.  For example, when the kit was developed, testing was accomplished using
relatively pure test materials.  In a "real-world" situation the time between
the occurrence of a spill and its analysis will have a significant effect on
oxidation, hydrolysis, and decomposition of the contaminant.   The results can
be shifts in color response, multiple TLC spots, and loss of sensitivity.  It
is evident that analysis should commence at the site as quickly as possible and
with neat material -- solid or liquid -- if possible.  If neat materials are
not present, then the order in which samples should be taken are:  (1) water,
(2) clean sandy soil, (3) sand-clay mix, and (4) black, loamy soil.   Interfer-

                                     49

-------
ence, in the-last case, is significant and should be avoided when possible.

     The ID Kit is largely designed to function alone.  However, if the EPA's
Detection Kit for Spilled Hazardous Materials is available, additional  informa-
tion can be gained, particularly in regard to inorganic contamination.   Ac-
cordingly, a section of this manual describes supplementary identifications,
which can be accomplished using the Detection Kit.
                                      50

-------
                                  SECTION 2

                         EQUIPMENT AND METHODOLOGY


     The components of the ID Kit are shown in Figure A-l (the inverter/lamp
unit) and Figure A-2 (the reagents and equipment loading diagram).  The test
procedures used in the kit are described in the analytical section; special-
ized processes or procedures are presented in the following paragraphs.

     Sampling and Recovery.  The contaminant at the spill site may be found
under a variety of conditions.  That is, it may be suspended or dissolved in
water, intimately mixed with the soil, or neat, as a pile of dry material or a
puddle of liquid.

     Water samples to be analyzed by TIC procedures require that the contami-
nant be extracted in an organic solvent.  The apparatus shown in Figure A-3
is used to accomplish this task.

     Remove the upper sealing cap and snip off the sealing tip from a column.
Press-fit a reservoir cup to the top of the column and suspend as shown from
the side of the case.  Pass 20 ml of water containing dissolved or suspended
contaminants through the column.  (Swirl the sample to obtain a representative
sample of suspended matter.)  Water that passes through the column may be dis-
carded.  Excess water may be forced through by applying pressure to the reser-
voir cup with the palm of the hand.

     Press-fit a filter cartridge to the lower end of the column.  Add 10 ml
of chloroform to the reservoir and loosen the upper cotton plug with tweezers.
The filter cartridge contains a phase filter, which allows the solvent to pass
through while retaining excess water.  Collect the chloroform extract in the
small bottle and use it for the TLC tests.

     In many instances described in the analytical procedures, it is necessary
to make an extraction of the soil sample.  Use the filtering apparatus de-
scribed below to filter out all suspended matter prior to analysis.

     When neat samples of contaminants are encountered, they may be dissolved
directly in water or chloroform as specified in the test procedures.

     Filter Assembly.  It may be desirable in some instances to filter the ex-
tract from soil samples.   For this purpose, the apparatus shown in Figure A-4,
for which glass fiber filters are provided, is used to obtain clear samples.
A sample of the extract is drawn into the syringe; a filter holder with a
filter in place is press-fitted to the tip of the syringe; and a sample of
extract is forced thorugh the Tygon tube into a 50-ml beaker for analysis.

                                     51

-------
en
ro
BLACK TERMINAL
 (NEGATIVE)

            RED TERMINAL
             (POSITIVE)
                                                   SHORT WAVE
                                                    UV LAMP
                                                                                                               POSITION FOR
                                                                                                               UV IRRADIATION
                                                                                                     LOW TEMPERATURE
                                                                                                        HOT PLATE
                                            Figure A-l.   Inverter/lamp unit.

-------
                                              TOP
                                            BOTTOM
EFFERVESCENT
  JAR
MCTHYUNE
CHLORIDE

EXTRACTION
 COLUMNS


ETMANOLAMI
                     WIPHENYL-
             FILTER    CARBAZONE
             ASSEMBLY
                    ARSINE
                   , TUMI \ 0. ItNZfHf
                                                  CHLORANIL   SROMOCRESOL
                                                             GREEN
PHENOLS/HYDROGEN
PEROXIDE CHEMETS
                                      QIQIQIQIQ
                     HYDROXYIJUWNE
                     HYDROCHLORIOE

                            HYOROCHUORIC
                              ACIO
          SANDWICH CHAKWER
           ASSEMBLr
    SODIUM
    NITROPRUUIOE
                                                                              EXTRACT
                                                                              SOTTLIi
                                 30 & 90 ML SEAKERS
                                 COLUMN RESERVOIRS
                                                VANADIUM IV)
                                                9-HYO RO XVQUINOUNE
                                              TRAY
                                            STIRRING ROD
                                            SCOOPULA
                                            TONGS
                                            FILE
                                            Mf TRIC RULER
                                            NO. 2 PENCIL
                                            EDGE ICRAPER
                                            TEST TU*E 9HUSH
                                                                      CMROMATOGRAPMIC
                                                                      TROUGH » COVER
                                                                  N
                                                                                 „ MICROPIPETTES
                                                          \        \
                                                          9ALUUT        TWEEZERS
                                                          DROPPERS       SNAPPER
                                                          TYGON PITTINOS  0.1 GM MEASURING SCOOP
                                                                       COLUMN HOLDER
              Figure  A-2.   Reagents  and equipment  loading  diagram.
                                               53

-------
   STEP
   (A)
STEP
 (8)
             AQUEOUS
             SAMPLE
                       EXTRACTING    \	
                         SOLVENT  ^~-U  	.
                COTTON PLUG
               AMBERLITE XAD-2
                 CARTRIDGE
                COTTON PLUG
                     COLUMN
                     SUSPENDED
                     FROM SIDE
                     OF CASE
Figure  A-3.  Extraction  system  for water.

                     54

-------
               FILTER HOLDER
                    TYGON TUBE                            SYRINGE


                        Figure 4.  Filter assembly.


     Inverter/Lamp Unit.  The high-intensity shortwave UV lamp is required for
photochemical  reactions.  The lamp operates on 110-V AC, which is provided by
the inverter.   The inverter, in turn, derives power from a common 12-V DC bat-
tery, such as  is found in the transporting vehicle.

     The unit  is attached to the vehicular battery with provided clips and
cable.   Pay particular attention when attaching the positive (red)  and negative
(black) clips  to the proper poles of the battery.  Once hookup is made, both
the lamp and inverter switches should be in the off position until  needed.  It
should be noted that the inverter, when on, will draw current even if the lamp
is off.

     The inverter, which emits considerable heat, has been modified to act as
a low-temperature hot plate for several tests.  The positioning of TLC plates
for heating and shortwave irradiation is shown in Figure A-l.  The TLC plates
are most easily positioned by removing the lamp from its retaining handle.

     The inverter/lamp unit can be used to accomplish the required tests in
about 10 minutes.  In order to minimize battery drain, it is recommended that
the tests be planned to run simultaneously.  If the vehicular battery is in
good condition, the power drain is insignificant.  If there is some question,
the inverter/lamp unit can be functioning while the transporting vehicle is
idling.

     Be certain that both the inverter and lamp switches are turned off when
tests are completed.

     CHEMets.   CHEMets are glass ampoules containing color-forming reagents
under vacuum;  some ampoules have additional reagents coated on the surface of
the tapered tip.  To conduct a test with a CHEMet, collect a sample of water
in a beaker.  Place a CHEMet in the barrel of the Snapper, tapered tip first
(Figure A-5).   Immerse the end of the Snapper below the water level and press
the CHEMet with your thumb to break the tip.  The ampoule will fill with water;
invert several  times to assure complete mixing.  Note the color and/or precipi-
tate that may  form.

     Detector  Tubes.  Some contaminants can be expelled from water by an ef-
fervescent material and, thereby, detected by detector tubes normally used for

                                     55

-------
                                                             CHEMet
                                                             AMPOULE
                                                           SNAPPER
                      TEST
                      SAMPLE
                           Figure A-5.   CHEMets.


airborne contaminants.   The test is performed using the  assembly  shown  in  Fig-
ure A-6.  Depending on  the test being conducted,  effervescence  can  be provided
using Alka-SeltzerR tablets or a zinc/potassium bisulfate/cupric sulfate mix.

     The test is performed with 80 ml of water in a jar.   A  file  is used to
score and remove the tips of a detector tube.  The tube  is inserted in  the
stopper top with the arrows pointing away from the jar.   Several  sizes  of
Tygon tubing are provided to accommodate the different size  detector tubes.

     The effervescent material is dropped in the  water and the  stopper  quickly
forced into the jar. The Draeger tubes have little resistance  and  the  top
need be held only lightly in place.  The Bendix/Gastec tubes  require that  the
stopper be held firmly.   Caution should be taken  to preclude water  from enter-
ing the tubes, since it  may interfere with the reaction.   When  effervescence
                                     56

-------
                               SOIL
                             TESTING
WATER TESTING
         «	DETECTOR TUBE
              CONNECTOR
                  4 OZ. JAR
              Figure A-6.  Effervescent jar assembly.

                                 57

-------
is complete, color forming on any part of the tubes is noted.

     It is found that when this apparatus is used to test for contamination
in soil, considerable foaming can occur, particularly with soils having a high
organic content.  To preclude water from entering the tubes, a ballast tank
(Figure A-6) is inserted in the system between the tubes and the stopper top.

     Spray Reagents.  Several of the reagents used in the kit are best applied
in a spray form.  Two of them, PAR and Bromocresol Green, are self-contained
and ready for use as received.

     The other reagents require that the propel!ant and the reagent be pack-
aged separately.  When they are needed, the Chromist Spray Unit is screwed
onto the reagent reservoir and applied in the conventional manner.  To avoid
cross-contamination, the spray units should not be interchanged with the re-
agents.  To extend the life of the reagents, they should be resealed as soon
as possible with the original lids.

     Thin-Layer Chromatography.  The TLC procedures are the most complex and
sophisticated portion of the ID Kit.  At the same time, they can provide some
of the most useful data to effect identification.

     Prepare a TLC plate as shown in Figure A-7.  Using a ruler, draw a light
line with a soft lead pencil 10 mm from the bottom of the plate; designate
this the origin.  Try not to damage the surface of the silica gel.  Draw
another line 80 mm from the bottom of the plate and designate it as the sol-
vent front.  With the scraper, remove a 15-mm width of silica gel from each
end of the plate.

     Spot the plate as follows:  Ignore the first segment; spot 8 microliters
Of chloroform extract in the center of each of the 2nd, 3rd, 4th, 5th, and 6th
segments on the pencil line that represents the origin.  On the 7th segment,
spot 3 to 4 microliters (estimated) of Sudan Yellow 3G standard.  Apply the
extract and standard in small increments blowing to dryness in between to
assure a tight, dense spot.

     Assemble the sandwich chamber as follows:  Align the vinyl spacer on the
silica gel s'ide of the plate so that it conforms with the sides and top,
which have been scraped clean of silica gel.  Cover with a clear glass face
plate (a spare may be found with the TLC plates package) and hold the "sand-
wich" together with binder clips as shown in Figure A-8, one at each of the
side and top edges.

     Fill the trough with 40 ml of chloroform (3-4 mm depth) and press on the
slotted plastic cover.  Place on a level, stable surface and insert the "sand-
wich" assembly; by proper location of the binder clips, the assembly can be
made to stand upright.  A small leaning forward or backward is acceptable.
When the solvent reaches the line for the solvent front, remove the "sandwich"
from the trough, disassemble, and allow to dry.

     Break the prescored plate as follows:  Turn the adsorbent surface up.
Grasp the clear end of the plate with the thumb and forefinger of each hand,


                                     58

-------
                                                          
                                                         O
                                                         o
                                                         (13
                                                         c:
                                                         o
                                                         ro

                                                         03
                                                         CL
                                                         
-------
                                                -Q

                                                5
                                                Wl
                                                R3
                                                 S-
                                                 O)
                                                 u
                                                 CO
                                                  OJ
                                                  s_
                                                  3
                                                  cn
60

-------
on either side of the appropriate score mark.  Twist hands sharply downward
to open score mark on bottom surface and snap plate.

     Use the proper visualization procedures to locate any contaminants.  When
a contaminant is found, note its color and mark the center of its apparent
mass or the leading and trailing edges of a streak.  Measure these distances
from the origin in mm.  The Rf value is the distance moved by the material,
divided by the travel of the chloroform (70 mm).  The Rs value is defined as
the distance traveled by the unknown, divided by the distance traveled by
Sudan Yellow 3G.  Record this data.

     Instant Data System (IDS).  The IDS system contained in the kit has cards
with data for specific contaminants (Figure A-9).  Data acquired at the scene
of an incident may be entered on a form provided for this purpose.  This in-
formation is then compared with that contained in the cards.  When the number
of data cards is small, this is easily done by observing the notches that are
evident on the edges of the stack.  However, as the card collection becomes
greater, it may be desirable to use sorting needles.

     Depending on the number of cards and the quality of data, it should be
possible to identify some compounds specifically, and narrow others to general
classes.  It is evident that many factors will affect the data, e.g., age of
the spill, background interference, concentration, color definition, etc.  It
is at these times that the skill and experience of the operator will be a
significant factor.
                                     61

-------
/°
o
o
o 8
o 8
o S
o jq
o a
O  2
0 <2
0 2
0 P
0 £
O «>
O oo
0 r-
O <°
o &
O f
O n
O  W N
>• 1
TS «=
O)
"Tn
T3
'3
cr
c
34. MALATHION, yellow to dark brow
14. Light brown

cx
"0

/
SUDAN YELLOW 3 G
o
a.
wl
2
=* n  s
Sim £V
rioo'o'
-t-CM
oooooooooooooooo
0 O
o
0
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
0
o
0
o
o
o
o
o
o
o
0
o
o
0 0
                                                         o

                                                         c
                                                         o
                                                         ta
                                                         o
                                                         
62

-------
                                  SECTION 3

                            ANALYTICAL PROCEDURES
     This section contains detailed procedures for using the ID Kit in the
analysis of contaminants in water, soil, or in a neat condition.  The ana-
lytical scheme is shown in Figure A-10 and the types of materials detected in
Table A-l.

     Before starting any analytical work, it is useful to record data regard-
ing the contaminant, such as color, physical form, solubility if in water,
odor, etc.

     As the analytical data is accumulated it should be entered in the form
shown in Figure A-ll.   When analysis is completed, the data should be compared
with that contained in the data cards.

DETECTOR PAPERS

Water Samples
                                J..L
     Test No. 1:  Immerse the Mn   test strip in the water sample until  wet,
then remove.  After 15 seconds, note the color.  Colors observed may range
from dark blue to dark green.
                                Ll,
     Test No. 2:  Immerse the Ni   test strip in the water sample until  wet.
Remove and note the color after 1  minute.  Colors may range from orange to
pink to red.

Soil Samples

     Prepare the sample by adding about 25 ml of loosely packed soil  to the
50-ml beaker.  Add distilled water to the 50-ml mark, stir well, and filter
with the filter assembly.  Complete tests 1  and 2 above.  Hold the sample for
tests by CHEMets, inorganic sprays, and spot tests.

Neat Samples

     Place 0.1 gm solid (level measuring spoon) or 2 drops liquid (eyedropper)
in the 50-ml beaker.  Add 50 ml distilled water, stir well, and allow to stand
5 minutes.  Complete tests 1 and 2 above.  Hold the sample for tests by CHEM-
ets, inorganic sprays, and spot tests.
                                    63

-------
CO

o












en
tn -;
w •<
Cl ^
5 ^ ^*
en c jg
a <
en w














V)
I SAMPLE
K
Cd
6-
1-^
















S '
Ci 1
O 1

D '
i £
uj oL
>• 0
H 0
£ 0 p-
U .
1 ' 1
1 1 1
1 ' 1
I i " J
2 i

1
1
1
1
1

1


I
1
—
1



1
1
t
I
1
j

|

|
I
1
1

|

1
""*
|

1
1
L




IN
1

^


cn
a
gj


u
	 • tn

.
i-4




cn
r*

W
~" ~0 EH
	 *H
0
cu
cn


z
"~ **O ^


z
H-i




0)
S
~ ^J »g
% w
s
u
a!

• .
H
a
_ _ _ __,
g i


•j
1
N
H
a



cn
a
a.
£ Ed
u a
> cn
O S OS
s: u cd
S9S
z < u

05 0\ O
rH

1
»-3 M &3
Hz z
N N
oi c ca P
cu in u a
• • «
to ^o rs


cn
i-l
. o
O Cd

                                        tn tn 3 iJ
                                        Cd Ed    M
                                        ai ei + z

                                        O O   <*BS   CM
                                        £ £ O O i-l
                                        a o 2 ~ cj
                                        Ct! OS  ftfl 3i "U
                                        O (O < O CU
                                                                    CU

                                                                    cu
                                                                   _C
                                                                    u




                                                                    o
                                                                   •^»
                                                                   4J
                                                                    >,


                                                                    c
                                                                    I
                                                                   <:
                                                                    0)
         64

-------
                TABLE A-l.   MATERIALS DETECTED BY THE ID KIT
   Test
 Process
             Test
                                 Materials Detected
Detector
Papers

CHEMets
Spot
Tests
Arsenic/
Gutzeit

Detector
Tubes
 1.
 2.

 3.

 4.
Inorganics   5.
Spray        6.
Reagents     7.
 8.
 9.
10.
12.
13.
14.
15.

16.
17.
Thin-Layer  18.
Chroma-
tography    19.
            20.
Hydrogen peroxide

Phenols
PAR
s-Di phenylcarbazone
Dithizone

Nitro compounds
Aldehydes
Esters
11.   Arsenic/Gutzeit
Draeger benzene
Draeger olefin
Ben/Gas benzene
Ben/Gas acetone

Bromocresol green
Bromocresol green
 + UV

Silver nitrate + UV

Chloranil + heat
Palladium chloride
Alcohols
21.  Alcohols
Certain inorganic metal  compounds
Certain inorganic metal  compounds

Certain inorganic metal  compounds;
inorganics
Certain inorganic metal  compounds.
organics, phenol

Certain inorganic metal  compounds
Certain inorganic metal  compounds
Certain inorganic metal  compounds

Organic nitro compounds
Acetaldehyde
Esters

Arsenic, antimony, aliphatic
arsenicals

Certain volatile organics
Certain volatile organics
Certain volatile organics
Certain volatile organics

Amines, certain organic  acids
Certain organochlorine materials
Amines, organic acids, organochlor-
ine materials, thiophosphates, phenol
Amines, certain organic acids, phenol
Amines, acids, thiophosphates

Alcohols
                                    65

-------
Test Process
Test
Response
DETECTOR
PAPERS
CHEMecs
INORGANICS
SPRAY
REAGENTS
SPOT
TESTS
ARSENIC /GUTZEIT
DETECTOR
TUBES
THIN
LAYER
CHROMATOGRAPHV
ALCOHOLS
1 . Mn 	 , . . .
2 - N't"1""" . , , . .

* Uy^™ p.™,^. .
L PVionols ,, M

S- PAR 	

1 p-ffh-f^ne



10. Eg ''«'''•'' T-. --.-.—._

11. Arf»n^c/f»iit2fiit __..




15, Ben/^flf Ac«":ri"»

lf ?r 	 Tocn1 nr..n


19, rh1nTa"'H + Heat _ . ,


n A1/.^,,1a

                    Figure A-11.  Data  sheet  for ID kit.
                                     66

-------
CHEMETS

Water Samples

     Test No. 3:  Using a snapper, break the tip of a hydrogen peroxide am-
poule beneath the surface of a water sample collected in a beaker.   Invert
several times to assure mixing.  A response may be a brown or yellow color or
even the formation of a precipitate.

     Test No. 4:  Place 25 ml of sample in a beaker.  Stir the sample with the
coated tip of a phenol CHEMet until all the crystals of potassium ferricyanide
have dissolved.  Place the CHEMet in the barrel of the snapper and break the
tip below the surface of the water.  Invert several times to assure mixing.
A response may range in color from orange to brown to purple, and may even
include a precipitate.

Soil Samples

     Using the sample prepared for testing with detector papers, perform tests
3 and 4.

Neat Samples

     Use the sample prepared for testing with detector papers and complete
tests 3 and 4.  Stir the sample to obtain a representative mix of suspended
material.

INORGANICS SPRAY REAGENTS

Water Samples

     Snap off three segments of TLC plate using the procedure described in
Section 2, Equipment and Methodology, Thin-Layer Chromotography.  Spot 8
micro!iters of water sample on each segment using a micro-pipette.   The
pipettes fill easily with water, but may require pressure with the rubber bulb
to completely empty.  It is not necessary to keep spots compact; the water may
be allowed to come out in a continuous flow.  Allow the samples to dry.

     Test No. 5:  Spray one segment of TLC plate with PAR reagent.   Note the
color after 5 minutes.  Colors may range from orange to purple.

     Test No. 6:  Spray a second segment of TLC plate with s-diphenylcarbazone
reagent.  Note the color after 5 minutes.  Colors include bright red-purple
and deep purple.

     Test No. 7:  Spray the third segment of TLC plate with dithizone reagent.
Note the color immediately.  Colors can include yellow, yellow-green, orange,
and red-orange.

     If any results with the above tests are in doubt, they should be com-
pared with a distilled water blank.
                                     67

-------
Soil Samples

     Using the sample prepared for testing with detector papers, spot samples
as described above and complete tests 5, 6, and 7.

Neat Samples

     Use the sample prepared for .testing with detector papers,  and complete
tests 5, 6, and 7.  Stir the sample to obtain a representative  mix.

SPOT TESTS

Water Samples

     Test No. 8, Nitro compounds:  Crack off a segment of TLC plate.   With an
eyedropper, place one drop of the water sample on the plate, followed by one
drop of ethanolamine.  A deep yellow color is obtained with nitro compounds.
Since there is always some yellow color present due to the reagent,  this test
should be compared with a distilled water blank.

     Test No. 9, Aldehydes:  Crack off a segment of TLC plate.   In turn, add
the following materials:  one drop of piperidine, one drop of sodium nitro-
prusside, and with an eyedropper, one drop of water sample.  A  deep  purple
color is obtained with aldehyde.

     Test No. 10, Esters:  To 1 ml of test water in a test tube, add 0.5 ml
(10 drops) of hydroxylamine hydrochloride and 0.5 ml  (10 drops) of sodium
hydroxide.  Allow to stand 2 to 3 minutes.  Acidify with hydrochloric acid
solution (the test solution turns purple to red to orange to yellow).  Stop
when the soltuion just turns yellow.  Add 2 drops of ferric chloride.  Esters
may give a color ranging from brownish-amber to a deep violet.   Compare with a
distilled water blank if there is any doubt.

Soil Samples

     Use the sample prepared for testing with detector papers.   Complete tests
8, 9, and 10.

Neat Samples

     Use the sample prepared for testing with detector papers,  and complete
tests 8, 9, and 10.  Stir the sample to obtain a representative mix.

ARSENIC/GUTZEIT

Water Samples

     Test No. 11:  Add 80 ml of water sample to the effervescent jar.  Prepare
the stoppered cap with a Draeger Arsine Tube in place, with the arrow pointing
away from the assembly.  Add the contents of a Gutzeit packet to the jar and
stopper quickly and firmly.  Allow effervescence to continue for at  least 5
minutes.  A positive test is a gray purple color.  Only arsenic and  antimony


                                     68

-------
respond to this test.  Some aromatic arsenicals may not be detected.

Soil Samples

     Add 40 ml of loosely packed soil to the effervescent jar.  Fill with dis-
tilled water to 80 ml and stir well.  Insert the ballast between the stopper
and arsine tube and follow the procedure for analysis of water (Figure A-5).

Neat Materials

     Add 0.1 gm of the solid (a level measuring scoop) or 2 drops of the liquid
(eyedropper) to the effervescent jar and fill to the 80-ml mark with distilled
water.  Stir well and continue the procedure as in the analysis of water, using
the arsine tube.

DETECTOR TUBES

Water Samples

     For each of the following tests, add 80 ml of water to the effervescent
jar.  Add one Alka-SeltzerR tablet and close firmly with the cap which has the
proper detector tube affixed in place.  (Use the correct size Tygon tubing to
match the detector tube.)  Color formation on any part of the tube is consid-
ered a response.  Allow the assembly to stand until effervescence is complete.
Water must not be permitted to wet the contents of the detector tube.  If this
happens, discard the tube and repeat the test with a new one.

Tubes to be tested:

     Test No. 12.  Oraeger Benzene Tube.  Color can form on the precleanse
layer or the detection portion proper.  Colors can include reds, browns, blues,
and purple.

     Test No. 13.  Draeger Olefin Tube.  Color may range from light tan to
brown.  This tube is particularly sensitive to water and must be discarded if
the crystals become wet.

     Test No. 14.  Bendix/Gastec Acetone Tube.  Colors can include yellows,
browns, greens, and blacks.

Soil Samples

     Add 40 ml of loosely packed soil to the effervescent jar.  Fill with dis-
tilled water to the 80-ml mark and stir well.  Insert the ballast between the
stopper and the detector tube and follow the procedure for analysis of water.

Neat Materials

     Add 0.1 gm of the solid (a level measuring scoop) or 2 drops of the
liquid (eyedropper) to the effervescent jar and fill to the 80-ml  mark with
distilled water.  Stir well and continue the procedure as in the analysis of
the water sample.


                                     69

-------
THIN-LAYER CHROMATOGRAPHY

Water Samples

     Recover the contaminant from water using the procedure given in Section
2, Equipment and Methodology, Recovery from Water.  Prepare a complete TLC
plate as described in Thin-Layer Chromatography (Section 2) and spot with the
extract from the water sample.   Assemble the TLC "sandwich" chamber as de-
scribed and complete the separation.   Afterwards, break the plate into indivi-
dual segments and spray and develop with the following reagents.   Compute Rf
values when spots are found, using 70 mm as the solvent travel.  Also compute
the Rs values by dividing the distance traveled by the unknown, by the dis-
tance traveled by the standard dye (Sudan Yellow 3G).

     Test No. 16.  Bromocresol  green.  Crack off a segment of TLC plate and
spray evenly with Bromocresol green reagent.  Hold the plate by the clear glass
end using the plastic-tipped forceps.  Spots when found will be yellow; note
their colors and compute the Rf and Rs values.

     Test No. 17.  Bromocresol  green + ultraviolet light.   Take the plate pre-
pared above and place under the UV lamp as shown in Figure A-l  of Section 2.
Do not turn on the lamp yet.

     Test No. 18.  Silver nitrate + ultraviolet light.  Assemble the silver
nitrate spray reagent.  Take another segment of TLC plate and spray evenly,
again holding the plate with the forceps.  Install the plate under the UV lamp
adjacent to the Bromocresol green plate.  Do not turn  on the lamp yet.

     Test No. 19.  Chloranil.  Assemble the chloranil  spray reagent.  Hold
another segment of TLC plate with forceps and spray evenly with reagent.   Place
the plate on one of the heating surfaces of the inverter designated in Figure
1 of Section 2.

     Hook up the inverter cable to the 12-V DC vehicular battery, paying  par-
ticular attention to the polarity.  Turn on the inverter and lamp switches and
allow the tests to continue for 10 minutes.  After 10  minutes,  examine each of
the three plates for spots.  DO NOT LOOK AT THE PLATE  OR THE LIGHT SOURCE DUR-
ING IRRADIATION, AS SEVERE EYE DAMAGE IS PROBABLE.

     The Bromocresol green plate may have yellow spots in addition to those
that were originally present.  These must be recorded  separately.  Determine
their R,r and R  values.

     The silver nitrate spots range from white, light  browns, and grays,  to
dark browns and blacks.  With high concentrations of some contaminants, white
spots may be ringed with gray or brown.  Determine the Rf/R$ values.

     Test No. 20.  Palladium chloride.  Prepare the Palladium chloride spray
assembly.  Holding a segment of TLC plate with the forceps, spray it evenly
with the reagent.  Colors will  generally be light brown.  In very high concen-
trations, however, yellow spots with brown rings are found.  Determine the R,;
and R  values.

                                    70

-------
Soil Samples

     Add 10 ml of loosely packed soil to the 30-ml beaker..  Add 10 ml  of
chloroform, stir well, and allow to settle.  Filtering the extract with the
filter assembly if necessary, spot the TLC plates with the extract and chroma-
tograph and complete tests described above.

     Extreme caution must be used in this test.  Most organic soils will have
large quantities of highly colored extractable materials.  Some of these re-
sponses may be confused with those found with certain contaminants.  In all
cases, findings should be compared with results from other tests for possible
confirmation.

     When the soils are lightly colored, as in the case with some sand and clay
mixes, interferences are not nearly as severe.

Neat Materials

     Using a micropipette, estimate a solution of 10 micro!iters of liquid
contaminant in 10-ml chloroform.  In the case of solid samples, estimate 10 mg
of unknown (spatula tip) in 10 ml of chloroform.  In either case, use the
solution to spot the TLC plates and complete the tests described above.

ALCOHOLS

     This analysis is applicable only to soil samples and neat materials.

Soil Samples

     Add 10 ml of loosely packed soil to the 30-ml beaker.  Add 10 ml  of methy-
lene chloride, stir well, and allow to settle.  Filter if necessary with the
filter assembly.

     Test No.  21.  To 1 ml of the methylene chloride extract, add 2 drops of 1%
vanadium (V) 8-hydroxyquinoline and 2 drops of acetic acid.  After 1 minute a
deep red color develops in the presence of alcohols.  Again, as with the
chloroform extraction described earlier, care must be taken with soils contain-
ing organic materials, as interferences may be found.

Neat Samples

     Using a micropipette, estimate a solution of 10 microliters of liquid con-
taminant in 10 ml of methylene chloride.  In case of solid samples, estimate 10
mg of unknown (spatula tip) in 10 ml of methylene chloride.  Complete Test No.
21.
                                      71

-------
                                  SECTION 4

                    SUPPLEMENTARY IDENTIFICATION WITH THE
                      HAZARDOUS MATERIALS DETECTOR KIT


     The ID Kit is designed to function completely independently.  However, if
the EPA Hazardous Materials Detector Kit is available, additional information
may be gained to support identification of various materials.

     If time is available, the best approach is to run through the test pro-
cedures in the detection kit.  Overall, it is more sensitive than the ID Kit,
and as such can serve as a guide to optimum sampling location.  (It will be
remembered, however, that the detection kit is designed solely for water samp-
ling and analysis.)

     When contamination is indicated by the detection kit, analysis is then
completed with the ID Kit ~ paying particular attention, but not limiting
testing, to those procedures referenced in Table A-2.  This is not to suggest
that .if all tests with the detection kit are negative,, that contamination may
be presumed to be absent.  The detection kit can serve as a guide in exploit-
ing certain tests in the ID Kit; however, all potential contaminants have not
been tested with either kit.  Obviously, there may be situations in which a
compound undetectable by the detection kit may be found with the ID Kit.

     It is suggested that if both kits are going to be used, tests for phenol
and heavy metals in the detection kit be omitted in favor of tests 4, 5, 6, and
7 in the ID Kit.  The tests in the ID Kit are generally simpler to perform and
more reliable.


              TABLE A-2.  SUPPLEMENTARY IDENTIFICATION WITH THE
                      HAZARDOUS MATERIALS DETECTOR KIT
Positive Response to
Detector Kit Test for:	Remarks	

Cholinesterase            Could indicate very acidic materials (confirm by
                          pH meter).  Possible organic inhibitor; check tests
                          12 through 20 in ID Kit.

Benzene                   Generally volatile organics; check tests 3, 4, 8
                          through 9, and 12 through 20 in ID Kit.

                                                                 (continued)

                                     72

-------
                           TABLE A-2 (Continued)
Positive Response to
Detector Kit Test for:
                    Remarks
Heavy Metals
Phenol
Cyanide

PH
Conductivity
Nitrate N2
Color


Sulfate
Phosphate

Ammonia, ^
Chloride
Fluoride

Turbidity
Omit.  Use tests 1 through 7 and 11 in ID Kit.
Omit.  Use test 4 in ID Kit.
Blue color response is definite for cyanide.
Other colors may indicate organic or inorganic
contamination.
Could indicate organic or inorganic contamination.
General indicator of inorganic materials.
May indicate organic or inorganic materials.
Could be organic or inorganic materials.   However,
in TLC procedures, organic materials will gener-
ally show some movement, whereas the inorganics
will remain at the origin.
This test is very selective for sulfate.   No  inter-
ference is known.
May indicate phosphates, arsenates, arsenites, and
bromates.  Arsenic may be confirmed by test 11 in
the ID Kit.
May indicate a wide range of organic and  inorganic
materials containing nitrogen.
May indicate a wide range of organic and  inorganic
materials.
Fairly selective for fluoride compounds,  unless
very high concentrations of other materials are
present.
May indicate a wide range of low solubility organic
or inorganic compounds.
                                     73

-------
                                             SECTION 5

                                       CONSUMABLE MATERIALS
Cat.  No.           Description

P-1126-4    Quant Dip Strip, EM Labs,
            Manganese
 Unit

pkg/10
P-1126-5    Quant Dip Strip, EM Labs,         pkg/10
            Nickel

            Chemetrics, Type HP for           pkg/30
            Hydrogen Peroxide
            Chemetrics, Type P-12 for         pkg/30
            Phenols

            PAR: 0.63 gm 4-(2 Pyridylazo)
            resorcinol in 1-liter methanol
            s-Diphenylcarbazone: 5 gm in
            1 liter chloroform

            Dithizone: 0.5 gm Diphenylthio-
            carbazorie in 1 liter chloroform

5-718-50    Gelmari Chromist propellant        pkg/5
D-1008      Microliter pipettes, 8            vial/100
            microliters
 Unit
Price ($)

 9.25
                   9.25
                  15.00
                  15.00
                  17.00
                   5.00
     Source

Scientific Products
8855 McGraw Road
Columbia, MD 21045

Scientific Products
           Chemetrics, Inc.
           Mill Run Drive
           Warrenton, VA 22186

           Chemetrics, Inc.
                             Spray cans may be prepared by
                             Case-Mason Filling, Inc.,
                             Joppa, MD 21085

                             Laboratory
           Laboratory


           Fisher Scientific Co.
           7722 Fenton Street
           Silver Spring, MD 20910

           Analtech,  Inc.
           75 Blue Hen Drive
           Newark, DE 19711

-------
Cat. No.          Description                  Unit

10521       Hardlayer (organic binder)  TLC    box/25
            Uniplates, Silical Gel  HL,
            Pre-scored

            Gutzeit Bags                      ea
CH 25001    Draeger detector tube,  Arsine     box/10
            0.05/a
6718801     Draeger detector tube,
            Benzene 5/a

CH31201     Draeger detector tube,
            Olefins 0.05%/a

121         Bendix/Gastec detector tube,
            Benzene
151         Bendix/Gastec detector tube,
            Acetone

            Alka-Seltzer  (without aspirin)

3500010-1   Sample Reservoirs
3500021-6   Absorbent Cartridges

3500030-5   Filter Cartridges (snip off
            lower tips to fit carrying case)
box/10


box/20

pkg/100



pkg/100

pkg/100
                  Unit
               Price ($)

                  40.00
                   0.50
                  17.00
box/ 10
box/10
box/10
17.00
16.00
8.48
 8.48


 0.80

16.75



45.00

32.00
                Source
           Anal tech, Inc.
           Midwest Research Institute
           425 Volker Blvd.
           Kansas City, MO 64110

           National Mine Service Co.
           3001  Koppers Bldg.
           Pittsburg, PA 15219

           National Mine Service Co.
National Mine Service Co.


Bendix Corp., Environmental
& Process Instr. Division
P.O. Box 831
Lewisburg, WV 24901

Bendix Corp., Environmental
& Process Instr. Division

Drugstore

Brinkmann Instruments, Inc.
Cantiague Road
Westbury, NY 11590

Brinkmann Instruments, Inc.

Brinkmann Instruments, Inc.

-------
Cat.  No.            Description                 Unit

6801150-7   Bromocresol Green, aerosol  can    ea

61693       Prefliters, Type A Glass Fibers,   pkg/500
            25 mm
            Silver Nitrate: 5 gm in
            1 liter water

            Chloranil: 5 gm in 1 liter
            chloroform

            Palladium Chloride: 5 gm
            in 1 liter methanol

            Sodium Nitroprusside: 5% in water

            Sodium Hydroxide: 12.52! in
            methanol

            Hydroxylamine Hydrochloride:
            12.5% in methanol

            Hydrochloric Acid Solution: 10 ml
            HC1 and 10 nig Sodium Alizarin
            Sulfonate diluted to 100 ml with
            water

            Ferric Chloride Solution: 5 gm
            Fed 3 and 5 drops HC1 in 50 ml
            water

            Vanadium (V) 8-hydroxyquincline:
            1% in methylene chloride
  Unit
Price ($)

 12.00

 15.00
     Source

Brinkmann Instruments, Inc.

Gelman Instrument Company
600 So. Wagner Road
Ann Arbor, MI 48106

Laboratory
            Laboratory


            Laboratory


            Laboratory

            Laboratory


            Laboratory


            Laboratory




            Laboratory



            Laboratory

-------
Cat. No.           Description

            Sudan Yellow 3G: 0.1% in
            chloroform
            Ethanolamine
            Glacial Acetic Acid
            Chloroform
            Methylene Chloride
            Piperidine
            Distilled Water
Unit
   Unit
Price ($)
Source

Laboratory

Laboratory
Laboratory
Laboratory
Laboratory
Laboratory
Laboratory

-------
                                 REFERENCES


Silvestri, A., A. Goodman, L.M. McCormack, M.  Razulis, A.R.  Jones,  Jr.,  and
M.E.P. Davis.   Development of a Kit for Detecting Hazardous  Material  Spills in
Waterways.  Environmental  Protection Series.   EPA-600/ 2-78-055.  March  1978.
                                      78

-------
                                  TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing)
1. REPORT NO.
  EPA-600/2-81-///
                             2.
                                                         3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
   Development of an Identification  Kit for Spilled
   Hazardous Materials
             5. REPORT DATE
              September 1981
             i. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
   A.  Silvestri, M. Razulis, A  Goodman,  A.  Vasquez,
   and A.R.  Jones, Jr.
                                                         8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
                                                          10. PROGRAM ELEMENT NO.
   Chemical  Systems Laboratory,  U.S.  Army Armament
   Research and Development  Command,  Aberdeen Proving
   Ground, Maryland 21010
             11. CONTRACT/GRANT NO.
              IAG-D6-0098
12. SPONSORING AGENCY NAME AND ADDRESS
  Municipal  Environmental Research  Laboratory-Cin.
  Office of Research and Development
  U.S.  Environmental Protection Agency
  Cincinnati, Ohio  45268
        OH
13. TYPE OF REPORT AND PERIOD COVERED
  Final, July 1976 - March 197
             14. SPONSORING AGENCY CODE
              EPA/600/14
15. SUPPLEMENTARY NOTES
Joseph P. Lafornara, Project  Officer
16. ABSTRACT
     The  Chemical  Systems Laboratory (CSL) has developed  a  field  kit to identify
spilled hazardous  materials in.jn.land waters and on  the ground.   The Hazardous Materi-
als Spills  Identification Kit is a two-component kit consisting of an inverter/short-
wave UV lamp unit  for  photochemical reactions and a  larger  package containing reagents:
and auxiliary  equipment.   The identification kit was designed  as  an adjunct to EPA's
Hazardous Materials  Detector Kit to utilize existing 'instrumentation, equipment, and
Drocedures.  Thirty-six materials, representative of those  with the greatest hazard
DOtential,  were  selected and commercial sources were screened.  Procedures selected fo
the kit include: thin-layer chromatography, detector tubes, detector papers, CHEMets,
an arsine/  Gutzeit test, and a number of color development  procedures for use with the
thin-layer  chromatography.   In addition, methods were developed for recovery of con-
taminants from water and soil.  All information pertinent to identification of 36
specific  materials was designed into a compact data  retrieval  system, which is include
in the  kit.  Two prototype kits were delivered to EPA, along with a supply of consum-
able materials for evaluation.  In addition, manuals, engineering drawings, and parts
1ists were  provided.

This report was  submitted in fulfillment of Interagency Agreement No. EPA-IAG-D6-0098
and covers  the period  July 1976 to Marrh iq?q	Work was  r.nmpl pt.pri as. nf March 1979.
                                            d
                               KEY WORDS AND DOCUMENT ANALYSIS
                 DESCRIPTORS
                                             b./O£NTIFI£RS/OPEN ENDED TERMS
                          c. COSATl field/Group
   Thin-Layer Chromatography, Developing
   Reagents,  Soil  Extraction, Detector
   Tubes,  Test Papers, CHEMets, Arsenic,
   Inorganics, Organophosphates," Hydro-
   carbons,  Organic Acids, Pesticides,
   Nitriles,  Amines
  Method Development,
  Method Assessment,
  Adjunct to EPA HM
  Detector Kit
13. DISTRIBUTION STATEMENT
    Release to public.
                                             19. SECURITY CLASS (This Report/
	Unclassified	
20. SECURITY CLASS (This page)
                          21. NO. OF PAGES
                              87
                                                                       22. PRICE
                                                Unclassified
EPA Form 2220-1 (9-73)
                                      79

-------