EPA-660/2-73-00!
January 1974
                        Environmental  Protection Technology Series
     Evaluation of a
     Microwave-Induced
     Plasma  Spectrometer
     for Trace Analysis
                                      I
                                      55
                                      \
\
 UJ
 O
                                  National Environmental Research Center
                                  Office of Research and Development
                                  U.S. Environmental Protection Agency
                                  Corvallis, Oregon 97330

-------
                                EPA-660/2-73-009
                                January 1974
     EVALUATION OP A MICROWAVE-INDUCED
   PLASMA SPECTROMETER FOR TRACE ANALYSIS
                    by

           William Rudolf Seitz
Southeast Environmental Research Laboratory
           College Station Road
           Athens, Georgia 30601
               ROAP 16ADN-32
          Program Element 1BA027
  NATIONAL ENVIRONMENTAL RESEARCH CENTER
     OFFICE OF RESEARCH AND DEVELOPMENT
   U. S. ENVIRONMENTAL PROTECTION AGENCY
          CORVALLIS, OREGON 97330

-------
                        ABSTRACT

A low pressure microwave-induced plasma in helium was
evaluated as an excitation source for spectrochemical
analysis of trace metals in water.  The sample is
introduced into the plasma by thermal atomization from
a wire filament.  Although the system is sensitive,
calibrations vary in slope as a function of matrix and
filament material.  Evaluation of the microwave spec-
trometer was terminated because the instrument required
too much further developmental work to make it analyti-
cally useful and because other spectrochemical tech-
niques are better adapted to EPA needs.
                           11

-------
                        CONTENTS

                                                   Page
Abstract                                            ii
Acknowledgments                                     iv
Sections
I          Conclusions                               1
II         Recommendations                           2
III        Introduction                              3
IV         Experimental                              5
V          Results                                   8
VI         Discussion                               13
VII        References                               17
                           iii

-------
                     ACKNOWLEDGMENTS

The author thanks Ted Martin of EPA's Analytical Quality
Control Laboratory, Cincinnati, for helpful discussion.
                            XV

-------
                        SECTION I
                       CONCLUSIONS

The low pressure microwave plasma spectrometer delivered
to EPA needs further developmental work before it can be
successfully applied to trace metal analysis in natural
waters.

-------
                       SECTION II
                     RECOMMENDATIONS

The EPA should sponsor research on other spectrochemical
methods for simultaneous multielement analysis.  Atomic
fluorescence with rapid sequential detection, and RF
plasma emission with multiple detectors appear to be the
best possibilities at present.

If a reliable atomization system for introducing elements
into the plasma is developed, then the low pressure
microwave spectrometer should be reconsidered as an
emission source for simultaneous multielement analysis.

-------
                       SECTION III
                      INTRODUCTION

The EPA needs to be able to analyze for all elements in
a water sample simultaneously with sensitivities better
than 1 yg/liter.  One possible approach is atomic
emission spectrometry using a microwave-induced plasma
as an excitation source.  Microwave-induced plasmas have
high electronic temperatures that can efficiently excite
emission from elements with high excitation potentials.
The plasma itself has relatively few emission lines.
An emission source can be adapted to multielement
detection by using multiple detectors.

Most work using microwave plasmas for trace metal
analysis has been done in argon at atmospheric pressure
using a 2450 MHz discharge operating between 25 and 100
watts (1-6).  The biggest problem is getting the sample
into the plasma.  Introducing the sample as a wet
aerosol, as in atomic absorption spectrometry, causes
background emission from H_0 species and may extinguish
the plasma or cause it to be unstable (1).  Drying the
aerosol before introducing it into the plasma improves
analytical performance at the price of greater instru-
mental complexity (2-4).

Satisfactory analytical performance with good
sensitivity has been reported for thermal atomization
from loop-shaped filaments (5,6), but other people have
failed to achieve satisfactory results with thermal
atomization devices (1,3,7).

-------
Although operating a plasma at atmospheric pressure is
more convenient, reduced pressure offers several
advantages:

     •  A larger number of gases can be used for the
        the plasma.

     •  Greater sensitivity can be achieved at low
        pressures.

     •  The problem of metals plating out on the walls
        of the discharge tube can be avoided.

     •  Sample introduction by thermal atomization from
        a filament is facilitated by low pressure.

Good analytical performance has been reported for a
low-pressure argon plasma into which the sample is
introduced as a partially desolvated aerosol (8).
Introducing dry aerosol into a low pressure plasma has
not been reported and would probably be difficult.  Poor
reproducibility was observed for thermal atomization of
Hg from a copper filament into a low-pressure argon
plasma (7).

In this report a microwave plasma emission spectrometer
designed to operate at low pressures using thermal
atomization from a metal filament is found to be subject
to matrix effects that make it unsatisfactory for water
analysis.

-------
                       SECTION IV
                      EXPERIMENTAL
APPARATUS
The microwave plasma spectrometer was built on contract
by Monsanto Research Corporation, Dayton, Ohio, and
delivered to the EPA at the conclusion of the contract
period.  The spectrometer is described in detail in the
final report on the contract (9).

The spectrometer operates at low pressure with sample
introduction by thermal atomization from a heated metal
filament.  The plasma is confined in an 8 mm i.d. quartz
tube.  The position of the plasma relative to the view-
ing port can be varied.  Normal operating conditions for
the plasma were 30-40 watts power, 2-3 mm pressure and
a flow rate of approximately 250 cm  helium/minute.

A 0.025-cm. diameter Ta filament is located about 2 cm
above the plasma.  The filament geometry recommended by
Monsanto was an acute V.  A drop of sample is placed at
the point of the V.  After the water evaporates leaving
solid on the probe, the solid sample is thermally atom-
ized by discharging a 0.21-farad capacitor charged to
7-10 volts through the filament.  In other thermal
atomization systems, the filament heating current is
controlled by a rheostat (5-7).   The purpose of using a
capacitive discharge is to increase sensitivity by
introducing the entire sample into the plasma over a

-------
very short time period.  Emission-peaks viewed on the
oscilloscope had a half-width of less than ten milli-
seconds .

The emission signal is too fast to be recorded on a
strip chart recorder.  Instead, a peak-reading circuit
holds the peak value of the emission signal/ which is
printed by a digital printer.

The spectrometer can be programmed to do the following
operations automatically in sequence at preselected
time intervals:

     (1)   Turn on carrier gas  (helium)

     (2)   Evaporate sample drop on filament by passing
          current (0-1 A) through the filament

     (3)   Reduce the pressure

     (4)   Turn on plasma

     (5)   Atomize sample

     (6)   Print out peak height

     (7)   Turn off and vent system

PROCEDURES

Measurements were performed by removing the sample probe
with the filament from the spectrometer, placing a
five-microliter drop on  it, and replacing the probe  in

-------
the spectrometer.  The automatic program was then
initiated.  Analysis time was eight minutes.  Approxi-
mately five minutes were required to evaporate five
microliters at the maximum desolvation current.

-------
                        SECTION V
                        RESULTS
SENSITIVITY
The plasma spectrometer built by Monsanto provides
greater sensitivity than other microwave plasma spec-
trometers previously reported (1-8).  Table 1 lists
detection limits for the Monsanto spectrometer (9).  The
superior excitation efficiency of the low-pressure
helium plasma and the use of a capacitive discharge to
atomize the sample both contribute to the increased
sensitivity, but the relative contributions of each are
not known.

CALIBRATION CURVES

Table 2 lists the slopes of log concentration-log peak
height calibration curves under various conditions.
The slope is a function of filament material and matrix.
Since the slope changes with matrix for a given filament
material, the problem is probably with the process of
evaporating the drop onto the filament rather than with
the plasma itself.  This effect is reflected in the
data from Monsanto and was confirmed for Cd in river
water vs. Cd in deionized water.  Because of this,
natural water samples cannot be analyzed using calibra-
tion curves prepared in deionized water.

The sample properties that cause this effect are
unknown.  It could be related to surface tension,

-------
Table 1.  DETECTION LIMITS OBSERVED FOR  THE  MICROWAVE
          EMISSION PLASMA SPECTROMETER USING A
          LOW-PRESSURE HELIUM PLASMA  (9)
Metal
As
Cd
Hg
Pb
Detection Limit (yg/liter)
5
0.
0.
5

2
5

       Monsanto projects that these detection
       limits can be reduced 10X by optimization.

-------




to
a
i
a
CH
m
w
CO

Pi
O
Cn

S3
O X
H H
EH tf
< EH
K rij
S3
H O
rj p3
S3 ^
O

3
U H
O ftf
Hi H

* ^
w|S
EH
EH S3
ffi H
O S
H  O -P
o H! fd
I_J^ J^
. -p
m rale:
0 > 0)
0
Q) • C
&-P O
OKU
rH
CO

c
OCT
•HX.
-p 1 1.
fd p
SncS
jj
pj ...
Q) 2i
0 ?
d !rt
O (V
U



.p
c
0)

i-H
•H
h








^
•H
in
-P
g




H
(0
-P
0)
g




CTl







O
O
•
rH






r-.
•
CM

0
-p

in
i
o
rH
^
in


ti
(D
0

T3
o
+J
CJ
H








rH
m
-P
0)
g

tr>
m


tn
M

























O
U

ty
3.

n
I
o
rH
^




^
O
a
(0

ra
rd
























CTi







CM
TJ«
•
O









o
H

O
•P

rH
O
•
0







!S






O
(N
ffi

rrj
0)

•H
C
O
•H

i^


&i
K





cn







n
in
•
o










0
H

O
-P

H
•
O






fd
EH






O
CM
a

T3
(U
N
-H
d
0
•H
(D
Q


_Q
HI





0>







H

.
O









o
rH

O
-p

H
O
•
o







IS






O
CM
a

T5
(U
N
-H
d
o
••H
0)
Q


T5
U













p^
U)
•
o












CN
1
CM
O
O
•
O






fd
EH






O
CM
K

T3
0)
N
•H
0
•H
(U
Q


•d
u













o
o
•
rH












CM
1
CM
O
O
•
O






fd
EH







J_)
(U
^
-H
W

0)
(U ^
d (U
O -P
u fd
o is


•o
u





CTi







CM
CO
.
O










0
H

O
•P

rH
•
O







IS






o
CM
|U

•o
0)
N
-rl
d
o
-H
(D
Q


ra

-------
solubility of the metal to be analyzed, dissolved solids
content of the sample, or the tendency of the metal to
adsorb on the filament.  It may be possible to find a
"constant matrix" that could be added to both standards
and sample to suppress the effects leading to changes in
the slopes of log-log calibration curves with matrix.
However, because the origin of this effect is not under-
stood, it is very difficult to predict what, if anything,
might be a successful constant matrix.

EFFECT OF DISSOLVED SOLIDS

In looking for a constant matrix it was found that when
water containing a high concentration of dissolved
solids is evaporated onto the filament, discharging the
capacitor through the filament causes the plasma to
momentarily shrink in size, interfering with the emission
measurement.  Apparently firing the filament generates a
small cloud of gas that interferes with the normal
                                    _2
helium flow.  Five microliters of 10   molar solution
deposited on the filament would produce 0.4 cm  of gas at
2 mm pressure and room temperature, assuming 1 gas mole-
cule per 1 solid molecule.  This volume of gas would be
expected to produce the observed effect.

This effect limits the low pressure microwave spectrometry
to freshwater samples and restricts the possibilities for
a constant matrix that might lead to satisfactory analy-
tical performance.
                            11

-------
PRECISION

Monsanto reports 2% standard deviation in peak height
measurements at 10    grams of sample in 5 microliters
(0.02 mg/£).  Because log peak height-log concentration
curves have slopes less than one, 2% standard deviation
in peak height corresponds to a larger standard devia-
tion in concentration.

Satisfactory precision (about 10%) was obtained only
when the drop of sample did not wet the filament.
Immediately after firing, a Ta filament is wettable by
aqueous solutions.  After standing in air for 10-60
seconds however, it is no longer wettable.

The plasma length at each firing varied 2 mm out of a
total plasma length of 8 cm.  Since intensity of the
emission signal varies with position in the plasma,
this leads to a change in peak height.  This was
corrected by adjusting the plasma power to compensate
for any changes in plasma length.
                            12

-------
                       SECTION VI
                       DISCUSSION

The evaluation of the microwave plasma spectrometer was
terminated earlier than originally planned for the
following reasons:

     o  It was evident that the thermal atomization
        system for introducing sample into the plasma
        needed to be modified so that emission would be
        independent of matrix.  It was uncertain whether
        an improved atomization system is feasible and
        how long it would take to develop.

     •  Other spectrochemical methods, which have demon-
        strated performance, require less time for
        analysis and can be readily automated, better
        fill the EPA need for simultaneous multielement
        analysis.  Pulsed source atomic fluorescence
        determines up to 12 elements in less than one
        minute and has already been automated (18).  RF
        plasma emission spectrometry determines over 20
        elements in less than one minute and can be
        readily automated (19).  The microwave spec-
        trometer requires 8 minutes per analysis and is
        not readily automated.  Table 3 summarizes work
        on spectrochemical multielement analysis using
        flame and plasma excitation.  References 11-18
        come from a review article by Busch and Morrison
        that gives a detailed analysis of the problems
        and possibilities for multielement flame spectro-
        scopy (10).
                            13

-------
































u
EH H
S g
W O
g EH
H <
U hi II
H H fa
OS K PC*"—
EH EH g
W r^ »O
E P ZH
O S Oin
EH Ken
O U) EH H
BB P ftS
CL| O K W
OHO
PS 12! CO W
EH H < 01 W
W O 5H«
> PS h4O
C4 ft < D
P ft 2 Hi
CO  J>
0 4-1
•H -rl
X! rH
O -H
r^C XI
id
O d
4-1 id
u
•d
0 4J
01 C
P 0
g
W 0
rH rH
ft 0
•rH -fH
O 4-1
C rJ
^ S
ft

















rH
C tn
^"B
N >,
>t-P m
rH -rl 4-1
(d 4-1 -H
SSI
T3 rl
01 I-H
-P C
C - C

E 0 I
0 .Q 0
rH E 0)
W 3 4-1
2 0
o



q

E

c
t
a
0
o
c
0
a
0

rH
id
O
-rl
C
•rl
rH
U
















01

0
4-1
O
0
4->
0


0
rH
ft
•H
4-1
rH
3

4-1
•rl
rH
ro

0
i-H
ft
•H
4-1
rH
3
g








CM
O
.
O

u

I.
(N CM
• O
O •
—•0

M
-^m
CM -
O ^
O VTJ

0 O 0

-H (d tn
fa 13 S






fa



H
rH


I-H ro
fl H-I T> -
u o Mra
•rl Id Id 0 0 M
fi rH M XI > C I)
-H -rl H-) O Id -rl 4-1
rH OX M 0 M <0
u co w o K! p is

rH "J jj
0 01 C
C N 0)
01 "O in c -H in
13 CMHOllBfi 0
O OIU0001XIO >H
XI 4J4J-rH0CJM HC CCO SOCCJrrj j-l>
•H C-rl-rl 00IdOO 0ld0
4J 04-1-P Hft"CXI T3S>
XI "O -rl -H 4-1 C 0 W O 4-1 -H
CT>00i 4-1UCO MEld 0) 4-14-1
-rIC C Id>, Ofi 4-lIfl 3X! TJ4-1
Oi-HrH C01H0 ftCD1 - -H
0 I-H Id r^r^J OE r*l£ O 0O -P MH|S
4-ltd rH 01O0 X4(O O 014-1 -rl O
0>O4-> jStTlrH i-H-H rH 01
TJ-rl U r-l-PCO) C 13 rrj(D m gj)
4-) 0 idid-H o0-H^^0ro O-ri
0ft ft OOCXI -HtTl >rlgrH 0 -HO
rHO m -rHOO 4JM 0)E3 rH JJC
ft cs-Hid iBid MHNidft ft idoi
•HM >d IdO4-10 4->Xl O>(M -rl rH3
4-1 ft -rl XI rH -rl -rl O rH tH XI -P 3 tT
rHE ft OrHralH Om 0ldO-P rH rrJOl
33 id (BOOO X -H roCVn-ri 3 OM
grH PS gXIft'H W13 pldftS g g>H

1-^
^^^^"^^~^^^~^~^
f-*** — ^^~1
pq f '
P4
^^
>HJ ^J1 -— •
^Q O *-*- «~t *— -
0 t^ O i-t
,^. t O O O
00 O fe • • O
O — ' rtj O O •
°* s ^ d S1^
-" XI S E <"
(d — ~ ^ u
U CO ,~^~ rH » -
o r~. I-H r^ o t~~ *~~ — - *•
OrH OO OCM inco^^
ff ^~, .0 «o .0 ocDin
f^ fa M1 O O O * O • ...
p4 • , of? 30 3 ^ fa "a ^ -^~— -^
xi>i ^ <-> ~ oc O-H ^fi*-1
Xl«>i tn 63 >i & USU
tJ> rQ ~g - XI -
g 18 - —^ — - ~- -^~^-

^ VOCM O^^ Or-Hr-- OrH (MlHrH
-H** OO OrH OOO OO 0OOO >t
Sid •• •• ..o •• >..-4-i
U Moo Moo oo« Moo Hooo c

•riM' oidtri Otn0 -Hide--' OCO - scais 5 0
COUfcS fagS far^CO CT^O
.HrHrHrH rHi-HrH i-HCN


-------
     •  The principle advantages of the low-pressure
        microwave plasma spectrometer are not useful to
        the EPA.  The small sample size requirement
        (1-10 microliters) is not significant for water
        analysis because more sample is usually available
        The improvements in sensitivity for single
        element analysis can also be obtained by flame-
        less atomic absorption, which does not have the
        same problem with calibration curve slopes as
        low-pressure microwave spectrometry.

If the operational problems are solved to make low
pressure microwave plasma emission spectrometry a
reliable analytical technique, it will be a significant
advance in analytical methodology because it will combine
the advantages of flameless atomization (sensitivity and
small sample size) with the advantages of working with
emission rather than absorption (no need for a light
source and easier adaptation to multielement analysis).
However, fundamental instrument development is not
normally a mission of the NWCCRP.   If a satisfactory
thermal atomization system for introducing the metal
sample into the plasma is developed, then the microwave
plasma spectrometer should be reconsidered.  The improved
thermal atomization system must inject the sample into
the plasma rapidly.  Otherwise broadening of the emission
peak will reduce sensitivity and the principle advantage
of the low pressure microwave plasma spectrometer will be
lost.
                           15

-------
This evaluation of the low-pressure microwave plasma
emission spectrometer is restricted to metal analysis
in aqueous samples requiring atomization.  Microwave
plasmas have been successfully used with vapor phase
samples, e.g., arsenic analysis as AsH_ gas (21) and
detection of GC effluents (22-25).
                            16

-------
                       SECTION VII
                       REFERENCES

1.  Runnels, J. H.,  and J. H.  Gibson.   Characteristics
    of Low Wattage Microwave Induced Argon Plasmas in
    Metals Excitation.  Anal.  Chem. 39^ (12):  1398-1405,
    October 1967.

2.  Kawaguchi, H., M. Hasegawa, and A. Mizuike.  Emission
    Spectrometry of Solutions  with a Low-Wattage Micro-
    wave Discharge.   Spectrochim. Acta 27B; 205-210,
    1972.

3.  Fallgatter, K.,  V. Svoboda, and J. D.  Winefordner.
    Physical and Analytical Aspects of a Microwave
    Excited Plasma.   Appl. Spectrosc.  ,25 (3):  347-352,
    March 1971.

4.  Lichte, F. E., and R. K. Skogerboe.  Analysis of
    Solution Samples by Microwave Induced Plasma
    Excitation.  Anal. Chem. £5(2): 399-401,  February
    1973.

5.  Aldous, K. M., R. M. Dagnall, B. L. Sharp, and T. S.
    West.  A Microwave-Induced Argon Plasma System
    Suitable for Trace Analysis.  Anal. Chim.  Acta
    (Amsterdam) 5_4:  233-243, 1971.

6.  Kawaguchi, H., M. Hasegawa, and A. Mizuike.  Thermal
    Atomization of Samples from a Wire Loop followed by
    Excitation in a Microwave  Electrodeless Discharge
    for Emission Spectrometry.  Bunko Kenkyu (Tokyo) 21;
    36-42, January 1972 (thru  CATT^, 121760W) .

7.  Busch, K.  Evaluation of Low-Pressure Microwave-
    Induced Plasmas  as Excitation Sources for  Spectro-
    analytical Chemistry.  Florida State University,
    Tallahassee, 1971. 209 p.

8.  Hingle, D. N., G. F. Kirkbright, and R. M. Bailey.
    A Simple, Low-Power, Reduced-Pressure Microwave
    Plasma Source for Emission Spectroscopy.   Talanta,
    16: 1223-1225, 1969.
                           17

-------
 9.  Monsanto Research Corporation.   A Microwave-Induced
     Emission Detector for Pesticides and Trace Metals.
     Final Report on EPA Contract No. 68-01-0085, 1973.

10.  Busch, K. W.,  and G. H. Morrison.  Multielement
     Flame Spectroscopy.  Anal. Chem. 45 (8):  712A-722A,
     July 1973.

11.  Vallee, B. L., and Margoshes, M.  Instrumentation
     and Principles of Flame Spectrometry.  Multichannel
     Flame Spectrometer.  Anal. Chem. 28_ (2):  175-179,
     February 1956.

12.  Mavrodineau, R. and R. C. Hughes.  A Multichannel
     Spectrometer for Simultaneous Atomic Absorption and
     Flame Emission Analysis.  Appl.  Opt. 7 (7): 1281-
     1285, July 1968.

13.  Dawson, J. B., D. J. Ellis, and R. Milner.  An
     Automatic High Speed Scanning Multichannel Spectro-
     photometer for Spectrochemical Analysis.   Spectro-
     chim. Acta 23B: 695-708, 1968.

14.  Mitchell, D. G., and A. Johansson.  Simultaneous
     Multielement Analysis Using Sequentially Excited
     Atomic Fluorescence Radiation.   Spectrochim. Acta
     25B; 175-182,  1970.

15.  Dagnall, R. M., G. F. Kirkbright, T. S.  West, and
     R. Wood.  Multichannel Atomic Fluorescence and Flame
     Photometric Determination of Calcium,  Copper,
     Magnesium, Manganese, Potassium, and Zinc in Soil
     Extracts.  Anal. Chem. 43_  (13):  1765-1769, November
     1971.

16.  Norris, J. D., and T. S. West.   Applications of a
     Wavelength Scanning Technique to Multi-Element
     Determinations by Atomic Fluorescence Spectrometry.
     Anal. Chem. £5 (2): 226-230, February 1973.

17.  Aldous, K. M.  and D. G. Mitchell.  Multichannel
     Spectroscopy Using a Vidicon Detector.   (Presented
     at Pittsburgh Conference.  Cleveland.   March 5-9,
     1973).
                            18

-------
18.  Malmstadt, H. V., and E.  Cordos.   Automated Atomic
     Fluorescence Spectrometer for Rapid Multielement
     Determinations.  Amer. Lab. _4: 35-42, August 1972.

19.  Fassel, V. A.  Inductively Coupled Plasma - Optical
     Emission Spectroscopy, A New Analytical System for
     the Simultaneous Multielement Determination of
     Trace Metallic Pollutants in Water.  (Presented at
     the Symposium on Recent Advances  in the Analytical
     Chemistry of Pollutants.   Athens.  May 14-16, 1973).

20.  Spectrametrics, Inc., Technical Literature,
     Burlington, Mass. 1972.

21.  Lichte, F. E., and R. K.  Skogerboe.  Emission
     Spectrometric Determination of Arsenic.  Anal. Chem,
     4_4 (8): 1480-1482, July 1972.

22.  McCormack, A. J., S. C. Tong, and W. D. Cooke.
     Sensitive Selective Gas Chromatography Detector
     Based on Emission Spectrometry of Organic Compounds,
     Anal. Chem. 37  (12): 1470-1476, November 1965.

23.  Bache, C. A., and D. J. Lisk.  Determination of
     Organophosphorus Insecticide Residues Using the
     Emission Spectrometric Detector.   Anal. Chem. 3J7_
     (12): 1477-1480, November 1965.

24.  Bache, C. A., and D. J. Lisk.  Selective Emission
     Spectrometric Determination of Nanogram Quantities
     of Organic Bromine, Chlorine, Iodine, Phosphorus
     and Sulfur Compounds in a Helium Plasma.  Anal.
     Chem. 3ji  (7): 786-789, June 1967.

25.  Moye, H. A.  An Improved Microwave Emission Gas
     Chromatography Detector for Pesticide Residue
     Analysis.  Anal. Chem. 39 (12): 1441-1445, October
     1967.
                            19

-------

-------
  SELECTED WATER
  RESOURCES ABSTRACTS
  INPUT TRANSACTION FORM
                   w
EVALUATION OF A MICROWAVE-INDUCED PLASMA
SPECTROMETER FOR  TRACE ANALYSIS
                   S. Re,
                   *.
                   *.
Seitz, William  Rudolf
Southeast Environmental Research Laboratory
Athens , Georgia
                                                         16ADN-32
     Environmental Protection Agency report number,
     EPA-660/2-73-009, January 1974.
     A low pressure microwave-induced plasma  in  helium was evaluated as
an excitation  source for spectrochemical analysis  of trace metals in
water.  The sample  was introduced by evaporating a 5-microliter drop of
solution on a  wire  filament and atomizing the solids by the discharge of
a capacitor through the filament.  The height of the resulting pulse from
the photo detector  was automatically recorded.   Repeatability was about
10% and detection  limits were 1 to 25 pg for  As, Cd, Hg, and Pb.  Slopes
of calibration curves depended strongly on matrix  and filament effects and
the maximum solids  loading of the filament was about 5 micrograms.
  17a. Descriptors
*Analytical Techniques,  *Spectroscopy, Microwaves,  *Trace Elements,
Arsenic, Cadmium,  Lead,  Mercury
  17b. Identifiers
Plasma, Excitation
  17v. COWKR Field & Group   05A
                                              Send To:
                                              WATER RESOURCES SCIENTIFIC INFORMATION CENTER
                                              U.S. DEPARTMENT OF THE INTERIOR
                                              WASHINGTON. D. C. 2O24O
          T. B. Hoover
tion EPA, SERL, Athens/  Georgia 30601

-------