NBS
EPA
U.S. Department
ol Commerce
National
Bureau of
Standards
Center for
Analytical Chemistry
Washington. DC 20234
United States
Environmental Protection
Agency
Office of Energy, Minerals, and
Industry
Washington DC 20460
EPA-600/7-79-174
August 1979
Research and Development
Quantitative
Ultratrace Transition
Metal Analysis of
High Salinity Waters
Utilizing Chelating
Resin Separation
Interagency
Energy/Environment
R&D Program
Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies^relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
QUANTITATIVE ULTRATRACE TRANSITION METAL ANALYSIS OF
HIGH SALINITY WATERS UTILIZING CHELATING RESIN SEPARATION:
APPLICATION TO ENERGY-RELATED ENVIRONMENTAL SAMPLES
by
Howard M. Kingston
Center for Analytical Chemistry
National Bureau of Standards
Washington, DC 20234
Interagency Agreement No. EPA-IAG-D5-E684
Program Element No. EHA-553
EPA Project Officer: J. Stemmle
Environmental Protection Agency
Washington, DC 20460
This study was conducted
as part of the Federal
Interagency Energy/Environment
Research and Development Program
Prepared for
OFFICE OF ENERGY, MINERALS, AND INDUSTRY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, DC 2Q46Q
-------
DISCLAIMER
This report has been prepared and reviewed by the Center for Analytical
Chemistry and the Office of Environmental Measurements, National Bureau
of Standards, and reviewed by the U. S. Environmental Protection Agency,
and approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the U.S. Environmental
Protection Agency. In order to adequately describe materials and experimental
procedures, it was occasionally necessary to identify commercial products
by manufacturer's name or label. In no instance does such.identification
imply endorsement by the National Bureau of Standards or the U. S.
Environmental Protection Agency nor does it imply that the particular
products or equipment 1s necessarily the best available for that purpose.
ii
-------
FOREWORD
The role of the National Bureau of Standards (NBS) 1n the Interagency
Energy/Environment R&D program, coordinated by the Office of Research and
Development, U. S. Environmental Protection Agency, is to provide those
services necessary to assure data quality In measurements being made by a
wide variety of Federal, state, local, and private industry participants in
the entire program. The work at NBS is under the direction of the Office of
Environmental Measurements and is conducted In the Center for Analytical
Chemistry, the Center for Radiation Research and the Center for Thermodynamics
and Molecular Science. NBS activities are in the Characterization, Measurement,
and Monitoring Program category and addresss data quality assurance needs in
the areas of air and water measurement methods, standards, and Instrumentation.
NBS outputs in support of this program consist of the development and description
of new or improved methods of measurement, studies of the feasibility of
production of Standard Reference Materials for the calibration of both field
and laboratory Instruments, and the development of data on the physical and
chemical properties of materials of environmental importance In energy
production. This report Is one of the Interagency Energy/Environment Research
and Development Series reports prepared to provide detailed information on
an NBS measurement method or standard development.
C. C. Gravatt, Chief
Office of Environmental Measurements
National Bureau of Standards
-------
CONTENTS
Foreword ...................................... • ........................... i i i
Abstract [[[ .yl
Fi gures [[[ yi ' '
Tables [[[ ix
Acknowledgment ...................................... * " " ' ............ ' ---- x1'
1 . Introducti on ...................................... • ............... 1
2 . Summary and Concl us1 ons ........................................... 2
3. Literature Review ................................................. 6
4. Experimental Procedures ....... .............. . ....... ...... ........ 12
Materials ........... ........... '• .............. '•• ................. 12
Apparatus ........ ........................ • ....................... 1 3
Procedure ............ • ........................................... 1 6
5. Results and Discussion ............................................ 21
Separating Agent Selection ....................................... 21
Concentration and Separation Parameters .......................... 33
Aspects of the Separation ...................... « « « ............... 36
Physical Parameters and Extension of the Separation of
Larger Vol umes ......... ,...,.,, ...... .............. ........... ... 38
Quantitativeness of the Total Recovery of Selected Transition
Elements Using Radiochemical Tracers
Evaluation of the Compatibility of the Chelex 100 Concentration
and Separation with Spark Source Mass Spectrometry .............. 50
Evaluation and Practical Use of the Chelex 100 Sample Prepara
tion Method Utilizing Graphite Furnace Atomic Absorption
Spectrometry Analysis ........................................... 5'
References
Appendices
A. The Abundances of the Chemical Elements 1n Sea Water (1) .......... 62
B. The Selectivity Coefficients for Chelex 100 Resin Based on __
-------
CONTENTS
(Continued)
Appendices
D.
The Experimental Conditions for Analysis of the 2.5 M. HN03
Fraction Containing the Trace Metals Using A Perkin-Elmer
Model 603 Atomic Absorption Spectrophotometer Equipped with
a Heated Graphite Atomizer, HGA-2100 -69
E. A List of Resin Volume for Chelex 100 1n Different Ionic Forms
Based on Na as 1.00 (58) 70
F. A Crystal Residue Observed 1n Frozen Neutral Sea Water Samples•••71
-------
ABSTRACT
In order to accurately evaluate the impact of energy related activities,
such as offshore drilling, on the sea water system, it is necessary to
measure trace element concentrations in the presence of considerably higher
levels of alkali and alkaline earth elements.
This report describes a technique which was developed for the elimination
of the alkali and alkaline earth elements Na, K, Ca and Mg from the trace
transition elements in environmental samples. Sea water is an environmental
sample where the alkali and alkaline earth elements are those of the highest
concentrations while the trace transition elements are present in the lowest
concentrations; it is an extreme example of this type of matrix. The separation
was accomplished by passing an ammonium acetate solution through a column of
Chelex 100 resin after a sea water sample had been chelated. The alkali and
alkaline earth elements were eluted from the column by ammonium acetate and
the trace transition elements were then collected using nitric acid.
The quantitative concentration, separation and removal of selected
transition elements was tested using radiochemical tracers. The study
revealed >99.9 percent recovery of Cd, Cu, Mn, Ni and Zn, using a 100 mL
sample, and >99 percent with a 1 liter volume. Cobalt and Pb exhibited
>99 percent and >98 percent recoveries, respectively, from a 100 mL sample,
and >97 percent from a 1 liter volume. Iron was found to be recovered only
approximately 92 percent in either volume.
The concentration and separation technique was applied to Chesapeake
Bay and Alaskan sea water samples. The samples were introduced into a
graphite furnace and were analyzed by graphite furnace atomic absorption
spectrometry. Analysis of concentrations below ng/mL for the trace elements
mentioned was possible using this combination of sample preparation and
instrumental analysis and no interelement interferences occurred. The
combination of this sample preparation technique and the graphite furnace
atomic absorption sensitivity enables extreme detection limits to be achieved
for the elements mentioned, and gave measurements that were reliable and
reproducible. The compatibility of the separation method when applied to
spark source mass spectrometry was also demonstrated for thirteen transition
elements, concentrated and separated from the sea water matrix.
The technique for the elimination and concentration of trace transition
elements from the salt matrix has advantages in sample preparation for other
analytical instrumental methods, in addition to the ones tested. Several of
these instrumental methods could not be utilized without such a technique to
remove the alkali and alkaline earth elements from sea water (neutron acti-
vation analysis and optical emission spectrometry) while others could reduce
these interferences and be used more effectively with the separation and
concentration of the sample prior to analysis (flame atomic absorption and
x-ray fluorescence).
vi
-------
Ultra-purification techniques were employed in the preparation of
the reagents used in this method; the analytical blank was found to be
below detectable limits of graphite furnace atomic absorption spectrometry.
A technique was developed for redissolving neutral sea water samples
after freezing or evaporation, utilizing the addition of C0? which is
lost in freezing and evaporation of the sea salts.
vii
-------
FIGURES
Number
1 The apparatus used for holding and delivering large quantities
of sea water at a controlled rate to Chelex 100 resin. The
apparatus (excluding column and clamp) was fabricated from
Teflon FEP (*) or Teflon TFE (t), which has desirable anti-
wetting and non-contaminating properties 15
2 The comparison of the removal of the alkali elements Na and K
by ammonium acetate and ammonium nitrate from a Chelex 100
resin column which has previously chelated 100 mL of sea
water 30
3 The comparison of the removal of Ca by ammounium acetate and
ammonium nitrate from a Chelex 100 resin column which has
previously chelated 100 mL of sea water 31
4 The comparison of the removal of Mg by ammonium acetate and
ammonium nitrate from the Chelex 100 resin column which
has previously chelated 100 mL of sea water 32
5 The elution of Ca, Mg, and Mn from a preconcentrated 50 mL
sample of sea water on a Chelex 100 resin column using
1.0 M ammonium acetate separating agent at pH 3.0, 4.5,
and 5.0 and 2.5 M nitric acid stripping agent 34
6 The elution of Na, K, Ca, and Mg from a preconcentrated 100 mL
sample of sea water on a Chelex 100 resin column using 1.0 M
ammonium acetate separating agent and 2.5 M nitric acid
stripping agent 36
7 The elution of Na, K, Ca, and Mg from a preconcentrated 1 Liter
sample of sea water on a Chelex 100 resin column using 1.0 M
ammonium acetate separating agent and 2.5 M nitric acid
stripping agent 44
viii
-------
TABLES
Number
1 The Gradient Elution of Na and K from a 100 mL Sea Water
Sample Preconcentrated on a Chelex 100 Resin Column Using
1.0 M (pH 5.2) Ammonium Nitrate Separating Agent and
2.5 M Nitric Acid Stripping Agent ................ 23
2 The Gradient Elution of Ca and Mg from a 100 mL Sea Water
Sample Preconcentrated on a Chelex 100 Resin Column Using
1.0 M (pH 5.2) Ammonium Nitrate Separating Agent and
2.5 M Nitric Acid Stripping Agent ..... ........... 2^
The Elution of Mn from a 100 mL Sea Water Sample Preconcen-
trated on a Chelex 100 Resin Column Subjected to 1.0 M
(pH 5.2) Ammonium Nitrate Separating Agent and 2.5 M
Nitric Acid Stripping Agent ....... ....... ..... 25
The Gradient Elution of Na and K from a 100 mL Sea Water
Sample Preconcentrated on a Chelex 100 Resin Column using
1.0 M (pH 5.2) Ammonium Acetate Separating Agent and 2.5 M
Nitric Acid Stripping Agent ................... 26
The Gradient Elution of Ca and Mg from a 100 mL Sea Water
Sample Preconcentrated on a Chelex 100 Resin Column Using
1.0 M (pH 5.2) Ammonium Acetate Separating Agent and 2.5 M
Nitric Acid Stripping Agent . ............ ...... 27
The Elution of Mn from a 100 mL Sea Water Sample Preconcen-
trated on a Chelex 100 Resin Column Subjected to 1.0 M
(pH 5.2) Ammonium Nitrate Separating Agent and 2.5 M
Nitric Acid Stripping Agent ........ ........... .28
The Quantity of Na, K, Ca, and Mg Found in Various Effluent
Fractions Substituting Water for a Separating Agent in
the Elution from a Preconcentrated 100 mL Sample of Sea
Water on a Chelex 100 Resin Column ....... .......... 37
The Amount of Na, K, Ca, and Mg Remaining in the Final 2.5 M
Nitric Acid Volume After Elution of the Resin Column with
40 mL of Water, Ammonium Nitrate or Ammonium Acetate. The
Column had Previously Preconcentrated a 100 mL Sea Water .... 38
ix
-------
TABLES (Continued)
Number
9 The Elution of Na from a Preconcentrated 1 Liter Sample of
Sea Water on a Chelex 100 Resin Column Using 1.0 M (pH 5.2)
Ammonium Acetate Separating Agent and 2.5 M Nitric Acid
Stripping Agent
10 The Elution of K from a Preconcentrated 1 Liter Sample of Sea
Water on a Chelex 100 Resin Column Using 1.0 M (pH 5.2)
Ammonium Acetate Separating Agent and 2.5 M Nitric Acid
Stripping Agent ......
11 The Gradient Elution of Ca from a Preconcentrated 1 Liter
Sample of Sea Water on a Chelex 100 Resin Column Using
1 . 0 M (pH 5 . 2) Ammonium Acetate Separating Agent and
2.5 M Nitric Acid Stripping Agent . . . . ............ 42
12 The Gradient Elution of Mg from a Preconcentrated 1 Liter
Sample of Sea Water on a Chelex 100 Resin Column Using
1.0 M (pH 5.2) Ammonium Acetate Separating Agent and
2.5 M Nitric Acid Stripping Agent ................ 43
13 Summary of the Radio Tracer Study Results Showing the Percen-
tage of Elution in Each of the Effluent Fractions for the
100 mL Samples ......................... 47
14 Summary of the Radio Tracer Study Results Showing the Percen-
tage of Elution in Each of the Effluent Fractions for the
1 Liter Samples . . ....................... 48
15 The Concentration of Cd, Co, Cu, Fe, Mn, Ni, Pb, and Zn Found
in Ten Identical 100 mL Sea Water Samples from the Chesapeake
Bay as Determined by GFAAS After Chelex 100 Concentration and
Separation ........................... 54
16 The Concentration of Na, K, Ca, and Mg Determined in the 2.5 M
HN03 Effluent of the Chesapeake Bay Samples After Analysis
of the Eight Transition Elements; Sample 3 was Spiked,
Sample 8 was Unspiked and Sample 13 was a Blank ......... 55
17 The Concentration of Cd, Mn, Nl, and Pb Found in Fifteen
Different Sea Water Samples of 100 mL Each, Taken from
Glacier Bay, Alaska as Determined by GFAAS After Chelex
100 Resin Column Concentration and Separation .......... 56
-------
ACKNOWLEDGMENT
The author is grateful to the U. S. Environmental Protection Agency for
partial support of this work under the Interagency Energy/Environment Agreement
EPA-IAG-D5-E684. This work is from a dissertation submitted and accepted by
the graduate school, The American University, by H. M. Kingston, in partial
fulfillment of the requirements for the degree of Doctor of Philosophy in
Chemistry.
-------
1. INTRODUCTION
One of the major emphasis areas of the measurements and instrumentation
component of the Interagency Energy/Environment Program is the development
of the capability to make baseline measurements in environmental systems
where future energy related activities are expected to occur which may lead
to changes in the state of the environment. The offshore area has been
identified as one where the impact of energy related activities may occur
and where extensive baseline measurements will be required. These measurements
will consist of determination of levels of trace elements and organic
constituents in the seawater, in the sediment, and in the biota. The
Proceedings of the Workshop on Measurement Methods and Standard Reference
Materials for Offshore Drilling-Petroleum, published as part of this
Energy/Environment Series, recommends a number of high priority measurement
method and Standard Reference Material needs related to the ocean waters,
sediments, and biota. The current interest and expansion of offshore
mining and oil exploration has the potential to alter the concentration
of many of the trace transition metals found in seawater. The determination
of the levels of these metals prior to any offshore energy related activities
requires analysis techniques which are extremely sensitive. In addition,
accurate analytical data on the current concentrations of the trace
elements is essential for future monitoring of the metal concentration
trends in sea and coastal waters. These measurements are extremely difficult
since it is necessary to measure trace toxic element concentrations in
the presence of considerably higher levels of alkali and alkaline earth
elements.
In environmental samples, Na, K, Ca and Mg are usually present many
orders of magnitude above the trace elements of interest; the most extreme
example of such a matrix is sea water. Sodium, K, Ca and Mg are present in
concentrations of 10.7, 0.399, 0.412 and 1.29 grams per liter, respectively.
While the salinity of the water may change with location, depth or season, the
ratio of these ions to one another is "conservative" and stays remarkably con-
stant. However, the trace transition metals are present in microgram and
submicrogram quantities and do not display the conservative relationship of the
major elements (salts) of sea water (1). Appendix A contains a table of average
concentrations of elements in sea water (1).
The major salts of saline waters are the normal ionic constituents of all
natural waters, whether they are streams, lakes or the oceans into which they
all eventually accumulate. The usual changes in their concentrations do not
disrupt aquatic life forms since their major contribution is a salt gradient.
This is not true of the minor cations which may cause undesirable effects on
many biological systems. The transition metals are of great interest biologi-
cally because small shifts in their concentration can cause significant biolog-
ical effects; they may also have adverse effects on humans who consume the
products of the sea, as many transition metals bioaccumulate to much higher than
ambient levels (2).
-------
This paper describes a technique which was developed for
the elimination of the alkali and alkaline earth elements from the trace
transition elements in seawater samples by the use of a Chelex 100 Resin
Seawater sample chelation. This technique was found to have excellent recoveries
of most trace toxic elements and was tested on a variety of seawater samples,
with the analysis being performed by a variety of common techniques. In
addition, in order to evaluate methods for preparing and storing a seawater
Standard Reference Material, a technique was developed for redissolving
neutral seawater samples after freezing or evaporation, utilizing the addition
of COwhich is lost in freezing and evaporation of the sea salt.
-------
2. SUMMARY AND CONCLUSIONS
The overall purpose of this study was to develop a technique by
which complete separation of the alkali and alkaline earth elements (Na,
K, Ca and Mg) from the trace transition metals (Cd, Co, Cu, Fe, Mn, Nl,
Pb and Zn) could be accomplished In a sea water matrix.
In fulfilling this purpose, five specific objectives were considered:
-Develop a separating agent technique capable of stripping the Na,
K, Ca and Mg from a column of Chelex 100, which has chelated trace
transition metals from sea water, without removing the transition
metals of interest.
-Determine the parameters (pH, resin volume, flow rate) which will allow
the maximum efficiency for the preconcentration and separation.
-Evaluate the efficiency of the preconcentration and separation for the
transition metals of interest.
-Establish conditions which will reduce the analytical blank of these
preconcentration and separation procedures to an acceptable value.
-Preconcentrate and separate actual sea water samples and determine the
actual usefulness on instrumentation normally suffering from interelement
effects rendering them incapable of this application, specifically,
graphite furnace atomic absorption spectrometry (GFAAS) and spark source
mass spectrometry (SSMS).
The attainment of these objectives provides a procedure enabling the
elimination of alkali and alkaline earth elements from trace transition
elements. This procedure not only provides the means for sea water analysis
by instrumental techniques previously inhibited by the sample matrix, but also
establishes a basis for further research by applying this combination of
preconcentration and separation to other samples presently prohibited from
instrumental analysis by similar matrix interferents.
A 1,0 M ammonium acetate solution has proven capable of stripping Na, K,
Ca, and Mg selectively from a column of Chelex 100 resin which has chelated up
to a 1 liter sea water sample. During the removal of the alkali and alkaline
earth elements the trace transition elements chelated by the Chelex 100 resin
are unaffected by the ammonium acetate separating agent.
The pH range of 5.0-5.5 appears to be the most suitable for both the con-
centration of the trace transition elements onto the Chelex 100 resin column and
for the removal of the alkali and alkaline earth elements from the resin using
ammonium acetate. The pH in this range affords ample separation between the
elution of the salt matrix and the transition elements attaining a complete
separation between these two elemental classes.
-------
The concentration and recovery is quantitative for Cd, Cu, Mn, Ni, and Zn,
being >99 percent efficient. The Mn ion can only be collected quantitatively
from volumes less than 500 mL using the present resin volume. The recovery of
Co and Pb is >99 percent and >98 percent, respectively, at the 100 ml sample
volume and >97 percent at the 1 liter level. These recoveries would appear to
be sufficiently quantitative for most instrumental techniques. Many instrumental
techniques have uncertainties in measurement of 5 percent. Both the SSMS and
GFAAS have uncertainties greater than the small losses detected in the concen-
tration step of the present procedure. The SSMS would, however, not be affected
by any losses of the magnitude involved in the ions investigated here, including
Fe. The SSMS technique would not have the sensitivity to measure all the
elements tested by GFAAS at the levels in the samples utilized in this study.
For many of the elements in these sea water samples the GFAAS operated near its
detection limits and would introduce more uncertainty into the measurement than
the recoveries of the analytes during the concentration technique. Iron was
not reproducibly retained by the Chelex 100 resin and varied in its recovered
efficiency. Instrumental procedures which depend on total recovery for quanti-
zation should not rely on the recovery of Fe, being more reproducible than about
2 percent due to air oxidation alteration of the Fe 2/Fe 3 ratio.
The ammonium acetate can be eliminated from the column with only several
resin volumes of water excluding it from any addition to the final sample
volume. The ammonium ion can be eliminated from the final volume, thus ren-
dering the separating agent innocuous to all instrumental applications. The
elimination of the ammonium ion from the final volume, when used in conjunction
with GFAAS, proved to be unnecessary as it produced no interference as ammonium
nitrate in the final volume.
The analytical blank, as determined by GFAAS for the entire procedure, was
low enough to be undetectable for every element tested, with the exception
of Fe (which was discussed). The blanks are well within acceptable
limits to avoid any interference with even pristine Alaskan sea water samples.
Perhaps, as the instrumental techniques are improved and lower detection
limits are achieved, other steps may be necessary to decrease these levels.
At the present time, however, the reagent purification procedures, the laboratory
environment, and procedural precautions used are adequate to eliminate the
analytical blank as a source of error in the analysis.
The use of this separation method has enabled the GFAAS and SSMS instrumen-
tal techniques to become compatible with the Chelex 100 preconcentration method
for high salinity samples. The final samples produced no interelement inter-
ferences when applied to GFAAS and enabled the maximum instrumental sensi-
tivities to be used. The method was also compatible for multielemental analysis
using SSMS. These successful instrumental trials can be extrapolated to other
instruments which require similar matrix elimination and also to those which, to
the present time, have made use of the Chelex 100 transition metal concentration
with moderate interferences. The method in its present form appears to be
ready for practical use, and should enable analysis such as those presented here
to become of only routine complexity.
-------
The significance of the development of this technique has two important
aspects. First, the separation and concentration technique enables the use of
instruments previously unable to measure the trace transition metals in saline
waters. Several of these instruments such as GFAAS possess extremely low detec-
tion capabilities for the elements. Others, such as neutron activation analy-
sis, possess multielemental character and superior sensitivity. Second, using
instrumental techniques which possess these low sensitivities, studies of the
extremely low levels of the trace transition elements in saline waters can be
done with accuracy. Measurement of the trace elements in pristine saline
waters can be achieved and true baseline data collected for environmental
assessment of man's influence on the oceans.
-------
3. LITERATURE REVIEW
The literature of marine water analysis reflects the considerable diffi-
culty in establishing an accurate and precise method of analysis for trace
metals in saline matrices. The complex and highly concentrated saline matrix
defies a simplified approach for trace metal analysis.
There are no analytical techniques currently available for the direct
determination of elements in sea water at concentrations below 5 yg L~l (3).
Usually it is essential to concentrate the trace elements from a large volume
and separate the transition elements from the alkali and alkaline earth ele-
ments. In such sample preparations the efficiency of concentration, complete-
ness of separation, and total analytical blank become critical to the final
instrumental method (3-6).
Current preconcentration techniques include co-precipitation (7), chelation
and extraction (8,9), and chelation ion-exchange resin '(4,10). These techniques
used for the preconcentration of trace elements from sea water have been
reviewed by Riley and Skirrow (3).
The preconcentration technique which has come into prominence during the
past seven years has been the chelating ion-exchange resin technique, utilizing
Chelex 100 resin (refined by Bio-Rad Laboratories from Dowex A-l resin). The
use of Chelex 100 to concentrate trace transition elements from sea water has
been found to be highly efficient, while exhibiting low analytical blanks
(3,4).
Lai, Callahan and co-workers first demonstrated the practical use of Chelex
100 resin for the concentration of Ag, Co, Fe and Zn from quantities of saline
water exceeding 50 liters (11,12). A column of Chelex 100 (50-100 wet mesh) was
contained in a 2.2 cm X 7.6 cm bed and the flow rate was 33 mL/minute, or approxi-
mately 2 liters per hour. The retention on the Chelex 100 resin was 54 percent
for ilO'Vg, 92-95 percent for 59Fe, 95-99 percent for 60Co and 100 percent for
65Zn. The elution of the ions from the elution of the ions from the column was
quantitative, with high concentrations of HC1 (3-5 M). The acid elution con-
tained quantitatively all of the trace metals chelated by the column, within
their experimental error (except for Ag). A quantity of the major salts Na, K,
Ca and Mg were also eluted in the acid effluent. The concentration of the major
ions in the final fraction was several orders of magnitude greater than that of
the heavy metals. There was no attempt to separate the ions collected and all
ions were eluted simultaneously.
-------
Quantitatively, 95-99 percent of the Co was found to be retained from sea
water by the Chelex 100. This was attributed to the existence of both Co 2
and Co ^ in sea water; the Co 2 was chelated quantitatively, while the Co 3
was not quantitatively retained. Cobalt exists in sea water principally in the
divalent state (about 96%). A similar incomplete chelation was observed for
Iron, for the same reason. Zinc was quantitatively retained, being present as
Zn 2 in natural sea water. The optimum results were obtained at pH 5.0.
In 1968 Riley and Taylor wrote a series of three closely related papers
investigating the use of Chelex 100 resin for the preconcentration of trace
transition metals in sea water and the utilization of this technique for sample
preparation in flame atomic absorption and spectrophotometry (13). The resin
was used in the hydrogen form and 50-100 mesh size. The column resin was packed
in a 1.2 cm glass tube to a height of 6 cm. The flow rate was not allowed to
exceed 5 mL/min. A 1 liter sample of 0.5 y membrane-filtered sea water was
spiked with the radiochemical tracers or natural elemental tracers, adjusted in
pH from 5.0 to 9.0, and concentrated on the resin. Evaluation of the quan-
titation was made either radiochemically or photometrically. The investigation
reported 99-100 percent retention and subsequent elution of Bi, Cd, Co, Cu, In,
Mn, Mo, Ni, Pb, Re, Sc, Th, W, V, Y and Zn. The eluting agent was 30 mL 2 N_
HN03 for all elements except Mo, Re, W and V, which were chelated as anion
oxides and were removed with 20 mL 4 N NH^OH. Prior to elution, the column was
washed with 20 mL of distilled water. The preconcentration eliminated most of
the alkali and alkaline earth elements, leaving 80 and 1.8 rag Na, 1 and 0.3 mg
K, 2.3 and 0.2 mg Ca, and 0.3 and 0.02 mg Mg remaining on the column, which were
eluted with 2 N HN03 and 4 ^ NH^OH, respectively. There was no attempt to
separate the alkali and alkaline earth elements that remained in the trace metal
effluent.
In two later works in 1968, Riley and Taylor recommended the ammonium form
of Chelex 100 (14,15) rather than the hydrogen form used in their original
paper. They found that low pH retention of the resin in the lower portion of
the column caused losses in the first 100 mL of sea water until the pH of the
resin increased to at least pH 5.0.
Since the developmental work of Riley and Taylor using Chelex 100 for the
concentration of trace metals from sea water, this technique has been used by
many researchers without significant alteration. The preconcentration technique
has been utilized for sample preparation to increase the concentration of trace
elements to detectable limits for flame atomic absorption spectrometry, spectro-
photometry, pulse polarography, anodic stripping voltammetry, and x-ray fluores-
cence .
The flame atomic absorption analysis technique has been utilized for Cd,
Co, Fe, Mn, Ni, Pb and Zn, collected from marine water using Chelex 100 (4,13,
15-19).
Spectrophotometry has utilized this concentration technique for Mo, V (14),
Cd, Cu, Fe, Mn, Ni and Zn (14,20).
-------
Pulse polarography was employed to determine Cd, Co, Cu, Ni and Zn in
marine samples using Chelex 100 to preconcentrate the trace metals (21).
Anodic stripping voltammetry has utilized Chelex 100 for the concentration
of Cd, Cu, Pb and Zn before analysis (22,23).
X-ray fluorescence has likewise utilized the concentration of Cu, Mn, Ni
and Zn from a 3 liter sample prior to irradiation to increase the concentration
of these metals to within detectable limits (24).
Neutron activation analysis has used Chelex 100 to concentrate the trans-
ition metals Ba, Cd, Ce, Co, Cr, Cu, Fe, La, Sc, U, V, and Zn (25).
The possibility of other naturally occurring chelators competing with the
Chelex 100 resin has been investigated to evaluate the effect they could have if
they occurred in high concentration in a natural water sample. Glycine, histi-
dine and photoplankton were added to marine water in concentrations as high as
10 5 M. No competition with Chelex 100 was found. Only EDTA at concentra-
tions of 10 6 M or higher showed any effective competition for 61+Cu (4) . Other
organic acids were also introduced in 2 to 250 mg/liter concentrations (citric
acid, malic acid, palmitic acid, p-hydroxybenzoic acid, glycine, humic acid,
aspartic acid, leucine, 1-cystine, NTA and EDTA). Only the synthetic chelator
EDTA exhibited any competing chelation for Cu or Zn (23). The natural chelators
do not compete favorably with Chelex 100, thus explaining in part the excellent
recoveries reported for transition metals concentrated from natural water
systems.
The ability of Chelex 100 to chelate the transition metals from a solution
that contains Na and K 108, and Ca and Mg 106 times that of the+trace elements
stems from a large range in selectivity coefficients. Using.Zn 2 as a standard
point and giving it a value of 1.000 in selectivity. Na , Ca 2 and Mg 2 would
chelate with a strength of 10~£, 1.3xlO"2 and 9x10 5, respectively. This is in
comparison with 1.3xl02 for Cu*, 4.4 for Ni 2, 3.9 for Pb 2, 4X10"1 for Cd and
2.4x10 2 for Mn (26). (Appendix B contains a more complete selectivity series.)
The order of selectivity for cations in nitrate or chloride media is:
+2 +2 +3 +3 +3 +2 +2
Cu » Pb > Fe J > Al J > Cr > Ni Z > Zn* >
Ag+ > Co+2 > Cd+2 > Fe+2 > Mn+2 > Ba+2 > Ca+2 >»
Na+ (26).
The selectivity series for cations in an acetate buffer system at pH 5.0
is:
v ,+2 > +2 » _ +2 > „ .+2 > _.+2 > «_+2 » „ +2
Pd Cu Fe Ni Pb Mn Ca »
Mg+2 >» Na+ (26).
-------
The selectivity of the iminodiacetate resin has been shown to be highly pH
sensitive. The divalent metal ions experience both ion exchange and chelation.
Both of these effects are pH dependent because they depend upon the number of
available ionic sites inside the resin particle and on the shape of the func-
tional group, which also changes with pH. Titration of Chelex 100 produces the
following zwitterionic forms as a function of pH (27).
CH_COOH CH-COOH CH-COO CH0COO
2 f 2 / 2 / 2
'N03 -CH0NH' (j)-CH2NH+
CH2COOH CH2COO CH2COO CH2COO
pH 2.21 >• 3.99 * 7.41 >- 12.30
At lower pH, the main effect is on ion exchange with each divalent ion
neutralizing the charge on two ionic sites, each of which is furnished by a
different diacetate group. Ion exchange increases with pH as more ionic sites
become available by migration of hydrogen ions out of the resin until a pH of
approximately 4 is reached. At this point chelation can start. The half
neutralization point (50% dissociated) inside the resin has been shown to occur
at a pH of approximately 5.6 + 0.2 (28). It has been found, however, that
chelation becomes effective at a lower pH as ionic strength increases (29).
When the distribution coefficient was plotted vs. the pH, the transition metals
exhibited curves similar to those of titration curves having the familiar "S"
shape. They begin to rise in distribution coefficient value at a pH of approxi-
mately 3.0 and have leveled off at a maximum value by pH 5.0. The distribution
coefficients appear to remain at a maximum value with increasing pH, with the
exception of Co and Cu, which begin a decline of the distribution coefficient at"
approximately pH 6.0 (28,30). The curve for the alkaline earth elements Ca and
Mg was found to have a plateau from pH 4 to 5.8 (31). When studied in a high
ionic strength medium, this' non-increasing region of the plot of the distribu-
tion coefficient vs. pH was found to decrease to a minimum between pH 4.0 and
5.8 and then began rising rapidly to a maximum distribution coefficient value at
pH 5.8 (29). When compared, these several studies suggest that a defined
working range of pH 5 to 6 is available to take advantage of the maximum distri-
bution coefficients of the transition metals and also to utilize the minimum
value for Ca and Mg in higher ionic strength solutions.
It has been concluded by other researchers that the differences in stabili-
ties for divalent ions are not significant,enough to allow a complete separa-
tion of one element from another (3). In a study of Chelex 100, Van Willigen
et al. (32) concluded that because of insufficient stability differences,
swelling with pH and insufficient equilibration, time, efficient separation of
metals was not feasible. In an attempted separation of Ca from Cu, complete
dseparation was not accomplished.
-------
At present the trace transition metals of saline waters are being concen-
trated and eluted with acid without any further attempt to separate the large
quantities of alkali and alkaline earth elements before stripping them collec-
tively from the Chelex 100 column (3).
The origins of these alkali and alkakine earth elements are the residual
sites on the resin not taken up by the trace transition metals; these residual
sites concentrate the salt matrix. The concentration of Na, K, Ca and Mg
remains many orders of magnitude above those of the trace heavy metals after
preconcentration (3,13). Further separation of the alkali and alkaline earth
elements beyond the salt that passes through the column has not been achieved.
Thus, the amount of salt eluted with the heavy metals is governed by the minimum
amount of resin required for the concentration of heavy metals. The reduction
of Na, K, Ca and Mg from the original amount in the sea water is accomplished
by using much less resin than can accommodate all of the ionic content of the
saline water. The transition metals, being held several orders of magnitude
more strongly by the Chelex 100 than the major salts, are quantitatively
chelated while the bulk of the salts pass through the column. Thus, the major
salts normally eluted with the trace metals are those occupying the remainder of
the resin capacity above the requirements of the heavy metals (3,13,21).
The use of this trace metal concentrate with a high salt content is limited
to instrumental techniques which can tolerate these large quantities of Na, K,
Ca and Mg eluted with the trace metals. These techniques which utilize the
Chelex 100 resin preconcentration have been previously described. They are
not, however, the most sensitive, nor are they multielemental techniques, with
the exception of neutron activation analysis and x-ray fluorescence. (X-ray
fluorescence has not been well quantified using conventional techniques.)
The method of x-ray fluorescence quantification described by Campbell et al,
utilizing the SA-2 resin loaded filter paper disks, cannot be Implemented while
even small quantities of Na, K, Ca or Mg are present in the sample matrix. The
alkali or alkaline earth elements, in combination with the trace metals, destroy
the analytical application of this technique through competition for sites in
the resin loaded paper used for quantification (33-35).
Spark source mass spectrometry, although independent of recovery for
analytical analysis, cannot tolerate the alkali or alkaline earth elements in
concentrations much above those of the trace elements being analyzed, making
necessary a separation selectively removing the higher concentrations of these
salts while retaining the trace elements of interest (36).
Optical emission spectrometry, using inductively coupled plasma or elec-
trode plasma (D.C. arc), exhibits interelement interferences when Na, K, Ca or
Mg are introduced at levels much higher than those of the trace elements under
study (37-39).
In neutron activation analysis the problem of interference from the salt
matrix is not so severe since Na is the only major interferent.
Sodium, upon radiation, produces 24Na, necessitating radiation shielding
and remote handling, as well as interfering with short-lived isotope counting.
10
-------
Two techniques using Chelex 100 have been employed tq overcome this problem.
Forberg and Lundgren (40), placed a sample on Chelex 100 after irradiation,
replaced 2IfNa with natural Na, and counted the column. Most recently, Lee,
et al. (25) used a mixture of Chelex 100 and ground glass to preconcentrate the
trace metals from 1 to 4 liters of sea water, which had been filtered with a
0.5 v filter and stored acidified to pH 1 with nitric acid. After adjusting the
pH to 8.0 and preconcentrating the sample, they eluted the Na, Br, K and Cl
using 100 mL of 0.01 M nitric acid and eluted the trace metals with 50 mL of
4 M nitric acid. They were able to recover 115mCd, 6l4Cu, 56Mn and 61+Zn from
99.7 to 100.2 percent. This technique did not remove Ca or Mg prior to the
final elution, but due to their non-interfering radiations, these large quan-
tities of alkaline earth metals did not interfere with the counting, utilizing
both the Nal(Tl) and Ge(Li) detection equipment.
One of the most sensitive methods of elemental analysis is graphite
furnace atomic absorption spectrometry (GFAAS). It is possible to detect 10~12
to 10~9 g of many of the trace elements in marine water. It would appear from
the detection limits alone that this technique could be used directly on marine
water samples for some of the trace metals. However, the high salt content of
sea water (35g/kg) makes it difficult to effectively volatilize the matrix
without loss of the analyte. The major component of sea water is sodium
chloride, which has a relatively high volatilization temperature. Also, the
trace metals in sea water are present as chlorides, which have a lower vola-
tilization temperature. Therefore, it is difficult to volatilize the sodium
chloride during the ashing step without losses of the analyte. Calcium and
magnesium chlorides are present in large quantities and a temperature greater
than 2000 °C is required to volatilize these elements. Thus, even if the sodium
chloride is removed during the ashing step using matrix modification (41), which
requires interference corrections, calcium and magnesium chloride remain to
interfere with the analyte during atomization (42,43). When the Chelex 100
column preconcentration of the trace elements is employed, the Na, K* Ca and Mg
are present in the final acid solution with the trace metals at levels almost
equal to the original sea water matrix (13). Even if the alkalis are removed,
the calcium and .magnesium cause depression of the signal from two to ten times,
depending on the analyte (43). This effectively decreases the normal detection
advantage of the GFAAS method. Thus, the interelement effects from the salt
matrix in both the original marine water or the Chelex 100 concentrate render
the GFAAS technique incompatible for these samples.
In summary, the application of the Chelex 100 trace metal concentrate from
marine waters has been instrumentally limited. The residual salt content eluted
with the trace metals approximates the concentrations in the original sea
water. These high concentrations of alkali and alkaline earth elements have
denied the use of instrumental techniques which suffer interelement effects
from this salt matrix. The present use of Chelex 100 has achieved an increased
concentration of trace metals for instrumental techniques able to utilize
samples with a high salt content. If the preconcentration were extended to a
complete separation, eliminating the alkali and alkaline earth elements, the
matrix free samples would permit the utilization of a variety of instrumental
techniques presently excluded from trace metal analysis of saline samples. Many
of these instrumental techniques have characteristics that would enable the
analysis of trace metals with greater sensitivity and multielemental efficiency.
11
-------
4. EXPERIMENTAL PROCEDURES
MATERIALS
Chemical Reagents
Water, nitric and glacial acetic acids were prepared using sub-boiling
distillation at the National Bureau of Standards (NBS), as utilized in ultra-
purification procedures at NBS (44). All reagents used in the separation
process were prepared and stored in clean FEP Teflon, unless otherwise stated.
Water and acid, unless otherwise specified, are these high purity reagents.
(Appendix C lists typical analyses for HNOa and HaO prepared using this
technique.)
Ammonium hydroxide was prepared by bubbling filtered ammonia gas through
sub-boiling purified water until room temperature saturation was achieved.
A 1.0 M solution of ammonium acetate was prepared by mixing 60 g of
purified glacial acetic acid and 67 g of saturated ammonium hydroxide and
diluting to 1 liter in a polypropylene volumetric flask. The acidity was
adjusted to pH 5.0-5.3 by dropwise addition of HNOs and/or NHi+OH.
All reagent and sample preparations were done in a class 100 clean air
laboratory (45).
Chelex 100 chelating resin, 200-400 mesh size, was purchased from Bio-Rad
Laboratories.
The elemental tracers 59Fe, 54Mn, and 65Zn in 0.5 N HCl were purified
reagents obtained from the Chemical and Radioisotope Division of ICN. The
60Co, and the short lived isotopes, 6t*Cu and 65Ni, were made by the Neutron
Activation Analysis Section at NBS from five-9's pure metals and dissolved in
nitric acid. The 109Cd and 210Pb were obtained by the Activation Analysis
Section from other sources and counted for purity before use.
All standard stock solutions for atomic absorption spectrometry (AAS) were
prepared from high purity metals or salts in sub-boiling distilled NBS acids,
as described by Dea,n and Rains (46). Working solutions were prepared as
needed.
12
-------
The buffer solutions used for pH standardization were 0.050 m potassium
hydrogen phthalate (pH = A.008 at 25 °C) and 0.025 m disodium hydrogen phos-
phate and 0.025 m potassium dihydrogen phosphate (pH = 6.865 at 25 °C) . These
reagents were prepared from ACS analytical grade reagents.
Chesapeake Bay Sea Water
The sea water used for method development was obtained during high tide at
the Virginia Institute of Marine Science (VIMS), at Gloucester Point on the
Cheaspeake Bay. The source was a submersible pump and plastic tubing perma-
nently submerged approximately 100 meters off shore from the Institute. This
sampling system has been in continuous operation for three years; the con-
tinuous sea water supply is normally used to support live laboratory experi-
ments at VIMS. The sea water was pumped directly into a conventional
polyethylene drum which had been cleaned first with hydrochloric and then with
nitric acid and then with purified water prior to use (47). After filtration
filtration '
through a 0.45 y millipore filter using an all polypropylene filter apparatus,
the sea water was collected in a clean polyethylene carboy and acidified with
Ultrex HN03 to prevent bacterial growth and to stabilize the trace element
concentrations (48,49).
Glacier Bay Alaskan Sea Water
A collection of pristine sea water samples was obtained by NBS personnel
for analysis. The ship "Surveyor" was used to transport men and equipment
from Juneau to the selected sample site, Glacier Bay, Alaska. Samples were
collected using a Teflon sampler and sampling technique developed at NBS (50).
One liter Teflon (FEP) bottles were prepared prior to sampling by cleaning in
hot (1:1) HC1 for twenty-four hours and then in hot (1:1) HNOa for a second
twenty-four hour period, after which they were thoroughly rinsed in sub-boiling
distilled water (44). A 44 g volume of NBS high purity HN03 was added to the
sea water sample to bring a 1 liter sample to 0.5 N in HNOa (48). The sea
water sample was then placed directly in the bottles at the sample site to
avoid contamination and the bottles were individually sealed in polyethylene
bags. The samples were transferred to a large freezer, maintained at -40 °C,
and were kept there until needed for analysis. These samples were designated
"NBS, 500 through 503, a and b".
A second group of samples also taken at Glacier Bay, Alaska were collected
by other researchers under varying conditions and transported to NBS. These
frozen samples arrived at NBS in varying states of acidity and filtration.
These samples were designated "Alaska, 1 through 5, a and b". The samples were
in polyethylene containers without polyethylene bags.
APPARATUS
Column Separation Apparatus
An Isolab QS-Q polypropylene column with porous polyethylene resin
support was used for 100 mL and 1 liter sample volumes. Although the same
column was used for both sample volumes, the amount of resin and reservoir
13
-------
systems was entirely different. For the 100 mL sample, the QS-S 25 mL conven-
tional polyethylene extension funnel was attached to the column to act as a
reservoir for the sample.
For a 1 liter sample the reservoir was a 1 liter Teflon (FEP) bottle
inverted and modified with a machined Teflon (TFE) closure insert containing a
microbore venting tube and outlet tube. The outlet was connected to a valve
(TFE) by a 1.59 mm i.d. Teflon (FEP) tubing and (TFE) connector linked to the
reservoir with a specially machined mount (TFE) which sealed the column into
the closed system. The mount contained a vent (sealed with a nylon screw,
allowing the removal of air from the system) as well as an inlet and was
tightly clamped to the column using the lip on the column at point B (see
Figure 1). The clamp (a modified glass joint clamp) and mount provided a seal
hich allowed the reservoir to be raised above the column to obtain enough
pressure to control the flow rate using the pressure of the raised reservoir
and the valve (Figure 1).
Atomic Absorption Spectrometry Apparatus
The initial experiments to determine the parameters necessary to separate
the alkali and alkaline earth elements (Na, K, Ca and Mg) from the transition
metals were analyzed utilizing flame atomic absorption. Manganese, being the
most weakly held of the transition metals, was used to trace their location.
Both atomic absorption and atomic emission were used in these determinations.
The Perkin-Elmer Model 403 Atomic Absorption Spectrophotometer, equipped with
Automatic Burner Control System, was used. Use was made of a Perkin-Elmer
3-slot Boling Burner Head with premix chamber and pneumatic nebulizer for
burner requirements. The expanded scale in the concentration mode, utilizing
either 10 or 100, averaging reading selection was used. The gas mixture was
either nitrous oxide-acetylene or air-aceytlene. All instrumental parameters
were adjusted to those suggested in the Perkin-Elmer instrument manual (51).
A Jerrell Ash Czerny-Turner Scanning Spectrometer, with wavelength modula-
tion for background correction (0.75 meter) was also used. It was equipped
with a Perkin-Elmer single slot burner having a pneumatic nebulizer and premix
burner chamber. A Perkin-Elmer concentration readout DC R2B was used for quan-
titative readout. A Princeton Applied Research lock-in amplifier with Fluke
high voltage power supply was used for the photomultiplier tube. An air-
acetylene or nitrous oxoide-acetylene flame was used and the wavelength and
burner parameters were those supplied by the Perkin-Elmer atomic absorption
manual (51).
The Hewlett-Packard Model 9830B Desk Top Computer was used to perform a
polynomial regression on the calibration data and to calculate the concentra-
tions of Na, K, Ca, Mg and Mn in the test solutions. The Hewlett-Packard
Calculator Plotter Model 9830A was used to plot the data, and the concentration
values were printed on a Hewlett-Packard 936613 Printer. The program was Poly-
nomial Regression File 3, as listed in the technical manual (52).
14
-------
Reservoir Bottle *
Modified Closure f
Microbore Tubing Air Ventt-
Valvet
1.59mm i.d. Tubing*
Connector t
Nylon Screw
Vent
Modified Clamp
Polypropylene Column
Porous Polyethylene Plate
A to B 2.0 cm
B to C 4.5 cm
A to C 6.5 cm
9.5 mm i.d. B 13.5 mm i.d.
2 mm Radius Step
C 8.5 mm i.d.
Figure 1.
The apparatus used for holding and delivering large quantities of sea
water at a controlled rate to Chelex 100 resin. The apparatus
(excluding column and clamp) was fabricated from Teflon FEP (*) or
Teflon TFE (t) which has desirable anti-wetting and non-contaminating
properties.
15
-------
pH Apparatus
The pH determinations were done using a Beckman Model SS-2 Expanded Scale
pH Meter, equipped with Beckman glass and reference electrodes.
Counting Apparatus
The gamma ray counting of the radiochemical tracers was done utilizing a
sodium iodide-thallium activated 7.5 cm crystal coupled to a Fluke high voltage
power supply and a Canberra Model 810 and 835 Energy Discriminator equipped
with Ridl Model 27101 Timer and 27104 Counter.
PROCEDURE
Column Preparation and Purification Procedure
The column preparation procedure consisted of precleaning the columns in
1:4 HC1 and then in 1:4 HNC-3 (ACS reagent grade acids) for one week in each
bath and then rinsing the columns with water after each acid wash. The column
was loaded with a slurry of Chelex 100 resin, 200-400 mesh size (in sodium
form). For 100 mL sea water samples 3.2-3.4 mL of resin was used which covered
the lower barrel of the column from point B to point C in Figure 1. For 1
liter samples 5.8-5.9 mL of resin was used which filled the column from point A
to point C in Figure 1. The resin was washed with a total of 15-20 mL of 2.5 M
HN03 (for the small and large resin volumes, respectively), in 5 mL portions to
elute any trace metal contamination present in the resin. Then two 5 mL
volumes of water were used to rinse the resin of excess acid. To transform the
resin to the NH4 form, 10-15 mL of 2.0 M NH4OH was added in 5 mL volumes.
After checking the pH of the effluent to assure the basicity, the column was
then rinsed with 10-15 mL of water to remove the excess NH^OH.
Procedure for the Comparison of Ammonium Nitrate and Ammonium Acetate as
Separating Agents
Ammonium acetate and ammonium nitrate were compared for the separation of
alkali and alkaline earth elements from a column of Chelex 100 which had pre-
concentrated the trace metals from 100 mL of sea water at pH 5.0. The reagents
were prepared from high purity sources as described and adjusted to a pH of 5.2.
The reagent was preweighed into a Teflon (FEP) beaker on a top loading balance
and added to the column 5 g at a time. After 50 mL of either ammonium acetate
or ammonium nitrate had been added, each column was treated with 15 mL of 2.5
M HNOa in 5 g aliquots. The effluent was collected in preweighed 7 mL poly-
ethylene bottles throughout the experiment. The bottles had been cleaned one
week in 1:4 HCl and one week in HN03, rinsed with sub-boiling distilled water
and dried on a clean air bench. The bottles were capped and reweighed on a top
loading balance. The effluent was analyzed for Na, K, Ca, Mg and Mn by atomic
emission and atomic absorption spectrometry. Instrumental parameters were
followed from the Perkin-Elmer atomic absorption instrumental manual (51).
16
-------
Evaluation of the Influence of pH on the Separation
Six columns of Chelex 100 resin were prepared in the ammonium form. A
300 mL sea water sample was spiked with 10 mL of 10.079 yg/mL Mn. The sample
was adjusted to pH 5.0 and divided into six identical 50 mL samples. Each sea
water sample was passed through a separate Chelex 100 resin column and the
effluent collected in a 60 mL polyethylene bottle. Five separate 70 mL volumes
of 1.0 M ammonium acetate were adjusted, using nitric acid, to pH 3.0, 3.5,
4.0, 4.5, and 5.0. The different ammonium acetate fractions were added to the
column in 6.0 mL volumes and one was treated with neutral water as a control.
The effluent was collected in 6.0 mL portions in 7 mL precleaned polyethylene
bottles. After 54 mL of the ammonium acetate had passed through the column,
6 mL of 2.5 M_HNOa was added to each column and this effluent collected as
before. The individual fractions from each column were analyzed for Na, K,
Ca, Mg and Mn by atomic absorption spectrophotometry or emission spectroscopy
using the Perkin-Elmer Model 403 Atomic Absorption Spectrophotometer (AAS).
Column Preconcentration and Separation Procedure
After changes in the procedure during exploratory experiments, the pro-
cedure was standardized for further quantitative experimental investigation.
For the 100 mL sea water sample 101.3 ± 0.2 g was weighed directly into a clean
250 mL Teflon (FEP) beaker. If a spike (natural or radiochemical) was to be
added, it was added and then the pH adjusted to pH 5.0-5.5 with the dropwise
addition of NHi+OH or HNOs. Then 0.5 mL of purified 8 M ammonium acetate was
added to aid in buffering the system in this range. Any necessary agitation of
the solution was done with a Teflon stirring rod. A small amount of the sea
water was added to the reservoir and column to allow the resin to undergo its
natural shrinkage as it changed ionic form and pH. This shrinkage results in
a resin volume of approximately one half of its original volume. After the
completion of this transformation was observed (two to three minutes) the
remaining sea water was added to the reservoir as needed to keep it filled.
The flow rate was approximately 0.8 mL/min for the duration of the separation.
To selectively elute Na, K, Ca, and Mg, and replace them with NHi, , 40 mL of
1.0 M ammonium acetate was added to the column in 10 mL volumes. At the com-
pletion of the ammonium acetate addition, a 5 mL.volume of water was added to
remove any residual ammonium acetate. The elution of the transition metals was
then accomplished with 7 raL of 2.5 M HN03 collected into clean preweighed 10 mL
conventional polyethylene bottles. The bottles were capped with clean poly-
ethylene lined caps and reweighed to accurately determine the weight of
effluent.
For the 1 liter samples the procedure.was essentially the same as the
100 mL samples, with minor alterations due to the apparatus requirement. The
sample, 1,013 g ± 0.5 g, was weighed into a 1 liter Teflon (FEP) bottle and
the pH adjusted in the same manner as previously described. The bottle became
the reservoir and was fitted with a modified closure (see Figure 1). The
bottle was inverted and the air purged from the system by means of the vent on
the column mount. The flow rate was adjusted using the valve and the height of
the reservoir and was kept to less than 0.2 mL/min until the shrinkage of the
resin was complete. Then the flow rate was increased to 1.0 mL/min and left
overnight to flow through the column (approximately sixteen hours required).
17
-------
After passing the sample through the column, the valve and tubing were
removed at the connector above the column mount and replaced with a smaller
reservoir containing 70 mL of 1.0 M ammonium acetate. The flow rate was
adjusted to 0.5 mL/min and allowed to flow until the reagent was exhausted.
The column was then removed from the mount and the resin washed with 10 mL of
water. The transition metals were eluted with two 5 mL portions of 2.5 M HNOs
into preweighed polyethylene bottles as previously described.
Radiochemical Procedure
Radiochemical tracers were used to gain specific information about the
behavior of each transition metal ion in the preconcentration and separation
with the Chelex 100 resin. The tracers were added to the sea water one radio-
isotope per sample prior to the pH adjustment, using disposable polypropylene
micro-pipettes. The column procedure was identical in all respects to the
preparation of the analytical samples previously described. However, all
effluent from the column was collected, including the sea water. The sea water,
ammonium acetate buffer, acid effluent, and column' resin were collected in
250 mL bottles in the 1 liter experiments, and the 100 mL samples were
collected in 125 mL bottles. The level of fluid in each bottle was adjusted to
that of the largest member of that elemental series by the addition of dis-
tilled water on a top-loading balance, allowing normalization of the samples
for counting purposes. All samples contained the same amount of fluid in iden-
tical containers. The bottles were counted for ten minute periods for gamma
radiation only.
This technique was checked for total recovery using 65Zn. The tracer was
added to 250 mL of the sea water sample and counted prior to manipulation.
This volume was then added to 750 mL of sea water and treated as a 1 liter
65Zn spiked sea water sample. The final acid volume was again adjusted to
250 mL and counted at the end of the separation as previously described. The
counting statistics for each element were optimized by energy discrimination.
The statistical error was obtained using the following equation (53,54,55):
Q = 100 X .. K .. /N + N,
^c N - N, s b
s b
where
Q = Percent experimental error corrected for background
K = Number of standard deviations
N = cpm = counts per minute or period unit time
s » sample including background
b = background
Evaluation and Application of the Separation Technique by Spark Source Mass
Spectrometry
Eight 101.3 g sea water samples were weighed into 150 mL Teflon (FEP)
beakers. Samples #1 and //2 were not spiked. Samples #3 and #4 were spiked
with 10~7 g of the stable isotopes 26Mg, t+1K, ^Ca, 47Ti, 53Cr, 5ltFe,
18
-------
"Cu, ^zn, 7iGa 82S 86S 91Zr> 97M 109Ag} 110pd> lllcd§ H3In 117Sn,
!25Te> l37Baj lt2Ce) l^5Nd> 203T1) and 26epb. Samples #5 and #6 were spiked
with 10 6 g of the aforementioned stable isotopes. The pH of each sample was
adjusted to between pH 5.2 and 5.3. A column of Chelex 100 was prepared in
ammonium form and each sample was passed through the column under gravity feed.
After the collection of the trace elements by the resin, each column was washed
with 5 g of water and 40 g of 1.0 M ammonium acetate. The column was again
washed with 10 g of water. The column was then stripped using 7 mL 2.5 M HN03.
Each of the acid effluents was collected in a Teflon beaker. Several drops of
HC1 were added to each beaker and the sample evaporated to dryness. The sam-
ples were treated in various ways and presented to Dr. P. J. Paulsen of the
Analytical Spectrometry Section, the Analytical Chemistry Division in the
Institute for Materials Research at the National Bureau of Standards. The
spark source mass spectrometer utilized for analysis was a Consolidated Electro-
dynamics Corporation (CEC) Model 21-110 utilizing the photoplate technique.
Sample #5 was analyzed directly after the elimination of the ammonium ion.
Sample #6 was plated onto gold wire at 3.00 volts for twenty-four hours and
then analyzed. Sample #4 was heated in a quartz crucible at 208-350 °C for
twenty-four hours before analysis.
Instrumental Evaluation and Application Procedure Using Graphite Furnace Atomic
Absorption Spectrometry
Portions of this research were done in conjunction with the Analytical
Spectrometry Section, Analytical Chemistry Division of the National Bureau of
Standards. The research in preconcentration and separation of trace transition
elements from the sea water matrix was within the scope of ongoing trace analy-
sis projects involving graphite furnace atomic absorption Spectrometry (GFAAS).
The usefulness of the final matrix for trace elemental samples of sea water was
evaluated by Mr. T. C. Rains and Mr. T. J. Brady for compatibility with current
high sensitivity atomic absorption equipment. The Perkin-Elmer Model 603
Atomic Absorption Spectrometer with heated graphite furnace with the AS-1 auto
sampler was used. The accuracy of the preconcentration technique was evaluated
by employing standard additions during sample preparation. To aid in the iden-
tification of the matrix effects of the major cations of sea water on the
analysis of Cd, Co, Cu, Fe, Mn, Ni, Pb and Zn, the analysis of these elements
was attempted from sea water directly and with the use of matrix modification
as described by Ediger et al. (41). The amount of Na, K, Ca and Mg remaining
with the trace metals was evaluated using ammonium acetate for the separation
of these salts from the Chelex 100 resin. Analysis of the separation pro-
cedure using water and ammonium nitrate substituted for ammonium acetate was
also accomplished. The homogeneous Chesapeake Bay sea water sample was used to
evaluate the quantitation of the method for the transition metals. Four sam-
ples were untreated, while six samples of this sea water were spiked with yg
quantities of the transition elements of interest: 50 yg Cd, 100 yg Co, 100 yg
Cu, 200 yg Fe, 200 yg Mn, 200 yg Ni, 100 yg Pb, and 50 yg Zn. Four total
process blanks of 100 mL each of sub-boiling distilled water were carried
throughout the entire process to obtain a blank level for each element. These
samples were then analyzed by GFAAS.
Alaskan sea water samples were prepared using the Chelex 100 resin column
Preconcentration and separation technique. Fifteen samples were analyzed in
19
-------
triplicate, with one replicate of each sample spiked for Cd, Mn, Ni, and Pb.
These samples were analyzed using the previously described GFAAS apparatus.
Experimental conditions for the analysis by GFAAS of the artificial matrix of
2,5 M HNOs and residual NHit were developed and optimized for each element.
They appear in Appendix D.
20
-------
5. RESULTS AND DISCUSSION
SEPARATING AGENT SELECTION
Characteristics Desired in the Separation
In developing a separation technique there exist a number of objectives
beyond the separation which determine the usefulness of the technique. The
parameters developed and reagents utilized must not themselves become hin-
drances to the compatibility of the final sample and the applications intended
to benefit from the separation.
These characteristics were considered essential to the successful utili-
zation of a separation technique developed for the removal of Na, K, Ca and Mg
from a Chelex 100 resin column:
—A reagent should be capable of selectively removing Na, K, Ca and Mg from
a Chelex 100 resin column without removal of transition metals or disrup-
tion of the resin. This would be the most effective approach.
—The reagent itself must be innocuous to the instrumental applications
desired. Thus, it would be desirable if the reagent could be completely
eliminated from the final matrix containing the trace transition metals.
This would be most satisfactory to all applications simultaneously.
—Contamination from the reagent should be minimal. A reagent which has
ultrapurification procedures available is most desirable.
These characteristics serve as a starting place for the research.
Early Experiments
In recalling the selectivity series for cations chelatad on Chelex 100
resin, the alkali and alkaline earth elements are the two most weakly held
groups. Sodium and K are held many orders of magnitude more weakly and are the
most easily removed. Calcium and Mg, however, are chelated with the same order
of magnitude in selectivity as some of the transition metals (Appendix B).
Manganese is the most weakly held transition metal, only 0.024 in comparison
with 0.013 for Ca selectivity. Since some of the more weakly held transition
elements may also be removed in an attempt to remove Ca and Mg, Mn is a logical
choice to act as a model for the transition elements in early experiments with
various separating agents.
21
-------
Dilute nitric acid was evaluated as a possible separating agent. A
0.05 M solution of acid was passed through a column of Chelex 100 resin which
had chelated 100 mL of sea water. Sodium and K were removed with 6 to 8 mL,
but Ca and Mg were only slightly removed, with 2,000 ug of Ca remaining in the
final volume when 2.5 M HNOs was used to strip the column. The hydrogen ion,
although an efficient stripping agent for the Chelex 100 resin, has several
drawbacks as a selective eluting agent. If the acid solution is too concen-
trated or too much is added, the pH of the entire column is lowered below
pH 5.0 and all elements chelated are eluted from the column simultaneously.
This totally defeats the purpose of the separation. Although nitric acid is
useful as a stripping agent to totally remove the ionic content of a Chelex 100
resin column, its use as a selective eluting agent is limited to Na and K only,
as described by Lee et al. (25).
The ammonium ion has many of the high purity and non-interfering charac-
teristics of the hydrogen ion. (It is the preferred form of the resin for the
concentration of transition metals, over that of the hydrogen ion, preventing
low pH problems in the resin.) An initial experiment using ammonium nitrate
was prepared to attempt the elution of Na, K, Ca and Mg from the resin. Ammo-
nium nitrate (0.4 M) was placed on a Chelex 100 resin column after chelation of
100 mL of sea water at pH 6.0. After the collection of 10 mL of ammonium
nitrate in 2 mL aliquots, the column was stripped with 2.5 M_ HN03. Analysis of
the effluent by atomic absorption spectrometry (AAS) indicated a tenfold
increase in the amount of Ca removed from the column during the ammonium
nitrate fraction. The Na and K showed the same pattern of elution as with the
dilute nitric acid, being eliminated in approximately 6 mL of the ammonium
nitrate. However, the amount of Ca remaining in the 2.5 M HNOs was approxi-
mately 2,000 yg.
The Comparison of Ammonium Nitrate and Ammonium Acetate as Separating Agents
The preliminary experiment with ammonium nitrate indicated some Ca and Mg
removal using the ammonium ion. Ammonium nitrate of high purity was prepared
as previously described. The solution concentration was increased to 1.0 M.
A second ammonium salt, ammonium acetate, was prepared, also at 1.0 M, The
acetate form was chosen for two reasons. First, it forms metal complexes with
Ca and Mg but does not form any extremely strong complexes with transition
metals (56). Secondly, it buffers in the pH range 5.0-5.5 and serves to stabi-
lize the critical pH during the elution. Because the effect of these ammonium
salts on the transition metals chelated by the resin was not known, the sea
water was spiked with 100 yg of Mn prior to chelation to allow detection in the
final volume. The experimental results for ammonium nitrate as an eluting
agent for Na and K appear in Table 1. Its effect on Ca and Mg is shown in
Table 2. The evaluation of the transition metals is modeled by Mn in Table 3,
The results of ammonium acetate for the elution of Na and K are given in
Table 4 and its effects on Ca and Mg are grouped in Table 5. The elution of
Mn is shown in Table 6.
Since the effluent was collected by weight (to prevent contamination) as
previously described in the Procedure, it is necessary to identify the density
of 1.0 M ammonium acetate to correct the weight to the volume. The density of
1.0 M ammonium acetate was therefore determined, using a pycnometer, to be
22
-------
N>
TABLE 1. THE GRADIENT ELUTION OF Na AND K FROM A 100 mL SEA WATER SAMPLE PRECONCENTRATED ON A CHELEX
100 RESIN COLUMN USING 1.0 M (pH 5.2) AMMONIUM NITRATE SEPARATING AGENT AND 2.5 M NITRIC ACID
STRIPPING AGENT
Sample
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Total
Cumulative
Volume
Reagent (mL)
NH4NO_ 4.91
NH4N03 9.70
NH4NO« 14.75
NH Wft 14 11
nn« m/^ jLf • **A.
HH.HO, 23.87
4 3
NH4H03 28.79
«R MM OO QQ
nn,FnJo J^.T?
NH.NO. 38.77
ft 3
NH4N03 44.02
NH.NO- 49.00
H_0 52.57
HN03 58.91
HN03 63.83
HNO 68.52
Na K
Concentra- Concentra-
tion Per- tion Per-
yg/mL yg centage Ug/mL yg centage
510 2500 99.3 26.8 132 99.4
3.9 18.7 0.7 0.15 0.7 0.6
<0.1 ~ ' "•- <0 . 1
<0.1 ^0.1
<0.1 <0.1
<0.1 <0»1
<0.1 <0.1
<0.1 ^0.1
-i
<0. 1 <0. 1
-------
NJ
TABLE 2. THE GRADIENT ELUTION OF Ca AND Mg FROM A 100 mL SEA WATER SAMPLE PRECONCENTRATED ON A CHELEX
100 RESIN COLUMN USING 1.0 M (pH 5.2) AMMONIUM NITRATE SEPARATING AGENT AND 2.5 M NITRIC ACID
STRIPPING AGENT
Sample
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
TOTAL
Cumulative
Volume
Reagent (mL)
NH4N03 4.91
NH4N03 9 . 70
NH4N03 14.75
NILNO 19.31
NH4N03 23 . 87
NH4NO_ 28.79
NH4N03 33.99
NH4N03 38.77
NH4NO 44 . 02
NH4N03 49 . 00
H20 52.57
HN03 58.91
HN03 63.83
HN03 68.52
Concentra-
tion
yg/mL
341
255
70.0
38.9
25.5
37.7
33.7
85.3
46.1
39.8
81.0
83.6
<0.1
<0.1
Ca
yg
1670
1220
354
117
126
186
175
408
242
198
289
530
5515
Per-
centage
30.28
22.12
6.42
2.12
2.29
3^.37
3.17
7.40
4.39
3.59
5.24
9.61
100.0
Concentra-
tion
yg/mL
347
299
96.3
36.8
23.6
26.2
21.7
23.9
22.5
20.2
10.6
13.0
<0.1
<0.1
Mg
yg
1700
1432
486
168
116
129
113
114
118
101
37.8
82.4
4597
Per-
centage
36.98
31.15
10.57
3.65
2.52
2.80
2.46
2.48
2.57
2.20
0.82
1.80
100.0
-------
TABLE 3. THE ELUTION OF Mn FROM A 100 mL SEA WATER SAMPLE PRECONCENTRATED ON A
CHELEX 100 RESIN COLUMN SUBJECTED TO 1.0 M (pH 5.2) AMMONIUM NITRATE
SEPARATING AGENT AND 2.5 M NITRIC ACID STRIPPING AGENT
Sample
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
TOTAL
Reagent
NH4N03
NH.NO,
4 3
NH4N03
NH.NO.
4 3
NH.NO.
4 3
NH4N03
NH4N03
NH4N03
A ^
A ^
H20
HN03
HN03
HN03
Mn
Cumulative Concentra-
Volume tion
(mL) yg/mL yg Percentage
4.91 <0.1
9.70 <0.1
14.75 <0.1
19.31 <0.1
23.87 <0.1
28.79 <0.1
33.99 <0.1
38.77 <0.1
44.02 <0.1
49.00 <0.1
52.57 <0.1
58.91 17.0 108 100
63.83 <0.1
68.52 <0.1
ins inn
25
-------
NJ
TABLE 4. THE GRADIENT ELUTION OF Na AND K FROM A 100 mL SEA WATER SAMPLE PRECONCENTRATED ON A CHELEX
100 RESIN COLUMN USING 1.0 M (pH 5.2) AMMONIUM ACETATE SEPARATING AGENT AND 2.5 M NITRIC ACID
STRIPPING AGENT
Sample
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Total
Cumulative
Volume
Reagent (mL)
MII r»r\APTi A QO
rl£l/ ViUUvitl^ *r..7fc
A J
WH PfV^PTT Q (\*)
ft 3
"KIU PfV^PTI 1 A ^A
all, v»UvH>Il_ J.'t . Jt
vrti frvw"1!! i o no
Vtfi * V^wvlOXln JL.7 . vlU
VTTl ' f^rtrtrf"1!! O/i *7 A
Nil . LUUL.M.™ Z4 . Z't
VJTI PA/IPW *7R Q1
Sid t \j\j\j\jfLt\ AfO^yj*
NH4COOCH3 33.53
VTTT r*t\r\f*v 1 Q 01
PlEl/ UVfWUllM JO . 4UJ_
Mf>f\f\ftJ f. O Q 7
COUCil— *ti . o /
NH. COOCH_ 47 . 44
H20 50.85
HN03 56.52
HN03 61.37
HN03 65 . 87
Na K
Concentra- Concentra-
tion Per- tion Per-
yg/mL ug centage yg/mL ug centage
510 2509 99.25 20.1 98.9 99.13
3.94 18.5 0.73 0.18 0.87 0.87
0.07 0.35 0.02 0.1
<0.1 <0.1
<0.1 <0.1
<0.1 <0.1
<0.1 <0.1
<0.1 <0.1
<0.1 • <0.1
<0.1 <0.1
<0.1 <0.1
<0.1 <0.1
<0.1 <0.1
<0.1 <0.1
2528 100.0 99.8 100.0
-------
TABLE 5. THE GRADIENT ELUTION OF Ca AND Mg FROM A 100 mL SEA WATER SAMPLE PRECONCENTRATED ON A CHELEX
100 RESIN COLUMN USING 1.0 M (pH 5.2) AMMONIUM ACETATE SEPARATING AGENT AND 2.5 M NITRIC ACID
STRIPPING AGENT
Sample
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
TOTAL
Cumulative
Volume
Reagent (mL)
WH f^^^f^f^TT ft Q *}
ft J
WTTT f^f\f\f^'O Q (\^
nu / v»V/\_H_/Xi« J • v**-
tt _j
NH4COOCH3 14.34
NH.COOCH- 19.08
4 3
NH , COOCH- 24 . 24
NH4COOCH3 28.91
xru of\om ^^ ^^
nn. %_iv/v/v>n^ j j • j j
\ra /^rtrtPTi TR 01
£%iif \j\J\J\sHf* JO •
-------
TABLE 6. THE ELUTION OF Mn FROM A 100 mL SEA WATER SAMPLE PRECONCENTRATED ON A
CHELEX 100 RESIN COLUMN SUBJECTED TO 1.0 M (pH 5.2) AMMONIUM NITRATE
SEPARATING AGENT AND 2.5 M NITRIC ACID STRIPPING AGENT
Sample
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
TOTAL
Mn
Cumulative Concentra-
Volume tion
Reagent (mL) pg/mL yg Percentage
NH4COOCH3 4.92 <0.1
NH.COOCH, 14.34 <0.1
4 3
NILCOOCH, 19.08 <0.1
NH4COOCH3 24 . 24 <0 . 1
NH. COOCH« 28.91 <0.1
wu ff^of^u 1 Q i%^ ^r\ i
Wn. LUULcl^ j J • DJ
-------
1.0079 g/mL. The correction from weight to volume was found to be unnecessary
for 1.0 M ammonium acetate. However, the 2.5 M nitric acid used in eluting the
elements does require a correction using the density 1.08 g/mL (57).
A comparison of the separation performed with ammonium nitrate and ammo-
nium acetate for Na and K can be made by evaluating Tables 1 and 6, respec-
tively. It can be seen that both of these ammonium salts eliminate Na and K at
an extremely rapid rate. Approximately 9 mL of either ammonium salt reduces
the concentration of Na and K to <0.1 yg/mL. It is also evident that this is a
complete separation, with neither element appearing in the 2.5 M HN03 used to
strip the column. Figure 2 is a graphical representation of Tables 1 and 4.
In comparing the elution of Ca and Mg from the column with ammonium
nitrate and ammonium acetate (Tables 2 and 5, respectively), a decidedly dis-
similar set of results is observed.
The ammonium nitrate did remove a large portion of the Ca from the column.
However, it was not a complete separation, even when using 50 mL of 1.0 M
reagent. Ten percent of the Ca was eluted with 2.5 M HNOa. This represents
over 0.5 mg of Ca which would be present with the trace transition metals in
this volume. The elution curve can be seen graphically represented in
Figure 3. It appears to be a slow continuous process throughout the 50 mL of
reagent, with an exception in the initial 10 mL. This first 10 mL of reagent
also contains Ca from the residual resin volume. (Further discussion of this
volume is necessary later in this Section.)
The ammonium acetate elution of Ca assumes an appearance much like that of
a chromatographic peak, with much less tailing than when using ammonium nitrate.
It is a more complete separation; a quantity of less than 0.1 yg/mL Ca remains
in the 2.5 M HN03 fraction. Complete separation of Ca is accomplished by 30 mL
°f 1.0 M ammonium acetate. The ammonium acetate and ammonium nitrate separa-
tion of Ca can be compared graphically in Figure 3.
Both ammonium nitrate and ammonium acetate exhibit the removal of Mg in
patterns similar to those of the Ca removal, with the exception of a reduced
amount of Mg in the final volume of ammonium nitrate. These separations can be
compared graphically in Figure 4.
Neither ammonium nitrate nor ammonium acetate produces the elution of Mn
(Tables 3 and 6). However, 2.5 M HNOs completely strips the Mn from the column
in less than 6 mL.
These findings support several conclusions. Since the nitrate ion is a
neutral ion without an affinity for forming Ca and Mg complexes, the separation
°f Ca and Mg from the Chelex 100 resin using ammonium nitrate is primarily due
to the ammonium ion. Using ammonium acetate, there appears to be assistance
from the acetate ion in the separation of Ca and Mg, in addition to that of the
ammonium ion. Therefore, ammonium acetate, rather than ammonium nitrate,
emerges as the most efficient separating agent for the removal of Na, K, Ca and
MS from the Chelex 100 resin.
29
-------
100
3 80
'c
D
X
1_
o
£ 60
.a
w
o
e
o
c
o
u
C
o
u
40
20
• Na, K NH4COOCH3
D Na, K NH4NO3
10
20
30
40
50
60
70
ml
or
1.0.M NH4COOCH3 PH 5.2
l.OM. NH4NO3 pH 5.2
H2o[— 2.5M. HNO3—|
Figure 2. The comparison of the removal of the alkali elements Na and K by ammonium acetate and ammonium
nitrate from a Chelex 100 resin column which has previously chelated 100 mL of sea water.
-------
100
• 80
>
k.
o
I 60
i_
<
c
o 40 -
c
«
o
u
20
Co, NH4COOCH3
— Co, NH4NO3
10
20
30
40
50
60
70
ml
or
l.OM. NH4COOCH3 pH 5.2
l.OM. NH4NO3 pH 5.2
HO
- 2.5M HN03 -j
Figure 3. The comparison of the removal of Ca hy ammonium acetate and ammonium nitrate from a Chelex 100
resin column which has previously chelated 100 mL of sea water.
-------
u>
to
100
Mg, NH4COOCH3
Mg, NH4N03
£lfc*
30
40
50
60
70
m
l
or
1.0.M NH4COOCH3 pH 5.2
l.OM NH4NO3 PH 5.2
120\— 2.5M HN03-|
Figure 4. The comparison of the removal of Mg hy ammonium acetate and ammonium nitrate from
the Chelex 100 resin column which has previously chelated 100 mL of sea water.
-------
CONCENTRATION AND SEPARATION PARAMETERS
Evaluation of the Influence of pH on the Separation
The pH of the ammonium nitrate and ammonium acetate used to separate the
Na, K, Ca and Mg in the previous study was pH 5.2. Since the reduction of pH
is also a method for removing ions from the resin, a reduction in the pH of
the ammonium acetate could perhaps increase the efficiency of the separation,
require less reagent to achieve the separation, and lower any blank that might
be contributed by the ammonium acetate. A study of the influence of pH on the
separation was initiated as described in the Procedure. The results have been
graphed in Figure 5.
The Mn is eluted by the ammonium acetate at any pH below pH 5.0. The
Mn overlaps the Ca peak from pH 3.0-4.5 and does not provide enough separation
from the Ca to obtain a complete separation. The Ca and Mg are eluted very
rapidly at pH 3.0, but from 3.5-5.0 they appear to change very little in place-
ment of the elution or the tailing towards the Mn. To obtain a complete
separation using ammonium acetate a pH of 5.0 or greater is necessary to pre-
vent Mn from being stripped with the Ca and Mg. This permits the retention of
Mn and presumably the other transition metals until stripped simultaneously in
a small volume of 2.5 M HN03.
Parameters Fixed for Further Study
The parameters used in the ammonium acetate separation were standardized
and repeated throughout all subsequent 100 mL sea water sample preparations.
These conditions are given in the Procedure. A composite of Tables 4, 5, and 6
was constructed to provide Figure 6. This figure shows the conditions for
the separation of the alkali and alkaline earth elements from a column of
Chelex 100 resin. It also shows the elution of Na, K, Ca and Mg from a resin
column that has preconcentrated a 100 mL sea water sample. The location of the
transition metals was also influenced by preliminary qualitative studies of the
radiochemical tracers 54Mn, 109Cd and 6=Zn.
The conditions specified in Tables 4 through 6 were repeated in each use
of this technique hereafter to eliminate an alteration in procedure affecting
the experimental results. Figure 6 can be viewed as a generalization of the
separation treatment applied to any sea water sample introduced.
The use of the water prior to stripping of the column of the transition
elements was not previously discussed due to its inconsequential effect on the
separation. It was originally placed in the separation to prevent the buffer-
ing action of the residual ammonium acetate remaining on the column prior to
stripping with 2.5 M nitric acid. However, some preliminary experiments
utilizing the 2.5 M nitric acid concentrate by GFAAS indicated that the acetate
ion caused suppression of the signal of some elements. As an organic, acetate
was also not favorable for SSMS and was undesirable in the final volume.
Therefore, the small water wash remained in the procedure to remove the acetate
anion from the column. It is not chelated by the resin and is effectively
removed with the residual aqueous volume prior to stripping of the column by
2.5 M nitric acid.
33
-------
IA
'c
o
o
h
c
8
c
o
-------
U)
Ui
100 r
Na, K
Mg
10
20
Transition
Metals
*-**•
•—I
30
40
50
60
70
ml
l.OM. NH4COOCH3 pH 5.2
•JH2Oh- 2.5M HN03 —H
Figure 6. The elution of Na, K, Ca, and Mg from a preconcentrated 100 mL sample of sea water on a
Chelex 100 resin column using 1.0 M ammonium acetate separating agent and 2.5 M nitric
acid stripping agent. ~~
-------
ASPECTS OF THE SEPARATION
The Use of Water as an Aid in Understanding the Separation
A discussion of the inital volume in the separation is necessary to gain a
complete understanding of the separation. A large initial amount of Na, K, Ca
and Mg appears in the separating reagent effluent (in approximately the first
10 mL), as observed in Tables 1 through 6 for both the ammonium nitrate and
ammonium acetate. It was assumed that much of this initial volume was the
expulsion of the residual volume in the resin and that the ions in this volume
were not only being stripped chemically but also displaced physically. A
sample was prepared using the standard conditions for preconcentrating a 100 -mL
sea water sample onto the standard resin column described in the Procedure.
Approximately 50 mL of water (neutral pH) was substituted for the separating
agent to investigate the initial removal of all four ions from the column with-
in the first 10 mL volume of the separating agent. The sea water was analyzed
concurrently as the study was done. The concentration of Na, K, Ca and Mg was
found to be 6,200, 267, 283 and 742 yg/mL, respectively, in the original sea
water sample (Chesapeake Bay Water). The ratio of the ions in the initial
volume approximates the ionic ratio of these elements in the original sea
water (see Table 7). The initial high amounts of the alkali and alkaline earth
elements in the separating agents, ammonium acetate and ammonium nitrate, are
partially contributed by this physical displacement. However, the ion
exchanged Na and K is also almost completely removed in this first 10 mL
volume. The amount remaining on the column after water elution is the amount
of Na and K not displaced physically, but which is chemically removed by the
ammonium ion using either ammonium salt.
A comparison of the Ca and Mg removed by physical displacement exhibits a
similar phenomenon. They are physically removed in the first 10 mL volume;
chemical elution also contributes to the Ca and Mg in this initial volume. The
removal throughout the remainder of the 50 mL of separating agent is due almost
totally to chemical removal by ammonium acetate or ammonium nitrate. The
amount of these ions actually removed by water in the remaining volume is neg-
ligible for both the alkali and alkaline earth elements, as seen in Table 7.
The amount of alkali and alkaline earth elements in the 2.5 M nitric acid
stripping agent, as illustrated in Table 7, indicates the amount of these
elements that would be left with the trace transition metals if these salts
were not first separated from the column: Na, 2,375; K, 74.7; Ca, 2,599 and
Mg, 466 yg. It is this quantity which is actually removed by the ammonium
acetate.
Selective Chelation of the Alkali and Alkaline Earth Salt Matrix During Pre-
concentration
The ratio to one another of Na, K, Ca and Mg removed by physical displace-
ment was similar to that in the original sea water, but the ratio of these ions
in the acid removal of chemically held ions is quite different, as shown in
Table 7. This same information can be obtained from Tables 1, 2, 4 and 5 when
the total quantity of the elements in the separating agent fractions is added.
The ratio of Na, K, Ca and Mg could be reduced to roughly 620:2.7:2.8:7.4 for
36
-------
TABLE 7. THE QUANTITY OF Na, K, Ca, AND Mg FOUND IN VARIOUS EFFLUENT FRACTIONS SUBSTITUTING WATER FOR A
SEPARATING AGENT IN THE ELUTION FROM A PRECONCENTRATED 100 mL SAMPLE OF SEA WATER ON A CHELEX
100 RESIN COLUMN
Na
Sample No. Reagent
1¥J f\
Elj\ vl
2TTT f\
«n \J
3 H20
4 H_0
5 H_0
6 HN03
7 HN03
8 HN03
Total
Cumulative Concentration
Volume yg/mL yg
8.90
18.68
28.81
38.31
48.19
55.62
60.36
65.05
859
0.788
1.78
1.01
0.63
319.8
-
<0.01
7,645
7.71
17.5
9.92
6.22
2,375
10,061
K
Concentration
yg/mL - yg
40
0.138
0.667
0.875
0.779
10.06
0.17
<0.01
356
1.35
6.54
8.59
7.70
74.71
.08
454.9
Ca
Concentration
yg/mL yg
35.2 313.3
0.030 0.29
0.024 0.24
0.019 0.19
0.059 0.58
350.0 2,599
-
0.31 1.4
2,915
Mg
Concentration
yg/mL yg
93.8 834.8
0.011 0.11
0.007 0.07
0.002 0.02
0.001 0.01
63.1 466.
-
0.027 0.13
1,308
-------
the natural sea water, while fraction #6 of Table 7 indicates a ratio of
24:0.7:26:4.7. This exhibits a noticeable reduction in the ratio of Na and an
increase of Ca by an order of magnitude. This is due to the amount of
each ion chemically held by the Chelex 100 resin. The Ca, although present in
a smaller amount than Na, K, or Mg in the sea water, has been selectively
chelated by the resin and actually occupies the most sites on a column of
Chelex 100 resin after the chelation of a sea water sample of at least 100 mL.
This serves to illustrate that without the removal of Ca by a separating
agent prior to stripping of the column, the matrix would be dominated by Ca.
This becomes important when several of the instrumental techniques are con-
sidered. The significant instrumental ramifications of this observation will
be discussed specifically in a following section, with reference to Graphite
Furnace Atomic Absorption Spectrometry.
The amount of Na, K, Ca, and Mg remaining in the final 2.5 M nitric acid
purging of a Chelex 100 resin column which has preconcentrated a 100 mL sea
water sample and has been eluted with 50 mL of either water, ammonium nitrate
or ammonium acetate, is presented for comparison in Table 8.
TABLE 8. THE AMOUNT OF Na, K, Ca, AND Mg REMAINING IN THE FINAL 2.5 M NITRIC
ACID VOLUME AFTER ELUTION OF THE RESIN COLUMN WITH 40 mL OF WATER,
AMMONIUM NITRATE OR AMMONIUM ACETATE. THE COLUMN HAD PREVIOUSLY
PRECONCENTRATED A 100 mL SEA WATER SAMPLE
Na K Ca Mg
Pg Pg Pg Pg
H20a 2,375 74.71 2,599 466
NH4N03b 0.0 0.0 530 82.4
NH.COOCH_C 0.0 0.0 0.0 0.0
4 3
aThe data for H20 as the eluting agent was taken from Table 7.
The data for NH,NOg as the eluting agent was taken from Tables 1 and 2.
°The data for NH,COOCH« as the eluting agent was taken from Tables 4 and 5.
PHYSICAL PARAMETERS AND EXTENSION OF THE SEPARATION TO LARGER VOLUMES
Chelex 100 resin is a dynamic resin and in the ammonium form at a pH of 7
to 14, the resin shrinks to approximately one half of its original volume when
subjected to sea water at pH 5.0-5.5. Appendix E lists different resin volumes
with different ionic forms (58). This shrinkage is acceptable for the original
38
-------
3.2-3.4 mL of resin (200-400 mesh) used for a 100 mL sea water sample. The
flow rate, influenced only by gravity, is approximately 0.8 mL/min. A 100 mL
sample can be placed on the column in four 25 mL portions in approximately
two hours, using the QS-S 25 mL column reservoir.
When the use of larger volumes is necessitated, these characteristics
become intolerable. When the column is filled with resin to its 6 mL capacity,
the flow rate is less than 0.3 mL/min and 1 liter of sample requires fifty
hours to percolate through the resin. Therefore, the Teflon apparatus dia-
grammed in the Apparatus section was developed to surmount these physical
characteristics. Employing this apparatus, a 1 liter sample was preconcen-
trated in sixteen hours, with only initial adjustment of the flow conditions.
The flow rate of 1 mL/min was attained by simultaneously adjusting the Teflon
valve and reservoir height to control the pressure, as described in the Pro-
cedure. The residual volume of the entire apparatus which may retain sample
was determined to be less than 0.1 mL. Constructing the apparatus of Teflon in
all areas where contact is made with the sample utilizes the non-wetting and
non-contaminating characteristics of this fluorocarbon, which can be scrupu-
lously cleaned in acid (47).
A further benefit of this apparatus is that during the preconcentration
onto the resin, the sample and column are protected from contamination from the
environment; the only entrance into the system is a microbore tube which can be
filtered to exclude particulate contamination. In addition to the apparatus'
required use with large volumes, it possesses desirable qualities for field or
shipboard use.
Using the Teflon apparatus, a 1 liter sea water sample was adjusted to
pH 5.0 and preconcentrated as described in the Procedure. The ammonium acetate
effluent was collected in approximately 5 mL volumes as the Na, K, Ca and Mg
were stripped from the resin. Table 9 is the quantltation of Na in the ammo-
nium acetate elution from this column. Table 10 shows the K eluted in this
effluent. The elution of Ca is presented in Table 11 and the Mg is presented
in Table 12. These results were used to typify the separation of the alkali
and alkaline earth elements using 1 liter samples and are graphed in Figure 7.
(Radiotracer data, discussed later in this chapter, was used to justify the
location and quantity of the transition metals in Figure 7).
The pattern of elution is identical to that of the 100 mL volumes. It
is extended only to the amount equal to the increase in the total quantity
of resin, which is approximately double. These results were used to establish
the procedure for preconcentrating and separating the trace transition metals
from 1 liter sea water samples. The procedure was not altered from that indi-
cated in the Procedure for any additional tests to avoid procedural differences
in the comparison of results.
39
-------
TABLE 9. THE ELUTION OF Na FROM A PRECONCENTRATED 1 LITER SAMPLE OF SEA WATER
ON A CHELEX 100 RESIN COLUMN USING 1.0 M (pH 5.2) AMMONIUM ACETATE
SEPARATING AGENT AND 2.5 M NITRIC ACID STRIPPING AGENT
Sample
No . Reagent
jn Nrl/ v^wUwCl-A
ft X
£ Ntl/ C^ LJC/VvifT/^
j NH * wUUl-»rl«
A 3
A NH COOCH
5 NH4COOCH3
6 NH4COOCH
7 NH4COOCH3
o NHiCOOCH*.
H 3
9VJTT f^f\^\f*T3
VtLl f OvJV/UtlM
A x
10 NH4COOCH
11 NH4COOCH
± £. JNtlj V.fWUv'O-n
/I T
13 H20
14 H20
15 HN03
16 HN03
17 HN03
TOTAL
Cumulative
Volume
(mL)
5.85
11.75
17.36
22.80
28.87
35.09
40.97
46.48
52.43
58.56
65.25
69.50
78.01
80.27
87.93
93.37
98.18
Concentra-
tion
yg/mL yg Percentage
2200 12870 96.1
90 531 3.9
0.09 0.5 <0.01
<0.1
<0. 1
-------
TABLE 10. THE ELUTION OF K FROM A PRECONCENTRATED 1 LITER SAMPLE OF SEA WATER
ON A CHELEX 100 RESIN COLUMN USING 1.0 M (pH 5.2) AMMONIUM ACETATE
SEPARATING AGENT AND 2.5 M NITRIC ACID STRIPPING AGENT
Sample
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
TOTAL
Reagent
JNH* v'Wv/VjLiM
ft 1
NH4COOCH3
NH4COOCH3
NH4COOCH3
NH4COOCH3
NH4COOCH
NH4COOCH3
NH4COOCH
NH4COOCH3
NH4COOCH3
NH4COOCH3
NH4COOCH3
H20
H2°
HN03
HN03
HN03
Cumulative Concentra-
Volume tion
(mL) Wg/mL
5.85 80
11.75 . 4.7
17.36 <0.1
22.80 <0.1
28.87 <0.1
35.09 <0.1
40.97 <0.1
46.48 <0.1
52.43 <0.1
58.56 <0.1
65.25 <0.1
69.50 <0.1
78.01 <0.1
80.27 <0.1
87.93 <0.1
93.37 <0.1
98.18 <0.1
yg Percentage
468 94.4
28 5.6
496 100.00
41
-------
TABLE 11. THE GRADIENT ELUTION OF Ca FROM A PRECONCENTRATED 1 LITER SAMPLE OF
SEA WATER ON A CHELEX 100 RESIN COLUMN USING 1.0 M (pH 5.2) AMMONIUM
ACETATE SEPARATING AGENT AND 2.5 M NITRIC ACID STRIPPING AGENT
Sample
No. Reagent
1VTTJ f^f\f\f^"LJ
iNii j w\_/v/L»n ,-.
£* W ti / \-*UCJ Ljll.«
fl X
ft X
*\ r*j ti » i_*u\jk_/ttrt
ft ~t
5XTTT f^/*^^f*TT
IN n / L* wu u n «
H 3
6 NH,COOCH3
/ Wrr* \j \J\j\~tLLn
ft j
o T^W t*C\C\r*\k
fi j
9 NH4COOCH3
^ i
"1 O "fcTTT ^1^1/\OTT
ft X
13 H20
14 H20
15 HN03
16 HNO_
17 HN03
TOTAL
Cumulative
Volume
(mL)
5.85
11.75
17.36
22.80
28.87
35.09
40.97
46.48
52.43
58.56
65.25
69.50
78.01
80.27
87.93
93.37
98.18
Concentra-
tion
Ug/mL
310
200
110
93
72
60
42
27
16
14
2.7
1.1
0.1
0.4
<0.1
<0.1
yg
1812
1180
617
505
438
373
247
149
95
85
18
4.7
0.9
3
5527
Percentage
32.8
21.4
11.2
9.1
7.9
6.8
4.5
2.7
1.7
1.6
0.3
<0. 1
<0. 1
<0. 1
100.0
42
-------
TABLE 12. THE GRADIENT ELUTION OF Mg FROM A PRECONCENTRATED 1 LITER SAMPLE OF
SEA WATER ON A CHELEX 100 RESIN COLUMN USING 1.0 M (pH 5.2) AMMONIUM
ACETATE SEPARATING AGENT AND 2.5 M NITRIC ACID STRIPPING AGENT
Sample
No. Reagent
JL JNtl * vvvV/Hn
ft 3
^ Nrl * \j \j\J\jti^
A T
j JMH . O\JUt>ri*»
A x
ft 3
5 NH4COOCH3
o NH * COOCIi«%
A X
A 3
8 NH4COOCH3
A Q
1 ^ TvTU rT^^^lT
iw 1NI1/ wx/L/wilM
11 NH^COOCH
13 H20
14 H20
15 HN03
16 HN03
17 HN03
TOTAL
Cumulative
Volume
(mL)
5.85
11.75
17.36
22.80
28.87
35.09
40.97
46.48
52.43
58.56
65.25
69.50
78.01
80.27
87.93
93.37
98.18
Concentra-
tion
yg/mL
772
493
190
87
35
10
2.2
0.76
0.31
0.15
0.08
0.02
0.01
0.01
0.20
0.22
0.04
yg
4512
2910
1066
473
213
62
13
4.2
1.8
0.9
0.5
0.1
0.1
<0.1
1.5
1.2
0.2
9259.5
Percentage
48.7
31.4
11.5
5.1
2.3
0.7
0.2
0.1
<0. 1
<0.1
<0, 1
<0. 1
<0. 1
<0. 1
<0.1
<0.1
100.0
43
-------
100 r
£ 8°
'c
o
I
6°
c
o
c
0
u
c
o
u
40 -
20
Transition
Metals
10 20 30 40 50 60 70 80 90 100
ml
l.OM NH4COOCH3 pH 5.0 H2o 2.5& HNO3
Figure 7. The elution of Na, K, Ca, and Mg from a preconcentrated 1 liter sample of sea water on a
Chelex 100 resin column using 1.0 M ammonium acetate separating agent and 2.5 M nitric acid
stripping agent.
-------
QUANTITATIVENESS OF THE TOTAL RECOVERY OF SELECTED TRANSITION ELEMENTS USING
RADIOCHEMICAL TRACERS
"When developing analytical methods for the deter-
mination of trace elements in sea water, it is essential
to check the efficiency of the proposed concentration
and separation process. This can be done chemically by
spiking stripped sea water samples with known amounts of
the appropriate species of the element in question and
then checking the recovery by physico-chemical methods.
However, if a suitable radio-nuclide of the element is
available the recovery can be determined more simply
radiochemically (it is more important that the radio-
nuclide added should have a high specific activity so
that the amount of the inert form of the element added
is kept to a minimum). In both the chemical and
radiochemical methods of checking the separation
process it is, of course, vital that the spike should
be equilibrated with that present naturally in the
sample" (3).
J. P. Riley has defined well the purpose of the tracer study on this analy-
tical technique. Other researchers have used Chelex 100 resin to preconcentrate
the trace transition elements from marine waters and have reported recoveries of
>99 percent for many of these elements. When altering an analytical procedure
the yield may also be altered with the introduction of the separation procedure.
Although no evidence of losses has occurred, the process would be unusable until
a quantitative study was done on the elements that are to be determined using
this method. The alkali and alkaline earth elements have thus far been the
primary target of analysis and only a model of the trend of the transition
elements has been followed through Mn. The quantitative recovery of the tran-
sition elements was obtained using radiochemical tracers. This technique allows
the most accuracy with the least alteration of the actual sample. Most of the
radiotracers used were carrier-free sources of less than 1 yg total elemental
content, allowing the minimum alteration of the actual elemental ratios and
total quantities.
Due to their varying specific activities and elemental concentrations, a
different amount of each tracer was used. The elemental tracers 109Cd, 59Fe,
2lOPb 54Mn and 65Zn were carrier-free reagents. The elements 50Co, 6^Cu
and 65Ni were made at the NBS reactor facility and had concentrations of 3.55,
2.29 and 3.52 yg/mL, respectively. The extreme sensitivity for .these radio-
tracers allowed the measurements of extremely minute total spike additions.
The radiochemical isotopes behave identically in chemical reactions and
can be treated as the normal stable isotopes in chemical consideration (59,60,
61). "Disturbing isotope effects arise only rarely" (61) and for these tracers
chemical considerations are identical to the naturally occurring isotopes until
the instant decay occurs and transformation of the radiochemical element into
another element takes place. This transformation presents no problem as long
as the decay state is a non-radioactive form, or at least one with a distinct
45
-------
mode and energy of decay different enough from the element being detected that
they can be distinguished (61).
Addition of the tracer to the sea water sample occurred just after the
sea water sample was weighed for analysis and prior to manipulation of the pH.
Both the sea water and the tracers were 0.5 If in HN03 (with the exception of
59Fe in HCl). The pH was then adjusted to between pH 5.0-5.5, allowing the
natural and tracer elemental species to be altered simultaneously if any alter-
ations in form occurred. During this procedure a Teflon coated stirring bar
(with remote magnetic stirrer) was substituted for the Teflon stirring rod used
on the samples destined for instrumental analysis. This was done to promote
the most homogeneous conditions in every aspect.
The data obtained during these radiochemical tracer experiments are given
in Tables 13 and 14.
Counting each column fraction had two distinct advantages. First, the
amount of tracer was not only quantified in the acid volume where it can be
utilized by instrumentation (thus establishing useful recovery), but the
presence of the tracer in other specific fractions could be identified and could
pinpoint the exact source of the "loss" for the total final efficiency to the
concentration onto, separation and stripping of the Chelex 100 resin column.
Secondly, the technique of counting all effluent from the column, resin and
QS-Q column (physical column shell) offers a closed system that lends its appli-
cation to counting of all the tracers used in this study, short and long lived,
with a minimum of counting error introduced due to decay of the tracer. Count-
ing the tracers before and after the experiment could not be done with the
short lived tracers 6tfCu and 65Ni without a large introduction of error from
decay during the experiment. However, the counting of all fractions involved in
any one separation was accomplished in forty minutes, eliminating the error due
to rapid decay. Only the 65Ni exhibited two fractions possessing detectable
gamma radiation and thus eliminated any correction. The 65Ni, having a half
life of 2.5 hours, decayed beyond useful detection (essentially background)
during the 1 liter experiment and gave no results for that application.
The 5lfMn in the 1 liter experiment also gave no results. This occurred
because the "breakthrough" point was reached at approximately 500 mL of the
initial loading of the sea water and therefore nullified the remaining results
for 5ltMn. To quantitatively hold Mn from a full 1 liter volume of sea water a
slightly larger volume of resin would be required. This did answer the ques-
tion of whether Mn was a good model for the trace transition metals during the
earlier work. It was the only transition element tested that reached its
"breakthrough" point. Thus, under these conditions and in this matrix it is the
most weakly held of the transition elements, and a good indication of the first
of this group to appear from the column. No attempt was made to increase the
volume of resin to concentrate the Mn from a 1 liter volume, which would give an
excellent recovery on a longer column, judging from the quantitative recovery in
the 100 mL experiment. The parameters were established to compare the recov-
eries under a set of identical conditions. If the procedure were altered, the
continuity of the other work would have been reduced. It can be suggested that
46
-------
TABLE 13. SUMMARY OF THE RADIO TRACER STUDY RESULTS SHOWING THE PERCENTAGE OF ELUTION IN EACH OF THE
EFFLUENT FRACTIONS FOR THE 100 mL SAMPLES3
Element
109Cd
60Co
6ItCu
59Fea
5"lfa
65Ni
21°Pb
65Zn
Sea Water
Effluent
<0.07
<0.06
0.30 +0.04
0.03 + 0.001
0.012 + 0.0005
5.3
7.21
<0.048
0.08 + 0.007
0.09 + 0.004
1.42+0.13
0.020 + 0.008
0.059 + 0.006
Loss Percentage
Ammonium
Acetate
Effluent
<0.07
<0.06
<0.03
<0.002
0.001 + 0.0004
0.25
0.57
<0.02
<0.006
<0.003
<0.01
<0.03
<0.02
Column
and
Resin
<0.07
<0.06
0.16 + 0.03
0.03 + 0.001
0.011 + 0.0005
0.22
0.20
0.01 +0.006
<0.06
<0.003
0.18 + 0.13
<0.02
<0.02
Recovery
Percentage
Effluent
99.99
99.99
99.54
99.94
99.976
94.23
92.02
99.99
99.92
99.91
98.40
99.98
99.94
+ 0.064
+ 0.079
+ 0.30
+ 0.036
+ 0.024
+ 0.11
+ 0.099
+ 0.067
+ 0.48
+ 0.11
+ 0.087
jj - •• - - "- • ........ . .- _,...,. . . ......
xhe upper error limits have been used where zero was obtained based on one standard deviation in the
computation of the error.
The error on this determination is reserved and described in the text.
-------
js
oo
TABLE 14. SUMMARY OF THE RADIO TRACER STUDY RESULTS SHOWING THE PERCENTAGE OF ELUTION IN EACH OF THE
EFFLUENT FRACTIONS FOR THE 1 LITER SAMPLES3
Element
109Cd
60Co
Sea Water
Effluent
<0.024
1.74 -«- 0.005
Loss Percentage
Ammonium
Acetate
Effluent
<0.06
0.03 + 0.01
Column
and
Resin
<0.012
0.89 + 0.02
Recovery
Percentage
Nitric
Acid
Effluent
99.99 + 0.063
97.34 + 0.17
0.17 + 0.006
6.22
<0.007
0.90
0.01 + 0.003
0.31
99.82 + 0.069
92.57
210Pb
65Zne
2.13 + 0.14
<0.01
<0.056
<0.03
0.11 ± 0.081
<0.04
97.76 + 0.39
99.99 + 0.12
100.1 + 0.12
The upper error limits have been used when zero was obtained based on one standard deviation in the
computation of the error.
The error on this determination is reserved and discussed in the text.
exceeded the capacity of the column at 500 mL and was not evaluated.
Ni decayed below useful levels and yielded no data for the 1 liter experiment.
Zn was determined using two different techniques as discussed in the text.
j
65
65
-------
a larger column for 1 liter samples would collect Mn quantitatively. In volumes
over 1 liter, other elements may exhibit phenomena similar to the Mn if the
column is kept at its present size. Thus, a 1 liter limit is suggested for the
larger sea water volumes.
The technique of counting each fraction was checked using 65Zn by another
counting technique. The 65Zn was placed in 250 raL of the sea water and counted
prior to manipulation; the 250 mL was added to 750 mL of sea water and the usual
procedure was followed. The count (corrected counts in ten minutes) prior to
manipulation was 742,152. The corrected count obtained after adjustment of the
acid fraction to 250 mL, was 743,179 corrected counts in ten minutes. Done in
the same manner as the other tracer studies, the recovery in the acid fraction
was calculated to be 99.99 + 0.12, when counted before and after manipulation,
100.1 Hh 0.12 was calculated as the recovery. This indicates that losses which
were not accounted for were not encountered and act as a check on the technique.
Using Table 13 for the results of the 100 mL sample volumes and Table 14
for the results of the larger 1 liter sea water volumes, the results for all the
elements tested and all the fractions can be compared. Of the elements tested,
none (with the exception of 59Fe) showed any significant amount of the respec-
tive elements in the ammonium acetate fraction. This fraction is the total
additional manipulation of separation added.to the established procedure. The
remaining volumes represent only the concentration of the transition elements by
the column, the removal of the elements from the column and the resin and column
remaining after removal. This effectively demonstrates that the separation of
Na, K, Ca and Mg is obtained with no cost to the total overall efficiency,
beyond concentration and elution from the column. The elements of interest are
not sacrificed within detectable limits at the expense of the addition of the
separation step.
The losses that occurred with 5^Fe are synonymous with other problems of
Fe. The small amount of loss in the ammonium acetate fraction can be explained
by residual loss from the previous sea water fraction which is still present in
the residual volume. At least some contribution from this factor is certain
since there is no wash between the sea water and the start of the ammonium
acetate. The non-quantitative retention of Fe and Co on Chelex 100 was a major
par_t of the paper by Callahan et al. (12). They reported that both Fe 3 and
Co 3 were not quantitatively retained by Chelex 100, and tried several reducing
agents on these ions to reduce and stabilize their total complement in sea water
to the +2 state. The Co was found to be naturally retained, 97-98 percent on
the Chelex 100 resin from sea water. This high retention rate for Co was
attributed to the fact that approximately 96 percent of.the Co in sea water
naturally exists as Co 3 and the remainder exists as Co 3. The Fe exhibited
these same phenomena, but the recovery was usually only 92-95 percent; No data
on the e/fective^ species ratios were given, but Fe is easily oxidized by air
from Fe 2 to Fe 3. A small difference in the general amount of stirring could
then alter this Fe 2/Fe 3 ratio and cause a different amount of the Fe to be
initially absorbed onto the Chelex 100 resin column. The error limits on Fe'
were therefore withheld from the final summary. The error as calculated is a
function of radiological counting uncertainty, not an experimental chemical
alteration, and therefore does not apply in this case. For the other elements
49
-------
(Ca, Mg, Cd, Co, Cu, Mn, Ni, Pb and Zn) the +2 state Is the normal species in
sea water (1).
The error limits represent very conservative limits for the data. The
equation described in the Procedure was used and takes into account not only
the error in the sample count, but also the background count. Several of the
experiments were repeated to verify the results. The repeated experiments
usually used a different amount of tracer (usually at least double) to alter
this parameter, which should not alter the results if the values are true. In
each case (with the exception of 59Fe previously discussed) the uncertainty in
the recovery was more than adequate to encompass both determinations. Table 13
contains these comparisons.
In Tables 13 and 14 the zero, negative and negligible counts were replaced
by the unit greater than the upper detection limit. The determination was made
that the error was approximately one order of magnitude more significant than
the negligible counts and that this error bracketing the zero numbers was a more
significant number and represented an upper limit more meaningful than the use
of zero or ND (Not Detected) (55).
The use of both 100 mL and 1 liter samples in this quantitative study was
intended to establish quantitative boundaries that would encompass a wide range
of sample volumes. The 100 mL sample size was the lower or smaller limit and
the 1 liter sample was the larger or upper limit. This was set up to establish
a flexible range of sample volumes from 100 to 1,000 mL which had already been
quantified. Any volume in between these two extremes would fall quantitatively
between the recoveries obtained. This range is intended to encompass a large
range of sample concentrations and instrumental needs which change with changing
samples and instrumentation.
EVALUATION OF THE COMPATIBILITY OF THE CHELEX 100 CONCENTRATION AND SEPARATION
WITH SPARK SOURCE MASS SPECTROMETRY
The spark source mass spectrometric (SSMS) technique has several unique
features which make it a desirable technique for \ise in conjunction with multi-
elemental samples of aqueous origin. It possesses the unique benefit of being
both multielemental and independent of recovery for analytical measurement. The
quantitative results are obtained from the isotopic ratio measurements of stable
isotopes of the elements. The technique uses the addition of enriched stable
isotopes, high in the lower naturally occurring abur.-lance isotopes of the
element of interest. Equilibration of these added spikes (enriched isotopes)
and the naturally occurring isotopes of the elements in the sample prior to
manipulation is of primary importance to the analytical determination. The
quantitative determination is based only on the ratio of spike isotope to
natural isotope determined using the mass spectrometer. Thus, the sample is
independent of losses of both spike and natural isotope of an element after
equilibration as the analysis depends on a ratio and not on the total amount.
However, enough of the sample and spike is required to determine this ratio
and prevent the blank from becoming significant. The technique will yield
analysis if approximately 10~7 g of the element is present. This amount will
yield a confidence of approximately five percent in the actual analysis.
50
-------
The isotopic ratio measurements require a sample free of organic matter and
low in other matrix ions. The organic ions fragment with unit charges and
superimpose values on the elemental masses, causing uncertain results. The
matrix ions, if they are in quantities high above those of the elements of
interest, can dilute the sample and also undergo polymerization into multiele-
mental forms possessing charges. The alkali and alkaline earth elements are
especially notorious for this and join in many molecular combinations possessing
charges (36).
Sea water samples were spiked with a multielemental isotopic spike; its
elemental components appear in the Procedure. The samples were equilibrated,
adjusted to pH 5.0-5.5, concentrated on a Chelex 100 resin column, separated
from Na, K, Ca and Mg using ammonium acetate, and stripped from the column as
described in the Procedure. A few drops of hydrochloric acid were added to the
nitric acid fraction containing the spiked Isotopes and the natural elements,
and the solution warmed gently and then taken to dryness. This removed the
residual ammonium ion (62). A sample was examined by the Analytical Spectro-
metry Group of the Inorganic Analytical Research Division at the National Bureau
of Standards, for analysis by SSMS. The investigation showed interference of
organic molecular origin. These organics were presumed to originate during the
acid elution from the Chelex 100 resin and were probably break-down products
from the resin. Sample #6 was plated onto gold wires (to eliminate any organic
molecules) as described in the Procedure, and both the anode and cathode were
analyzed by SSMS. A second technique was also utilized to eliminate the organic
material. Sample #4 was heated with nitric acid in a quartz crucible at 208-
250 °C for twenty-four hours and was then analyzed.
The samples were then checked for "good measurable signal" which would lead
to routine analysis for the element. The spiked isotope was used as an indica-
tor of measurability if the normal isotopes were below detectable limits. The
following elements were reported to have "good measurable signal" for sample #4:
Pb, Nd, Tc, Cd, Sn, Ga, Zn, Cu, Nl, Fe, Cr, and Ti. The following elements were
reported to have "good measurable signal" for sample #6: Pb, Nd, Cd, Sn, Ag, Ga,
Zn, Cu, Ni, Fe, Cr (36).
Neither the organic nor matrix problems presented any difficulty in measure-
ment when these techniques were used. The compatibility of the Chelex 100 resin
concentration with the separation steps added, was confirmed. The addition of a
post column step to eliminate any resin degradation products was necessary, but
proved to be a problem which was quickly solved. The actual analysis of the
elements was not pursued due to the amount of time and expense required to
quantify the ratios from the photoplates. The evaluation of the technique as
applicable for the instrument was achievable without complete quantitation of
the sample due to the nature of the instrumental technique.
EVALUATION AND PRACTICAL USE OF THE CHELEX 100 SAMPLE PREPARATION METHOD
UTILIZING GRAPHITE FURNACE ATOMIC ABSORPTION SPECTROMETRY ANALYSIS
Unlike the SSMS, the graphite furnace atomic absorption spectroraetry
(GFAAS) cannot be qualitatively evaluated for its useful compatibility with the
preconcentration separation technique. Only through the quantitative achieve-
51
-------
ment of analysis could the compatibility of the sample preparation and instru-
mental GFAAS analysis technique be determined.
The GFAAS technique has a distinct advantage in its great sensitivity for
elemental analysis. It also has a disadvantage; it is highly susceptable to
matrix interelement effects.
The evaluation and analytical determinations made using the GFAAS were
done in cooperation with the Analytical Spectrometry Group of the Inorganic
Analytical Research Division of the National Bureau of Standards. The analysis
utilized the Perkin-Elmer Model 603 Graphite Furnace Atomic Absorption Spectro-
meter.
Effects of Direct Injection of Sea Water Into the Electrothermal Device
From the detection limits published in the literature for GFAAS, it could
be assumed that several of the heavy metals in sea water could be determined by
direct injection of the sample into the electrothermal device. However, in
reality this has not been proven to be true unless the samples are taken from
heavily polluted areas. A sample from the Chesapeake Bay sea water samples was
analyzed for Cd, Co, Cu, Mn, Ni, Pb, and Zn by direct injection into the graph-
ite furnace by AAS. Only Pb and Ni produced absorption signals of any analy-
tical value. The other elements could not be detected. This is due in part to
the highly depressing effect of the matrix on the analyte signal, which can vary
by a factor of two to ten, depending upon the analyte. Also, when the sample is
evaporated, a small amount of solution may be trapped in the salt crystal
lattice, which could result in losses due to splattering during the atomization
cycle.
The absorbances obtained for Pb and Ni were very erratic due to the smoke
produced during atomization. Ediger et al. (41) used matrix modification with
ammonium nitrate to assist in the removal of sodium chloride; however, the
method of standard addition was necessary to correct for interferences. In
applying their method of matrix modification to the Chesapeake Bay sample, Cd,
Co, Cu, Fe, Mn and Zn were still not detected (43).
Suppression Study of Ca on Mn and of Reagent Effects on Other Transition Metals
Because Ca is concentrated by Chelex 100 resin, a study was done of Ca
suppression on the Mn signal using GFAAS. Standards were prepared containing
30 yg/mL Mn and varying amounts of Ca were added; 2 mg/mL, 1 mg/mL, 500 yg/mL,
100 yg/mL, 10 yg/mL and 1 yg/mL. Complete suppression of the Mn signal was
observed for 2 mg/mL through 100 yg/mL. At 10 yg/mL approximately 33 percent
of the signal was observed. With 1 yg/mL Ca the normal signal for 30 yg/mL Mn
was observed. If the solution was one percent in HNOs also, the pattern
changed drastically. The 1 and 2 mg/mL Ca additions produced enhancement of
12 percent in the Mn signal, and the Ca concentrations of 500 yg/mL and less gave
normal Mn signals.
These tests showed the extreme effects that just one of the alkaline earth
elements produces if remaining in the sample.
52
-------
When th& preliminary separation products using ammonium nitrate, as the
separating agent were placed in the. graphite furnace, results similar to those
for 500 pg/mL of Ca were observed. The sample was not compatible with the
instrument with the residual Ca and Mg present at these levels.
The reagents used in the separation which, could cause suppression or
enhancement of the signal were checked using each element to Be analyzed (Cd,
Cu, Co, Fe, Mn, Ni, Pb and Zn). The nitric acid caused no significant enhance-
ments or suppressions of the elements of interest. The nitric acid was also
added to the standard solutions when they were prepared to keep the elements in
solution.
The acetate ion (using ammonium acetate) was also evaluated in the initial
experiments and found to cause some small suppressions or enhancement of the
absorbance signal with several of the elements. It tended to enhance Cu, Pb and
Co signal between two and five percent, and to suppress Cd and Fe between 15 and
50 percent, respectively. This problem was easily eliminated early in the
development of the separation technique by the addition of a 5-10 ml water wash
prior to stripping with nitric acid. This volume effectively removed the resid-
ual acetate which does not ion exchange or chelate onto Chelex 100 resin.
In addition to the suppression experiments, the nitric acid and ammonium
acetate were checked for contamination of these reagents with the analytes of
interest. Any level of contamination was below the detection limits of the
GFAAS.
GFAAS Determination of Trace Elements in Sea Water
Evaluation of the preconcentration and separation methods was accomplished
using GFAAS.
Prior to the preparation of the samples the density of the homogeneous sea
water sample was determined. This enabled the samples to be weighed directly
into a Teflon beaker without utilization of volumetric ware. The density was
determined to be 1.0135 g/mL.
Six 100 raL samples of the Chesapeake Bay. water were spiked with varying
amounts of the elements of interest which are given in the Procedure and were
used to check the recovery by GFAAS. Four samples were not spiked and all
samples were processed as described in the Procedure;. To determine the total
blank, four blank columns were carried through the entire process of prepara-
tion, concentration, separation and stripping, using 2.5 M HN03. The 8-10 roL of
2.5 M HN03 effluent collected was analyzed for the trace elements of interest by
GFAAS. Table 15 shows the results of these analyses. The instrumental con-
ditions for drying, charring and atomizing appear in Appendix D. The conditions
were optimized for each analyte to obtain/the maximum.sensitivityand precision
with the minimum of interferences. As a check for chemical interferences, each
sample was tested by the single standard addition method (63) and no chemical
interferences were encountered. Complete recovery of the trace elements of
interest from spiked sea water samples was obtained, as shown in Table 15.
53
-------
TABLE 15. THE CONCENTRATION OF Cd, Co, Cu, Fe, Mn, Ni, Pb, AND Zn FOUND IN TEN
IDENTICAL 100 mL SEA WATER SAMPLES FROM THE CHESAPEAKE BAY AS DETER-
MINED BY GFAAS AFTER.CHELEX 100 CONCENTRATION AND SEPARATION
Sample No.a'b CONCENTRATION, ng/mL
Cd Co Cu
1
2
3
4
5
6
7
8
9
10
11
12
13
14
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
<0.
<0.
<0.
<0.
05 <0
06 <0
03 <0
03 <0
02 <0
05 <0
04 <0
05 <0
05 <0
05 <0
01 <0
01 <0
01 <0
01 <0
.1 1
.1 1
.1 1
.1 1
.1 1
.1 2
.1 2
.1 1
.1 3
.1 2
. 1 <0
.1 <0
.1 <0
.1 <0
.9
.9
.8
.8
.8
.0
.0
.9
.1
.1
.1
.1
.3
-1
Fe
2
1
2
2
1
1
1
1
2
2
0
0
0
0
.1
.9
.5
.4
.4
.5
.6
.9
.1
.8
.1
.3
.1
.1
Mn
2
2
2
2
2
2
2
2
2
2
<0
-------
Cobalt was not detected in the ungpiked 1QQ ml aaraple. To obtain an analy-
tical value for Co in the Chesapeake Ray samples, it would be necessary to
concentrate and separate a 1 .liter sample.
Some difficulty was encountered in the GFAAS determination of Fe. Iron is
known to form carbides in the graphite furnace, which produce erratic results.
Also, a high reagent blank was obtained. This could have been introduced by the
environment at the time of analysis. Although the samples were prepared in a
particulate-free environment, the analysis by GFAAS was carried out in a lab-
oratory module constructed mainly of iron and steel.
The Pb values are close to the detection limits using a 100 mL sample.
With a 0.5-1 liter sample the precision of the Pb analysis could be improved.
The largest spread between spiked and unspiked sea water samples was in the
Zn analysis. This was due to the low concentration of Zn added to the spiked
samples. Since Zn is so sensitive by GFAAS, a 1:10 dilution of the 2.5 M HNOa
effluent had to be made. As a result, the original spike added to the sea water
samples was too low, producing a slightly low recovery for the spiked samples.
After the samples had been analyzed by GFAAS for the elements of interest,
samples #3, #8 and #13 were checked for Na, K, Ca and Mg content to verify the
amounts in a typical analysis which exhibited no chemical interferences. Table
16 shows the results of this analysis.
TABLE 16. THE CONCENTRATION of Na, K, Ca AND Mg DETERMINED IN THE 2.5 M HN03
EFFLUENT OF THE CHESAPEAKE BAY SAMPLES AFTER ANALYSIS OF THE EIGHT
TRANSITION ELEMENTS; SAMPLE 3 WAS SPIKED, SAMPLE 8 WAS UNSPIKED AND
SAMPLE 13 WAS A BLANK
Sample No.
3
8
13
Na
0.68
1.47
0.12
Concentration, yg/mL
K Ca
1.1 0.13
2.7 0.36
<0.01 <0.01
Me
<0.05
0.06
<0.05
Other sea water samples were then analyzed using the Chelex 100 separation
method and GFAAS analysis. These samples were sea water samples from Glacier
Bay, Alaska. They possessed two characteristics distinguishing them from the
Chesapeake Bay samples. First, they were more representative of open ocean
water, with the average salinity being somewhat higher than in the Chesapeake
Bay samples. Secondly, they were collected for their pristine character.
These samples should approach a baseline on the concentration levels in un,con-
taminated oceanic samples. The analysis of these samples is given in Table 17.
(An interesting discovery about frozen unacidified samples was made during the
course of these analyses, but is not relevant to the continuity of the text at
this point and has been placed in Appendix F.) The samples were indeed lower
55
-------
TABLE 17. THE CONCENTRATION OF Cd, Mn, Ni, AND Pb FOUND IN FIFTEEN DIFFERENT
SEA WATER SAMPLES OF 100 mL EACH, TAKEN FROM GLACIER BAY, ALASKA AS
DETERMINED BY GFAAS AFTER CHELEX 100 RESIN COLUMN CONCENTRATION AND
SEPARATION
Sample
NBS 500a
NBS 500b
NBS 501a
NBS 501b
NBS 502
NBS 503
Alaska la
Alaska Tb
Alaska 2a
Alaska 2b
Alaska 3a
Alaska 3b
Alaska 4a
Alaska 4b
Alaska
Q
Sample
.Treatment
Fc, UA
UF, UA
Fc, UA
UF, UA
UF, A
UF, A
F, A
UF, A
F, A
UF, A
F, A
UF, A
F, A
UF, A
F, A, T
Concentration ng/mL
Original pH
7.3
7.2
8.3
8.2
0.4
1.3
3.5
3.6
5.2
6.5
6.6
5.4
5.1
5.4
1.7
Cd
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.11
0.10
0.12
0.13
0.05
0.06
0.11
0.12
0.10
0.14
0.13
0.12
0.07
0.13
0.09
0.08
0.05
0.14
Mn
0.45
0.37
0.59
0.67
. 0.40
0.46
0.71
0.64
1.3
1.4
1.4
1.3
0.05
0.06
0.02
0.02
0.03
0.03
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
0.22
0.28
0.18
0.18
0.68
0.72
Ni
1.2
1.7
0.90
0.70
0.49
0.53
0.66
0.71
0.45
0.59
0.41
0.45
0.24
0.32
0.49
0.40
0.27
0.29
0.36
0.38
1.7
0.8
0.90
0.78
0.36
0.34
0.41
0.45
0.65
0.78
Pb
0.18
0.17
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
4.5
5.8
<0.02
<0.02
0.02
0.05
<0.02
<0.02
0.08
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
<0.02
0.20
0.32
<0.02
<0.02
17
11
All samples stored frozen except T which was stored at ambient temperature.
F "filtered, Fc = filtered after thawing, A = acidified, UF •- unfiltered,
UA » unacidified. Less than values are AAS detection limits on 100 mL test
portions.
56
-------
in concentration for each of the elements tested; Cd, Mn, Ni and Pb in compari-
son with these same elements in the Chesapeake Bay sample. These samples were
also concentrated from a 100 mL sample volume and can be directly compared not
only for concentration but for detection limitation of the GFAAS used in con-
junction with the separation procedure.
The concentration and separation proved to be compatible with the GFAAS
technique. It enabled greater sensitivity through concentration of the analyte
and, by virtue of the separation, the interferences were eliminated, enabling
optimization of the instrument for maximum sensitivity and accuracy.
57
-------
REFERENCES
1. Riley, J. P., and G. Sklrrow. Chemical Oceanography. Vol. 1, Academic
Press, New York, 1975. pp. 393-324, 416-421.
2. Natelson, S., D. Leighton, and C. Galas. Microchem. J., 4:239., 1962.
3. Riley, J. P., and G. Skirrow. Chemical Oceanography. Vol. Ill, Academic
Press, New York, 1975. pp. 269-408.
4. Davey, E. W., and A. E. Soper. Chapter in Analytical Chemical Methods in
Oceanography. T. R. P. Gibb, Jr., ed., Advances in Chemistry Series 147,
American Chemical Society, Washington, B.C., 1975. pp.16-21.
5. Mancy, K. H. Instrumental Analysis of Water Pollution Control. Ann Arbor
Science Publishers, Inc., Michigan, 1971.
6. Strickland, J. D., and T. R. Parsons. A Practical Handbook of Sea Water
Analysis. Fisheries Research Board of Canada, Ottawa, 1972. pp. 111-120.
7. Boyle, E. A., and J. M. Edmond. Chapter in Analytical Chemistry Methods
in Oceanography. T. R. P. Gibb, Jr., ed., Advances in Chemistry Series
147, American Chemical Society, Washington, B.C., 1975. pp. 44-55.
8. Morris, A. W. Anal. Chim. Acta, 42:397, 1968.
9. Stolzberg, R. V. Chapter in Analytical Chemistry Methods in Oceanography
T. R. P. Gibb, Jr., ed., Advances in Chemistry Series 147, American Chemi-
cal Society, Washington, B.C., 19.75. pp. 30-43.
10. Blunt, C. W., W. R. Morgan, and B. E. Leyden. Anal. Chim. Acta, 53:463,
1971.
11. Lai, M. G., and H. A. Goya. U.S. Naval Radiological Befense Laboratory,
Bept. of Commerce, National Technical Information Service, AD-648485, 1966.
12. Callahan, C. M., J. M. Pascual, and M, G. Lai. U. S. Nayal Radiological
Befense Laboratory, Bept. of Commerce, National Technical Information
Service, AB-647661, 1966.
13. Riley, J. P., and B. Taylor. Anal. Chim. Acta, 4Q:479_, 1968.
14. Riley, J. P., and B, Taylor. Anal. Chim. Acta, 41:175, 19.68.
58
-------
15. Riley, J. P., and D. Taylor. Deep Sea Res., 15:629, 1968.
16. Riley, J. P., and D. Taylor. Deep Sea Res., 19:307, 1972.
17. Knauer, G. A., and J. H. Martin. Limnology and Oceanography, 18:597,
1973.
18. Windom, H. L., and R. G. Smith. Deep Sea Res., 19:727, 1972.
19. Davey, E. W., and A. E. Soper. Limnology and Oceanography, 20:1019, 1975.
20. Chester, R., and J. R. Stoner. Mar. Chem., 2:17, 1974.
21. Abdullah, M. I., and I. G. Royle. Anal. Chim. Acta, 58:283, 1972.
22. Florence, T. M., and G. E. Batley. Talanta, 22:201, 1975.
23. Florence, T. M., and G. E. Batley. Talanta, 23:179, 1976.
24. Morris, A. W. Marine Pollution Bulletin, 5:54, 1974.
25. Chul Lee, Nak Bae Kim, Inn Chong Lee, and Koo Soon Chung, Talanta, 24:241,
1977.
26. Bio*Rad Laboratories. Separating Metals Using Chelex 100 Chelating Resin.
Rockville Center, New York, 1976.
27. Bio*Rad Laboratories. Materials for Ion Exchange, Adsorption, Gel Chroma-
tography, Gel Electrophoresis and Hollow Fiber Technique Columns and
Stable Isotopes. Rockville Center, New York, 1972.
28. Leyden, D. E., and A. L. Underwood. J. Phys. Chem., 68:2093, 1964.
29. Luttrell, Jr., G. H., C. More, and C. T. Kenner. Anal. Chem., 43:1370,
1971.
30. Holynska, B. Radiochem. Radioanal. Letters, 17:313, 1974.
31. Sides, J. L., and C. T. Kenner. Anal. Chem., 38:797, 1966.
32. VanWilligen, J. H. J. G., and R. C. Schonebaum. Rec. Trav. Chim. Pays-Bas,
85:35, 1966,
33. Campbell, W., E. Spano, and T. Green. Anal. Chem., 38:987, 1966.
34. Campbell, W. J., T. E. Green, and S. L. Law. Amer. Lab., 28, 1970.
35. Van Grieken, R. E., C. M. Bresselers, and B. M. Vanderbraught. Anal.
Chem., 49:1326, 1977.
36. Paulsen, P. J. Personal Communications. The National Bureau of Standards,
Washington, D.C., 1977.
59
-------
37. Boumans, P. W. J. M., and F. J. De Boer. Spectrochim. Acta, 31B:355, 1976-.-:
38. Larson, G. F., V. A. Fassell, R. H. Scott, and R. W. Kniseley, Anal. Chem.,
47:238, 1975.
39. Larson, G. F., and V. A. Fassel. Anal. Chem., 48:1161, 1976.
40. Forberg, S., and S. Lundgren. Anal. Chem., 23:1202, 1960.
41. Ediger, R. D., F. W. Peterson, and J. D. Kerber. Atomic Absorption News-
letter, 13:61, 1974.
42. Douglas, A. S., and A. Y. Cantillo. Chapter in Analytical Methods in
Oceanography. T. R. P. Gibb, Jr., ed., Advances in Chemistry Series 147,
American Chemical Society, Washington, D.C., 1975.. pp. 56-81.
43 i Rains, T. C. Personal Communications. The National Bureau of Standards,
Washington, B.C., 1978.
44. Keuhner, E. C., R. Alverez, P. J. Paulsen, and T. J. Murphy, Anal. Chem.,
44:2050, 1972.
45. Useller, J. W., NASASP-5074. Office of Technology Utilization. NASA,
Washington, D.C., 1969.
46. Dean, J. A., and T. C. Rains. Flame Emission and Atomic Absorption Spec-
trometry: Components and Techniques, Vol. 2. Marcel Dekker, New York,
1971. .
47, Moody, J. R., and R. M. Lindstrom. Anal. Chem., 49:2264, 1977.
48, Moody, J. R., H. L. Rook, P. J. Paulsen, T. C. Rains, I. L. Barnes, and
M. S. Epstein. NBS Special Publication 464. W. A. Kirchoff, ed., U.S.
Government Printing Office, Washington, D.C., 1977.
49. Maienthal, E. J., and D. A. Becker. NBS Special Technical Note 929. U.S.
Government Printing Office, Washington, D.C., 1976.
50. Harrison, S. H., P. D. LaFleur, and W. Zoller. Accuracy in Trace
Analysis: Sampling, Sample Handling, and Analysis. P. D. LaFleur, ed.,
U.S. Government Printing Office, Washington, D.C., 1976.
51. The Perkin-Elmer Company. Analytical Methods for Atomic Absorption Spec-
troscopy. Atomic Absorption Dept., Norwalk, CT, 1973.
52. The Hewlett-Packard Company. Hewlett-Packard Calculator 9830A Plotter
PAC for Model 30. Loveland, Colorado, 1974.
53. Kayser, A. C. J. Chem. Educ., 36:128, 1959.
54. Lapp, R. E., and H. L. Andrews. Nuclear Radiation Physics. Prentice-Hall,
Inc., 1954. pp. 456-459.
60
-------
55. Currie, L. A. Personal Communications. The National Bureau of Standards,
Washington, D.C., 1978.
56. Meites, L. Handbook of Analytical Chemistry. McGraw-Hill Book Company,
New York, 1963. pp. 1-39.
57. Lange, N. A., and G. M. Forker, eds.. Handbook of Chemistry. Handbook
Publishers, Inc., Sandusky, Ohio, 1952. p. 1145.
58. Dow Chemical Company. Dowex Chelating Resin A-l, Midland, Michigan, 1959.
p. 21.
59. Bleuler, E., and G. J. Goldsmith. Experimental Nucleonics. Holt, Rine-
hart, and Winston, New York, 1952. p. 236.
60. Friedlander, G., and J. W. Kennedy. Introduction to Radiochemistry. John
Wiley and Sons, Inc., New York, 1949. p. 290.
61. Broda, E., and T. Schonfeld. The Technical Application of Radioactivity,
Vol. 1, Pergamon Press, New York, 1966. p. 64.
62. Lundell, G. E. F., H. A. Bright, and J. I. Hoffman. Applied Inorganic
Analysis: With Special Reference to the Analysis of Metals, Minerals, and
Rocks. John Wiley and Sons, Inc., New York, 1953. pp. 133-134.
63. Dean, J. A., and T. C. Rains, ed. Flame Emission and Atomic Absorption
Spectrometry: Elements and Matrices, Vol. 3. Marcel Dekker, New York,
1975.
64. Sidgwick, N. V. The Chemical Elements and Their Componds. Oxford Univer-
sity Press, London, England, 1950. pp. 252-254.
61
-------
APPENDIX A
THE ABUNDANCES OF THE CHEMICAL
ELEMENTS IN SEA WATER (1)
ELEMENT
H
He
Li
Be
B
C
N
0
F
Ne
Na
Mg
Al
Si
P
S
Cl
Ar
K
Ca
Sc
Ti
V
Cr
TOTAL
(MOLAR)
55
1.7 x 10"9
2.6 x 10"5
6.3 x 10"10
4.1 x 10~4
2.3 x 10"3
1.07 x 10~2
55
5
6.8 x 10D
7 x 10~9
4.68 x 10"1
2
5.32 x 10
7.4 x 108
s
7.1 x 10
2 x 106
2.82 x 10
i
5.46 x icr
1.1 x 107
1.02 x 102
1.02 x 102
11
1.3 x 10
2 x 108
5 x 108
5.7 x 109
CONCENTRATION
(yg
1.1
6.8
180
5.6
4440
2.8
1.5
8.8
1.3
1.2
10.77
12.9
2
2
60
9.05
18.8
4.3
3.8
4.12
6
1
2.5
0.3
L-1)
x 108
x 10"3
x 10"3
x 104
xlO5
x 108
3
x 10J
x 10"1
x 106
5
x 10
g
x 10b
x 105
f.
x 10b
x 105
x 10
A
x 10
62
-------
APPENDIX A (Continued)
ELEMENT
Mn
Fe
Co
Ni
Cu
Zn
Ga
Ge
As
Se
Br
Kr
Rb
Sr
Y
Zr
Nb
Mo
Tc
Ru
Rh
Pd
Ag
Cd
In
Sn
Sb
(MOLAR)
3.
3.
8
2.
8
7.
4.
6.
5
2.
8.
2.
1.
9.
1.
3.
1
1
4
1
0.
8.
2
6
5
8
6
3
9
5
4
4
4
1
5
3
8
4
x
x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
TOTAL CONCENTRATION
-------
APPENDIX A (Continued)
ELEMENT
Te
I
Xe
Cs
Ba
La
Ce
Pr
Nd
Pm
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
Hf
Ta
W
Re
Os
Ir
Pt
Au
TOTAL CONCENTRATION
(MOLAR) (yg L"1)
5
3.8
3
1.5
2
1
4
1.9
3
9
4
9
6
1
4
8
5
9
4
1
5
2
2
_7
x 10 '
x ID"10
x 10~9
_7
x 10 '
x lO'11
x 10-10
x 10-12
x ID"11
x ID'12
x 10-13
x 10~12
x 10-13
x ID'12
x 10-12
x ID'12
in-13
x 10
x ID'12
x 10-13
x 10
x ID"11
x ID'10
x 10"11
x ID'11
60
5
0.4
2
3
1
6
3
0.5
0.1
7
1
9
2
8
2
8
2
7
2
0.1
4
4
x 10~2
x 10~3
x 10"3
x 10
x 10~3
x 10~4
x 10~4
x 10
x 10~4
x 10~4
x 10~4
x 10~4
x 10~4
x 10~4
x 10~4
x 10~3
x 10"3
x 10~3
x 10~3
64
-------
APPENDIX A (Continued)
ELEMENT
Hg
Tl
Pb
Bi
Po
At
Rn
Fr
Ra
Ac
Th
Pa
U
TOTAL CONCENTRATION
(MOLAR) (pg L"1)
1.5 x
5 x
2 x
1 x
2.7 x
3 x
4 x
2 x
1.4 x
io--10
ID'11
io-10
io-10
io-21
ID'16
io-11
ID'16
io-8
3 x
1 x
3 x
2 x
6 x
7 x
1 x
5 x
3.2
io-2
io-2
io-2
io-2
io-13
io-8
io-2
io-8
65
-------
APPENDIX B
THE SELECTIVITY COEFFICIENTS FOR CHELEX 100
RESIN BASED ON Zn+2(26)
IONIC SPECIFIC SELECTIVITY COEFFICIENT
Hg+2 1060
Cu*"2 126
+2
UO 5.70
Ni+2 4.40
+2
Pb 3.88
Zn+2 1.000
Co+2 0.615
Cd+2 0.390
Fe+2 0.130
Mn+2 0.024
Ba+2 0.016
Ca+2 0.013
+2
Sr 0.013
Mg+2 0.009
Na+1 0.000,000,1
66
-------
APPENDIX C
TYPICAL SUB-BOILING DISTILLED WATER
IMPURITY CONCENTRATIONS PPB BY WT (ng/g) (44)
ELEMENT
Pb
Tl
Ba
Te
Sn
Cd
Ag
Sr
Zn
Cu
Ni
Fe
Cr
Ca
K
Mg
Na
WATER FROM SUB-BOILING STILL
0.008
0.01
0.01
0.004
0.02
0.005
0.002
0.002
0.04
0.01
0.02
0.05
0.02
0.08
0.09
0.09
0.06
Z 0.5 ppb
67
-------
APPENDIX C (Continued)
TYPICAL SUB-BOILING DISTILLED NITRIC ACID
IMPURITY CONCENTRATIONS PPB BY WT (ng/g) (44)
ELEMENT
Pb
Tl
Ba
Te
Sn
In
Cd
Ag
Sr
Se
Zn
Cu
Ni
Fe
Cr
Ca
K
Mg
Na
ACID FROM SUB-BOILING STILL
0.02
—
0.01 .
0.01
0.01
0.01
0.01
0.1
0.01
0.09
0.04
0.04
0.05
0.3
0.05
0.2
0.2
0.1
1
E 2.3 ppb
68
-------
APPENDIX D
THE EXPERIMENTAL CONDITIONS FOR ANALYSIS OF THE 2.5 M HN03 FRACTION CONTAINING THE
TRACE METALS USING A PERKIN-ELMER MODEL 603 ATOMIC ABSORPTION SPECTROPHOTOMETER
EQUIPPED WITH A HEATED GRAPHITE ATOMIZER, HGA-2100
VO
P & E 603
Element
Cd
Co
Cu
Fe
Mn
Ni
Pb
Zn
Wavelength
nm
228.8
240.7
324.7
248.3
279.5
232.0
283.3
213.9
SBW
nm
0.7
0.2
0.7
0.2
0.7
0.7
0,7
0.7
Scale
Expansion
1
2
1
2
2
5
3
0.5
Drying
T-sec
100-30
100-30
100-40
100-30
100-30
100-30
100-40
100-30
HGA-2100
Charring
T-sec
200-20
500-30
700-30
600-30
300-30
1000-30
400-30
500-20
Atomization
T-sec
2100-7
2700-7
2500-6
2700-7
2700-7
2700-6
2200-7
2000-7
Gas
Ara
Ara
Ara
Ara
Ara
Ara
Arb
Arb
Interrupt mode.
Normal mode.
-------
APPENDIX E
A LIST OF RESIN VOLUME FOR CHELEX 100
IN DIFFERENT IONIC FORMS BASED ON Na AS 1.00 (58)
Resin volume In different ionic forms
Na+ 1.00
H+ 0.45
Te 0.45
Zn"*"*" 0.55
Ca4"*" 0.53
K+ 1.06
Li+ 0.98
Ag+ 0.70
70
-------
APPENDIX F
A CRYSTAL RESIDUE OBSERVED IN FROZEN
NEUTRAL SEA WATER SAMPLES
While analyzing the NBS and Alaskan samples, a sample inhomogeneity was
observed. Some samples which had been unacidified and frozen had white crystals
at the bottom of the container when thawed. Shaking and heating did not aid in
redissolving these crystals and they did not appear in the moderate to strongly
acidified samples.
A sample of these crystals was extracted and analyzed by spark source mass
spectrometry. They were found to contain Ca almost exclusively. The crystals
produced a violent evolution of gas upon contact with (1+1) HN03.
Several samples of unacidified sea water were evaporated to dryness. Dis-
tilled water of equal volume was heated to reduce the C0£ gas dissolved in the
water. The evaporated salts were mixed with the distilled water and a crystal
residue similar to the NBS and Alaskan samples resulted.
Using the data obtained, the crystals were hypothesized to be CaC03 com-
pounds in various forms of hydration. The formation of these crystals was
suspected to originate in both the frozen and the evaporated sea water by this
mechanism:
Ca(HC03) * CaC03* + H20 + C02t.
The calcium carbonate is much less soluble in water in the absence of carbon
dioxide (64). Although acid was found able to put the crystals into solution,
an alternative method was tested where the previous equilibrium was reversed:
CaC03 + H20 + C02 >• Ca(HC03)2.
Two methods of redissolving these crystals in their original solutions
were successful. First, a 2 cm* piece of dry ice was sufficient to place the
crystals back into solution in a 1 liter sea water sample from which they had
been frozen out. The time required was thirty to sixty minutes. Secondly, the
sample was placed in an ice bath and a positive pressure of C02 filtered gas
was applied for twenty-four to thirty-six hours, which totally redissolved the
crystals in a 1 liter sample.
The only sample analyzed which contained any crystals was Alaska 3a. The
sample was decanted into the three 100 mL samples necessary for analysis,
leaving the crystals behind. Sample Alaska Ib also exhibited the crystals when
thawed, but they dissolved slowly upon standing, perhaps due to the higher acid
content.
The technique developed for redissolving the crystals in neutral sea water
samples was not used on samples Alaska 3a and Ib. Further study was desired on
the possibility of contamination from dry ice, and adsorption by the container
71
-------
walls of warm, mobile, unacidifled samples was a possibility (48). The remain-
der of the samples exhibiting the "crystal freeze out" phenomenon were duplicate
samples and analysis was unnecessary.
72
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Uuantitatiye Ultratrace Transition Metal Analysis pf
high Salinity Waters Utilizing Chelating Resin Separatioi
Application to Energy-related Environmental Samples
5. REPORT DATE
June 1979
* PERFORMING ORGANIZATION CODE
. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO,
Howard M. Kingston
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Center tor Analytical Chemistry
National bureau of Standards
Washington, DC 20234
10. PROGRAM ELEMENT NO.
EHA-553
11. CONTRACT/GRANT NO.
1NE625C
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Research & Development
Office of Energy, Minerals & Industry
Washington, DC 20460
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/17
15. SUPPLEMENTARY NOTES
This project is part of the EPA-planned and coordinated Federal Interagency
Energy/Environment K&U Program.
18. ABSTRACT
In order to accurately evaluate the impact of energy related activities, such as
offshore drilling on the sea water, it is necessary to measure trace element concen-
trations in the presence of considerably higher levels of alkali and alkaline earth
elements.
This report describes a technique which was developed for the elimination of the
alkali and alkaline earth elements Ha, Kv Ca and Mg from the trace transition elements
in sea water samples. This was accomplished by passing an ammonium acetate solution
through a column of Chelex 10U resin after a sea water sample had been chelated. The
alkali and alkaline earth elements were eluted from the column by ammonium acetate and
the trace transition elements were then collected using nitric acid.
Trie quantitative concentration, separation and removal of selected transition
elements was tested using radiochemical tracers. The study revealed >99.9 percent
recovery -of Cd, Cu, Mn, Ni and Zn, using a 100 raL sample, and >99 percent with a 1 litei
volume. Cobalt and Pb exhibited >99 percent and >98 percent recoveries, respectively
from a 100 mL sample, and >97 percent from a 1 liter volume. Iron was found to be
recovered only apprpximately 92 percent in either volume.
The concentration and separation technique was applied to Chesapeake Bay and
Alaskan sea water samples. The samples were introduced into a graphite furnace and
were analyzed by grapnite furnace atomic absorption spectrometry. Analysis of concen-
trations below ng/mL for the trace elements mentioned was possible using this
combination of sample preparation and instrumental analysis and no interelement inter-
ferences occurred. The combination of this sample preparation technique and the
graphite furnace atomic absorption sensitivity enables extreme detection limits to be
achieved for the elements mentioned, and gave measurements that were reliable and
reproducible.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
-heiating Resin
sea Water
Salt water
Separation
Trace Metals
Trace Element Analysis
£ leetrothermal
Atoraization
Atomic Absorption
Spectrometry
Neutron Activation
Analysis
Spark Source Mass
Spectrometry
Processes & Effects:
Ecological Effects
Fuel;
Hydroelectric
18. DISTRIBUTION STATEMENT
Release to public
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
$6.00
Form 22200
.S. GOVERNMENT PRINTING OFFICEi .1979-311-046/295
------- |