EPA-600/1-78-015
February 1978
Environmental Health Effects Research Series
EVALUATION AND RESEARCH OF METHODOLOGY
FOR THE NATIONAL ENVIRONMENTAL
SPECIMEN BANK
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL HEALTH EFFECTS RE-
SEARCH series. This series describes projects and studies relating to the toler-
ances of man for unhealthful substances or conditions. This work is generally
assessed from a medical viewpoint, including physiological or psychological
studies. In addition to toxicology and other medical specialities, study areas in-
clude biomedical instrumentation and health research techniques utilizing ani-
mals — but always with intended application to human health measures.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/1-78-015
February 1978
EVALUATION AND RESEARCH OF METHODOLOGY FOR THE NATIONAL
ENVIRONMENTAL SPECIMEN BANK
by
T.E. Gills, H.L. Rook and P.O. LaFleur, Editors
Analytical Chemistry Division
Institute for Materials Research
National Bureau of Standards
Washington, D.C. 20234
Contract No. EPA-IAG-D5-0568
-------
DISCLAIMER
This report has been reviewed by the Health Effects Research
Laboratory, U.S. Environmental Protection Agency, and approved
for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the U.S. Environmental
Protection Agency, nor does mention of trade names or commercial
products consitute endorsement or recommendation for use.
ii
-------
FOREWORD
The many benefits of our modern, developing, industrial society are
accompanied by certain hazards. Careful assessment of the relative risk of
existing and new man-made environmental hazards is necessary for the estab-
lishment of sound regulatory policy. These regulations serve to enhance
the quality of our environment in order to promote the public health and
welfare and the productive capacity of our Nation's population.
The Health Effects Research Laboratory, Research Triangle Park,
conducts a coordinated environmental health research program in toxicology,
epidemiology, and clinical studies using human volunteer subjects. These
studies address problems in air pollution, non-ionizing radiation, environ-
mental carcinogenesis and the toxicology of pesticides as well as other
chemical pollutants. The Laboratory develops and revises air quality
criteria documents on pollutants for which national ambient air quality
standards exist or are proposed, provides the data for registration of new
pesticides or proposed suspension of those already in use, conducts research
on hazardous and toxic materials, and is preparing the health basis for
non-ionizing radiation standards. Direct support to the regulatory function
of the Agency is provided in the form of expert testimony and preparation
of affidavits as well as expert advice to the Administrator to assure the
adequacy of health care and surveillance of persons having suffered imminent
and substantial endangerment of their health.
This report documents one aspect of an International effort, supported
by EPA, to provide a comprehensive environmental monitoring program to
assess the relative risk of environmental hazard to the health and well-being
of our population and to aid in the improvement of our environmental quality.
This program, the National Environmental Specimen Bank, will serve as an
environmental warning system by providing real time chemical analysis of
collected specimens. In addition, this system would permit the use of
tomorrow's more sensitive and more specific methods of chemical analysis on
stored samples. The advantages of such a program will permit us to assess
the effectiveness of our present environmental control techniques by
monitoring pollutant trends, as well as establishing environmental baseline
levels of new pollutants or pollutants of current concern not previously
investigated.
John H. Knelson, M.D.
Director,
Health Effects Research Laboratory
iii
-------
ABSTRACT
This report is a summary of analytical methodologies developed or
adopted by NBS to insure proper procedures for sampling, storage and analysis
of biological and environmental type matrices. Protocols established in
these procedures will insure that the measurement of an analyte can be
made within known and/or required uncertainty levels and that samples
can be stored for retrospective analysis.
The contributions in this report are divided in seven sections which
includes analytical methods for trace element analysis and techniques for
sampling and storage of specific matrices. This report is submitted in
partial fulfillment of EPA Interagency Agreement IAG-D5-0568 by the National
Bureau of Standards. This is the final report for the 1976 contract year.
iv
-------
CONTENTS
Page
Disclaimer ii
Forward iii
Abstract iv
List of Figures vi
List of Tables vii & viii
Acknowledgment ix
Section 1 - The Determination of Iodine in Biological and
Environmental Standard Reference Materials 1
Section 2 - Determination of Chromium in Biological Matrices
by Neutron Activation Avalysis: Application to
Standard Reference Materials 12
Section 3 - Method for the Determination of Platinum in
Biomaterials 21
Section 4 - A Method for the Determination of Arsenic,
Antimony, and Copper in Biomaterials by Neutron
Activation Analysis 26
Section 5 - The Selection and Cleaning of Plastic Containers
for Storage of Trace Element Samples 32
Section 6 - Evaluation by Activation Analysis of Elemental
Retentions in Biological Samples After Low
Temperature Ashing 48
Section 7 - Stability of Elemental Components in Multi-Trace
Element Water Standard 58
-------
LIST OF FIGURES
Section 1:
Figure 1. Biological combustion apparatus 4
Section 6:
Figure 1. Block Diagram of LTA Apparatus 51
Figure 2. Elements retained during LTA 54
vi
-------
LIST OF TABLES
Section 1:
Table 1. Percent Recovery for Iodine/Chlorine Separated
Using HMD [[[ 8
Table 2. Concentration of Iodine in Standard Reference
Materials [[[ 9
Table 3 . Comparative Results on Iodine in Standard Reference
Materials [[[ 9
Section 2:
Table 1. Concentration of Chromium in Orchard Leaves, SRM 1571,
Cr/gram) ................................................ 18
Table 2. Concentration of Chromium in Brewers Yeast, SRM 1569
(yg Cr/gram) ................................................ 19
Table 3. Concentration of Chromium in Brewers Yeast, SRM 1569,
Using INAA (yg Cr/gram) ..................................... 20
Section 3:
Table 1. Platinum Doped Fe20s with Fused Silica ................. 24
Section 4:
Table 1. Method of Analysis: Ion Exchangers with NAA ........... 30
Table 2. A Comparison of NAA Results Using the Described
Separation Procedure to Other Independent Methods ........... 31
Section 5 :
-------
Section 5, Continued: Page
Table 3. Impurities Leached from Plastic Containers in One
Week by (1+1) HC1 (ng/cm2) Determined by IDMS ............... 42
Table 4. Impurities Leached from Plastic Containers in One
Week by (1+1) HN03 (ng/cm2) Determined by IDMS .............. 43
Table 5. Concentrations of Trace Elements in Plastics Determined
by NAA (ng/g) ............................................... ^
Table 6 . Removal of Sodium from Plastics ....................... 45
Table 7. Sequential Leaching of Trace Elements from Plastics
(ng/cm2) .................................. ................. 46
Table 8. Suggested Method for Cleaning Plastic Containers ...... 47
Section 6:
Table 1. Typical Weight and Carbon Losses for Ashing Periods of
15 Hours and RF Power of 70 Watts ......................... 53
Table 2. Percent of Element Retained (Experimental) ........... 55
Section 7:
\
Table 1. Trace element loss study using radioisotopes .......... 62
Table 2. Trace Element Stability Analytical Using Independent
Analyses .... ....................... ...... .................. 63
viii
-------
ACKNOWLEDGMENTS
The authors of this report wish to acknowledge their appreciation
to the Environmental Protection Agency for their funding. We especially
want to thank Dr. George Goldstein, Project Officer of the Clinical
Pathology Branch for his consultations and program guidance. Finally, we
extend sincere thanks to the operating staff of the NBS Research Reactor
and the NBS Electron Linac Facility for the fine service provided.
ix
-------
SECTION 1
THE DETERMINATION OF IODINE IN BIOLOGICAL AND ENVIRONMENTAL
STANDARD REFERENCE MATERIALS
H.L. Rook
Analytical Chemistry Division
National Bureau of Standards
Washington, D.C. 20234, U.S.A.
-------
INTRODUCTION
The accurate determination of iodine in biological samples has been
of interest in many medical and nutritional studies. . The importance of
iodine to thyroid function is a well understood biochemical relationship;
however, the role of iodine in other biological functions is not well
understood. One of the main obstacles to this understanding has been the
inability to measure accurately iodine at the concentration levels needed.
Exclusive of thyroid tissue, iodine levels in human soft tissue and serum
are in the range of 0.0001 to 0.1 pg iodine per gram wet tissue.
Few analytical techniques have the demonstrated sensitivity or accuracy
for the determination of iodine at this level. Thermal neutron activation
analysis is one technique that does have the required inherent sensitivity.
The combination of a large thermal neutron cross section and a relatively
short half life for the product iodine-128 would normally yield more than
adequate sensitivity. However, in biological systems, one complicating
fact has confounded analytical chemists for years. The very high levels
of chlorine, and to a lesser extent bromine, coupled with the similar
chemistry of these elements has made the practical limit of detection much
higher than the theoretical limit.
The determination of iodine in biological material by neutron activa-
tion analysis has received considerable attention. It was observed that
nondestructive techniques were inadequate except in samples with unusually
high levels of iodine such as thyroid tissue or kelp (1). During this
period, numerous publications appeared detailing radiochemical separations
for iodine. Most, however, were of a similar procedural nature. Samples
were rapidly dissolved, iodine carrier added, and the iodine separated by
either liquid-liquid extraction or by liquid-ion exchange (2,3). Methods
have been reported where the iodine was distilled directly from the
matrix (4,5) but even in these procedures, a subsequent separation was
performed to obtain the necessary decontamination from chlorine.
The best indication of the state of iodine analysis in biological
matrices is the limited number of publications containing data on iodine
analyses. A review of published iodine values on Bowen's kale indicated a
limited data set with a variability so large as to invalidate any con-
sensus mean value 06). To the author's knowledge, no values have been
reported for iodine on either NBS SRM's 1571, Orchard Leaves, or 1577,
Bovine Liver. A recent publication on environmental matrices reported
-------
iodine results for NBS SRM 1632, coal. These results were achieved using
instrumental neutron activation in a matrix where chlorine was not a severe
problem and were at a significantly higher iodine concentration than nor-
mally found in biological samples (7). In spite of the favorable analytical
conditions, these results appear to be high when compared to those obtained
in this work.
This paper presents a new method for the determination of iodine in
biological matrices using a unique adaption of the general gas phase
separation of volatile elements as developed at NBS (8,9,10). The procedure
involves controlled combustion of the sample followed by direct gas phase
separation of the chlorine and bromine from iodine using hydrated manganese
dioxide. This procedure has four major advantages over conventional
chlorine/iodine separations.
The procedure is rapid, requiring less than one half-life of
iodine-128 for complete separation.
The procedure is simple, as it involves no extractions or other
complex chemical manipulations, thereby minimizing technique
related errors.
The separation is quantitative, thereby eliminating the need for
chemical yield determinations and the associated additional errors.
The procedure gives better than a 10 to 1 separation of chlorine
from iodine. This is far superior to liquidextraction procedures
and essentially eliminates the Compton baseline from chlorine-38.
The accuracy of the described analytical procedure has been verified by
independent photon activation analysis on selected samples.
-------
APPARATUS AND REAGENTS
COMBUSTION TUBE
The combustion tube consists of a Vycor tube, 50 cm long with a 19/38
standard taper joint at the downstream end. The inner portion of the stan-
dard taper joint contains a 5 cm long column of hydrated manganese dioxide
inorganic ion exchange material. The downstream portion of this column has
an outer 5/12 standard taper joint welded to it. The inner portion of the
5/12 joint contains silver coated glass wool which is used to trap all re-
maining halides. The entire combustion tube and gas phase separation assem-
bly is shown in figure 1.
QUARTZ
WOOL
02 INLET
SILVER
HMD WOOL
Figure 1: Combustion Apparatus
IRRADIATION SOURCE
All samples were irradiated in the pneumatic transfer tube RT-3 of
the NBS research reactor. The thermal neutron flux was approximately
6 x 1013 n-cm 2 sec * and the Cu/Cd ratio was
-------
COUNTING EQUIPMENT
All samples were counted on a 25 cm3 volume Ge(Li) detector with a
2.1 keV resolution for the 1332 keV photo peak of 50Co. The detector was
coupled to a Hewlett Packard Computer-Analyzer.* Quantitation of the
individual photopeaks was performed directly using the total peak area
method.
GAS PHASE CHLORINE TRAP
The gas phase chlorine trap consisted of a column of Carlo Erha
manufactured hydrated manganese dioxide (HMD). The trap had a minimum
dimension of 5 cm length and 0.8 cm diameter. The ion exchange material
was held in place by two plugs of quartz wool. Columns were prepared by
simply pouring the material, as received, into the column with very gentle
side tapping for packing. The column was capable of passing a gas flow of
100 cm3 per minute with negligible build up of pressure.
IODINE TRAP
The glass wool was coated with a heavy silver deposit. Any convention-
al silvering process is adequate. After silvering, the material was washed
with three successive portions of distilled water and oven dried. The
silver wool was packed tightly into a column 4 cm long and 0.5 cm in dia-
meter. The portion of the trap utilized for iodine collection is clearly
visible due to the formation of yellow silver iodide. Trap lengths should
be adjusted to accommodate the quantity of iodine carrier used.
*
Certain commercial equipment, instruments, or materials are identified in
this paper in order to adequately specify the experimental procedure. In no
case does such identification imply recommendation or endorsement by the
National Bureau of Standards, nor does it imply that the material or equip-
ment identified is necessarily the best available for the purpose.
-------
EXPERIMENTAL
PROCEDURES
Samples of biological materials were weighed, as received, into pre-
cleaned quartz tubes. Sample weights of 200-500 mg were used for analysis.
Replicate samples of each material were vacuum dried for 24 hours to obtain
dry weight values. Standard solutions of 250 and 500 ml were also weighed
and sealed in quartz tubes. Individual samples were irradiated for 15
seconds, separated, and counted for two successive 10 minute counts. Stan-
dards were processed in the same manner as the samples and were run a
minimum of one per every five samples. All analysis sets were begun and
finished with a standard sample.
Following irradiation, the quartz vials were washed in 6 M HC1, rinsed
with water and wiped dry. The samples were then cooled in liquid nitrogen
and opened. The bulk material was transferred to a combustion boat and
100 yl of iodine carrier solution containing 2 mg of KI was added. The
combustion boat plus quartz irradiation vial were inserted into the combus-
tion tube and an oxygen flow of ^100 cc/min was passed over the sample.
The sample was ignited by heating the combustion tube directly over the
sample with an oxygen-gas torch. Most biological samples ignited readily
as the quartz tube reached a dull white heat. Once ignited, the samples
were allowed to burn freely with the intensity of the burn controlled by
the oxygen flow. During the combustion process, the entire assembly was
heated moderately with the gas torch to prevent the condensation of water
vapor. When the self-maintained combustion ceased, the sample was heated
to a dull red heat with the gas torch to insure complete combustion. Then,
a resistance furnace was placed over the entire combustion tube and the
sample heated to 1000 °C to drive all of the iodine through the trap and
onto the silver wool collector. The collector was also heated gently to
drive off all residual condensed water. Care must be used in this drying
step as silver iodide is volatized at slightly over 500 °C.
Following distillation, the silver wool collector was removed, sealed
in a polyethylene bag and placed 5.0 cm from the detector. Two successive
ten minute counts were taken to check the half life of the iodine-128. The
441 keV photopeak of iodine-128 was used for analysis. When sufficient
iodine-128 activity was ovtained, the 528 keV photopeak was also quantified
as a validation. Standard solutions were processed and counted with exactly
the same procedure as the samples. Quantitative results were obtained by
the comparator method.
6
-------
RESULTS AND DISCUSSION
Gas phase separations to isolate selectively radiochemical species
have been used successfully in activation analysis. This type of separation
has been applied most successfully for the determination of mercury in com-
plex matrices (8). More recent work applied the technique to the separation
of selenium and to a group separation of Hg, Se, Zn, As, and Ce (9,10). All
of the methods had one underlying principle in common, however, that the
isolation of the total volatile component of the sample was sufficient
separation for elemental identification and quantitation. One method for
the determination of mercury developed by Kosta, et al., introduced a subtle
new concept (11). In this method, a silver wool adsorption trap, heated to
300 °C, was placed in the gas stream to remove halides while allowing
mercury to pass to a final collector. This process of selective gas phase
adsorption offers one of the most rapid, simple procedures for radiochemical
separations.
In 1974, work at NBS was begun on the analysis of iodine by gas phase
distillation. Iodine had previously been observed to distill quantitatively
in an oxygen atmosphere. This procedure was marginally attractive for longer
half life iodine isotopes such as iodine-131 and iodine-133, but in almost
all cases, the predominance of chlorine-38, which codistilled with the iodine,
precluded the direct analysis of total iodine. Attempts to separate the
halides by conventional wet chemical procedures, after distillation, did not
prove satisfactory. Separation efficiencies were generally not high enough
to remove sufficient chlorine in one step and the additional time required
severely degraded the sensitivity of the technique.
In an attempt to accomplish the entire separation in the gas phase,
various inorganic ion exchange compounds were studied as selective adsorp-
tion traps. These compounds were attractive in their ability to withstand
high temperature gas streams. To facilitate the early work, the longer
half life iodine species were studied. It was found in this study that
HMD, under carefully controlled temperature conditions, exhibited selective
adsorptive properties for bromine while quantitatively passing iodine. The
culmination of this work was a procedure for the determination of tellurium
and uranium using their iodine daughters (12).
Following this work, the use of HMD to separate chlorine from iodine
was studied. Initial separations using iodine-126 and chlorine-38 traces
indicated that HMD was a superior material for the selective adsorption of
-------
chlorine. To verify this, one hundred microliters of standard solution of
each isotope were pipetted onto 250 rag of nonradioactive orchard leaves.
The orchard leaf was burned and the gaseous products were distilled through
a warm HMD trap and collected quantitatively on the silver collector. No
evidence of chlorine-38 activity was observed on the silver collector.
Separation efficiencies of greater than 1014 were estimated and a series of
these experiments indicated respectable collection efficiencies of greater
than 95 percent for iodine (Table 1).
TABLE 1. PERCENT RECOVERY FOR IODINE/CHLORINE SEPARATED USING HMD
Run
% Recovered 126I Separation Efficiency for 38C1
1
2
3
4
5
Mean
97.4
99.0
96.3
98.8
98.2
97.8
> 103
> 103
> 5 x
10*
2 x
103
10*
tc _' «• • f °A 38C1 on HMD
Separation efficiency for
38
C1 on Silver
After this initial work, samples of orchard leaves and bovine liver
were irradiated for 15 seconds. Aliquots of the standard iodine-126
solution were added, and the iodine was separated. Again the iodine-126
was recovered quantitatively, and iodine-128, induced from the iodine in
the sample, was observed. In the case of bovine liver, chlorine decon-
tamination of 105 was obtained.
With the separation satisfactory in every respect, a detailed pro-
cedure for the analysis of total iodine was devised. This procedure is
given in the Experimental section. Replicate samples of NBS SRM's 1571,
1577, and 1632, orchard leaves, bovine liver, and coal, were analyzed
using this procedure. In all cases, the iodine-128 was easily observed
after the separation. Freshly prepared standard solutions, made from high
purity potassium iodide and cesium iodide were used as primary standards.
These solutions were carried through the same procedure as the samples.
Results of these analyses are given in Table 2.
-------
TABLE 2. CONCENTRATION OF IODINE IN STANDARD REFERENCE MATERIALS
Iodine
Orchard Leaves
Sample SRM 1571
1 0.176
2 0.165
3 0.171
4 0.180
5 0.160
6 0.166
7 0.153
Mean 0.167
fs 0.01
s is the standard deviation
Concentration in yg/g
Bovine Liver
SRM 1577
0.187
0.176
0.208
0.199
0.173
0.181
0.183
0.180
0.012
of the mean.
Coal
SRM 1632
2.52
3.01
2.54
2.60
2.71
2.68
0.20
As a final verification of the method, samples of these materials
were independently determined by nondestructive photon activation analysis
(13). A summary of these results is given in Table 3. The agreement among
results is relatively good in all cases, but the errors were large in the
cases of orchard leaves and bovine liver as the photon method was near its
limit of sensitivity.
TABLE 3. COMPARATIVE RESULTS ON IODINE IN STANDARD REFERENCE MATERIALS
Material This Work Photon Activation Analyses1
SRM 1571 0.17+0.01 .10 ± 0.05
Orchard Leaves
SRM 1632 2.7 ±0.2 3.3 ± 0.4
Coal
1
Results obtained in private communication from Dr. John Hislop, Environ-
mental and Medical Sciences Division, Harwell, England.
-------
In conclusion, the gas phase separation of iodine using combustion and
selective adsorption appears to be a simple, attractive method for the
analysis of total iodine in complex matrices. Separations can be completed
in 10-15 minutes with a total analysis time of less than 60 minutes. With
this procedure, accurate iodine analyses can be performed in samples which
were heretofore difficult or impossible to analyze.
10
-------
REFERENCES
1. Brues, A.M., and Robertson, O.H. AECD-2009, 1947.
2. Bowen, H.J.M. Biochem J., 73:381, 1959.
3. Greendale, A.E., Love, D.L., and Delucchi, A.A. Trans. Am. Nucl.
Soc. 7:334, 1964.
4. Meinke, W.W. Modern Trends in Activation Analysis. In: Proceedings
International Conference, College Station, Texas, 1961. pp.36-40.
5. Becker, D.A., and Smith, G.W. Modern Trends in Activation Analysis.
In: Proceedings, International Conference, College Station, Texas,
1965, pp. 230-235.
6. Lenihan, J.M.A., and Thompson, S.J., eds. Advances in Activation
Analysis, Vol. 1. Academic Press, 1969. p. 108.
7. Nadkarni, R.A. Radioanal. Letters 21:13,161, 1975.
8. Rook, H.L., Gills, T.E., and LaFleur, P.D. Anal. Chem. 44:1114, 1972.
9. Rook, H.L. Anal. Chem. 44:1276, 1972.
10. Orvini, E., Gills, T.E.,'and LaFleur, P.D. Anal. Chem. 46:1294, 1974.
11. Kosta, L., and Byrne, A.R. Talenta 16:1297, 1969.
12. Gladney, E.S., and Rook, H.L. Anal. Chem. 47:1554, 1975.
13. Hislop, J.S., private communication. Environmental and Medical
Sciences Div., AERE, Harwell, United Kingdom.
11
-------
SECTION 2
DETERMINATION OF CHROMIUM IN BIOLOGICAL MATRICES BY NEUTRON
ACTIVATION ANALYSIS: APPLICATION TO STANDARD REFERENCE MATERIALS
L.T. McClendon
Analytical Chemistry Division
National Bureau of Standards
Washington, D.C. 20234
12
-------
INTRODUCTION
Chromium has been recognized as an essential trace element (1) in
human nutrition for several years. At high concentrations or in different
chemical states it can also cause deleterious or toxic effects. There is a
need for reliable methods of analysis for chromium at trace levels in a
variety of matrices, especially biological matrices. A laboratory inter-
comparison on environmental materials (2) pointed up the sad state of
chromium determinations as performed by most of the analytical community.
Also, the author has observed a wide range of results reported for chromium
in the National Bureau of Standards (NBS) Bovine Liver Standard Reference
Material (SRM 1577). As a result of these observations and communication
with several analytical laboratories involved in the determination of
chromium in biological matrices, persons at NBS became convinced that more
biological materials needed to be analyzed and certified for chromium
content.
Neutron activation analysis, both instrumental (INAA) and destructive
(DNAA), has been used to determine the chromium concentration in a variety
of matrices. The results obtained for chromium in two biological Standard
Reference Materials (SRM1s)--Orchard Leaves, and Brewers Yeast--are des-
cribed in this paper. The chromium content has been certified by NBS in
these two SRM's.
EXPERIMENTAL
Samples Analyzed
The biological materials analyzed were NBS Standard Reference Materials
Orchard Leaves (SRM 1571), and Brewers Yeast, which was furnished to NBS by
the Nutrition Institute, U.S. Department of Agriculture, Beltsville, Maryland,
for preparation as a Standard Reference Material. The Brewers Yeast is
certified for total chromium concentration only and is currently being
prepared for sale by the NBS Office of Standard Reference Materials as SRM
1569.
Preparation of Standards and Carrier
Two standards were used in this work. One standard was prepared by
dissolving a weighed amount of chromium metal (99.99% pure) in high purity
HC1 and diluting to a specific volume with high purity H20 to obtain the
desired Cr concentration. The second standard was prepared by dissolving
a weighed amount of NBS-SRM 136C (K2Cr2Oy) in high purity H20 and diluting
to volume to obtain the desired Cr concentration. Chromium carrier solutions
were prepared from analytical reagent grade CrCls and K2Cr207 and normally
13
-------
contained 3 mg Cr/ml.
All reagents used in the analysis were of analytical grade, unless
stated otherwise.
Irradiation Conditions
Chromium was determined in the two biological materials using both
INAA and DNAA. The samples (200-350 mg) along with Cr standards were
encapsulated in clean Supersil quartz vials and irradiated for periods of
one to six hours in the NBS Reactor at a thermal neutron flux of Ixl13n cm 2*
sec 1. The samples were allowed to decay 3-6 weeks for INAA xrork and ^36
hours for DNAA work to reduce the matrix activity before processing.
Procedure for INAA
Following a three-week decay the sample vials were washed clean of
exterior contamination with 1:1 HN03 and E^O', frozen in liquid nitrogen
opened, and the material was transferred to clean polyethylene counting
vials. The amount of sample transferred was determined by weight. The
samples and standards were counted on a 75 cm3 Ge(Li) detector coupled to
a 4096 channel pulse height analyzer for measurement of 51Cr produced by
the 50Cr(n,y) 5*Cr reaction. The concentration of chromium was determined
by the direct-comparator method.
Several brewers yeast samples and standards were preweighed into
clean quartz vials, sealed, and irradiated as already described. After
allowing these samples to decay for ^6 weeks, the vials were washed clean
of exterior contamination and counted (material still contained in vial)
as already described.
Procedure for DNAA
Following a 36-hour decay, the sample and standard vials were cleaned,
frozen in liquid N2, opened, and transferred (by weight) to a 50 ml Erlenmeyer
-flask designed for dissolving materials in a closed system (3) and trapping
volatile material in a trap solution. Chromium carried (^5 mg of both
Cr III and Cr VI) and 5-10 ml of concentrated HClOij-HNOs mixture (1:3)
were added to the flask. The flask was heated on a Pyrex-top hot plate
until all the sample had dissolved and all the vapor fumes visibly trapped
in a 1:1 HN03 solution. The flask was removed from the heat and 5 ml of
hot Ce(SOtt)2-3M t^SOi^ solution (10% w/v) was added; the flask was heated
an additional 5-10 minutes to assure oxidation of and to maintain the chro-
mium as Cr VI. After cooling, the sample solution in the flask was transferred
to a 50-ml extraction tube with 15 ml of 1.5 M HC1. The sample was then
extracted with 10 ml of 1 percent (w/v) tribenzylamine-chloroform solution.
A 5-ml aliquot was taken from the organic phase and counted on a 25 cc
Ge(Li) detector coupled to a HP 4096 computer-analyzer. The trap solution
from the dissolution step was transferred to a suitable counting vessel
and counted on this system also. The 320 keV gamma ray peak of 51Cr was
integrated and the chromium concentration was determined by the direct-
comparator method using the two chromium standards subjected to the same
14
-------
procedure. All of the chromium from these standards remained in the
flask. The trap solution was checked for chromium but none was detected.
The procedure is the same for open-flask dissolution of samples except a
regular Erlenmeyer flask is used, eliminating the trap solution.
RESULTS AND DISCUSSION
Recent interlaboratory comparison studies (4,5) of analytical methods
and their results for the analysis of chromium in biological and environmental
materials showed very large differences in analytical values. During the
spring of 1974, a workshop on chromium analysis was held at the University
of Missouri in Columbia where these differences were discussed and recommen-
dations made to resolve them. There was consensus among the participants
that biological standard reference materials, certified for total chromium
content by NBS, would provide a starting point for the analytical community
to critically evaluate their methods.
Chromium has been determined and certified in several materials by
NBS. However, this manuscript describes the results of chromium analyses
in two biological materials—Orchard Leaves, Brewers Yeast—using neutron
activation analysis (NAA). Studies here in our laboratory and other
laboratories (6,7) on some biological materials had suggested possible
losses of chromium content during sample manipulation (e.g.> dissolution,
charring, ashing,, etc.). Instrumental neutron activation analysis (INAA)
and destructive neutron activation analysis (DNAA) were used to determine
the chromium concentration in these biological materials. The use of INAA
to determine the total chromium concentration in these biologicals provided
a check for the values obtained using DNAA and also other techniques which
required sample treatment. The results obtained for chromium in the two
biologicals mentioned above, using NAA and the NBS Reactor, are given in
Tables 1-4.
By employing two dissolution techniques—closed system and open
flask—for the irradiated biologicals, the loss of chromium in the dissolution
step could be evaluated. The results obtained for chromium in Orchard
Leaves (Table 1) were essentially the same using INAA and the two dissolution
techniques, indicating no chromium is lost from this material in the
dissolution procedures used.
A large amount of matrix radioactivity is produced when these biological
materials are irradiated. Much of this activity results in long-lived beta
radiation creating a bremsstrahlung continuum in the gamma spectra of each
sample. This continuum affects the low energy range of the spectra where
the chromium photopeak is located thus arge background corrections are
necessary. Radiochemical separation of Cr (DNAA) eliminates this background
interference. The results obtained from open-flask dissolution (DNAA) of
yeast (Table 2) were consistenlty lower (^25%) than those obtained instru-
mentally (INAA). The results obtained from dissolution of brewers yeast in
a closed system as described in the DNAA procedure are in good agreement with
the INAA results.
15
-------
The results in Tables 1-3 show chromium can be lost during sample
dissolution of brewers yeast. However, no loss of chromium was observed in
our experimental design for the analysis of chromium in orchard leaves.
The chromium loss observed in brewers yeast is probably due to a volatile
component during the acid attacks. Further studies to characterize this
component are underway in this laboratory.
To establish that chromium was not lost in the irradiation step, brewers
yeast was preweighed, packaged, irradiated, and counted in the same container.
The results in Table 3 show the amount of chromium obtained using a poly-
ethylene irradiation container (INAA-P.E.) are in good agreement with those
obtained from a quartz irradiation container (INAA-Quartz).
The radiochemical procedure described for chromium (DNAA) provides the
analyst with a simple, rapid, and selective technique for chromium determina-
tion in a variety of matrices. The procedure is also adaptable for use by
other techniques. Results of chromium analysis (from a variety of techniques)
reported in the literature in recent years show very large differences in
concentration on the same materials. Thus, it is hoped these two biological
materials, certified'for total chromium concentration, will help those
involved in chromium analysis to evaluate their methods and improve chromium
results reported in the literature and elsewhere for analytical quality
control samples.
16
-------
REFERENCES
1. W. Mertz, Physiol. Rev., 49 (1969) 163.
2. L. McClendon, Trace Subst. Environ. Health, JJ (1974) 255.
3. D.A. Becker and G.W. Smith, Mod. Trends in Activ. Anal. Conf. , College
Station, Texas (1965) 230.
4. J. Pierce, et al, Symp. on the Development of Nucl. Based Tech. for
Meas., Detection and Control of Environ. Pollutants, Vienna, Mar.
1976.
5. R. Parr, Mod. Trends in Activ. Anal. Conf., Munich (1976) 1414.
6. V. Maxim, et al, symp. Nucl. Activation Tech. in the Life Sciences,
Bled, Yugoslavia, April, 1972.
7. W. Wolf, Interface, 2 (1973) 31.
17
-------
TABLE 1. CONCENTRATION OF CHROMIUM IN ORCHARD LEAVES,
SRM 1571, (ug Cr/gram)
Spl #
1
2
3
4
5
6
INAA
2.590
2.574
2.563
2.567
2.569
2.582
DNAAA
2.514
2.489
2.499
2.495
2.501
2.472
DNAAB
2.472
2.447
2.484
2.433
2.463
2.481
X 2.574 ± 0.010 2.495 ± 0.014 2.463 ± 0.020
* NBS Certified Value: 2.6 ± 0.2 yg/g
Note: A - Closed system dissolution
B - Open flask dissolution.
* Based on results of two independent analytical methods-
This work and isotope dilution mass spectrometry.
18
-------
TABLE 2. CONCENTRAION OF CHROMIUM IN BREWERS YEAST,
SRM 1569 (ug Cr/gram)
Spl #
1
2
3
4
5
6
INAA
2.079
2.075
2.090
2.077
2.067
2.104
A
DNAA
2.070
2.068
2,062
2.068
2.088
2.090
R
DNAA
1.583
1.543
1.548
1.588
1.551
1.565
X 2.082 ± 0.013 2.074 ± 0.012 1.558 ± 0.015
NBS Certified value: 2.12 ± 0.05 pg/g Based on results of two
independent analytical methods—This work and
isotope dilution mass spectrometry.
Note: A - Closed system dissolution.
B - Open flask dissolution.
19
-------
TABLE 3. CONCENTRATION OF CHROMIUM IN BREWERS YEAST,
SRM 1569, Using INAA (ug Cr/gram)
Spl #
1
2
3
4
5
6
INAA - P.E.
2.079
2.105
2.117
2.081
2.120
2.075
INAA - Quartz
2.131
2.087
2.137
2.092
2.149
2.120
X 2.096 ± 0.020 2.119 ± 0.025
20
-------
SECTION 3
METHOD FOR THE DETERMINATION OF PLATINUM IN BIOMATERIALS
D.A. Becker and T.E. Gills
Analytical Chemistry Division
National Bureau of Standards
Washington, D.C. 20234
21
-------
INTRODUCTION
Heavy metal contamination of the environment has received much
attention as serious pollution and health problems. Among the metals,
platinum is receiving increasing attention because of its many uses as
a catalyst, particularly in the catalytic converters used in the auto-
motive industry. Neutron activation analysis has the sensitivity for
determining platinum at very low concentrations, however the most
sensitive and useful nuclear reaction involves isotopes of relatively
low gamma energies, thus requiring separation of these isotopes from
an activated matrix. Similar methods (1,2) have been used for the
determination of platinum in glasses and indium and copper in gold.
The technique involves the spontaneous deposition of one metal
from solution using a noble metal (i.e. , one above it in the electro-
motive series) in the metallic form. Specifically, 199Au (as the chloro
complex) formed from the decay of 199Pt is deposited using silver metal.
The analysis for platinum was made by irradiating samples with neu-
trons and subsequently separating the element of interest by spontaneous
deposition. The nuclear reaction most suitable for the analysis was
19&pt(n,Y)199Pt|199Au. The end product of the reaction, 199Au has a 3.15
day half-life and y-ray energies of 158 keV (37% abundant) and 208 keV (8%
abundant).
22
-------
EXPERIMENTAL
The encapsulated samples, along with the platinum standards, were
irradiated in the NBS Research Reactor at a thermal neutron flux of
~6xl013n«cm -sec 1 for one hour. After irradiation, the samples were
allowed to decay for 24 hours to reduce short half-life matrix activities.
The samples were then quantitatively transferred to 100 ml beakers, and
carriers of Pt, Na, Ba, and Mn were added. The samples were dissolved in
10 ml of hot aqua regia and reduced in volume, by evaporation, to 2 ml.
After cooling, the solutions were washed into 25 ml of water containing
50 mg of finely divided silver powder. These mixtures were shaken, in
capped polyethylene bottles, to allow the spontaneous reduction of the
platinum. The excess silver metal, containing the platinum, was filtered
onto 25 cm glass fiber-filter pads, mounted in 42 cm petri dishes, and
counted in a standard geometry.
INSTRUMENTATION
The counting was carried out using a 75 cc Ge(Li) detector coupled
to a 4096 Channel Pulse Height Analyzer. Spectra were read onto magnetic
tape and processed with the ALLSPICE and QLN1 computer codes. Both the
158 keV and 208 keV gamma peaks of 199Au were quantified.
REAGENTS
The platinum standard used was NBS Standard Reference Material (SRM)
680, High Purity Platinum. The SRM was in the form of 0.0508 cm (.020
inch) diameter wire, and had a purity of 99.998 percent.
All other reagents were of ACS analytical reagent-grade quality.
SAMPLE PREPARATION
Samples were prepared for reactor irradiation by weighing into poly-
ethylene micro vials. Carbonaceous materials such as the filter papers were
high-temperature ashed in porcelain crucibles before encapsulation.
Platinum standards were prepared by dissolving a weighed quantity of
high purity platinum metal with a minimum volume of hot aqua regia.
23
-------
RESULTS AND DISCUSSION
Analytical results are given in Table 1. Concentrations were calcu-
lated using the standard-comparator method. The concentration specified
are given in addition to the concentration found. The overall analytical
error is estimated to be ±]0 percent, even though certain individual
results may be significantly better. Both photopeaks of 199Au were
individually quantified with consistent results in all cases. The absence
of 199Au in significant amounts preclude the remote possibility of double
neutron capture by large quantities of stable gold, leading to 199Au
which was used in the analysis.
TABLE 1. PLATINUM DOPED FE203 WITH FUSED SILICA
Sample ID Specified PT yg/g Concentration Pt
Concentration Mg/g Found
4.4 4.7 - . .
V — £l I
3.6 3.8
1.5 , ,
1.8 x=1'6
.5 .42 -
.42 X=
24
-------
REFERENCES
1. Becker, D.A. Analytica Chimica Acta, 61, 1972.
2. Kinard, W.D., Becker, D.A., and LaFleur, P.D. Activation Analysis
Section Summary of Activities, July 1969 to June 1970, NBS Tech.
Note 548, 1970. p. 66.
25
-------
SECTION 4
A METHOD FOR THE DETERMINATION OF ARSENIC, ANTIMONY, AND COPPER
IN BIOMATERIALS BY NEUTRON ACTIVATION ANALYSIS
T.E. Gills
Analytical Chemistry Division
National Bureau of Standards
Washington, D.C. 20234
26
-------
INTRODUCTION
The environmental impact of metal pollutants that can lead to dele-
terious effects on human health have been well documented. Because of the
severe toxicity of some metals such as arsenic, antimony, selenium, etc.,
levels of less than 1 ppm must be routinely determined to ensure the safety
of both plant and animal life. Neutron activation analysis has the sensi-
tivity for determining many of these toxic elements and is an ideal technique
for determining trace metals in biological materials. However, in many
situations the most sensitive and useful nuclear reactions involves isotopes
of relatively low gamma energies; thus requiring separation of those isotopes
from a neutron activated matrix to obtain maximum sensitivity and accuracy.
Among possible reagents for the radiochemical separation of toxic
trace metals, inorganic ion exchangers have recently received some attention1,
By selecting the proper acidic media and inorganic ion exchangers, very
selective separations were achieved. A cuprous chloride column in series
with a tin dioxide column, equilibrated in a 6M HCIO^ media proved effective
in isolating 6t*Cu, 122Sb, 76As, respectively. This separation scheme was
tested using radioactive tracers and by reanalyzing NBS Orchard Leaves SRM
1577. The procedure was later applied to the certification of As, Cu, and
Sb in the following NBS proposed SRM's: Spinach, Pine Needles, and Tomato
Leaves (1573), (1570), and (1575).
This procedure involved the neutron irradiation of a biological sample
and subsequent mineralization of the material by nitric and perchloric
acids. After obtaining the irradiated material in a soluble form, the
solutions were adjusted to the desired acid concentrations and passed through
chromatographic columns of cuprous chloride and tin dioxide which had been
equilibrated with 6M HClOt,. Copper and antimony were sorbed on the cuprous
chloride and arsenic sorbed on the tin dioxide. Sodium, the major inter-
ference in the analysis of biological materials, was eliminated by eluting
with 6M HC1CV
1F. Giardi, R. Pietra, E. Sabbioni, "Radiochemical Separations by Retention
on Inorganic Percipitates (1968). Proceedings of the International
Conference on Modern Trends in Activation Analysis, pp. 642-650.
27
-------
EXPERIMENTAL
Samples of each material (weighing approximately 250 rag each) were
encapsulated in polyethylene microcentrifuge tubes. Solution standards of
Cu, As, and Sb, as well as NBS Orchard Leaves SRM #1571, were also en-
capsulated to quantitate the three botanicals for Cu, As, and Sb. The
samples were then irradiated for 30 minutes in RT-3 of the NBS reactor in a
flux of approximately 5xl013 n cm 2sec *. Because of the high matrix radio-
activities, the irradiated materials were allowed to decay for 24 hours to
reduce personnel radiation exposure. After decay the samples were post
weighed into 125 ml Erlenmeyer flasks containing carriers (21 mg) of each
element to be analyzed. The samples were then dissolved in a 1:3 mixture
of HCL04-HN03 with 3-4 drops of cone. HF to dissolve any siliceous material.
After dissolution the samples were taken to near dryness and allowed to cool.
Ten milliliters of 6M HCLO^ were added to each sample and the solution then
passed through 3 cm columns of cuprous chloride and tin dioxide. Copper
was absorbed on the cuprous chloride and antimony and arsenic were absorbed
on the tin dioxide column. The columns were then washed with two ten milli-
liter portions of 6M HCLO^. After the final wash the columns were emptied
separately into 45 cm petri dishes and counted.
INSTRUMENTATION
The counting was done on a 25 cc Ge(Li) detector in conjunction with a
2048 channel pulse height analyzer. The peak analyses were done by a
Hewlett Packard analyzer-computer.
REAGENTS
The standard solutions were made of high purity metals or oxide
dissolved in NBS high purity acids. All other reagents were of ACS
analytical reagent-grade quality.
SAMPLE PREPARATION
Samples of each of the proposed SRM's were freeze-dried at room
temperature. The nominal weight loss was experimentally determined to be
±2 percent.
28
-------
RESULTS AND DISCUSSIONS
The results are given in Table 1. Concentrations were calculated
using the standard-comparator method. The described procedure gave
precisions acceptable for certification. As to the accuracy of the
analysis, a comparison of these results to those obtained by other in-
dependent methods is shown in Table 2. The comparison shows relative
good agreement with other methods. Inorganic exchangers have proved
potentially useful for radiochemical separations. Such separations have
the following advantages:
Effective separations from strong acidic media (> 1 molar)
that can be easily adjusted from typical dissolution
procedures.
Quantitative separations, thus eliminating errors involved
in chemical yields.
High decontamination from activated biological matrices.
Multielement analysis consequently reduced analysis cost
With this procedure rapid analysis of biological matrices can be made
utilizing highly selective group separations, an attractive feature when
many samples are to be processed.
29
-------
TABLE 1. METHOD OF ANALYSIS: ION EXCHANGERS WITH NAA
Concentrations in PPM
SRM/SPL No.
Tomato - 1
Tomato - 2
Tomato - 3
Tomato - 4
Ave:
Pine Needles - 1
Pine Needles - 2
Pine Needles - 3
Pine Needles - 4
Ave:
Spinach - 1
Spinach - 2
Spinach - 3
Spinach - 4
Ave:
Cu
10.8
11.3
11.0
11.3
*
11.1 ± .2
2.96
2.98
2.80
3.20
*
2.98 ± .16
10.8
10.8
10.6
11.0
*
10.8 ± .2
As
.198
.249
.227
.261
*
.234 ± .028
.195
.193
.223
.185
*
.199 ± .016
.116
.109
.110
.139
A
.118 ± .014
Sb
.109
.147
.084
.128
*
.117 ± .027
.235
.218
.201
.210
*
.216 ± .014
.049
.040
.043
_ __
*
.043 ± .004
*
Errors are expressed as the standard deviation of the mean.
30
-------
TABLE 2. A COMPARISON OF NAA RESULTS USING THE DESCRIBED SEPARATION
PROCEDURE TO OTHER INDEPENDENT METHODS
Code Element Spinach Pine Needles Tomato Leaves
1
4
3
1
2
Cu
Cu
Cu
As
As
10.8 ±
12.7 ±
11.5
.12 ±
.17 ±
.2
.4
.01
.05
2.98 ±
3.2 +
2.9
.20 ±
.21 ±
.16
.4
.02
.04
11.1 ±
11.5 ±
10.6
.23 ±
.29 ±
.2
.2
.03
.05
1 Sb .043 ±.004 .22 ± .01 .12 ±.03
Codes: 1 - Neutron Activation Analysis with Chemical Separation.
2 - Colorimetry.
3 - Isotope Dilution Spark Source Mass Spectrometry.
4 - Atomic Absorption Spectrometry.
31
-------
SECTION 5
THE CLEANING, ANALYSIS AND SELECTION OF CONTAINERS FOR
TRACE ELEMENT SAMPLES
J.R. Moody and R.M. Lindstrom
Analytical Chemistry Division
National Bureau of Standards
Washington, D.C. 20234 .
32
-------
INTRODUCTION
The determination of trace elements at very low levels, particularly in
liquid samples, has been found to be biased by the analytical blank and can
often be attributed in large part to contamination from sampling and storage
containers (1,2,3,4). Synthetic polymeric materials are often employed as
containers in the expectation that contamination will be less than if glass
or metallic containers were used. The degree of improvement depends on the
plastic and on the cleaning procedure (if any) employed to clean the plastic
before use. The literature is contradictory on this topic (1). The present
work examines several materials and cleaning methods in an effort to define
a set of materials and procedures adequate for most work at the ng/g level.
For long term storage of liquid samples, an important consideration
together with low contamination levels is a low rate of loss of water vapor.
Factors to be considered include permeability, wall thickness, and closure
integrity and are best determined by experiment. The rate of loss which can
be tolerated is dependent upon both the accuracy desired and the amount of
time during which the sample must be stored. For high accuracy work over a
storage period of 10 years an annual rate of loss of 0.1 percent or less per
year is necessary.
33
-------
EXPERIMENTAL
A number of plastic materials were studied, including bottles constructed
of conventional polyethylene (CPE), linear polyethylene (LPE), polypropylene
(PP), polymethylpentane (PMP), polycarbonate (PC), polyvinylchloride (PVC),
and several types of Teflon (TFE, FEP). In addition specimens of ethylenete-
trafluoroethylene copolymer (Tefzel or ETFE), Teflon PFA, polystyrene (PS),
and Teflon pipe sealing tape (TPT) were examined. The containers used for
these studies were carefully selected to be free from visible occlusions
and were average in wall thickness, weight, and other characteristics. The
studies of these materials may comprise three types of experiments: a
gravimetric study of the rate of loss of water from bottles, isotope dilution
mass spectrometric (IDMS) analysis of the impurities leached from a bottle
during a cleaning process, and neutron activation analysis (NAA) of the
impurities contained within and leached from the plastics.
Gravimetric Studies. Two bottles each of CPE, PP, PVC, PMP, and PC
were obtained from commercial sources (Bel Art, Nalge or equivalent). The
bottles were 1 liter or 500 ml capacity, and except for rinsing with 95
percent ethanol followed by distilled water they were not subjected to a
special cleaning procedure. All of the containers were weighed, then were
filled to the same relative level with distilled water and re-weighed. The
second container of each type was retained as a weighing tare to correct for
the influence of changes in the barometric pressure, relative humidity and
temperature, and loss of plasticizer. The samples were re-weighed after 17
days and 66 days to establish a rate of loss of moisture.
IDMS Studies. Leach solutions from one liter or 500 ml size bottles of
FEP, CPE, LPE, and PC were examined by spark source IDMS (multi-element) and
by thermal emission IDMS (for lead only). The general principles and tech-
niques used have been given elsewhere (5,6). The bottles were rinsed with
distilled water to remove any surface contamination and three sets of bottles
were filled with a (1+1) mixture of ultra-pure HC1 or ultra-pure HNC>3 and
ultra-pure water (7). The bottles were allowed to stand at room temperature
for one week, with the exception of the FEP bottle which was heated to 80 °C
for one week.
Aliquots of the contents of one set of bottles leached by (1+1) HC1 were
spiked with 206Pb. Aliquots from two other sets of bottles, leached by (1+1)
HC1 and (1+1) HNC>3 respectively, were spiked with a 19 element multi-spike
(7) for the spark source mass spectrometer (SSMS). The spiked solutions were
t-hen wallevaporated to several drops under a Class 100 laminar flow clean
hood. The. sample solutions were analyzed in the SSMS (multi-element) or
thermal source mass spectrometer (for Pb only).
-------
The first set of bottles were then re-filled with a (1+1) mixture of
ultra-pure HNOs and ultra-pure water and the entire cleaning, sampling,
spiking, analysis sequence was repeated as before. Finally, the bottles were
filled with high purity 0.5 percent (wt./wt.) HNOs, sampled after one week
and again after two months, and then the aliquots were spiked with 2(^Pb and
subsequently analyzed.
NAA Studies. Specimens of 0.2-2 grams (4-12 cm2) were cut from new
0.5-1 H plastic bottles (CPE, LPE, PP, PMP, PC, PVC, and.FE?), sheet (TWE and
PFA), bottle caps (ETFE), boxes (PS) or tape (TPT). The samples were pre-
cleaned by rinsing lightly with 95 percent ethanol followed by distilled
water. All irradiations were performed at a thermal neutron flux of
1.3 x lO1^ n cm 2sec * and a gamma flux of 2 x 107 rad/hr (8). Other positions
in the reactor offer higher neutron flux, but at the expense of a much higher
gamma flux per neutron, and hence a much higher radiation damage to the
samples.
NAA Experiment 1; The samples packaged in high purity LPE film (9) were
irradiated for two minutes, the edges trimmed, and the induced radioactivity
determined by counting on a Ge(Li) gamma detector. The samples were weighed,
cleaned in (1+1) HC1 by heating at 100 °C for two hours, rinsed with high
purity water, dried, repackaged, and irradiated and counted again. After a
similar two-hour leach in hot (1+1) HNOs the plastics were analyzed a third
time.
NAA Experiment 2: Different samples of the same plastics were used in
measurements of long-lived activation products. After irradiation (4-6 hrs)
and decay of short-lived activities, the remaining radioactivation products
were counted. The samples were then leached in (1+1) HC1 at 100 °C, rinsed,
and then leached in hot (1+1) HNOs- Both leach solutions and the cleaned
plastics were then counted on a gamma-ray detection system.
35
-------
RESULTS AND DISCUSSION
Gravimetric Studies. The results are summarized in Table 1. Two of the
bottles (PP and CPE) have water loss rates which are compatible with the long
term storage of aqueous samples. In addition, data previously obtained by
Moody, et al, (10) indicated an annual rate of water loss of 'vO.OS percent
from oneTYiFer Teflon FEP bottles. PVC, PMP, and PC bottles may be adequate
for storage over short time periods. Losses due to transpiration may be
reduced by sealing the bottle in a commercially available vapor barrier, e.g.,
a polyethylene-coated film of aluminized mylar. The practice of using a
desiccator partially filled with water to act as a humidity chamber to slow
transpiration losses is recommended although one must be careful that high
humidity conditions do not cause a weight gain (absorption of moisture) .
None of the observed loss rates should be construed as permeability data
since the closure may have been responsible for a significant portion of the
total water loss. The results obtained in this work for CPE are similar to
those observed by Curtis, et al, (11) .
IDMS Studies. The results obtained for the leaching of lead from FEP,
LPE, CPE, and PC containers are listed in Table 2. Lead was chosen as a
model element for this study, partly because of the high accuracy obtainable
by thermal source isotope dilution mass spectrometry and partly because lead
is a common contaminant. The practice of cleaning bottles first with HC1 and
then with HN03 has been proven to be an effective procedure in our labora-
tories (7). Most of the cleaning of these containers is accomplished by
only one week of soaking in HC1. The additional week of soaking in (1+1)
HNOs removes additional lead only from FEP and PC. Patterson (2) has
suggested that 0.5 percent HNOs i-s more efficient for the long term cleaning
of Teflon containers from Pb contamination than is (1+1) HNOs. In these
experiments no further leaching of lead was detected after leaching with
(1+1) HNOs except from the PC container. After one month of leaching with
0.5 percent HNOs, no further lead was leached even from the PC container.
Data obtained by SSMS is compiled in Tables 3 and 4 for impurities
leached by (1+1) HC1 and (1+1) HNOs respectively. Again, the Teflon FEP
bottles were heated to 80 °C while the others were leached at room tempera-
ture. The (1+1) HC1 leached more than the (1+1) HNOs except for the FEP
containers. A single element, Ca, is responsible for most of the difference
between the (1+1) HC1 and (1+1) HN03 in leaching Teflon FEP. Values which
are below 2 ng/ctn2 or which are prefixed by < are upper limits. That is, the
concentration is below the optimum concentration range for the amount of
spike isotope used or the concentration is near the blank or detection limit.
The greater total quantity of various elements leached from FEP may be
partially explained by the higher temperature (80° vs room temperature) used
to clean the Teflon FEP container.
36
-------
NAA Studies. The amount of impurities found in the uncleaned plastics
as determined in NAA experiments 1 and 2 are summarized in Table 5. A
comparison of the impurities found in CPE in this study shows agreement with
that found by other workers (4,9,12-15) to approximately an order of
magnitude. It is interesting to note that there has been no clear change in
trace elemental impurities in over 25 years of manufacture (12). The ten
materials studied show striking differences in trace element composition.
The purest materials are TFE and CPE, in agreement with the experience of
numerous workers (1). Polystyrene was also very clean. Many materials
showed easily detectable amounts of a few elements. Some of these elements
are residues of polymerization catalysts (16).
Interpretation of the acid-leaching experiments is complicated,
particularly with the long irradiation in the second experiment, by the
fact that the experiments are imperfect models of a reagent or analyte stored
in a bottle. Those atoms of trace elements leached and detected in the
leachate in the second experiment are precisely those not representative of
undisturbed trace elements in an undisturbed matrix. The radioactive atoms
underwent recoil on absorption of a neutron and thus may have been more
labile than their inactive neighbors. In addition, the matrix suffered
radiation damage from some 108 rads of gamma radiation and the halocarbons
were damaged additionally by the absorption of nuclear recoil energy and beta
radiation induced in major constituents of the matrix. However, detailed
examination of the leaching results for sodium, the only element common to the
two experiments, shows that radiation damage does not seem to greatly affect
the general conclusions (Table 6) which are discussed below.
The results of the acid leaching experiments for all elements (Table 7)
are compiled from the differences in trace element composition of the plastic
before and after leaching in the first and second NAA experiments, and from
direct measurement of leached tracer in the second, in a manner similar to
the sodium results in Table 6.
With the exception of Na in pipe tape, (perhaps a special case because of
its high surface/volume ratio), and to a lesser extent Na in CPE, the trace
elements studied are in general not leached from the polymer matrix even with
the rather severe acid treatment used here. It may be inferred that the bulk
of most trace elements present is distributed throughout the matrix, and not
merely on the surface.
The hydrochloric acid leach in experiment 1 increased the chloride con-
centration in all the plastics corresponding to a 10 ^ cm thickness of 1+1 HC1
in the surface of polypropylene, and less in the other materials. This inward
diffusion of HC1 is in accord with the water loss measurements, both experi-
ments leading to a diffusion coefficient for aqueous solutions in polyolefin
of on the order of D=10 12 cm2/sec. With this value of D, the rms diffusion
length in one year is 0.1 mm.
37
-------
CONCLUSIONS
Differences were observed between NAA and IDMS results for trace elements
leached from the plastics. At least some of these differences may be in the
way the samples were leached. For IDMS work the leaching was carried out
within intact bottles. In contrast, the NAA studies examined the trace
elements leached from both sides of a small sample of plastic immersed in
acid. Since at least some containers are known to be blow molded, the higher
results obtained by NAA may simply reflect a difference in contamination
levels between inside and outside walls, the NAA data probably including
contamination from the mold.
With the exception of Teflon FEP, (1+1) HC1 has been found to be the
better cleaning agent. However, HC1 and HNOs appear to leach various elements
with different efficiencies, thus, the use of both acids one after the other
is recommended. The various Teflons and CPE bottles have been found to be the
least contaminating bottles once they have been cleaned. Many methods have
been suggested and several have been shown to be sufficient for the purpose
at hand (1,2,7).
Suggested Method for Cleaning Bottles. From the present work, that of
Karin et^ ad. (15), and earlier work in this laboratory (3,7), a procedure is
suggested which may be suitable for most contemporary trace element work.
The procedure is outlined in Table 8. Only reagent grade acid is necessary
for the acid leaching; but, ultrapure distilled water should be used for the
final step.
38
-------
ACKNOWLEDGEMENT
The authors wish to acknowledge the contributions of T. J. Murphy,
P. J. Paulsen, and J. W. Gramlich for the IDMS analysis of solutions stored
in plastic containers, and of D. A. Becker for numerous constructive
suggestions.
39
-------
REFERENCES
1. E. J. Maienthal and D. A. Becker, ''A Survey of Current Literature on
Sampling, Sample Handling and Long Term Storage for Environmental
Materials, NBS Tech Note 929 (1976).
2. C. C. Patterson and D. M. Settle, in "Accuracy in Trace Analysis:
Sampling, Sample Handling, and Analysis," Proc. of the 7th IMR Symp.,
P. D. LaFleur, ed., NBS Spec. Pub. A22, 321 (1976).
3. T. J. Murphy, in "Accuracy in Trace Analysis: Sampling, Sample Handling,
and Analysis," Proc. of the 7th IMR Symp., P. D. LaFleur, ed., NBS Spec.
Pub. 422, 509 (1976).
4. D. E. Robertson, Anal. Chem. 40, 1067 (1968).
5. P. J. Paulsen, R. Alvarez, and C. W. Mueller, Applied Spectroscopy 30, 42
(1976).
6. I. L. Barnes, T. J. Murphy, J. W. Gramlich, and W. R. Shields, Anal.
Chem. 4_5_, 1881 (1973).
7. S. C. Kuehner, R. Alvarez, P. J. Paulsen, and T. J. Murphy, Anal. Chem.
44, 2050 (1972).
8. H. E. Despain, private communication.
9. S. H. Harrison, P. D. LaFleur, and W. H. Zoller, Anal. Chem. 47^ 1685
(1975).
10. J. R. Moody, H. L. Rook, T. C. Rains, and P. J. Paulsen, unpublished
data from the preparation of SRM 1642.
11. G. J. Curtis, J. E. Rein, and S. S. Yamamura, Anal. Chem. 45, 996 (1973).
12. R. E. Thiers, "Methods of Biochemical Analysis," D. Click, ed., .5, 274-
309, Interscience, New York (1957).
13. J. U. Mitchell, Anal. Chem. 4_5, 429A (1973).
14. H. Sorantin and P. Patek, Z. Anal. Chim. 211, 99 (1965).
15. R. W. Karin, J. A. Buono, and J. L. Fasching, Anal. Chem. 47, 2296 (1975).
16. E. C. Kuehner and D. H. Freeman, in "Purification of Inorganic and
Organic Materials," M. Zief, ed., Marcel Dekker, New York 1969,
pp. 297-306.
40
-------
TABLE 1. ANNUAL RATE OF LOSS OF WATER FROM CONTAINER MATERIALS
Container
CPE
PP
PVC
PMP
PC
After 17 Days
0.116%
0.034%
0.429%
0.988%
1.65%
After 66 Days
0.109%
0.049%
0.601%
1.018%
2.00%
TABLE 2. LEAD LEACHED FROM CONTAINERS (ng/cm2)
DETERMINED BY IDMS
Bottle (1+1) HC1 (1+1) HN03 0.5% HN03 0.5% HN03
1 week 1 week 1 week 2 months
FEP
LPE
CPE
PC
0.41
0.20
0.18
0.36
0.014
0.023
0.023
no significant amount over blank level
41
-------
TABLE 3. IMPURITIES LEACHED FROM PLASTIC CONTAINERS IN
ONE WEEK BY (1+1) HC1 (ng/cm2) DETERMINED BY
IDMS
Elements
Pb
Tl
Ba
Te
Sn
Cd
Ag
Sr
Se
Zn
Cu
Ni
Fe
Cr
Ca
K
Mg
Al
Na
Teflon FEP
2
<1
2
2
1
0.6
<6
=
<1
0.8
4
6
0.8
16
4
2
1.6
1.0
4
2
LPE
0.6
£0.6
1
£l
0.2
0.2
0.4
9
1
0.8
1
0.8
60
1
0.4
4
6
CPE
18
3
0.3
0.7
£0.8
0.2
0.2
<0.3
-— :
1.0
0.7
0.3
1.0
0.3
0.8
0.7
0.7
10
42
PC
10
0.7
3
13
<8
0.3
<0.5
<6
=
0.3
£49
£5
£16
£5
0.8
3
8
Z58 £89 £81 £130
42
-------
TABLE 4. IMPURITIES LEACHED FROM PLASTIC CONTAINERS IN
ONE WEEK BY (1+1) HN03 (ng/cm2) DETERMINED BY
IDMS
Elements
Pb
Tl
Ba
Te
Sn
Cd
Ag
Sr
Se
Zn
Cu
Ni
Fe
Cr
Ca
K
Mg
Al
Na
Teflon FEP
2
-------
TABLE 5. CONCENTRATIONS OF TRACE ELEMENTS IN PLASTICS DETERMINED BY (ng/g)
CPE
Na 1 . 3E3
Al 500
Cl 7E3
K >5E3
Ca
Ti
Mn
" Co
Zn
Br >20
Sn
Sb 5
Lz
W
Au
LPE PP PMP PS PC PVC TFE FEP
15E3 4.8E3 200 2.2E3 2.7E3 20E3 160 400
30E3 55E3 6.2E3 500 3.0E3 230 200
30E3 180E3 50E3 major 800
>600 >200 90E3
800E3
5E3 60E3 5E3 1E3
10 20 10 20 60
40 6
520E3 33E3
800 >5 >2 >1 29E3 >6 >2
2.4E6
200 600
0.3 0.6
700
0.1 0.6 0.04 0.03 0.4
PFA TPT ETFE
100 2.3E3 600
29E3
50E3 7E3 1E6
3E3 1 . 1E3
2E6
20 20
90
14E3
16E3 240
1
0.2 0.4
Notation: 1. 3E3=1.3xl03 ng/g. When lower limits are quoted, the element sought was short-lived and
found only in the leachate or leached sample. Hence the concentration in the unleached plastic may have
been greater but was not visible in the presence of large quantities of short-lived interfering radio-
activities. Blanks in the table represent the extreme where the element sought for was not visible.
Detection limits may be roughly judged from values in adjacent columns.
-------
TABLE 6. REMOVAL OF SODIUM FROM PLASTICS
Matrix
CPE
LPE
PP
PMP
PS
PC
PVC
TFE
FEP
PFA
TPT
ETFE
Expt. 1 Short Irradiation Expt. 2 Long Irradiation
Before Post Post HC1 Before HC1 Leach HN03 Leach
Cleaning HC1 + HN03 Cleaning Liquid Liquid Post Cleaning
600
±100
17000
±3000
5000
±1000
220
±50
2000
±400
2400
±400
170
±40
180
±50
2900
±600
1290
±30
15000 14000 13200
±3000 ±1000 ±200
4640
±70
160
±5
1500 2450
±400 ±40
2860
±50
19500
±400
160
±10
400
±20
39 4
1900 2270
±200 ±50
575
±15
Concentrations are in ng/g plastic. Errors quoted reflect
estimate the total error of measurement. Note that the tw
amount of Na removed by the leach step; all other columns
sample.
Blanks imply
that Na was not detected.
1480
±10
273
±2
480
±3
52
±1
60
±1
68
±1
190
±10
111
±2
242
±3
15
±1
1890
±20
427
±3
counting
o "leach
give the
2.0
±0.3
5.9
±0.5
32
±1
1.0
±0.1
9
±1
4
±1
28
±3
144
±3
1.2
±0.1
statistics only, and
liquid" columns give
amount present in the
330
±20
13200
±100
4330
±60
120
±10
2430
±70
2850
±100
21000
±900
10
±10
380
±40
208
±3
so under-
directly the
plastic
-------
TABLE 7. SEQUENTIAL LEACHING OF TRACE ELEMENTS -FROM PLASTICS (ng/cm2)
CPE
LPE
PP
PMP
PS
PC
PVC TFE
FEP PFA TPT
ETFE
Na
Al
K
Co
Zn
Br
Sb
La
W
Au
200+0.4 70+1 100+7
50+
1000+ 100+
3+
1+
50+30
2+0.5 1+
3+
20+0.4 20+8
80+
20+1 30+10 80+ 60+
1+
0.03+0.01
+2
0.3+0.1 7+20 2+
0.1+
0.01+
0.01+
90+
20+ 50+4 300+1
400+
20+
10+
0.4+
0.3+
4+0.2
0.7+
0.01+ 0.3+0.01
Determined by NAA. The two numbers given (in ng/cm2) are respectively the amount leached
by 1:1 HC1 and by 1:1 HNOs in 2 hours at 100°C. A blank means that the constituent was not
detected. See also footnote to Table 5.
-------
TABLE 8. SUGGESTED METHOD FOR CLEANING
PLASTIC CONTAINERS
1. Fill with 1+1 HC1 (AR grade).
2. Allow to stand one week at room temperature
(80° for Teflon).
3. Empty and rinse, with distilled water.
4. Fill with 1+1 HN03 (AR grade).
5. Allow to stand one week at room temperature
(80° for Teflon).
6. Empty and rinse with distilled water.
7. Fill with purest available distilled water.
8. Allow to stand several weeks or until needed,
changing water periodically to assure
continued cleaning.
9. Rinse with purest water and allow to dry in
a particle- and fume-free environment.
47
-------
SECTION 6
EVALUATION BY ACTIVATION ANALYSIS OF ELEMENTAL RETENTIONS
IN BIOLOGICAL SAMPLES AFTER LOW TEMPERATURE ASHING
George J. Lutz, John S. Stemple and Harry L. Rook
Analytical Chemistry Division
National Bureau of Standards
Washington, D.C. 20234
48
-------
INTRODUCTION
Low temperature ashing (LTA) is the low temperature (^100 °C)
decomposition of organic or biological samples with atomic oxygen produced
by an electrodeless radio-frequency (RF) discharge. There are a number
of advantages in using LTA as the first step in an analytical method
whether by nuclear activation or otherwise:
1. Since the oxidation temperature is low, volatility losses are
minimized during the destruction of large quantities of organic material.
2. There are no electrodes to contaminate the sample.
3. A sample can be more readily dissolved after LTA than by
alternative methods. Wet ashing techniques with hot oxidizing acids or
strong alkalis are potentially hazardous, sometimes incomplete, and
usually susceptible to contamination because of the large amounts of reagents
required.
4. High temperature ashing in a muffle furnace introduces the
possibility of contamination from containers or the furnace walls as well
as loss of some elements by volatilization. The convenience of the LTA
method can be demonstrated by reference to a publication concerning the
spectrophotometric determination of boron extracted from ashed animal
tissues (1). Four grams of freeze-dried tissue were ashed 40-50 hours at
100-150 watt forward RF power. After the LTA treatment, the sample could
be dissolved in 2 ml of IN HC1.
5. In the X-ray fluorescence method, which is strongly matrix
dependent, LTA reduces the sample to a silicate-carbonate-oxide form,
which minimizes errors arising from matrix interferences.
6. In spark-source mass and emission spectrometry the ashed residue,
when mixed with powdered graphite, gives a uniform consistency which also
reduces matrix error.
With specific respect to activation analysis, one can identify several
advantages of beginning the analysis of a biological sample with freeze-
drying followed by LTA.
1. A very high neutron flux can badly char a biological or organic
sample leaving it with an undesirable consistency. That the LTA method is
very efficient for removing organic carbon is demonstrated by an American
Society for Testing and Materials procedure (2) which specifies the technique
49
-------
in removal and hence determination of organic carbon as opposed to carbonate
in sediment samples.
2. LTA results in a substantial reduction in the volume of the
sample. This can yield a corresponding reduction in electron accelerator
or neutron generator irradiation time in photon or 14-MeV neutron activation
analysis.
3. Since lengthy wet digestions are eliminated, a sample after
irradiation can be dissolved more quickly and hence short-lived nuclides
can be more quickly separated from the sample for counting.
4. In the instrumental photon activation analysis of samples containing
large amounts of carbon, it is usually necessary to delay counting for a
few hours after irradiation to allow the '1]-C produced by the reaction
12C(y, n)1]-C to decay. Some shorter-lived nuclides are therefore not
detected. Many of them could be detected by removal of the carbon prior
to irradiating.
The possibility exists in the LTA method of the loss of elements of
interest during the ashing process. Studies to evaluate elemental reten-
tion during ashing have been conducted with radioactive tracers (3). This
is not an entirely satisfactory procedure since an element may not be
present in a biological sample in the same chemical form as was used in
the tracer studies. This paper describes a study, using activation
analysis, of loss or contamination of trace elements during LTA. It is
hoped that this will be useful not only in analytical chemistry for the
reasons listed above, but also to evaluate LTA as a potential processing
method for the retention of biological samples by the National Environmental
Sample Bank over several decades of time for purposes of baseline studies
to evaluate increases or decreases in pollution.
EXPERIMENTAL
A commercial model LTA apparatus capable of 100 watts forward power
was used in these experiments. A schematic of the apparatus is shown in
figure 1. The sample was placed in a borosilicate glass ashing chamber
which was briefly rotated by hand to induce the organic sample to spread
as evenly as possible on the inside surface. The cold trap was filled
with crushed dry ice (liquid nitrogen would condense explosive ozone).
The system was pumped down to 1.3-2.6 Pa (10-20 millitorr). The oxygen
flow rate was adjusted so as to maintain a pressure of 13-40 Pa (100-300
millitorr). Forward power was usually set to the maximum of 100 watts.
Tuning to optimum was accomplished by observing the maximum intensity in
discharge glow in the tube.
50
-------
BOROSILICATE GLASS ASHING CHAMBER
VACUUM GAUGE I
'^i i
I
VALVE
TO
VACUUM
js s
1 v.
REGULATOR
DRY ICE
'COOLED TRAP
FLOW GAUGE -»
RF GENERATOR AND
TUNING CONTROLS
OXYGEN CYLINDER
Figure 1: Block Diagram of LTA Apparatus
Preliminary experiments involving addition of tracers to biological
materials and determining retention of each isotope after ashing verified
previously published data /3/ and served to establish and perfect the
basic ashing technique. After these experiments, it was decided to study
three NBS Standard Reference Materials in detail; 1571-Orchard Leaves,
1577-Bovine Liver, and 1632-Trace Elements in Coal.
Samples of about one'gram were ashed for a period of several hours.
It was impossible to remove all of the ashed material from the tube, thus
it was necessary to determine the exact amount of material removed. This
was first attempted by weighing of the tube which proved difficult, since
the only balance capable of accommodating the 200 g in weight, 30 cm in
length ashing tube was accurate to only + 2 mg. A second and more successful
technique involved the use of a nonvolatile radioactive tracer that wouldn't
51
-------
interfere with later activation analysis. The uniform distribution of the
tracer throughout the bulk sample was demonstrated by taking several small
samples from the bottle and measuring their specific activities. Subsequently,
it was found possible to use the activation product of an element demonstrated
to be nonvolatile as an internal standard for this purpose.
The relative amounts of the elements in ashed and unashed samples
were determined by nondestructive neutron activation analysis with the NBS
10 megawatt nuclear reactor and photon activation analysis with the NBS
electron linear accelerator.
For the reactor irradiation a paired experiment was carried out using
approximately one gram of unashed material compared to the ashed residue
of one gram of material. Three series of neutron irradiations were conducted:
1. Fifteen-second irradiation followed by three countings at few
minute intervals,
2. Twenty-minute irradiations followed by 2-3 countings over a
period of one hour to a few days,
3. A four-hour irradiation followed by 2-3 countings over a period
of several days to a few weeks after irradiation (depending on Na-24
levels).
The two shorter irradiations were conducted at a flux of 1.3xl013n*cm 2-
sec *, and the longer irradiation at a flux of 5.6xl013n'cm 2-sec 1.
•v*
Two series of photon irradiations were conducted with approximately
three grams of unashed material and the ashed residue of three grams of
material with an electron energy of 35 MeV and a beam current of about 50
yA.
1. Thirty to sixty minute irradiation followed by 2-3 countings
over a period of two hours to a few days after irradiation,
2. Two to six hour irradiation followed by 2-4 countings over a
period of two days to a few weeks after irradiation.
Irradiated samples were counted with a 75 cc Ge(Li) detector and a 4096
channel analyzer using one-half of the memory. Molybdenum, iodine, and
zinc by photon activation analysis were also determined with a low energy
photon detector (LEPD). Photopeak counting rates were corrected for times
of irradiation and decay and ratio of weights of unashed material and
ashed residue corrected for weight loss.
52
-------
RESULTS AND DISCUSSION
The experimental study of the observation of ashing losses was carried
out in three phases. The first consisted of spiking samples of interest
with radioactive tracers of known chemical states. This phase was simple
to conduct as the isotopes could be so added to avoid interference in
radioactive counting. This first phase served to assist in development
of handling techniques for subsequent work. The results with radioactive
tracers showed the retention of zinc, cadmium, arsenic, palladium, rhenium,
antimony, iron, platinum, iridium, gold, and silver. A previous paper (3)
had reported the loss of gold and silver but this phenomenon was not
confirmed here. Mercury and osmium were lost after several hours of ashing
in every case.
The second phase consisted of irradiating portions of the three
samples, counting them, then ashing and recounting. These experiments
added to the retention list the following elements: sodium, potassium,
rare earths, and selenium.
The third phase, which was the major part of this study, involved the
simultaneous activation, either by thermal neutrons or high-energy photons,
of an untreated portion of the samples and an ashed portion. This study
thus established loss data on elements in the chemical form they would
actually be found in three representative matrices. Table 1 gives typical
weight losses and carbon losses as measured by the photonuclear reaction
nC(Y,an)7Be for an ashing period of approximately 12 hours and power of
approximately 70 watts. In addition to carbon, the loss of chlorine,
bromine, and iodine was observed in all three matrices as well as the
expected loss of mercury. Table 2 gives experimental recoveries of
retained elements.
TABLE 1. TYPICAL WEIGHT AND CARBON LOSSES FOR ASHING PERIODS OF
15 HOURS AND RF POWER OF 70 WATTS
Coal
Beef Liver
Orchard Leaves
% Weight Lost
80
90
75
% Carbon Lost
75
98
80
53
-------
Figure 2 shows, in the format of the periodic chart, the elements
detected and not lost during ashing in all experiments. The symbols
in the individual square of each element, OL, C, and BL stand for
orchard leaves, coal, and beef liver, respectively, in which the
element was determined. In addition, some elements reported retained
in the literature determined by neutron activation analysis (4),
atomic absorption spectrometry, (5) or spectrophotometry (1) are
included. The numbers refer to references.
H
LI
NA
C
OL
BL
K
C
OL
RB
C
OL
BL
cs
c
OL
BL
FR
BE
MG
C
OL
BL
CA
C
OL
BL
SR
C
OL
BA
C
OL
BL
RA
SC
C
OL
BL
Y
LA
C
OL
BL
AC
Tl
c
OL
ZR
C
OL
HF
c
OL
V
C
NB
TA
C
CR
c
OL
BL
MO
OL
BL
w
c
MN
C
OL
BL
TC
RE
FE
C
OL
BL
RU
OS
CO
c
OL
BL
RH
IR
Nl
C
OL
PD
PT
CU
5
AG
4
AU
ZN
c
OL
BL
CO
5
HG
B
1
AL
c
OL
BL
GA
C
IN
TL
5
C
SI
GE
SN
PB
c
OL
N
P
AS
C
OL
SB
C
OL
BL
Bl
0
S
in -!_/
COUOm
TE
PO
F
CL
BR
1
AT
HE
NE
AR
KR
XE
RN
CE
C
OL
BL
TH
c
OL
BL
PR
PA
NO
U
PM
NP
SM
PU
EU
c
OL
BL
AM
GD
CM
TB
BK
DY
CF
HO
ES
ER
FM
TM
MD ,
YB
NO
LU
LW
1. J.W. Mair. Jr.. and H.G. Day. Anal Chem. 44 (1972) 2015. Mnlrtx was animal tissue.
4. O.BehneandPAMa'amba. Z. Anal Chem. 274 (1975) 195. Matrix was blood scrum.
5. B.B. Stafford. Proceedings 2nd Conf. Trace Subs, in Environmental Health (1968) Univ. of Mo.
Matrix was atmospheric paniculate.
Figure 2: Elements Retained During LTA
Although some of the results have been previously reported, we include
data here on many additional elements, including those particularly
susceptible to loss such as arsenic, selenium, and chromium, since they
are known to form compounds which are volatile at the temperatures of LTA.
54
-------
TABLE 2. PERCENT OF ELEMENT RETAINED (EXPERIMENTAL)
Coal Orchard Leaves Beef Liver
Sodium 98 100 10A
Magnesium 98 101 95
Aluminum 105 - 112
Potassium 101 106
Calcium 99 101 99
Scandium 100 101
Titanium 105 99
Vanadium 98
Chromium 99 106 94
Manganese 102 102 105
Iron 96 102 99
Cobalt 108 95 106
Nickel 99 99
Zinc 107 102 100
Gallium 91
Arsenic 110 92
Selenium 99 105
Rubidium 98 98 92
Strontium 98 106
Zirconium 104 108
Molybdenum 93 94
Antimony 104 108
Cesium 102 104
Barium , 105 109
Lanthanum 107 95 95
Cerium 99 98 100
Europium 98 - 94
Hafnium 96 98
Tantalum 109
Tungsten 97
Lead 106 • 97
Thorium 103 92
It is also significant to note that, equal in importance to loss of
trace elements, there was no pick-up of contamination during the ashing
process as would be evidenced by an increase in the amount of the element
after ashing. This is in spite of the fact that only very ordinary precautions
were taken to guard against contamination during cleaning of the borosilicate
glass ashing tube with ordinary laboratory detergent followed by rising
with water, hot dilute hydrochloric acid and finally deionized water and
drying in an oven at 100 °C.
55
-------
An application of immediate interest is those elements which upon
thermal neutron or photon activation yield radioisotopes with half-lives
in the range of 2-30 minutes and are not detectable in the typical biological
sample without separation. Since after ashing, biological samples, especially
animal tissue, are rapidly dissolved in dilute mineral acid, one may
consider radiochemical separations in the determination of magnesium,
aluminum, titanium, and vanadium in the case of thermal neutron activation
and potassium and iron in photon activation.
56
-------
REFERENCES
1. J.W. Mair, Jr., and H.G. Day, Anal. Chem. 44^ (1972) 2015.
2. V. Janzer, Private Communication.
3. C.E. Gleit and W.D. Holland, Anal. Chem. 34_, (1962) 1454.
4. D. Behne and P.A. Matamba, Z. Anal. Chem. 274, (1975) 195.
5. B.B. Stafford, Proceedings 2nd Conf. Trace Subs, in Environmental
Health (1968) Univ. of Missouri.
57
-------
SECTION 7
STABILITY OF ELEMENTAL COMPONENTS IN MULTI-TRACE ELEMENT WATER STANDARD
H.L. Rook, J.R. Moody, P.J. Paulsen and T.C. Rains
Analytical Chemistry Division
National Bureau of Standards
Washington, D.C. 2023A
58
-------
INTRODUCTION
The study of natural water samples which require trace element
compositional data poses a specially difficult analytical problem (1,2).
The extremely low concentrations of many toxic elements lead to all the
normal problems associated with ultratrace analysis. These problems are
compounded by the chemical mobility of the species being determined (3).
Also, unique problems are encountered in proper sampling and storage of
samples for laboratory analysis (4,5). The subject of valid sampling of
natural waters will not be discussed here although it must be emphasized
that improper sampling can invalidate any study of these waters.
Most analytical methods used for trace element analyses are relative
techniques; that is, the signals developed must be compared to those
obtained from standards of known compositions. Unfortunately, experience
has shown that pure, single element aqueous standards do not compensate
for interferences or changes in signal response caused by the analyte
matrix. In an effort to provide an adequate standard for trace metals
analysis in natural waters, the National Bureau of Standards is currently
certifying a mixed trace element in water Standard Reference Material
(SRM).
Two important considerations in the development of any standard are
the homogeneity of and stability of the certified species. In the case
of water standards, elemental homogeneity is easily achieved by careful
mixing of the bulk solution using a single large container. Long-term
stability, however, is a recognized problem for many elements at low
concentrations. Mineral acids have been employed as stabilizing agents
and are usually successful in the ppm range and above (1,6,7). However,
in the ppb concentration range, methods for long-term storage have not
been documented.
Two studies have been conducted to verify the long-term stability
of ten trace elements of wide interest to water chemists. The first
study was conducted with radioactive tracers and was designed to indicate
any elemental losses to container walls. This study was continued for
255 days and indicated acceptable stability for all elements included.
Following the initial study, a more comprehensive investigation was
carried out using simultaneous analyses by atomic absorption, neutron
activation, and isotope dilution spark source mass spectrometry. This
second study was designed to indicate any problems of concentration
elevation due to reagent blanks and leaching of trace elements from
contact materials. The results of this investigation indicated that
commercial grade polyethylene or Teflon containers were equally acceptable
59
-------
except for the element mercury. In the case of mercury, polyethylene
containers are very much more susceptible to contamination from external
mercury than are Teflon containers, because of higher permeability of
mercury through polyethylene.
The results of the stability studies indicated that the trace
elements of interest could be stabilized for more than one year using
0.5 Molar HNOs- With this assurance, a multitrace element water standard
containing twenty trace elements has been prepared.
EXPERIMENTAL
Initial stability studies were carried out using radiotracers of
five important elements, As, Se, Hg, Cd and Zn. A master solution was
prepared containing approximately 50 parts-per-billion (1 ppb = 10
g/g) of the elements of interest. The solution was stabilized with HN03
at a final concentration of 0.5 M HN03 and gold ion at a concentration
of 10 8 g Au/g water. The gold ion was added to stabilize mercury as
described previously (8). The final solution was stored in a one liter
Teflon bottle under normal laboratory conditions.
Concentration stability was determined by withdrawing 1 ml of the
standard solution and counting the solution at a fixed geometry with a
Ge(Li) detector coupled to 4096 channel pulse height analyzer. Individual
photopeak activities were quantified by the total peak area method and
decay corrected to the beginning of the experiment. Relative elemental
concentrations were obtained from the counting data for a period of 255
days. At this time, the errors in the decay corrections of the 73As
and the ^Hg became large and the experiment was terminated.
The tracer experiment gave information as to possible elemental
loss phenomenon. However, that type of experiment yielded no information
on potential blank problems, either from preparatory reagents or from
the storage containers. Thus, a concurrent study was conducted to
provide information on these problems. In this study, a trial solution
of a multiple element standard was prepared under clean room conditions
and analyzed over a 17-week period. The trial solution was stored in a
precleaned one liter Teflon and polyethylene bottles. The bottles were
cleaned and tested for blank using a previously described procedure (8).
To make the trial solution and also the final Standard Reference
Material, concentrated solutions of individual elements were prepared by
dissolving weighed quantities of spectrographically pure metals or
compounds of the elements of interest. By accurate dilution of these
standard solutions, a .test solution containing 18 trace elements plus
mercury, gold and the four matrix elements Na, K, Ca, and Mg was made.
In certain instances, most notably for Be, it was necessary to elevate
the concentration over the level normally found in fresh waters in order
to assure a reasonable chance for analysis by two independent methods.
Analysis of this preliminary sample by NAA and AAS indicated no potential
problems at these concentration levels.
60
-------
For the final solution a single 55 gallon polyethylene drum and
stirring paddle was cleaned in a manner similar to the storage bottles.
Carefully weighed amounts of specially purified water and HNOs (2) were
added to the 55 gallon drum with stirring. Three master solutions which
together contained all 24 elements were carefully prepared by weight
dilution of the original concentrated standard solutions. Aliquots of
these three dilute master solutions were then added to the water and
HN03 in the 55 gallon drum with constant stirring.
The resulting solution was stirred for several days and then to
ensure mixing vertically throughout the drum, aliquots were removed from
a spigot at the bottom and poured back into the top of the drum. Finally,
after complete mixing, the solution was transferred into 230 precleaned
one liter polyethylene bottles which were serially numbered. All of the
dissolutions, dilutions, and other manipulations were carried out in a
Class 100 clean laboratory using techniques designed to avoid contamination
of the samples. Selected samples were taken for certification analyses
and long-range stability studies.
RESULTS AND DISCUSSION
The initial stability study using radioactive tracers was continued
for a period of 255 days. In this study, the initial specific activity
of each isotope on day zero was defined as unit concentration, i.e.,
1.000 relative units. Changes in concentration were then determined by
a simple ratio of the specific activity at any time t, compared to that
at time zero. Initial data were taken at frequent time intervals to
detect any rapid changes in concentration. When no discernable changes
were noted, the time interval between data points was prolonged. After
250 days, the errors introduced by the half life correction of data
became unacceptably large for the shorter half lifed isotopes, thus the
experiment was terminated.
With completion of this experiment, the data set for each element
was analyzed by unweighted linear regression. Predicted final concentrations
and slopes were calculated with associated errors. The individual data,
predicted initial and final concentrations and slopes are given in
Table 1. Two hypotheses were tested to indicate elemental stability. First,
was to test equivalence between initial and final concentrations. In the
case of the predicted initial and final concentrations, the differences were
equal to or less than the standard deviations of this prediction, except in
the case of Zinc. Even in that case, the difference was small. The second
test was that of a nonzero slope for individual data sets. Again, only in
the case of Zinc was a significant nonzero slope observed and that was again
marginal.
In the case of Zinc, the positive slope gave a predicted final concentra-
tion which was slightly larger than the initial concentration. However,
a real increase in zinc-65 isotopic concentration is a physical impossibility.
The observed change is best attributed to either a small error in half
life correction of the zinc-65 data or simply a statistical fluctuation.
61
-------
Table 1. Trace element loss study using radioisotopes
Time in days Isotopic concentrations relative to day zero
73
As
75
Se
203
Hg_
109
Cd
65
Zn
0
9
14
22
35
79
94
255
1.000
1.006
1.041
0.997
1.027
1.080
0.977
1.000
1.012
1.041
0.991
1.007
1.000
1.049
0.981
1.000
0.999
1.026
1.031
1.008
0.983
0.997
1.0.32
1.000
0.997
0.965
0.976
0.938
0.973
0.984
1.000
1.012
1.032
1.033
1.036
1.012
1.031
1.057
Predicted Con.
Day 0
Std. Dev. of
Con. at Day 0
Predicted Con.
Day 255
Std. Dev. of
Predicted Con.
Slope
1.025
±0.017
1.001
±0.034
-0.000096
1.016
±0.011
0.992
±0.022
-0.00009
Std. Dev. of
Slope ±0.00017 ±0.00011
P(F) .59 .42
(P(F) significant if <0.05).
1.006
±0.009
1.021
±0.017
+0.00006
±0.00008
.50
0.978
±0.011
0.971
±0.021
-0.00003
±0.00010
.79
1.017
±0.006
1.055
±0.013
+0.00015
±0.00006
.052
The tracer experiment indicated that the five elements studied could be
stabilized at trace concentrations in water for at least one year.
A second study was initiated to provide information on possible
blank problems as well as additional information on stability. In this
study, a trial solution of a multiple element standard was prepared
under clean room conditions and analyzed twice over 17-week period. The
analytical results were obtained by three independent analytical methods,
neutron activation (NAA), isotope dilution spark source mass spectrometry
(ID-SSMS) and by atomic absorption spectroscopy (AAS). The individual
methods used where those developed for SRM certification over the past
ten years. The data obtained by the analyses were compared with elemental
concentration obtained by calculation from dilution of the master spike
solution. Data were also compared among analytical techniques for
consistency. The results of the analytical study are given in Table 2.
The analytical data presented are mean values obtained on a replicate
set of two to four analyses by a given technique. Error levels are
roughly indicated by the number of significant figures presented but
62
-------
insufficient data were obtained to statistically estimate analytical
errors. The comparison of data among techniques is a valid indicator of
changes in elemental concentrations.
Table 2. Trace element stability study using
independent analyses
Element
Concentration in yg/g
Calculated Analytical Initial
Concentration Technique Concentration
17-Week
Concentration
Zn
Cu
As
Pb
Mn
Cr
Be
Cd
Se
Hg
NAA
4.89 AAS
SSMS
NAA
0.807 AAS
IDMS
0.0937 NAA
AAS
0.0469 AAS
IDMS
0.0418 NAA
AAS
0.0473 AAS
SSMS
0.0397 AAS
Spectro.
NAA
0.0137 AAS
SSMS
NAA
0.0101 AAS
SSMS
0.00184 NAA
AAS
4.96
4.96
5.16
0.814
0.851
0.845
0,0603
0.064
0.0512
0.0468
0.0412
0.0400
0.051
0.0471
0.0416
0.0406
0.0146
0.0150
0.0185
0.0105
0.0112
0.0098
0.0018
0.0018
5.08
5.08
0.849
0.856
0.848
0.0630
0.060
0.0459
0.0465
0.0470
0.0455
0.053
0.0465
0.0416
0.0409
0.0149
0.0145
0.011
0.0118
0.0019
0.0017
63
-------
+ NAA = Neutron Activation Analysis
AAS = Atomic Absorption Spectrometry
SSMS = Isotope Dilution Spark Source Mass Spectrometry
IDMS = Isotope Dilution Thermal Source Mass Spectrometry
Spectro = Spectrofluorimetry
The results of the analytical study were consistent with those of
the tracer study. No indication of trace element loss was observed
during the 17-week period. However, two interesting effects did occur.
First it is obvious from the data that for arsenic, the calculated
concentration was grossly different from the measured concentrations.
There are two possible causes for this observed occurrence. First, it
is possible that a large, rapid decrease in arsenic concentration did
occur either by precipitation of the arsenic or by wall adsorption.
However, it is difficult to imagine such a mechanism for an immediate
large loss of arsenic which would not continue to be observed for a
longer period of time. No such continued change in concentration wa.s
observed in the 17-week analysis. A more likely explanation is that the
calculated concentration was in error due either to a mistake in weighing
or volumetric dilution.
A second, and more subtle effect was observed for manganese where a
slight increase in concentration over the 17-week period was observed.
This increase was noted both by neutron activation and by atomic absorption.
The effect is small, however, and subsequent analyses have,not indicated
a continued increase in manganese concentration. It is most probable
that a small amount of manganese was initially leached out of the containers,
even though those containers had been subjected to rigorous cleaning.
The manganese concentration has not been observed to increase significantly
in subsequent analyses. The element manganese will be carefully checked
for an extended period of time in the certified Standard Reference
Material.
The third observation from the analytical study was that some
analytical procedures were in error. Even though the procedures used
were considered to be the best available, more analytical research was
needed to accurately analyze all elements at the concentrations selected
for the standard. This problem is particularly evident with the cadmium
data as determined by isotope dilution spark source mass Spectrometry
and with the copper data as determined by neutron activation. Research
into the respective analytical methodologies was conducted and significant
improvement was noted in the 17-week analyses.
The two studies on trace element stability indicate that a mixed
trace element in water standard can be stabilized with 0.5 M HNOs
in conjunction with trace gold ion. The studies illustrate the type of
investigations which must be carried out before trace solution standards
can be assumed to give valid data. A final solution has now been prepared,
bottled, and analyzed for the certification of SRM 1643 Trace Elements
in Water.
64
-------
REFERENCES
1. A. Rattonetti, 1976, "Stability of Metal Ions in Aqueous Environmental
Samples", NBS Special Publication 422, National Bureau of Standards,
Washington, D.C. 20234; 633-648.
2. T. J. Murphy, 1976, "The Role of the Analytical Blank in Accurate
Trace Analysis", ibid; 509-514.
3. P. Benes and E. Stiennes, 1975, "Migration Forms of Trace Elements
in Natural Fresh Water and the Effects of the Water Storage, Water
Research, 9j 741-749.
4. S.H. Harrison, P.O. LaFleur, and W.H. Zoller, 1975, "Evaluation of
Lyophilization for the Preconcentration of Natural Water Samples
Prior to Neutron Activation Analysis", Analytical Chemistry, 47;
1685-1688.
5. J.R. Moody and R.M. Lindstrom, 1977, "The Cleaning, Analysis and
Selection of Containers for Trace Element Samples", Analytical
Chemistry, in press.
6. D.E. Robertson, 1976, "Analytical Chemistry of Natural Waters", NBS
Special Publication 422, National Bureau of Standards, Washington,
D.C. 20234; 805-836.
7. E. Goldberg, Ed., 1972, "Baseline Studies of Pollutants in the
Marine Environment and Research Recommendations", The IDOE Baseline
Conference, Scripps Institution of Oceanography, LaJolla, California.
8. J.R. Moody, P.J. Paulsen, T.C. Rains and H.L. Rook, 1976, "The
Preparation and Certification of Trace Mercury in Water Standard
Reference Materials", NBS Special Publication 422, National Bureau
of Standards, Washington, D.C. 20234.
9. E.G. Kuehner, R. Alvarez, P.J. Paulsen and T.J. Murphy 1972, "Production
and Analysis of Special High-Purity Acids Purified by Sub-Boiling
Distillation," Analytical Chemistry, 44_, 2050-2056.
65
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/1-78-015
3. RECIPIENT'S ACCESSION>NO.
4. TITLE AND SUBTITLE
Evaluation and Research of Methodology for the
National Environmental Specimen Bank
5. REPORT DATE
February 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
I.E. Gills, H.L. Rook and P.O. LaFleur, Editors
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Institute for Materials Research
National Bureau of Standards
Washington, D.C. 20234
10. PROGRAM ELEMENT NO.
1AA601
11. CONTRACT/GRANT NO.
EPA-IAG-D5-0568
12. SPONSORING AGENCY NAME AND ADDRESS
Health Effects Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park. N.C. 27711
13. TYPE OF REPORT AND PERIOD COVERED
RTP..NC
14. SPONSORING AGENCY CODE
EPA-600/11
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report is a summary of analytical methodologies developed or adopted
NBS to insure proper procedures for sampling, storage and analysis of
biological and environmental type matrices. Protocols established in these
procedures will insure that the measurement of an analyte can be made within
known and/or required uncertainty levels and that samples can be stored for
retrospective analysis.
The contributions in this report are divided in seven sections which
includes analytical methods for trace element analysis and techniques for sampling
a.nd storage of specific matrices.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
standards
chemical analysis ,
biological extracts
specimen bank
07 C
05 B
06 F
3. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS {ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
75
20. SECURITY CLASS (This page)
22. PRICE
EPA Form 2220-1 (9-73)
66
------- |