United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
EPA-600/1-79-017
May 1979
Research and Development
The National
Environmental
Specimen Bank
Research Program
for Sampling,
Storage, and
Analysis
-------
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-600/1-79-017
May 1979
THE NATIONAL ENVIRONMENTAL SPECIMEN BANK RESEARCH PROGRAM
by
T. E. Gills, H. L. Rook, R. A. Durst, Editors
Center for Analytical Chemistry
National Bureau of Standards
Washington, D. C. 20234
Contract No. EPA-78-D-X0105-1
Project Officer
G. M. Goldstein
Clinical Studies Division
Health Effects Research Laboratory
Research Triangle Park, North Carolina 27711
HEALTH EFFECTS RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
This report has been reviewed by the Health Effects Research Laboratory,
U.S. Environmental Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the views and policies
of the U.S. Environmental Protection Agency, nor does mention of trade names
or commercial products constitute endorsement or recommendation for use.
n
-------
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.
F. Gordon Hueter, Ph.D.
Director
Health Effects Research Laboratory
iii
-------
ABSTRACT
This work was performed under a joint NBS/EPA research program to develop
state-of-the-art protocols for sampling, storage, and analysis of biological
and environmental-type matrices. This report is a compilation of research
papers and/or efforts describing developed or adopted procedures for retrospec-
tive analysis of biological and environmental samples. Preliminary protocols
for sampling, sample handling, and sample storage are given for human liver
autopsy tissue in addition to methods for the accurate measurement of selected
toxic elements in biological and environmental materials. Analytical methods
employed were neutron activation analysis (NAA), polarography, and isotope
dilution spark source mass spectrometry (IDSSMS).
This report is submitted in partial fulfillment of EPA Interagency agree-
ment EPA-78-D-X0105 by the National Bureau of Standards for the 1978 contract
year.
IV
-------
CONTENTS
Page
Disclaimer ii
Forward iii
Abstract iv
List of Figures vi
List of Tables vii
Acknowledgments viii
Section 1 - Development of a Preliminary Protocol for Sampling,
Sample Handling, and Long-Term Storage of Human
Liver 1
Section 2 - Research on Freezer Storage 14
Section 3 - Container Materials for the Preservation of Trace
Substances in Environmental Materials 16
Section 4 - Design and Construction of the Pilot Bank Facility . . 22
Section 5 - Organic Mercury in Tissues 25
Section 6 - The Determination of Trace Elements in New Food Grain
SRM's Using Neutron Activation Analysis 27
Section 7 - Cadmium Analysis by Radiochemical Neutron Activation
Analysis 32
Appendices
I. Simultaneous Determination of Arsenic, Antimony, Cadmium,
Chromium, Copper, and Selenium in Environmental Material by
Radiochemical Neutron Activation Analysis 37
II. The Quantitative Determination of Volatile Trace Elements in
NBS Biological Standard Reference Material 1569, Brewers
Yeast 40
III. Chemical Preparation of Biological Materials for Accurate
Chromium Determination by Isotope Dilution Mass Spectrometry . 50
IV. The Determination of Zinc, Cadmium and Lead in Biological
and Environmental Materials by Isotope Dilution Mass
Spectrometry 54
v
-------
List of Figures
Page
Section 4:
Figure 1. Specimen Bank Laboratory/Storage Facility 24
VI
-------
List of Tables
Page
Section 2:
Table 1. Analyses of a Stored Bovine Liver, Storage and
Analysis Protocol Tests 15
Section 3:
Table 1. Annual Rate of Water Loss from Plastic Containers .... 17
Table 2. Trace Elements in Plastics Determined by Neutron
Activation Analysis 18
o
Table 3. Impurities Leached from Plastic Containers (ng/cm~). ... 19
Section 6:
Table 1. Irradiation Conditions and Data Used for Instrumental
Analysis 28
Table 2. Irradiation Conditions and Data Used for Radiochemical
Activation Analysis 29
Table 3. The Determination of Trace Metals in Rice and Wheat
Flour in yg/gm 30
Section 7:
Table 1. Cadmium in Various SRMs 34
Table 2. Copper in Various SRMs 35
Table 3. Sub-Bituminous Coal SRM 1635 35
Table 4. Bovine Liver SRM 1577 36
Table 5. Orchard Leaves SRM 1571 35
-------
ACKNOWLEDGMENTS
The authors of this report wish to acknowledge their appreciation to the
Environmental Protection Agency for its funding. We especially want to thank
the Project Officer, Dr. George M. Goldstein, Chief of the Clinical Pathology
Branch for his consultations and program guidance.
viii
-------
SECTION 1
DEVELOPMENT OF A PRELIMINARY PROTOCOL FOR SAMPLING,
SAMPLE HANDLING, AND LONG-TERM STORAGE OF HUMAN LIVER
by
E. June Maienthal
INTRODUCTION
In response to the increasing concern with environmental pollution, the
National Bureau of Standards (NBS) in collaboration with the Environmental
Protection Agency (EPA) has developed a preliminary protocol for sampling,
sample handling and long-term storage of human livers, which are to be moni-
tored for trace element and trace organic compositions for the Pilot National
Environmental Specimen Bank which will be located at NBS. The criteria for
arriving at these guidelines will be discussed, and the preliminary guidelines
up to the stage of subsampling will be given.
Considerations
From an earlier literature survey (1), it was obvious that much of the
work concerning environmental sampling, sample handling, and long-term storage
is of questionable value owing partially to sources of contamination from
various steps in the sampling and sample processing or to losses of some of
the trace constituents through adsorption, volatility, etc. It was also
obvious from a survey of existing environmental collections that few of the
collections surveyed would be of sufficient value for environmental trace
inorganic and organic determinations (2). It was necessary, therefore,
prior to the establishment of the Pilot National Environmental Specimen
Bank, to set up a valid working protocol.
The rationale and plans for the National Environmental Specimen Bank
(NESB) have been described by Goldstein, where he states "... regardless of
the outcome of the NESB, the methodology protocol development would provide
the scientific community with state-of-the-art standardized protocols for sam-
ple collections in a variety of ecologically important materials. The cost
benefit of this alone is incalculable." (3).
The protocol to be described was arrived at as the result of work con-
ducted at NBS, a past literature survey (1), and additional literature inves-
tigated and consultations with other workers in the field, including chemists,
biochemists, pathologists, cryobiologists, physicians, and equipment suppliers.
The necessity of obtaining and storing a sample valid for both trace organic
and inorganic constituents requires very rigid control of the sampling mater-
ials and methods, shipping methods, storage containers and storage conditions.
-------
Although shipping and storage are the last items in the protocol, they
will be discussed first in order to explain the reasons for certain other steps
of the procedure. It was generally concluded that it would be necessary to
freeze the specimen as soon as possible after sampling for long-term banking.
Frozen storage is usually done at about -20 °C (approximate home freezer tem-
perature) , -80 °C (approximate dry ice temperature), or at -196 °C (liquid
nitrogen temperature). At -20 °C, however, samples often may not be frozen,
and at -80 °C, molecular or enzymatic activity may still occur (4,5). Al-
though this might not affect the stability of the inorganic constituents, some
speciation and organic changes could occur. Some chemical changes are re-
ported to occur as low as -130 °C (5), and some physical changes such as
devitrification are reported to take place at temperatures substantially lower
than -130 °C (6). Meryman had earlier reported that no biochemical activity
should take place below -130 °C (7), but states that on the basis of Dowell
and Renfret's work (6), further study of this activity at low temperatures
should be investigated (7a). Mazur has reported that at -196 °C, no aqueous
reaction occurs and that a tissue sample should be preserved, unchanged
indefinitely at that temperature (8) . Cravalho also states that to arrest
biochemical and physical processes, the lower the storage temperature the
better, and that there is no lower temperature limit for storage of biomater-
ials (8a). It would appear therefore, that liquid nitrogen would be an ideal
storage as well as shipping medium, having the additional advantage of being
relatively inexpensive and readily available. Laessig, et al. describe some
of the earlier uses of liquid nitrogen for storing and shipping biological
samples in tanks by air freight with no difficulty (9). These tanks were used
by Laessig, et al. for epidemiclogical serum studies in remote field areas and
were found adaptable under the most severe circumstances. The tanks are
commercially available from several manufacturers and are being used routinely
for tissue shipping and storage by groups such as the Tissue Bank of the Naval
Medical Research Institute. The use of dry ice has not been as satisfactory
for shipping, in part because of frequent failures and thawing of samples, and
also because of possible pH changes of the sample owing to diffusion of carbon
dioxide into the container (4). Because of cell rupture and tissue leakage
during thawing, a sample cannot be salvaged for many analytical purposes by
refreezing once it has been thawed. Omang and Vellar have described the
concentration gradients occurring in biological samples during freezing and
thawing (10).
The choice of donor specimen must also be under some restrictions. For
the pilot study and to check the preliminary protocol, normal livers which show
no erratic concentrations of the constituents to be determined are desired.
For this reason, certain donors are to be excluded, such as those: having
liver weights less than about 1000 grams; stored at temperatures above 4 °C;
previously frozen; deceased more than 24 hours; a history of alcoholism, gross
sepsis, viral hepatitis, tuberculosis, cirrhosis, liver carcinoma, chronic
circulatory failure or congestion, chemical or drug overdose or exposure; or
previously embalmed.
Ideally the sample should be taken immediately after donor death since
significant enzymatic changes occur with time (11) , but at present this would
not be generally possible; so on a practical basis an upper limit of 24 hours
has been set.
-------
Ideally, the autopsy should be performed in a laminar-flow clean room
atmosphere, because the chances of contamination from the typical autopsy room
environment are very great. Again, this is not at present a practicality;
hence, as few manipulations as possible are to be done prior to sealing the
sample in the cleaned container.
Most autopsies must be performed in a sterile atmosphere requiring the
use of sterile surgeon's gloves which are typically latex, powdered both
inside and out. This type of glove would be prone to introduce trace element
contaminations, such as zinc from the latex itself, and other trace elements
from the powder. A producer of sterile, clean-room packaged, talc-dust free,
polyvinyl chloride gloves has been located, and these gloves have been found
acceptable for use at the autopsy facilities so far contacted. It must be
remembered, however, that there may be a chance of trace organic contamina-
tions of the liver surface from the plastic or plasticizers used in the PVC,
as will be discussed later. This organic contamination could occur with latex
gloves also. For the first samples taken, relatively standard surgical instru-
ments are to be used — unrusted, cleaned steel for the liver removal, and an
unrusted, cleaned, hardened carbon steel knife for slicing the left lobe into
two equal sections. It must be remembered that possible surface contamination
from the knives may occur, as discussed previously (1), particularly as shown
by Versieck, Speecke and coworkers (12,13), and by Maletskos, et al. (13a).
A number of other types of less-contaminating implement materials are being
considered and will be discussed later.
The liver section should be rinsed with pure distilled water to remove
gross external contamination, put into its container, sealed, and immediately
frozen in liquid nitrogen.
One problem arising from freezing a section of liver complete with its
blood and extracellular fluid is that the amount of fluid contained can vary
greatly in a normal human being, depending on time and a variety of physio-
logical factors, thus causing an apparent difference in compositions from
sample to sample because of varying amounts of fluid present at time of death.
The fluid can differ markedly in composition from that of the tissue. This
has been shown by LeBaron and coworkers by analysis of human kidney tissues
and fluids (13b). lyengar and Kasperek have used a very promising approach
which will be investigated here and possibly incorporated into the protocol
(13c). They rinse the liver sections three times, squeezing the sections
gently between clean polyethylene sheets (Teflon would probably be used here)
in between rinsing processes, thereby eliminating most of the blood. The sec-
tions were then frozen in liquid nitrogen and gently pressed between two
Perspex plates. This results in the fracture into small pieces which could be
picked out with plastic forceps, separating the remaining blood vessels and
homogenizing the rest by the brittle fracture technique using a special Teflon
vessel and a Teflon ball with a metal core (14) .
A very important item of concern is the choice of container and this is
discussed in more detail in another section of this report. This also was
discussed at length previously (1) and in the Proceedings of the 7th Materials
Research Symposium (14a). It would be desirable to have one type of sample
container suitable for both organic and inorganic constituents. Although some
plastics can be cleaned sufficiently for trace element analysis (1,15), the
contamination for organic constituents and migration of plasticizers has been
-------
often shown (1,16,17,18). Also, most of the plastics will not withstand
liquid nitrogen temperatures without breaking. One lot of polypropylene cryo-
genic tubes also showed the presence of sodium, aluminum, chloride, and tita-
nium (18a). Properly cleaned Pyrex containers are suitable for organic con-
stituents, but are notoriously bad for trace elements through contamination
and/or adsorption (1,19). The latter reference also found significant zinc
contamination from polypropylene containers (19).
The polyfluorinated hydrocarbons such as Teflon, FEP or PFA, after appro-
priate cleaning appear to be the most suitable for both organic and inorganic
constituents. They will also withstand liquid nitrogen temperatures if not
subjected to too severe physical stresses.
A preliminary experiment was done at NBS freezing pork liver in a heat-
sealed 5-mil Teflon FEP bag in liquid nitrogen with promising results (19a).
The bags are evacuated immediately prior to sealing to increase the sample
stability (20,21). The packaged sample should be put in a forming cylinder
prior to freezing in liquid nitrogen in order to obtain a more compact, easily
handled package for shipping and storing.
The liver sections are weighed separately in order that weight loss or
gain can be checked after they are received at NBS and periodically during
storage. One problem with Teflon is that it is porous to gases, and the
samples may be subject to moisture loss or gain. If this should prove to be a
problem, a vapor barrier such as aluminized Mylar or glass may be required in
the protocol.
The liquid nitrogen shipping containers can also be used as storage con-
tainers prior to shipping the livers to NBS. These containers have an absor-
bent around the inner walls, so that during shipping only vapor-phase nitrogen
need be present. The type which will probably be used for the first experi-
ments has a static holding time of 11 days after the absorbent is saturated.
The two halves of the left liver lobe are to be sent back to NBS in
order that different subsampling and storage techniques may be investigated.
One section will be stored untouched in the event that the subsampling pro-
cedures (as yet unfinalized) should in some way alter or contaminate the other
section. It must also be emphasized that the sampling protocol is preliminary
and will necessarily change as unforeseen problems are encountered and
improved methods arise. It is now thought that the subsampled section will be
stored in approximately 8 mL PFA bottles (which may also require an outside
vapor barrier). It is possible that glass might be used for the samples for
the organic determinations.
A major problem remaining is how to subsample and homogenize the liver
section without contaminating it. Various methods have been suggested, in-
cluding Waring blenders, meat grinders, ultrasonic tissue homogenizers, food
processors, cryogenic grinding and brittle fracture. Some preliminary studies
have been made at NBS on subsampling, homogenization, analysis by neutron
activation, freezing, storing, and reanalysis of beef liver (21a,21b). A
number of types of sampling tools have been used which seem to offer much less
chance of contamination than the steel knives mentioned earlier. lyengar and
Kasperek used a specially prepared Suprasil-quality-quartz knife for slicing
-------
the liver, and used the brittle fracture technique, as mentioned earlier, at
liquid nitrogen temperatures utilizing a specially prepared Teflon container
and Teflon ball with a metal core (13c) . Lievens, et al. also used an ultra-
clean spec-pure quartz tube to obtain liver subsamples (22). Thompson and
Bankston studied contamination arising from grinding and sieving devices made
from different types of materials (23). They found that a boron carbide
mortar introduced little or no contamination (except for boron) and that
stainless steel or brass sieves introduced appreciable levels of cobalt,
copper, iron, manganese, nickel, lead, tin, and zinc. Maletskos and coworkers
used a stainless steel knife to cut a frozen segment of tissue, then a special
high purity boron nitride knife to shave the surfaces which had been in contact
with the stainless steel (13a). Folsom developed several different types of
plastic sampling tools for frozen tissue sampling (24). These, however, might
not be suitable for trace organic samples. Another interesting sampling
possibility involves the use of the laser. Hislop and Parker used a carbon
dioxide laser for cutting bone with no trace element contamination (25).
Auth, Doty, and coworkers have developed a "laser blade" or "laser scalpel"
using high power argon laser radiation transported via a low-loss optical
fiber into a sharp transparent quartz knife (26). This device has proved very
useful for operations in which massive bleeding is a problem. This device has
the effect of cauterizing the incision and sealing the blood vessels. Of
course for the trace constituent analysis the several millimeters of cauter-
ized tissue would need to be removed by an implement which would be noncon—
taminating for the constituents of interest for the particular determination.
A number of additional references concerned with problems involved in
trace analysis of tissue or biological matrices are listed (27-44).
It must be emphasized again that the attached protocol is preliminary and
will change as improved methods are developed.
-------
LIVER SAMPLING PROTOCOL
I. DEMOGRAPHIC DATA
a. Record autopsy identification number and NBS number on label pro-
vided and attach to sample bag (Note 1).
b. Collect and include donor information with shipment and send a
copy under separate cover. Donor information should contain as a
minimum: the autopsy identification number; the NBS number; date and
time of death; date and time of autopsy; weight of whole liver and
separate left lobe sections; donor height; weight; age; sex; and
ethnic group (Note 2).
II. SAMPLE HANDLING
a. Sterile, cleaned PVC gloves will be provided and should be used for
liver removal and handling. Extreme precaution must be taken
throughout the autopsy procedure to reduce the risk of contaminating
the liver sample. Contamination may result during the autopsy from
the donor, the individual(s) performing the autopsy, the atmosphere
and/or the surgical instruments.
III. SAMPLING
a. The liver is identified and excised with instruments supplied by NBS.
The liver should be removed as early as is feasible after death (but
not later than 24 hours) and as close to the beginning of the autopsy
as normal routine permits (Note 3).
The liver will not at any time during the autopsy procedure be placed
on any surface other than cleaned Teflon sheets provided by NBS.
b. The liver is placed on an inclined surface (to provide rapid drain-
age) covered with a Teflon sheet. Washing is accomplished by pouring
approximately 250 mL of distilled water (Note 4) over the liver sur-
face. The liver is then turned over and the washing procedure
repeated.
c. Excess water is allowed to drain off the liver prior to weighing on
another Teflon sheet.
d. Left lobe is dissected with the knife provided by NBS.
e. Left lobe is cut into equal halves through an axis of symmetry with
the knife provided.
-------
f. A section is taken from a cut surface of the left lobe (with the
knife supplied) for the preparation of a histological slide. A
slide and copy of the pathological findings is sent under separate
cover as per Ib above.
g. Each liver section is immediately placed in separate Teflon bags
provided by NBS and weighed separately on Teflon sheet as in IIIc.
h. Double heat seal each bag (with the twin heat sealer supplied by
NBS) to within about 1 inch of one side. Insert a precleaned
Teflon tube, attached to a hand vacuum pump (both supplied by NBS),
into the unsealed corner of the bag. Evacuate each bag and double
heat seal across the remaining open corner while maintaining the
vacuum.
i. The sealed Teflon bags are then placed in the Teflon cylinders, pro-
vided by NBS, and frozen by immersion for at least 15 minutes in
liquid nitrogen contained in a plastic Dewar supplied by NBS
(Note 5).
j. Remove the Teflon bag from the cylinder and inspect the bag for
obvious damage. If loss of sample integrity is encountered, both
liver sections should be discarded.
k. Prefill the shipping container with liquid nitrogen so that there
is at least 4 inches of standing liquid. This may be estimated by
using a dipstick. The liquid nitrogen level must be maintained
(Note 6).
1. Transfer the Teflon-packaged frozen liver sections to the shipping
container. Check and maintain the liquid nitrogen level as directed
in Illk above until ready for shipment.
m. Store liver specimens no longer than 10 days or until 4 lobes
(8 sections) have been collected.
n. Draw off excess liquid nitrogen, cap securely, and ship to the
National Bureau of Standards by a carrier (to be designated later).
Include with the shipment a copy of the information listed in la and
Ib, as well as the Restricted Article Statement required by the CAB.
Copies of information from la and Ib, along with the histological
slide should also be sent under separate cover (mentioned in Ib,
Illf above) (Note 7).
-------
NOTES FOR PROTOCOL
1. Polyester tape, labels, and dispenser are to be supplied by NBS. Labels
are to be affixed to the sample bag by wrapping tape completely around
bag and overlapping ends one-half inch.
2. At a later date in this protocol, additional donor information will be
sought for a specified number of samples. This will include: smoking
and drinking habits, socioeconomic status, residential environment, and
employment history.
3. Liver samples are to be excluded if donors have: a total liver weight of
less than 1000 grams (35 oz); livers which have been ruptured or punc-
tured; been stored at temperatures above 4 °C, or have been previously
frozen; had a history of alcoholism, drug addiction, or long-term drug
therapy; had gross sepsis, viral hepatitis, tuberculosis, cirrhosis,
liver carcinoma; had a history of chronic circulatory failure or conges-
tion; died due to chemical overdose or exposure; or embalmed donors.
4. Water used in washing the liver will be provided by NBS or obtained from
a local source that has been previously analyzed by NBS and shown to be
of acceptable quality.
5. Precautions should be observed when working with liquid nitrogen. Liquid
nitrogen should not be stored in sealed containers. When transporting by
elevator, personnel should not accompany Dewar because of the possibility
of elevator and/or Dewar failure. Personnel handling liquid nitrogen are
cautioned to wear boots, cuffless trousers, non-absorbent apron, loose,
insulating gloves and face shields.
6. The cryogenic shipping container weighs 27 kg (59 Ibs) empty. With the
absorbent filled with liquid nitrogen, the weight increases to 38 kg (84
Ibs). The container requires approximately 14 liters to fill the absor-
bent. The static evaporation rate is approximately 1.2 liters/day.
7. A Government Bill of Lading (GBL) will be supplied to cover shipping
expenses. Torms for the Restricted Article Statements will be provided by
NBS. These forms must be included with each shipment.
-------
REFERENCES
1. Maienthal, E. J., and D. A. Becker. A Survey of Current Literature on
Sampling, Sample Handling, and Long-Term Storage for Environmental
Materials. NBS Tech. Note 929, Oct. 1976.
2. Becker, D. A., and E. J. Maienthal. Evaluation of the National Environ-
mental Specimen Bank Survey. EPA-600/1-77-015, U. S. Environmental
Protection Agency, 1977.
3. Goldstein, G. M., Plan for a National Environmental Specimen Bank.
EPA-600/1-78-022, U. S. Environmental Protection Agency, March 1978.
4. Strong, D. M. Tissue Bank Division, Naval Medical Research Institute,
Bethesda, MD 20014. Private communication.
5. Peterson, Jr., W. D., and C. S. Stulberg. Freeze Preservation of
Cultured Animal Cells. Cryobiol., 1:80, 1964.
6. Dowell, L. G., and A. P. Renfret. Low Temperature Forms of Ice as
Studied by X-Ray Diffraction. Nature, 188:1144, 1960.
7. Meryman, H. T. Mechanics of Freezing in Living Cells and Tissues.
Science, 124:515, 1956.
7a. Meryman, H. T. Review of Biological Freezing, In: Cryobiology,
H. T. Meryman, ed. Academic Press, London and New York, 1966. pp. 72.
8. Mazur, P. Preservation of Embryos and Organs. Talk presented at the 2nd
Annual Meeting of the American Association of Tissue Banks, May 15-18,
1978, Boston, Mass.
8a. Cravalho, E. G. The Cryopreservation of Living Cells. Technology Review,
Oct./Nov., 1975. pp. 31.
9. Laessig, R. H., F. P. Pauls, T. A. Schwartz. Long-Term Preservation of
Serum Specimens Collected in the Field for Epidemiological Studies of
Biochemical Parameters. Health Lab. Sci., 9:16, 1972.
10. Omang, S. H., 0. D. Vellar. Concentration Gradients in Biological Samples
during Storage, Freezing, and Thawing. Anal. Chem., 269:177, 1974.
11. Barker, E. D. Post-Mortem Loss of Hepatic Microsmal Mixed Function
Oxygenases. Res. Comm. in Chem. Path, and Pharm., 7:321, 1974.
12. Versieck, J., A. Speecke, J. Hoste, and F. Barbier. Trace Contaminations
in Biopsies of the Liver. Clin. Chem., 19:472, 1973.
-------
13. Speecke, A., J. Hoste, and J. Versieck. Sampling of Biological Material,
In: Accuracy in Trace Analysis: Sampling, Sample Handling, Analysis,
Vol. I, P. D. LaFleur, ed. Proc. 7th Materials Res. Symp., NBS Special
Publication 422, August 1976. pp. 299.
13a. Maletskos, C. J., M. D. Albertson, J. C. Fitzsimmons, M. R. Masureaker,
and Chung-Wai Tong. Sampling and Sample Handling of Human Tissue for
Activation Analysis. In: Trace Substances in Environmental Health — IV,
pp. 367. D. D. Hemphill, ed. Proc. of U. of Mo. 4th Annual Conf., 1970.
13b. LeBaron, G. J., W. H. Cherry, and W. F. Forbes. Studies of Trace-Metal
Levels in Human Tissues — IV. The Investigation of Cadmium Levels in
Kidney Samples from 61 Canadian Residents, p. 44. In: Trace Substances in
Environmental Health — XI. D. D. Hemphill, ed. Proc. of U. of Mo. llth
Annual Conf. 1977.
13c. lyengar, G. V., and K. Kasperek. Application of the Brittle Fracture
Technique (BFT) to Homogenize Biological Samples and Some Observations
Regarding the Distribution Behavior of the Trace Elements at Different
Concentration Levels in a Biological Matrix. J. Radioanal. Chem.,
39:301, 1977.
14. lyengar, G. V. Homogenized Sampling of Bone and other Biological Mate-
rials. Radiochem. Radioanal. Letters, 24:35, 1976.
14a. LaFleur, P. D., ed. Accuracy in Trace Analysis: Sampling, Sample Hand-
ling, Analysis. Proc. 7th Materials Res. Symp., Vol. I,II. NBS Special
Publication 422, August 1976.
15. Moody, J. R., and R. M. Lindstrom. The Cleaning, Analysis, and Selection
of Containers for Trace Element Samples. In: The National Environmental
Specimen Bank, pp. 19. H. L. Rook and G. M. Goldstein, eds. Proceedings
of the Joint EPA/NBS Workshop on Recommendations and Conclusions on the
National Environmental Specimen Bank held at the National Bureau of
Standards, Gaithersburg, Md., Aug. 19-20, 1976. NBS Special Publication
501, February 1978.
16. Missen, A. W., and S. J. Dickson. Contamination of Blood Samples by
Plasticizer in Evacuated Tubes. Clin. Chem., 20:1247, 1974.
17. Daun, H., and S. G. Gilbert. Migration of Plasticizers from Polyvinyl-
chloride Packaging Films to Meat. J. Food Sci., 42:561, 1977.
18. Jaeger, R. J., and R. J. Rubin. Migration of a Phthalate Ester Plasti-
cizer from Polyvinylchloride Blood Bags into Stored Human Blood and Its
Localization in Human Tissues. New bngl. J. Med., 287:1114, 1972.
18a. Lindstrom, R. M., National Bureau of Standards, Private Communication,
19. Subramanian, K. S., C. L. Chakrabarti, J. E. Sueiras, and I. S. Maines.
Preservation of Some Trace Metals in Samples of Natural Waters. Anal.
Chem., 50:444, 1978.
19a. Harrison, S. H., National Bxireau of Standards, Private Communication.
10
-------
20. Duncan, W. Clean Room Products, Bay Shore, NY. Private Communication.
21. Seidman, S. C., et al. Effect of Degree of Vacuum and Length of Storage
on the Physical Characteristics of Vacuum Packaged Beef Wholesale Cuts.
J. Food Sci., 41:732, 1976.
21a. Gills, T. E., S. H. Harrison, and H. L. Rook. Research on Freezer
Storage. In: Section 2, this publication.
21b. Harrison, S. H., and T. E. Gills. Investigation of Sampling and Storage
Techniques for Environmental Specimen Banking Using Neutron Activation
Analysis. Talk presented at the 5th Symposium on the Recent Developments
in Neutron Activation Analysis, July 1978, Oxford, England.
22. Lievens, P., J. Versieck, R. Cornells, and J. Hoste. The Distribution of
Trace Elements in Normal Human Liver Determined by Semi-Automated Radio-
chemical Neutron Activation Analysis. J. Radioanal. Chem., 37:483, 1977.
23. Thompson, G., and D. C. Bankston. Sample Contamination from Grinding and
Sieving Determined by Emission Spectrometry. Applied Spectroscopy,
24:210, 1970.
24. Folsom, T. R. Nonmetallic Tools for Sampling, Frozen Tissues for Trace
Metal Analyses. USAEC Health and Safety Laboratory Report, HASL-249,
1-19-24, April 1, 1972.
25. Hislop, J. S., and A. Parker. The Use of a Laser for Cutting Bone Samples
Prior to Chemical Analysis. Analyst, 98:694, 1973.
26. Auth, D. C., J. L. Doty, D. Neal, D. Heimbach, R. Wentworth, J. Colocousis,
and P. W. Curreri. The Laser Blade: A New Laser Scalpel. Proc. 2nd
International Surgical Laser Symp., Oct. 23-26, 1977, Dallas, Texas.
27. Koch, Jr., H. J., E. R. Smith, N. F. Shimp, and J. Connor. Analysis of
Trace Elements in Human Tissues. Cancer, 9:499, 1956.
28. lyengar, G. V., K. Kasperek, and L. E. Feinendegen. Determination of
Certain Selected Bulk and Trace Elements on the Bovine Liver Matrix using
Neutron Activation Analysis. Phys. Med. Biol., 23:66, 1978.
29. Persigehl, M., H. Schicha, K. Kasperek, and L. E. Feinendegen. Behavior
of Trace Element Concentration in Human Organs in Dependence of Age and
Environment. J. Radioanal. Chem., 37:611, 1977.
29a. Persigehl, M. , H. Schicha, K. Kasperek, and H. J. Klein. Trace Element
Concentration in Human Organs in Dependence of Age. Beitr. Path., 161:209,
1977.
30. Bernhard, U., K. Kasperek, A. Hock, K. Vyska, Chr. Freundlieb, and L. E.
Feinendegen. Changes in Trace Element Concentrations in Human Serum
after Parenteral Application of Amino Acids. J. Radioanal. Chem.,
37:383, 1977.
11
-------
31. Kollmer, W. E., G. V. lyengar. Normal Concentrations of Various Elements
in Organs and Body Fluids. Compilation of Data Published by Various
Authors. Gesellschaft fuer Strahlen - und Umweltforschung m.b.H.,
Neuherberg/Muenchen (F.R. Germany) Abt. fuer Nuklearbiologie, Dec. 1972.
32. Schicha, H., K. Kasperek, L. E. Feinendegen, V. Siller, and H. J. Klein.
Inhomogeneous but Partly Parallel Distribution for Cobalt, Iron, Sele-
nium, Zinc, and Antimony in Different Regions of Liver, Lung, Kidney,
Heart and Aorta, Measured by Activation Analysis. Beitr. Path. Bd.,
146:55, 1972.
33. Schicha, H., H. J. Klein, K. Kasperek, and F. Ritzl. Activation Analyti-
cal Estimation of Some Elements in Several Organs and in Cancerous Tissue,
Beitr. Path. Anat., 138:245, 1969.
34. Sumino, K., K. Hayakawa, T. Shibata, S. Kitamura. Heavy Metals in Normal
Japanese Tissues. Arch. Environ. Health, 30:487, 1975.
35. Velandia, J. A., and A. K. Perkons. Survey of 43 Constituent Elements in
Kidney Samples by Instrumental Activation Analysis. J. Radioanal. Chem.,
20:715, 1974.
36. Hock, A., U. Demmel, H. Schicha, K. Kasperek, and L. E. Feinendegen.
.Trace Element Concentration in Human Brain. Brain, 98:49, 1975.
37. Khera, A. K., and D. G. Wibberly. Some Analytical Problems Concerning
Trace Metal Analysis in Human Placentae. Proc. Analyt. Div. Chem. Soc.,
34D, 1976.
38. Sharma, R. P., and J. L. Shupe. Lead, Cadmium, and Arsenic Residues in
Animal Tissues in Relation to Those in Their Surrounding Habitat.
Science of the Total Environment, 7:53, 1977.
39. Langmyhr, F. J., and J. Aamodt. Atomic Absorption Spectrometric Deter-
mination of Some Trace Elements in Fish Meal and Bovine Liver by the
Solid Sampling Technique. Anal. Chim. Acta, 87:483, 1976.
40. Borjeson, H., and E. Fellenius. Towards a Valid Technique of Sampling
Fish Muscle to Determine Redox Substrates. Acta physiol. scand., 96:202,
1976.
41. Cornelis, R., A. Speecke, and J. Hoste. Neutron Activation Analysis for
Bulk and Trace Elements in Urine. Anal. Chim. Acta, 78:317, 1975.
42. Brune, D. Interference-Free Trace Element Analysis of Water, Biological
Specimens and Food Sources by Means of Nuclear Techniques. Science of
the Total Environment, 2:111, 1975.
43. Johnson, D. E., R. J. Prevost, J. B. Tillery, and R. E. Thomas. The
Distribution of Cadmium and Other Metals (Lead, Zinc, and Mercury) in
Human Tissue. EPA-600/1-78-035, Q. S. Environmental Protection Agency,
May, 1978.
12
-------
44. Tipton, I. H., M. J. Cook, R. L. Steiner, C. A. Boye, H. M. Perry, Jr.,
and H. A. Schroeder. Trace Elements in Human Tissue, Part I, Methods.
Health Physics, 9:89, 1963.
13
-------
SECTION 2
RESEARCH ON FREEZER STORAGE
by
T. E. Gills, S. Harrison, and H. L. Rook
An important part of the NESB research program has been a study and eval-
uation of freezer storage for long-term storage of tissues and/or other biolog-
ically active samples. The effects of microbiological action on trace constit-
uents concentrations and distributions are well documented. Freezing has long
been applied as a technique for storage, however no study has been previously
made to document the reliability with respect to inorganic ions of this method
of storage for more than a short period of time. Recently, the analysis and
subsequent reanalysis of a selected animal tissue (bovine liver) that had been
stored frozen at -80 °C for a period of one year gave important information on
the effects of freezer storage for select elements, namely iron, copper, zinc,
molybdenum, selenium, cobalt, and arsenic. The original organ had been subset
sampled using stainless steel scalpel blades and the remaining portion homog-
enized using a commercial food blender. The homogenate was poured into six
precleaned polypropylene ice cube trays which were sealed in 1 mil thick
polyethylene film and stored in a chest-type -80 °C freezer. Analyses were
performed on the subsets and homogenates to provide initial trace elemental
concentrations and to verify contamination-free homogenization of the sample.
For the one year reanalysis, one tissue cube was removed from each of the six
trays and the trays resealed for future studies. The selected tissue cubes
were freeze-dried and analyzed using the same analytical procedures utilized
in the initial analysis. A comparison of the results of the reanalysis to
those originally obtained is given in Table 1. For the elements determined,
the capability to procure a tissue and store it frozen for one year for retro-
spective analysis has been demonstrated. In addition, results indicate that
subset sampling can be a viable sampling strategy when applied to relatively
homogenous tissues such as liver, etc. Important elements such as chromium,
cadmium, iodine, lead, etc., though not analyzed in this set of samples, are
currently being evaluated for their stability in tissues during freezer
storage.
The sampling implements used in this work are not suggested for sampling
or homogenization but were utilized to provide a base for evaluating contamin-
ation with respect to elements such as chromium. Work is currently being done
to assess contamination from sampling implements and their effects upon analyt-
ical uncertainty of real data.
14
-------
Table I
Analyses of a Stored Bovine Liver
Storage and Analysis Protocol Tests
(Concentrations Reported on Dry Wt. Basis)
Ug/g
Sample Type
I. Subset
Samples (+)
II. Homogenate (*)
III. Homogenate (+)
IV. Subset
Samples (+)
AVERAGE
Cu
68±10
Zn
Mo
39±3 .89±.12
Se
Co
As
(ng/g)
.29±.05 .077±0.006 4±1
Analysis
Procedure Used
RNAA
7
5
6
79±4
74±6
75±4
69± 5
69± 3
—_
40±2
39±2
39±1
.76±.
.80±.
_—
04
05
.30±.
.30±.
.29±.
05
03
02
.077+0
.073±0
.075±0
.005
.003
.002
6±2
6±2
—
RNAA
RNAA
INAA
69± 1
39±1 .82±.07
,29±.01 .075±0.002 5±1
Error: Standard Deviations of Means
III. Reanalysis 6
after 1 yr (+,*)
69±7 70± 2 38±1 .82±.08 .27±.05 .076±0.003 4±1
RNAA
+ Fresh or Frozen Tissue
* Freeze-dried
-------
SECTION 3
CONTAINER MATERIALS FOR THE PRESERVATION OF
TRACE SUBSTANCES IN ENVIRONMENTAL MATERIALS
by
Richard A. Durst
INTRODUCTION
The success of the National Environmental Specimen Bank will be determined
in large part by the ability to preserve the integrity of the trace substances
in samples during long-term storage. It is widely recognized that the storage
of environmental materials is subject to a variety of uncertainties when one
is considering the identity and levels of trace elements and organic compounds.
Changes in the forms and concentrations of the numerous environmentally
important substances in specimens stored for extended periods may occur in
several ways. Processes such as surface adsorption and sample degradation may
reduce the concentrations of various components and/or produce species which
may not have been present in the original sample. On the other hand, contam-
ination from the container and evaporation of specimen fluids could lead to
apparent increases in trace substance concentrations or losses in the case of
volatile trace components. In addition, superimposed on these processes are
the factors which will affect their rates, such as, container material, contact
time and area, storage temperature, pH, and initial species concentration. All
of these factors are important considerations in evaluating the suitability of
various containers for long-term storage.
There have been numerous studies performed in recent years directed toward
the identification of suitable container materials and optimum storage condi-
tions, but much of this work is contradictory (1). In many cases, the analyt-
ical data were invalidated because of problems not associated with the storage
but with other links in the analytical chain such as sampling and procedural
contamination.
Container Materials
While certain container materials can be eliminated a prn-on, it is
usually necessary to consider the container composition vis-a-vis the type of
sample and/or the components of interest in the sample. Accordingly, it is
unlikely that samples intended for trace organic analysis would be stored in
plastic containers or acidified water samples in glass containers if trace
elements were to be determined. While these are obvious examples of
16
-------
incompatible sample/container combinations, even these may be acceptable
under certain storage conditions and for special requirements.
The National Environmental Specimen Bank is somewhat unique in that both
trace elements and trace organics are of environmental concern and a choice
must be made as to storing the samples in a single container material suitable
for both types of components or in two different container materials optimized
for the components of interest. In the former case, the choices are much more
limited and compromises must be made, whereas in the latter, the increased
complexity of separate sampling, sample handling and storage procedures may be
excessive. In the EPA/NBS Environmental Pilot Bank, this is one of the ques-
tions to be answered by careful evaluation of a variety of container materials
under different cleaning procedures and storage conditions.
At NBS, most of the research to date has been concerned with trace ele-
ment contamination and losses, while trace organic problems have largely been
avoided by the use of carefully cleaned glass containers and immediate freez-
ing of the samples. In a recent study (2), twelve different plastics were
examined by gravimetry, isotope dilution mass spectrometry, and neutron acti-
vation analysis in order to evaluate the rate of water loss, the levels of
impurities present in the plastics, and the quantities of trace elements
leached from the plastics during acid cleaning.
The annual rate of water loss, based on weighings at 17 and 66 days, are
given in Table 1. The polypropylene, Teflon, and conventional polyethylene
have water loss rates compatible with long-term storage of aqueous samples,
whereas the other plastics may be adequate for shorter storage periods (2).
TABLE 1. ANNUAL RATE OF WATER LOSS FROM PLASTIC CONTAINERS
Material % Loss/year*
Polypropylene 0.04
Teflon 0.05
Conventional polyethylene 0.1
Polyvinyl chloride 0.5
Polymethylpentane 1
Polycarbonate 2
X
Average values based on weight loss measured at 17 and 66 days.
Since the closure may have been responsible for a significant portion of the
water loss, these values should not be construed as permeability data but
merely typical losses from these types of containers. These losses could be
reduced in practice by sealing the container in a vapor barrier or storage at
an ambient relative humidity comparable to that of the aqueous samples.
The concentrations of trace elements contained in the plastics were deter-
mined by neutron activation analysis (2). The ten plastics studied showed
17
-------
striking differences in trace element composition as illustrated in Table 2
for several of the elements determined.
Table 2. Trace elements in plastics determined by Neutron
Activation Analysis (Concentrations given in ng/g)
Element CPE LPE PP PMP PS PC TFE FEP
Na
Al
Ti
Mn
Co
Zn
Br
Sb
Au
IxlO3 15xl03
500 3xl04
5xl03
10
5xl05
>20 800
5 200
5xl03
6xl04
6xl04
20
40
>5
600
0.1
200
6xl03
5xl03
10
3xl04
>2
0.6
2xl03
500
IxlO3
20
>1
0.04
3xl03 160 400
3xl03 230 200
60
6
3xl04 >2
0.03 0.4
CPE = conventional polyethylene; LPE = linear polyethylene;
PP = polypropylene; PMP = polymethylpentane; PS = polystyrene;
PC = polycarbonate; TFE and FEP = types of Teflon.
In agreement with the results of other workers, the purest materials appear to
be Teflon, PS, and CPE. Comparison of these data with those obtained in leach-
ing experiments indicate that the bulk of most trace element impurities present
is distributed throughout the matrix, rather than concentrated at the surface.
The acid-leaching experiments were performed on four types of plastic
using two acids prepared from ultra-pure reagents. In this study (2), the
plastic bottles were first rinsed with distilled water to remove any surface
contamination and then filled with a 1+1 mixture of ultra-pure HNO_ or ultra-
pure HC1 and ultra-pure water (3). The leaching process was allowed to proceed
for one week at room temperature (80 °C for the FEP bottle) before the analyses
were performed. The results obtained by isotope dilution mass spectrometry are
given in Table 3 for a selected group of the impurity elements measured.
The results of these experiments indicate that HN03 and HC1 leach various
elements with different efficiencies, and it is recommended that both acids, in
sequence, be used for cleaning these containers (2). After cleaning, the
Teflor and CPE bottles have been found to be the least contaminating. Based on
experience at NBS and reports of other studies, a recommended method for clean-
ing plastic containers has been proposed (2):
1. Fill container with 1+1 analytical reagent grade HC1.
2. Allow to stand for one week at room temperature (80 °C
for Teflon).
18
-------
3. Empty and rinse with distilled water.
4. Fill with 1+1 analytical reagent grade HN03.
5. Repeat steps 2 and 3.
6. Fill with the purest available distilled water.
7. Allow to stand several weeks or until needed, changing water
periodically to ensure continued cleaning.
8. Rinse with purest water and allow to dry in a particle- and
fume-free environment.
2
Table 3. Impurities Leached from Plastic Containers (ng/cm )
Element
Pb
Al
Na
Sn
Cd
Se
Zn
Cu
Ni
Fe
Cr
Mg
FEP
HN03
2
6
6
1
0.4
0.2
4
2
2
20
0.8
8
HC1
2
4
2
1
0.6
0.8
4
6
0.8
16
4
1
LPE
HN03
2
1
10
1
0.2
0.4
8
0.4
1.6
3
0.2
0.6
HC1
0.6
4
6
<1
0.2
0.4
9
1
0.8
1
0.8
0.4
CPE
HN03
0.7
1
8
<0.8
0.2
3
2
2
0.5
3
0.8
0.7
HC1
18
10
42
<0.8
0.2
<0.3
1
0.7
0.3
1
0.3
0.7
PC
HN03
0.3
5
3
0.2
0.3
0.5
0.8
0.8
0.7
3
0.3
2
HC1
10
3
8
13
<8
<0.5
-
<6
0.3
<49
<5
0.8
Values which are below 2 ng/cm2 or prefixed by < are upper limits.
For the storage of biological tissues and fluids, rapid freeze drying
immediately after sampling has been recommended but suffers from the disadvan-
tage that some volatile components may be lost (1). A more viable approach
would appear to be the immediate freezing of the sample or sub-samples to the
lowest conveniently attainable temperature. This approach serves two purposes:
1) it reduces or stops both chemical and biological processes which could
result in sample changes, and 2) it reduces the mobility of sample and con-
tainer components thereby lessening the possibility of contamination and/or
losses due to adsorption or volatility. However, because of concentration
gradients which may be produced during freeze-thaw cycles and physical changes,
e.g., cell destruction, caused by freezing, it is highly recommended that all
sub-sampling sites usually precludes this procedure because of the probability
19
-------
of sample contamination during the sub-sampling. Instead, the sample, which
has been frozen as rapidly as possible to avoid component fractionation, is
sub-sampled while still in the frozen state at a location which has a clean
facility.
When the sample is frozen, especially at cryogenic temperatures, the
reduced interaction between the sample and container permits greater flexi-
bility in the choice of container material. However, even under these condi-
tions, it is inadvisable to store samples intended for trace organic analysis
in any type of plastic container which incorporates additives such as plasti-
cizers, organometallic or other stabilizer antioxidants, colorants, or any
other components which may be leached or volatilized from the plastic. Teflon
is the only plastic which has been found suitable for trace organic storage at
cryogenic temperatures, and which, after proper cleaning, is also suitable for
the storage of samples intended for trace element analysis.
Glass (quartz or Pyrex), with Teflon (or aluminum foil) lined caps, is
also suitable for trace organic samples and may be satisfactory for frozen
trace element samples. At the present time, the key to long-term storage
appears to be storage at cryogenic temperatures to reduce reactions and inter-
actions to a minimum. Under these conditions, both glass and Teflon appear to
be suitable storage container materials.
20
-------
REFERENCES
1. Maienthal, E. J. and D. A. Becker, "A survey of current Literature on
Sampling, Sample Handling, and Long-Term Storage for Environmental Mate-
rials," NBS Tech. Note 929, U. S. Government Printing Office, Washington,
D.C. 20402, 1976.
2. Moody, J. R. and R. M. Lindstrom Anal. Chem. 49:2264, 1977.
3. Kuehner, E. C., R. Alvarez, P. J. Paulsen and T. J. Murphy Anal. Chem.
44:2050, 1972.
21
-------
SECTION 4
DESIGN AND CONSTRUCTION OF THE PILOT BANK FACILITY
by
H. L. Rook, T. E. Gills, J. Koskoris, R. A. Durst
A final proposal and design for the construction of the Pilot Bank low-
temperature storage facility has been made. The Pilot Bank facility will have
the capacity to house 20 replicate samples for each of the approximately 4000
input specimens planned for the five-year storage period. These will be
derived from four classes: human tissue, aquatic tissue, food grain, and air
accumulator. Nominal sample size will be five-to-ten grams per replicate
sample. Samples will be stored at approximately -196 °C liquid nitrogen tem-
perature, -80 °C, -25 °C, and freeze dried material stored at room tempera-
ture. The multiple storage conditions for the same sample will allow direct
evaluation of the candidate storage procedures. Protocols for sampling and
sample handling as discussed earlier in this report will be followed.
The pilot facility will be made up of an existing NBS clean room which
will be upgraded to a Class-100 clean lab, and a new Class-100 storage area
which will be constructed adjacent to the existing clean room (see figure 1).
The existing clean room will be modified to provide ultra-clean preparation
laboratory conditions equivalent to a trace constituent analytical laboratory.
Samples will be protected with ultra-clean air (99.97% efficient on all par-
ticles 0.3 micrometer and larger). Personnel will be protected by performing
all operations with potentially pathogenic samples (liver) in a biological
clean work station.
The air flow pattern and work function will be designed to eliminate or
minimize contamination caused by work functions. The preparation laboratory
will be divided into two principal work areas by means of a central dual air
return wall. A conventional fume hood and clean air module providing verti-
cal-flow, Class-100 air will be located in the inorganic analysis section.
A biohazard enclosure and clean air module providing Class-100, charcoal-
filtered air will be located in the organic analysis section of the labora-
tory. False walls will also be located at the ends of this laboratory to
provide additional air return to the clean air modules containing HEPA
filters.
The Class-100 storage area will consist of a room approximately 3.7 m by
7.3 m (121 x 24') connected to the existing clean lab via an air lock. The
cleanliness of this storage area will have a particle count of fewer than 100
22
-------
particles 0.3 micrometer and larger per cubic foot. Freezers will be located
in this area for the storage of specimens in liquid nitrogen freezers (approx-
imately -196 °C) and compressor-type freezers at -25 and -80 °C. Racks will
also be provided for freeze-dried samples. In order to avoid the possibility
of build up of potentially harmful levels of nitrogen which could lead to
oxygen starvation, provision will be made to vent and replace sufficient
storage room air to ensure safe oxygen levels. In addition, the freezer com-
pressors will be hermetically sealed in order to minimize potential contami-
nation from wear metals and/or organic vapors.
It is anticipated that this laboratory/storage facility will be completed
in mid-summer of 1979. Human liver specimens will be acquired immediately
upon completion of this facility.
23
-------
CM
Bio-Hood
Lab-Bench |
___Tii ___r a _,_... Timiji ^ , _-__ j i
I
1 !
Clean-Air Module I
1
, 1
Air Return Wall
,-
Lab-Bench i (|) 1
_______ i __ |
-Air Module i
I
I
Fume Hood
Storage Area
Air-Lock
r>'
i% h>
>
K
L—W *B^_ .^HV*
o
•
CM
Sff,
UJ
LU
23'-4'
Figure 1. Specimen Bank Laboratory/Storage Facility
24
-------
SECTION 5
ORGANIC MERCURY IN TISSUES
by
R. L. Zeisler, T. E. Gills
Increased interest of ecologists and analysts has been focused on the
ratio of organic mercury to total mercury content of biological and environ-
mental samples. The determination of the total mercury content of these
matrices using. Neutron Activation Analysis (NAA) procedures has been exten-
sively investigated in the past years, and some Radiochemical Neutron Activa-
tion Analysis (RNAA) methods are considered fully reliable for this purpose.
So far, the determination of organic mercury has been carried out by use of
other methods, e.g. gas-chromatography. Thus, two samples have to be analyzed
by different methods to yield the desired information on the ratio of organic
mercury to total mercury content.
Well developed extraction and volatilization techniques are currently
being investigated for the isolation of methylmercury from biological tissues.
Since the possible radiolytic damage to organic mercury compounds during irra-
diation cannot be fully assessed, we are developing and applying a modified
extraction technique to be used in a pre-irradiation and "normal" RNAA scheme.
This will yield information about the changes of the chemical form of organic
mercury during irradiation.
The separation of methylmercury is based on the volatilization of methyl-
mercury cyanide formed in the reaction of methylmercury in a sample with
hydrocyanic acid released by the interaction of a cyanoferrate with sulfuric
acid at elevated temperatures. The methylmercury cyanide released is captured
on cysteine paper in a microdiffusion cell. The paper is then placed in a
flux of neutrons and the mercury quantitatively determined by neutron activa-
tion analysis.
The feasibility of this method is currently being tested. Szilard-
Chalmers Reactions can change the chemical status of the mercury, and possible
losses of mercury from the sample have to be assessed. Test analyses have
proven that irradiation has to be performed in sealed quartz containers.
Samples can then be counted either "instrumentally" in the irradiation con-
tainer or after using the combustion methods. Both require comparable samples
and standard configurations. The use of mercury deposits on filter paper (com-
parable to the proposed cysteine paper) is being evaluated.
25
-------
In addition, the volatilization method for mercury determination after
the reduction of the inorganic mercury by SnCla will give the possibility of
cross-checking of the mercury contents. The reduction method is claimed to be
selective for the inorganic mercury (only 0.5% of the organomercury is
released). This is, however, critically dependent on the chemical treatment
of the samples. This has to be carefully investigated.
The procedure utilizes the volatility of mercury and its compounds, more
importantly the inherent sensitivity of neutron activation analysis along
with minimum use of reagents and solvents should provide increased accuracy
over currently used methods.
26
-------
SECTION 6
The Determination of Trace Elements in New Food Grain SRM's
Using Neutron Activation Analysis
by
Thomas Earl Gills, Mario Gallorini, Harry L. Rook
INTRODUCTION
Increasing concern with possible toxic elements in the food chain plus
growing recognition of the importance of a number of trace elements in nutri-
tion make the accurate determination of trace elements essential (1). The
National Bureau of Standards, in an interagency agreptnent wLth the Food and
Drug Administration, is involved in developing and certifying selected elements
in food grain as a part of the NBS Standard Reference Materials Program. The
food grains chosen, rice flour and wheat flour, are currently being analyzed
to determine trace element concentrations. Many of the elements considered to
be toxic to human health are present at the ppb level, and their analyses
require selective radiochemical separations. This paper presents results
obtained using radiochemical and instrumental neutron activation analysis to
determine thirteen trace elements whose effect on human systems ranges from
toxic to essential. The following elements were determined in the food grain
SRM's: Fe, Zn, Mn, Co, Na, K, Br, As, Sb, Se, Cu, Rb, and Hg.
EXPERIMENTAL
Samples and Standards:
Homogenized samples of Wheat and rice flours were vacuum dried at a pres-
sure of 0.55 mm Hg for 48 hours in order to obtain a moisture content correc-
tion factor. For elemental standards, primary solutions were made by the dis-
solution of high purity salts or metals in NBS high purity acids (2) .
Irradiations:
All samples and standards were sealed in Suprasil quartz vials and irra-
diated in the high flux pneumatic irradiation facility of the NBS Research
Reactor. In this position the nominal neutron flux is 5x10^ n-cm 2*sec *
with a Cu/Cd ratio of ^80. Within the irradiation container the radial flux
variation is < 2 percent (3).
27
-------
Reagents for Radiochemical Separations:
Chromatographic reagents used for radiochemical separations consisted of
inorganic ion exchangers developed by Girardi, et al. (4,5). Tin Dioxide (TDO),
Cuprous Chloride (CUC) and Hydrated Manganese Dioxide (HMD) were obtained from
Carlo Erba, Italy.
Counting:
All samples and standards were counted on a 66 cm volume Ge(Li) detector
coupled to a 4096 channel pulse height analyzer. For multiple analyses an
automatic sample changer system was employed. After each count the accumu-
lated data were transferred to a magnetic tape and were processed by the quan-
titative analysis program OLN (6).
INSTRUMENTAL NEUTRON ACTIVATION ANALYSIS (INAA)
The following elements were determined by INAA: Br, Co, Fe, K, Mn, Rb,
Zn, Na, and Se. The irradiation conditions and data used for the instrumental
analysis are shown in Table 1. Different irradiation times were used to ob-
tain the proper conditions for maximum sensitivity and accuracy. Approxi-
mately 300 mg of each food grain SRM, along with appropriate standards, were
encapsulated in Suprasil quartz and irradiated in the high flux pneumatic tube
facility of the NBS Reactor. After a]lowing the samples to decay for the
specified time, samples and standards were post-weighed into fabricated poly-
ethylene bags and counted. The resultant spectra were reduced to individual
photopeaks associated with specific isotopes of interest and peak areas com-
puted both by the total peak area method and by a peak fitting method. Ele-
mental concentrations were determined by the direct comparator method.
Table 1. Irradiation Conditions and Data Used for Instrumental Analysis
Element
Det'd
Fe
Co
Zn
Zn
Se
Rb
Mn
Na
Br
K
Isotope
Measured
59Fe
6°Co
65Zn
69MZn
75Se
86Rb
56Mn
24Na
82Br
42K
Half-Life
44.6 d
5.27 y
244 d
13.9 h
120 d
18.7 d
2.58 h
15.03 h
35.34 hr
12.36 hr
Irradia-
tion Time
8 hr
8 hr
8 hr
8 hr
8 hr
8 hr
3 min
1 hr
1 hr
1 hr
Decay
Time
30 d
30 d
30 d
30 d
30 d
30 d
2 hr
30 hr
30 hr
30 hr
Counting
Time (sec)
5 x IO4
5 x IO4
5 x IO4
5 x IO4
5 x IO4
5 x IO4
4 x IO3
4 x IO3
4 x IO3
4 x IO3
Peaks Used for
Evaluation (MeV)
1.099,
1.173,
1.115
.439
.136
1.077
.847,
1.369,
.776,
1.525
1.292
1.332
.265
1.811
2.754
.619
28
-------
RADIOCHEMICAL SEPARATIONS (RNAA)
The concentrations of arsenic, antimony, copper, selenium, and mercury
were determined by RNAA. The irradiation conditions and data used for the
radiochemical analysis are shown in Table 2. Inorganic ion exchangers were
used in the determination of As, Sb, Se, and Cu while a combustion distillation
procedure was used in the determination of Se and Hg. Complete descriptions
and details of these procedures are given elsewhere (7-9).
Table 2. Irradiation Conditions and Data Used for Radiochemical Activation
Analysis
Element
Det'd
As
Cu
Se
Hg
Sb
Sb
Isotope
Measured
76.
As
64Cu
75Se
1 Q7
122sb
124sb
Half-life
26
12
120
64
2
60
.4
.7
•
.1
.7
.3
hr
hr
d
hr
d
d
Irradia-
tion Time
(hr)
4
4
4
4
4
4
Decay
Time
(hr)
36
36
36
36
36
36
Counting
Time
(sec)
5
5
5
5
5
5
x 10
x 10
x 10
x 10
x 10
x 10
4
4
4
4
4
4
Peaks used for
Evaluation
(MeV)
.559,
.511,
.136,
.077
.564
.603,
.657
1.345
.265
1.691
, M
RESULTS AND DISCUSSION
The results of the instrumental analysis are shown in Table 3. The exper-
imental design of the conditions for irradiation and counting was optimized to
obtain the best statistical parameters. Under optimized conditions, the pre-
cision obtained was relatively good. The results of the analysis of NBS SRM
Orchard Leaves 1571 also indicates relatively good accuracy.
The results obtained using radiochemical separation procedures are also
summarized in Table 3. The values represent the averages obtained from the
different radiochemical separation procedures.
The results shown constitute only one independent method of analysis
(NAA) and are not considered NBS certified values. However, these results
give a preliminary trace elemental profile of typical food grains (rice and
wheat flour).
The procedures used in this work have been evaluated with respect to the
sources of error that can affect the accuracy and precision of the results
obtained.
29
-------
Table 3. The Determination of Trace Metals in Rice and Wheat Flour in pg/gm
0 INSTRUMENTAL ACTIVATION ANALYSIS
• RADIOCHEMICAL ACTIVATION ANALYSIS
Element
"Bromine
°Cobalt
"Iron
"Potassium
"Manganese
"Rubidium
"Zinc
"Sodium
"Selenium
Arsenic
Antimony
Copper
Mercury
Selenium
No. of Deter'
minations
6
6
6
6
12
6
6
6
6
4
4
4
4
4
NBS Orchard Leaves SRM (1571)
*
Certified
(10)
(0.2)
0.030% ± 0.002
1.47% ± 0.03
91 ± 4
12 ± 1
25 ± 3
82 ± 3
0.080 ± 0.01
10+2
2.90 ± 0.3
12 ± 1
0.155 ± 0.015
0.080 ± 0.01
Found
9.7 ± 1.1
0.14 ± 0.01
0.026% ± 0.002
1.49% ± 0.03
89 ± 5
11.7 ± 0.1
24.8 ± 1.1
84 ± 4
0.086 ± 0.01
10.0 ± 0.4
2.86 + 0.08
11.8 ± 0.30
0.154 ± 0.02
0.084 ± 0.008
No. of Deter- Wheat Flour
minations Found
7
5
5
7
12
8
8
6
6
4
4
4
4
4
(9.9 ± 1.5)
0.021 ± 0.004
17.2 ± 0.6
1392 ± 37
8.6 ± 0.4
0.99 ± 0.16
10.88 ± 0.56
(10.4 ± 2.5)
1.12 + 0.01
0.0054 ± 0.0005
0.038 ± 0.001
2.0 ± 0.2
0.0010 ± 0.0003
1.11 ± 0.05
Rice Flour
Found
(1.23 ± 0.08)
0.018 ± 0.002
8.85 ± 0.94
1125 ± 16
19.95 ± 0.69
7.27 ± 0.21
19.97 ± 0.69
(6.9 ± 0.4)
0.42 ± 0.03
0.40 ± 0.01
0.005 ± 0.001
2.2 ± 0.13
0.0064 ± 0.001
0.45 ± 0.03
'''Numbers in parentheses indicate information values only.
-------
REFERENCES
1. Tanner, J. T. and M. H. Friedman. Neutron Activation Analysis for Trace
Elements in Foods. Proceedings 1976 International Conference Modern
Trends in Activation Analysis, _!, p 244.
2. Kuehner, E. C., R. Alvarez, P. J. Paulson and T. J. Murphy. Production
and Analysis of Special High Purity Acids Purified by Sub-Boiling Distil-
lation Anal. Chem. 44:2050, 1972.
3. Becker, D. A. and P. D. LaFleur. Characterization of a Nuclear Reactor
for Neutron Activation Analysis J. Radioanal. Chem. 19:149-157, 1974.
4. Girardi, F., R. Pietra and E. Sabbioni. Euratom Techn. Rep. EVR 4287,
1969.
5. Girardi, F., R. Pietra and E. Sabbioni. Radiochemical Separations by
Retention on Inorganic Precipitates. 1968 International Conference on
Modern Trends in Activation Analysis. NBS Special Publication 372,
Vol. 1, pp 639.
6. Yule, H. P. (NUS), J. P. F. Lambert and S. K. Gerauson (EPA-NC). Auto-
matic Activation Analysis Code for Component Identification and Quantita-
tive Result Computation Transactions of the Amer. Nuol. Soc. 17:128-129,
1973.
7. Gills, T. E., M. Gallorini and R. R. Greenberg. Determination of
Selected Toxic Elements in Biological Matrices Using Radiochemical Acti-
vation Analysis. Proceedings of the International Conference Nuclear
Methods in Environmental and Energy Research, 11 October 1977, University
of Missouri-Columbia.
8. Rook, H. L., T. E. Gills and P. D. LaFleur. Method for Determination of
Mercury in Biological Materials by Neutron Activation Analysis Anal.
Chem. 44, No. 7, June 1972.
9. Rook, H. L. Rapid, Quantitative Separation for the Determination of
Selenium using Neutron Activation Analysis Anal. Chem. 44, No. 7, June
1972.
Certain commercial equipment, instruments, or materials are identified in
this paper in order to adequately specify the experimental procedure. In no
case does identification imply recommendation or endorsement by the National
Bureau of Standards, nor does it imply that the materials or equipment identi-
fied is necessarily the best available for the purpose.
31
-------
SECTION 7
Cadmium Analysis by Radiochemical Neutron Activation Analysis
by
Robert R. Greenberg, Mario Gallorini, Thomas E. Gills
Cadmium has been suspected of causing detrimental health effects in
humans even at very low levels. Since Cd is commonly found at the trace or
ultra-trace level, many analytical techniques do not possess the sensitivity
to accurately determine the concentration of this element in many environment-
ally important materials. In addition, there are some instances in which the
total amount of material available for analysis is very small, such as for
certain types of atmospheric particulate samples, or hair samples. Analytical
techniques capable of measuring very small amounts of Cd are therefore
required.
Radiochemical neutron activation analysis (RNAA) was used at the National
Bureau of Standards to analyze Cd in a variety of matrices. RNAA offers the
advantages of high sensitivity, excellent selectivity, and no chemical
blank. Furthermore, the ability to add carriers during the chemical dissolu-
tion and separation enables quantitative recovery of the element.
The concentrations of many elements can be determined instrumentally —
that is, without separating them from the activated matrix. Other elements,
however, such as Cd, usually have to be determined radiochemically, or iso-
lated from other neutron-activated products. Many different types of radio-
chemical separations can be used for RNAA, such as: solvent extraction,
distillation, precipitation, ion-exchange chromatography, and electrodeposi-
tion. The method we use to isolate Cd is solvent extraction. Zinc diethyl-
dithiocarbamate [Zn(DDC)2l in chloroform will quantitatively extract Cd from
an aqueous solution over a pH range of from 1 to 12.
In addition to Cd, Zn(DDC)2 also extracts Cu. In many matrices, Cu can
interfere with the determination of Cd by producing a high background level of
radiation. This problem can be eliminated by first extracting with bismuth
32
-------
diethyldithiocarbamate [Bi(000)3] in chloroform which removes Cu but not Cd.
Cu can then also be analyzed if desired.
Preparation of Metal DDC Compounds
The Zn(DDC)2 and Bi(DDC)3 compounds are prepared by mixing aqueous solu-
tions of NaDDC and either Zn(1503)2 or Bi(N03)3. The M(DDC) compound formed
is insoluble in water and precipitated. The precipitate is filtered, washed
with water, and dissolved in chloroform. An equal volume of ethanol is added
to the solution which is then set aside to allow the chloroform to evaporate
at room temperature. After the chloroform evaporates, the M(DDC)X compound
crystallizes in the remaining ethanol. The crystals are filtered and allowed
to dry at room temperature.
Procedure
Prior to irradiation, the samples and primary standards were encapsulated
in cleaned quartz vials. The standards consisted of a solution prepared from
high purity metals dissolved in high purity HN03. The samples and standards
were then irradiated together in the NBS reactor at a thermal neutron flux of
5 x 1013 n*cm 2-sec * for times ranging from one to four hours.
After irradiation, the radioactive samples were allowed to decay for
three days after which they were transferred to dissolution vessels. Three
tvpes of vessels were used depending on the matrix: Erlenmeyer flasks, Teflon
wet-ashing vessels and a Teflon-lined bomb. One hundred yg of Cu and Cd
carriers were added to each vessel and the samples were dissolved. Various
acid mixtures were used including HN03 alone, KN03 — l^SOi^ and HN03 — HC10i4
(not in the bomb). If silica was present, some HF was also added.
After dissolution, 100 mg of Zn 2 hold-back carrier was added to minimize
exchange between the radioactive Zn of the samples, and the organic Zn of the
Zn(DDC)2 solution used to extract the Cd. The pH of each solution was
adjusted to 1.5 with NH3 and the volume brought to about 60 mL with deionized
water. Each solution was then transferred to a 125 mL separatory funnel and
shaken for 30 minutes with 20 mL of 0.003 mol/L Bi(DDC)3 in chloroform using
a shaking machine. The organic fraction containing Cu was drained into a
120 mL polyethylene bottle and the aqueous phase was washed for 15 seconds
with an additional 10 mL of the Bi(DDC)3/CHCl3 solution. The wash was then
combined with the first Bi(DDC)3 fraction and retained for counting.
Twenty mL of 0.005 mol/L Zn(DDC)2 in chloroform was then added to the
aqueous phase still in the separatory funnel, which was then shaken for five
minutes. The organic fraction was drained into a second 120 mL polyethylene
bottle, and the aqueous phase washed for 15 seconds with an additional 10 mL
of the Zn(DDC)2/CHCl3 solution. The wash was combined with the first
Zn(DDC)2 fraction and was retained for counting.
Two different procedures were followed for the standards. One standard
was prepared by pipetting a known amount of irradiated solution into a disso-
lution vessel with carriers and some unirradiated sample material. This
material was dissolved in the same manner as were the samples used for analy-
sis. A second standard was pipetted directly into 1 mol/L HN03 along with
carriers. Both standards were then subjected to the same separation proce-
dure used for the samples.
-------
Counting
The samples and standards were counted on Ge(Li) detectors with active
volumes of from 60 to 75 cm3 coupled to 4096 channel pulse height analyzer
systems. The Bi(DDC)3 solutions containing Cu were counted immediately after
separation. The 511 keV peak produced from the annihilation of positrons
emitted by ^Cu was used for analysis.
The Zn(DDC)2 solutions containing Cd were allowed to decay for at least
24 hours to establish the equilibrium between ^5Cd and its daughter 115'ln.
The 336 keV line from 115mln, and the 527 keV line from 115Cd were both used
for analysis. Computer code QLN1 was used for data reduction supplemented by
hand integrations of poorly defined peaks.
Results and Discussion
A large number of National Bureau of Standards Standard Reference Mate-
rials (SRMs) were analyzed by the above procedure, indicating the applicabil-
ity of this method to a wide variety of matrices. The Cd and Cu results
obtained are compared with the certified values in Tables 1 and 2. Very good
agreement is observed demonstrating the accuracy of this technique.
Table 1. Cadmium in Various SRMs
Concentration-pg/g
Orchard Leaves (SRM 1571)
Pine Needles (SRM 1575)
Bovine Liver (SRM 1577)
Rice Flour (SRM 1568)
Wheat Flour (SRM 1567)
Sub-Bituminous Coal (SRM 1635)
River Sediment (SRM 1645)
Urban Particulate (SRM 1648)
This Work
0.116 ± 0.008
0.194 ± 0.009
0.295 ± 0.015
0.029 ± 0.005
0.030 ± 0.005
0.030 ± 0.002
10.2 ± 0.4
71.2 ± 3.7
Certified
0.11
< 0.5
0.27
0.029
0.032
0.03
10.2
75
± 0.01
± 0.04
± 0.004
± 0.007
± 0.01
+ 1.5
± 7
-
This two-extraction radiochemical separation scheme is very versatile and
can be used a;, a part of a larger multi-element analysis scheme. One such
schem^ used at NBS involves the use of an ion-exchange resin, HMD (Kydrated
Manganese Dioxide) prior to the extractions. HMD will quantitatively retain
As, Sb, Se, and Cr from a 1 mol/L HN03 solution. To perform this separation,
we normally dissolve the sample in HN03 + HCIO^ with some HF if silica is
present. The samples are heated to incipient dryness and then brought to a
volume of 20 mL with 1 mol/L HN03. The solutions are passed through plastic
chromatographic columns, each containing a bed of HMD 7 mm by 30 mm. The
columns are washed twice with 1 mol/L HN03 and dismantled for counting. Under
34
-------
Table 2. Copper in Various SRMs
Concentration-iJg/g
This Work Certified
Orchard Leaves (SRM 1571)
Pine Needles (SRM 1575)
Bovine Liver (SRM 1577)
Rice Flour (SRM 1568)
Wheat Flour (SRM 1567)
Sub-Bituminous Coal (SRM 1635)
11.6 ±0.4
3.04 ± 0.16
185 ± 7
2.12 ± 0.09
2.21 ± 0.10
3.56 ± 0.18
12+1
3.0 ± 0.3
193 ± 10
2.2 ± 0.3
2.0 ± 0.3
3.6 ± 0.3
these conditions Cu and Cd are completely eluted. The eluted fraction is then
subjected to the two-extraction procedure described above. Various SRMs were
analyzed using this procedure and the results obtained are compared with the
certified values in Tables 3-5. Very good agreement is observed.
Table 3. Sub-Bituminous Coal SRM 1635
Concentration-yg/g
This Work
As
Sb
Se
Cr
Cd
Cu
0.
0.
0.
2.
0.
3.
44
12
82
48
030
56
± 0.
± 0.
± 0.
± 0.
± 0.
± 0.
05
01
04
08
002
18
0
0
2
0
3
Certified
.42
(0.
.9
.5
.03
.6
± 0.
14)*
± 0.
± 0.
± 0.
± 0.
15
3
3
01
3
Values in parentheses are MBS information only values
35
-------
Table 4. Bovine Liver SRM 1577
Concentration-jjg/g
This Work
As
Sb
Se
Cr
Cd
Cu
0.
0.
1.
0.
0.
185
054
010
06
085
295
± 0.
± 0.
± 0.
± 0.
± 0.
± 7
004
002
06
009
015
Certified
(0.055)*
1.1 ± 0
0.088 + 0
0.27 ± 0
193 ± 10
.1
.012
.04
Values in parentheses are NBS information only values.
Table 5. Orchard Leaves SRM 1571
As
Sb
Se
Cr
Cd
Cu
9.
2.
0.
2.
0.
11.
This
7
8
09
67
116
6
C one en t r a t i on-y g/g
Work Certified
± 0
± 0
± 0
± 0
± 0
± 0
.4
.1
.01
.15
.008
.4
10
2.9
0.08
2.6
0.11
12
± 2
± 0.
± 0.
± 0.
± 0.
± 1
3
01
3
01
Conclusions
The two-extraction radiochemical separation scheme for Cu and Cd de-
scribed in this paper is both highly selective and extremely sensitive. The
simplicity of the method allows a quantitative recovery of both elements,
thus avoiding a Calculation of chemical yield. Furthermore, a relatively
short time is required to carry out the complete procedure. One person can
separate about 15 samples a day.
36
-------
ANALYTICAL CHEMISTRY, VOL. 50, NO 11, SEPTEMBER 1978 • 1479
Appendix I
Simultaneous Determination of Arsenic, Antimony, Cadmium,
Chromium, Copper, and Selenium in Environmental Material by
Radiochemical Neutron Activation Analysis
M. Gallorini,1 R. R. Greenberg, and T. E. Gills*
Canter for Analytical Chemistry, National Bureau of Standards, Washington, D.C 20234
A multielement radioanalytical procedure for the simultaneous
determination of As, Cr, Se, Sb, Cd, and Cu has been de-
veloped. An inorganic Ion exchanger coupled to a solvent
extraction system was used for the selective separation of
these elements from neutron activated matrices. The method
has been used for centrification of environmentaly related NBS
SRMs.
In studies of environmental pollution, elements such as
arsenic, antimony, chromium, cadmium, copper, and selenium
are among the most interesting because of their toxic nature.
These elements are usually found at very low concentration
in most samples of interest. The accurate analysis of these
elements can be accomplished only by techniques that are very
sensitive and selective. Neutron activation analysis usually
meets these requirements; however, in many cases the ele-
ments to be determined must be separated selectively from
neutron activated matrices. Utilizing radiochemical neutron
activation analysis (RNAA), one can obtain maximum sen-
sitivity, accuracy, and selectivity (1).
Usually, the choice of the radiochemical procedure to be
followed is dependent upon the elements to be determined,
the nature of the materials to be analyzed, and the simplicity
of the chemistry. Several radiochemical procedures such as
solvent extraction, distillation, ion-exchange chromatography,
precipitation and electrodeposition can be used to obtain
satisfactory results (2-6). However, when used alone, most
of these procedures are not suitable for simultaneous mul-
tielement analysis. Furthermore, when several chemical steps
are necessary, quantitative separation becomes extremely
'Guest worker from Laboratorio di Radiochiraica e Anahsi per
Attivazione del C.N.R., Pavia, Italy.
difficult; often the determination of chemical yields, which
can contribute to the analytical error, is required.
In this work, a multielement radiochemical separation
procedure has been tested and optimized to determine six
elements simultaneously in different NBS environmental
Standard Reference Materials (SRM's). Hydrated manganese
dioxide (HMD), an inorganic ion exchanger developed by
Girardi et al. (7,8). was used for the retention and subsequent
determination of As, Cr, Sb, and Se, while a solvent extraction
system using diethyldithiocarbamate compounds (9,10) was
used to determine Cd and Cu.
In an attempt to demonstrate applicability to matrices
having different chemical compositions, the following NBS
materials were analyzed: Orchard Leaves (SRM 1571), Bovine
Liver (SRM 1577), and Subbituminous Coal (SRM 1635).
EXPERIMENTAL
Reagents and Apparatus. The HMD inorganic exchanger
was obtained from Carlo Erba, Milan, Italy.
Bismuth and zinc diethyldithiocarbamates (DDC) were pre-
pared by mixing aqueous solutions of Na (DDC) with their re-
spective nitrate salts (9). The M(DDC)I compounds formed
precipitated from water. The precipitates were filtered, washed
with water, and dissolved in CHC13. An equal volume of ethanol
was added to the CHC13 solution which was then set aside to allow
the CHC13 to evaporate at room temperature. After evaporation,
the MCDDOj compound crystallized in the remaining ethanol.
The crystals were filtered and allowed to dry at room temperature.
Ion-exchange columns were made of polyethylene with an
internal diameter of 7 mm and packed with 3 cm of hydrated
manganese dioxide.
A Teflon-lined digestion bomb and Teflon wet-ashing vessels
were used for sample dissolutions.
Standards and Carriers. A multielement primary standard
was prepared from solutions of high purity metals dissolved in
NBS high purity HNO3 and/or H2SO4 (11). A multielement
carrier solution was prepared by mixing solutions of the salts or
Published 1978 by the American Chemical Society
37
-------
1480 • ANALYTICAL CHEMISTRY VOL 50, NO 11 - t'i-
1978
liable I. Determination of As, Sb, Se, Cr, Cd, and Cu in N'BS SRM Orchard Leaves (1571) and SRM Bovine Liver (1577)d
matnce
NBS Orchard Leaves 1571 NBS SRM Bovino Liver 1577
element
As
Sb
Se
Cr
Cd
Cu
certified values
10 t 2
2.9 i- 0.3
0.08= 0.01
2.60 ± 0.3
0.11 i 0.01
12 t 1
this work
9.7 .- 0.4
2.8 - 0.1
0.09 ± 0.01
2.67 ± 0.15
0.116 i 0.008
11.6 ± 0.4
certified values
(0.055)
( )
1.1 + 0.1
0.090 ± 0.015
0.27 i 0.04
193 ± 10
this work
0.054 i 0.004
0.010 ± 0.002
1.06 T 0.06
0.085 T 0.009
0.30 : 0.02
185 ± 7
a Concentration in Mg/g- The reported value for each element consisted of ten determinations. Values in parentheses are
NBS information values.
metals of interest dissolved in distilled water, HN03, and/or
H2SO4. The concentration of each element in the multielement
carrier solution was ~1 mg/mL.
The use of HC1 or chloride salts was avoided to prevent possible
losses of volatile chloride compounds during sample dissolution.
Sample Preparation. Special attention was given to sample
handling prior to irradiation in order to avoid any possible
contamination. Teflon-coated spatulas and forceps were used to
transfer the samples into cleaned silica vials. Sample sizes ranged
from 200-300 mg. Each irradiation capsule contained six vials,
one liquid multielement standard and five samples of the ma-
terials to be analyzed.
Irradiation. The standards and samples were irradiated in
the RT-3 pneumatic tube facility of the NBS Research Reactor.
In this position, the nominal neutron flux is 5 x 1013 n-cnT2 s~'.
The radial flux variation within the irradiation container is <2%
(12).
Sample Dissolution. After irradiation and two days cooling
time, the quartz vials were washed with coned HNOg and distilled
water, cooled in liquid nitrogen to reduce the internal pressure,
and opened. The contents were weighed and transferred to
chemical dissolution containers along with 500 nL of the mul-
tielement carrier. Two types of dissolutions were performed to
test for possible losses of volatile compounds. These were: (1)
Dissolution in glass Erlenmeyer flasks or open Teflon wet-ashing
vessels using 10 mL of coned HNO,, HC104, and HF in a 10:3:1
ratio. (2) Dissolution in Teflon-lined bomb using 10 mL of fuming
HNO3 and HF in 20:1 ratio.
These acid mixtures were tested and found satisfactory for the
subsequent radiochemical procedures.
Hydrated Manganese Dioxide (HMD) Separation. The
solutions obtained from the Erlenmeyer flasks or Teflon vessels
were evaporated to a final volume of approximately 1 mL, cooled.
and brought to a 20-mL final volume with 1 M HNO3. The
solutions obtained from the bomb were evaporated in Teflon
vessels to a volume of about 2 mL. One mL of coned HC104 was
then added and the samples were heated to fumes of HC104.
cooled, and brought to a final volume of 20 mL with 1 M HNO-,.
The 1 M HNOj solutions were then passed through HMD
columns (7-mm diameter and 3-cm height) which had been
preconditioned with 1 M HNOj. The flow rate was adjusted to
0.5 mL/min by varying the packing density of the HMD. Sel-
enium, chromium, arsenic, and antimony were retained on the
HMD bed. Sodium-24, a rr.ajor interference in both the biological
and environmental matrices, passed through in the eluate. The
columns were washed with 20 mL of 1 M HN03, dismantled, and
the HMD was transferred to polyethylene counting vials while
the eluted fraction was saved for the subsequent Cu and Cd
separations.
Metal Diethyldithiocarbamate Extraction. The pH of the
eluted fractions was adjusted to 1.5 with NH4OH, and 100 mg
of Zn holdback carrier was added. The solutions were then placed
in separatory funnels and 20 mL of 0.003 M bismuth diethyl-
dithiocarbamate [Bi(DDC)3] in chloroform was added in order
to extract the copper. The solutions were then shaken for 30 min,
utilizing a shaking machine. The organic fractions containing the
copper were placed in polyethylene bottles for counting. The
cadmium remained in the aqueous fractions and was extracted
with 20 mL of 0.005 M zinc diethyldithiocarbamate [ZnfDDC)2l
in chloroform. Five minutes of shaking time was found sufficient
Table II. Determination of As, Sb, Se, Cr, Cd, and Cu in
NBS SRM Subbituminous Coal (1635)"
Matrice
NBS SRM Subbituminous Coal (1635)
Element
As
Sb
Se
Cr
Cd
Cu
a Concentration in Mg/g- The reported value for each
element consisted of ten determinations. Values in paren-
theses are NBS information values.
certified values
0.42± 0.15
(0.14)
0.9± 0.3
2.5 - 0.3
0.03 ± 0.01
3.6 ± 0.3
this work
0.44 ± 0.05
0.12 ± 0.01
0.82 ± 0.04
2.48 : 0.08
0.029 ± 0.003
3.56 t 0.18
for a complete extraction of the cadmium complex into the organic
phase. The resultant extract was also placed in polyethylene vials
for counting.
The Zn holdback carrier was added to minimize exchange
between activated Zn from the sample and the organic Zn used
for the Cd extraction. Radioactive Zn can interfere with the
detemination of Cd by producing a high background level of
radiation
Standards. Standards for analysis were prepared in two ways.
The first was prepared using the same procedure as used for the
sample dissolution. HMD separation, and M(DDC), extraction.
The second standard was prepared by pipetting a known amount
of irradiated standard solution directly onto HMD m a poly-
ethylene counting vial for the determination of As. Sb, Se, and
Cr. Copper and Cd standards were pipetted into 1 M HN03 acid
and extracted in the same manner as the samples.
Counting. The samples and standards were counted on Ge(Li)
detectors with active volumes of 63 to 75 cm3 coupled to 4096-
channel pulse height analyzers. The HMD fractions were counted
twice; first, 4 days after irradiation for measuring short-lived
radioisotopes of arsenic and antimony, and then 3 weeks after
irradiation for measuring the long-lived radioisotopes of selenium
and chromium. Because of the very low concentration of
chromium in Bovine Liver, and the high background coming from
32P 3~ bremsstrahlung, counting times of from 5 to 10 h were
necessary. The organic fractions containing the Cu were counted
immediately after separation. The organic fractions containing
the Cd were allowed to stand for 24 h to establish the equilibrium
between 115Cd-115mIn. y rays from the decay of both these isotopes
were used for Cd analysis.
Data reduction was accomplished by a quantitative analysis
computer code (13) and hand integration of poorly defined peaks
RESULTS AND DISCUSSION
The results are shown in Tables I and II. The results
indicate very good agreement between certified and found
values, demonstrating that the procedure is accurate.
During the dissolution of the samples in the Teflon bombs
and in open Teflon containers, no appreciable losses of Ah,
Cr, Se, Cu. and Cd were found. Large losses of antimony (up
to 80% ) were observed in some cases during the dissolution
38
-------
in the open Erlenmeyer flasks due to the adsorption of Sb onto
the glass surface of the flasks. The quantitative recovery of
this element was impossible even with repetitive washing with
I M HN03. No losses of antimony were observed, however.
when the samples were dissolved in open Teflon vessels or
in the Teflon-lined bomb.
A previous study by L. T. McClendon (14) cited apparent
losses of organornetallic chromium compounds during initial
acid digestion. Because these volatile compounds were not
characterized, all Bovine Liver samples analyzed in this work
were digested in a Teflon-lined bomb. However, when the
digested samples from the Teflon-lined bomb were boiled with
HC1O4-HNO3, no appreciable losses were observed. These
observations give support to the theory that the first digestion
in the Teflon bomb converts the organochromium compounds
to inorganic chromium, which is not volatile in nonchlonde
systems. Furthermore, standards for analysis that were di-
gested in HC104-HNO3, and those pipetted directly were
always in total agreement.
The retention of arsenic, selenium, chromium, and antimony
on HMD in 1 M HNO3 is highly selective and quantitative,
while the elution of Cd and Cu is complete. The deconta-
mination of the elements of interest from sodium-24 and other
radioactive matrix interferences can be estimated to be a factor
ANALYTICAL CHEMISTRY, VOL 50, NO. 11, SEPTEMBER 1978 « 1481
CONCLUSION
The simplicity of the method permits a rapid isolation of
the radioisotopes of interest without many chemical steps.
The quantitative recovery avoids the calculation of the
chemical yield. The sensitivity and accuracy obtained in thK
work demonstrate the validity and the precision of this ra-
diochemical procedure.
LITERATURE CITED
(li T E Gills, M Gallormi, and R R Greenberg Proceedings of Third
International Conference on Methods in Environmental and Energy
Research, University of Missouri-Columbia, Mo , October 1977
(2) K Heydorn and E Damsgaad, Talanta, 20, 1 (1973)
(3) A Gaudry, B Maziere, and D Comar, J Radioanal Chem , 29, 77 (1976)
(4) H. M N H Irving, J Radioanal Chem , 33, 287 (1976).
(5) K Kudo, T Shigematsu, and K Kobayashi, J Radioanal. Chem , 36,
65 (1977)
(6) M. Galtormi, M DiCasa, R Stella, N Geneva, and E Orvini, J Radioanal
Chem , 32, 17 (1977)
(7) F Girardi, R Pietra, and E Sabbioni, J Radioanal Chem, 5, 141(1970).
(8) J Cuypers, F Girardi, and F Monsty, J Radioanal. Chem., 17, 115 (1973)
(9) A. Wytlenbach and S Bajo, Anal Chem., 47, 1813 (1975)
(10) S Bajo and A Wyttenbach, Anal Chem , 48, 902 (1976)
(11) EC Kuchner, R Alvarez, P J. Paulson, and T J. Murphy, Anal Chem .
44, 2050 (1972).
(12) D A Becker and P D LaFleur, J Radioanal Chem . 19, 149(1974)
(13) H P Yule and H L. Rook, J Radioanal Chem , 39, 255-261 (1977)
(14) EPA Technical Report EPA-600/1-77-020, April 1977, p 44
,,. April IS, 1978. Accepted June 13, 1'JTS.
39
-------
Appendix II
The Quantitative Determination of Volatile Trace
Elements in NBS Biological Standard Reference
Material 1569, Brewers Yeast
Harry L. Rook
Analytical Chemistry Division
National Bureau of Standards
Washington, D. C.
Wayne Wolf
Nutri t ion Institute
U.S. Department of Agriculture
Beltsvi1le, Maryland
ABSTRACT
In the past few years, a large body of analytical data has
been reported on trace levels of chromium in biological samples.
From data on materials such as NBS Standard Reference Material
1577, Bovine Liver, and IAEA standard materials, it is now
apparent that much of the reported Cr data are in error.
It has been suggested that some of the analytical problems
may be due to the presence of a volatile organic complex of Cr
in many biological matrices.
In an effort to resolve the question of Cr volatility, a
series of experiments have been conducted on a new NBS Standard
Reference Material--Brewers Yeast SRM 1569, which has been
certified for Cr content. The experimental design allowed for
the quantitative collection of volatile species in a thermally-
heated vacuum distillation system over a temperature range of
150-325°C. A small fraction of the total Cu (<1%), about 25%
of the total Hg and about 50% of the total Se were trapped and
determined quantitatively. Arsenic, Ag and Au were also
observed in the trapped fraction.
INTRODUCTION
The accurate determination of Cr in biological samples has histori-
•ly been difficult. Analytical methods such as atomic absorption
pectroscopy and neutron activation have the required sensitivity to
determine ng levels of Cr but reported analytical data on biological
materials, such as NBS Standard Reference Material (SRM) 1577 Bovine Liver,
have been widely divergent (5). More recent data have indicated that the
current status of Cr analysis has not greatly improved (6). It has been
suggested that part of the analytical error may be due to the presence of
a volatile organic species which may be lost in certain sample-preparation
procedures (7,8).
In general, analytical verification of volatile trace elemental
spe_ies in biological systems has been limited. Elements such as Hg and
Pb have been shown to form elemental or simple molecular species which
vaporize upon mild thermal treatment. The awareness and confirmation of
possible vaporization losses of other metals which are not normally
considered volatile \s becoming of increasing importance. Two recent
studies have reported no Cr losses during ashing and acid digestion of
Repnnlcd Iran Tr.u-e SuhMancco In F.nuronrnenldl Heallh-XI 1977 A symposium. D D Hemphill. Ed .© University of
Missouri. Columbu
40
-------
biological samples to which radiotracers of Cr-5' were added (5,6). A
more recent study has reported as much as 30Z volatilization loss of Cr
from neutron activated brewers yeast samples during dissolution with
concentrated acids (A).
In an effort to resolve the question of Cr volatility, a vacuum
distillation experiment has been designed and conducted on a new NBS
Standard Reference Material, SRM 1569 Brewers Yeast, which is certified
for total Cr content. The experimental design enabled volatile species
to be collected from samples heated in the range of !50-325°C, and for the
quantitative determination of those elemental species amenable to neutron
activation. In addition to Cr volatility losses were found and quantified
for mercury, Se and As. Other elements such as Br and I were observed to
volatilize but were not quantified due to experimental conditions of the
NA procedure.
MATERIALS AND METHODS
The basic experimental system is pictured in Figure 1. The system
consists of a tube furnace (Lindberg-, Type JJ035 Hevi-Duty, 300W, V/ater-
town, Whv) used to heat a 19 nm i.d. quartz combustion tube containing the
;c.~.p!s a^c 2 ^e^^^^efo r n,-e end of the auartz tube was closed with a
standard taper ground glass removable stopper. The other end was fitted
with a second standard taper joint, to which was sealed a 10-cm piece of
8 mm i.d. quartz tube. This extension was passed straight through a
liquid nitrogen (LM2) trap and was used to collect the distillate. The
trap was constructed of styrofoam with holes bored through the sides to
seal to the collection tube. The collection tube v describe materials and experimental procedures, it
wasoccasicnal ly necessary to identify commercial products by manufacturer's
name or label. In no instance does such identification imply endorsement
oy the National Bureau of Standards nor does it imply that the particular
oroducts or equipment is necessarily the best available for that purpose.
-------
contamination from the inner surface (3)- The combustion tube was also
cleaned and purged by heating to 800°C with a low flow of oxygen through
it. The collection tube was then sealed to the sample tube and the system
again purged. Following the precleaning procedure, approximately one g of
NBS SRM 1569 Brewers Yeast was weighed into an aluminum foil sample boat.
The sample was placed into the combustion tube along with a thermometer
whose bulb rested against the sample boat. The collection tube was then
connected to the vacuum system, liquid N was added to the collection trap,
and the system was evacuated.
The furnace was preheated to the desired temp and placed around the
sample tube. The entire system was then heated to the desired equilibra-
tion temp and maintained at that temp (j^5°C) for 15 min. During this 15"
min period the entire system, up to the LN2 condenser, was heated gently
with the oxygen-gas torch to prevent condensation of the vapors outside
of the collection trap. Care was taken to prevent decomposition of the
vapors due to excessive heat in the tranfer line. The volatilized
material from the sample was collected in the liquid N trap portion of the
collection tube. For several of the higher temp runs, the trap had to be
periodically moved several mm closer to the sample tube to prevent a solid
plug of condensed material from forming inside the collection tube. All
of the condensed material, however, remained at liquid N temp throughout
the run.
At the end of the 15-min collection period, the furnace was removed
and the combustion tube allowed to cool. The collection tube, while still
under vacuum, was sealed off between the combustion tube and the liquid N
collection trap. The other end of the collection tube was then sealed
viith the sample still frozen and under vacuum. The sealed samples were
then set aside for irradiation and analysis.
The analytical portion of the experiment was conducted by non-
destructive neutron activation. For analysis, A of the sealed quartz
tubes containing the condensed volatile material were irradiated together
with one quartz tube containing 300 M! of a mixed trace element water
standard. The k samples were placed in a single irradiation container so
that one sample from each of the temperature levels, ranging from 150-300°
C, were irradiated together. The samples and standard were irradiated for
^ hr at a thermal neutron flux of f- 6xl013 n-cm~2sec~!. After irradiation,
the samples were allowed to decay for 2 days to eliminate short half-
lived radioactivity and counted on a large volume Ge(Li) detector coupled
to a f»096 channel pulse height analyzer. The samples were then allowed
to decay for 2 wk and were recounted to determine those elements with
half lives greater than a few days. The resultant gamma-ray spectra were
recorded on magnetic tape and processed off-line by computer. The indi-
vidual photopeaks were quantified by the total peak area method. Quanti-
tative results were obtained by direct comparison of the specific activity
•f the samples and standard.
RESULTS
Detectable amounts of 6 trace elements were observed in trapped
fractions distilled from the brewers yeast. The amount of each element
trapped was dependent upon the maximum temp to which the sample was heated.
The quantities of those trace elements observed were normalized to the
weight of yeast used in individual experiments and are given in Table I.
The amount of each element in the trapped fraction increased with
increasing temp in all cases except Ag. However, the quantity of the
element volatilized and the dependence on temp varied among the individual
elements, as shown graphically in Figures 2-5-
The data in Table II reflect elemental concen-trat ions in the yeast
residue relative to those in the original unheated yeast sample. These
data are corrected for total wt loss of material from the sample due to
42
-------
vacuum distillation. The elemental data are given relative to initial
yeast concentration because the absolute trace element content of the SRM
Yeast has been determined only for Cr. Approximately 50% of the total Se,
about 25% of the total Hg and less than ]% of the total Cr content were
lost at the 300°C temp. The total weight loss of material was between 30
and 40% at that temp. For Cr these data are consistent with the observed
volatility data. Although a volatile Cr component was observed, the quan-
tity, ^.003--005 pg/g, was within measurement error of the total certified
Cr content in the yeast. Scandium, Fe and Co were not detected in the
trapped fraction and no differences were observed in any residue samples
compared to initial samples.
TABLE I. TRACE ELEMENTAL LOSS FROM BREWERS YEAST
BY VACUUM DISTILLATION
ug trapped/g sample
Temp °C
E lement
150
190
205
250
295
305
Hg
Se
Cr
As
Ag
Au
0.0011*
0.0013
0.0028
0.064
0.0045
0.00010
0.006
0.0027
0.0021
0.060
0.0005
---
---
0.0018
0.0021
0.050
0.0028
0.00006
0.017
0.036
0.0050
0.10
0.0027
0.00040
0.10
0. 16
0.0038
0.22
0.0010
—
0.055
0.23
0.0052
0.32
0.0042
TABLE II. RELATIVE ELEMENTAL COMPOSITION OF YEAST RESIDUE
Concentrations Relative to Unheated Yeast
Temp °C
£lement
200
265
290
305
Sc
Fe
Co
Cr
Hg
Se
% wt loss
0.
1 .
1 .
1 .
0.
1
996
00
.00
.01
.990
.04
0
1 .
0.
1.
0.
1
0
.01
.983
.00
.990
.01
.960
0
0
1
0
1
1
.992
.00
-905
.06
xxx
. 12
7%
1.
I.
1.
0.
1
0
,00
00
.01
.920
.01
.64
121
] .00
! .0!
0.910
1.02
1 .06
0.44
39*
1
0
0
1
0
0
.02
• 983
• 933
.06
.74
.57
32%
DISCUSSION
The appearance of detectable amounts of several trace elements in the
vacuum distillate of a sample of brewers yeast suggests the presence of
volatile elemental species in biological materials. For some metals such
as Hg, these species may volatilize at temps lower than 150°C, while for
the other elements observed in this work the volatilization began between
200 and 250°C. This is also the temp at which visible charring of the
brewers yeast occurred (Table II). Thus, the volatility and loss of these
elements appears in general to be more of a thermal degradation of the
natrix with concomitant release or decomposition of the naturally
occurring metallic species, resulting in a volatile product. The
43
-------
analytical significance of these losses, especially in methods which rely
upon thermal decomposition of the organic matrix prior to solubiIization
and analysis, is obvious. For example, the data from this study indicate
that up to 50$ of the total Se content can be volatilized upon moderate
heat ing.
ZOO
29O
3OO
TEMPEMATUftE
FIGURE 2- COLLECTED VOLATILE MATERIAL VS. TEMPERATURE-rtERCURY.
Figures 2-5 show some very interesting trends. If one assumes that
the ideal pattern for volatilized material would follow a curve as
depicted in Figure 6, there is. a temp, A, below which no loss occurs. At
temp A, losses begin due to natural volatility or thermal degradation. At
any temp above A, the amount of volatfliz-ed trace element will be deter-
mined b-' the rate of formation, vapor pressure of the volatilized species
and '.he length of time at that temp. In this experiment the length of
time at the maximum tetnp was held constant; therefore, the amount of
material volatilized was dependent upon the rate of formation of the
volatile species, up to the limiting total amount of that species or its
precursor. At point 8 in Figure 6, the limiting condition has been
reached and no more material is volatilized.
-------
s«
10
,0*
10"
SELENIUM
ISO
200 290
TEMPERATURE *C
3OO
FIGURE 3- COLLECTED VOLATILE MATERIAL VS. TEMPERATURE-SELENIUM.
If we examine the experimental data for the different elements in
terms of this idealized curve, a consistent trend is apparent. For Hg
(Fig. 2) we see a logarithmic increase throughout the temp range studied,
indicating that we are between transition points A and B for this element.
The initial volatilization temp for Ho is below 150°C. At 300°C, a
significant portion of the total volatile Hg was not vaporized in the 15-
min distillation period; thus point B had not been reached. This inter-
pretation is consistent with the Hg loss data in Table II.
The data for Se (Fig. 3) show the entire curve as idealized. The
initial volatilization point is around 175°C and we see that the amount
of trapped Se appears to be reaching a maximum at 300°C. Thus, a signif-
icant portion of the total volatile species has been vaporized in 15 min
at 300°C. Again, the loss data in Table II are consistent in that approx-
imately 50% of the total Se in the sample was lost. This implies that
there are at least 2 different fractions or species of Se in the brewers
yeast and that they are approximately equal in amount. V/ith the knowledge
that different chemical species or forms may have widely differing effects
-------
or functions in biological systems, this confirmed multiplicity of Se
species is of great significance.
ffl
A«
.0*
ARSENIC
ISO
200
290
—I—
900
TEMPERATURE *C
FIGURE k- COLLECTED VOLATILE MATERIALS VS. TEMPERATURE-ARSEN1C.
For As (Fig. k) we see the beginning of the idealized curve with
transition point A at about 200°C. At the maximum temp of 300°C we have
not yet volatilized a significant portion of the total As species as the
curve is still increasing logarithmically.
The amounts of trapped Cr (Fig. 5) are significantly higher for the
sanples collected above 200°C than for those collected below that temp
although the absolute amounts are very low. These data show that about
5-10 ng/g of Cr is volatilized by thermal degradation between 150°C and
300°C. Although this is a very small fraction of the total content, it
signifies a fraction of Cr in this sample which is chemically different
from t.ie bulk. It has been shown that determination of Cr in biological
materials by thermal heating and atomization in graphite furnaces is very
susceptible to chemical form and matrix in which it occurs (7,8). This
existence of a fraction of Cr that can be directly distilled supports
those earlier findings. In the studies which reported no losses of Cr
during ashing and acid digestion with added radiotracer (1,2) there was no
46
-------
assurance that the Cr-51 tracer was present in, or had exchanged with,
the organic forms of Cr in the samples. Also, the experiments were
designed to determine the amount of Cr retained in the sample after treat-
ment, but were not specifically designed to determine small Cr losses via
volatilization. In the recent study reporting loss of up to 39% of the Cr
during dissolution with concentrated HClOi, and HNOa , mixtures were
collected and the losses due to volatility were quantitatively verified
('() . Due to the prior neutron irradiation and the severe conditions of
acid dissolution, it is highly probable that the Cr species originally
present in the sample was changed during the procedure. In our studies,
the sample was subjected to only a thermal stress and no chemical treat-
ment was carried out. These studies do reemphasize the importance of
initial chemical speciation and matrix upon the potential for formation of
volatile species during thermal or chemical reactions.
i or
200
250
300
TEMPERATURE *C _
FIGURE 5- COLLECTED VOLATILE MATERIAL VS. TEMPERATURE-CHROMIUM.
In summary, we have shown that chemical species of several trace
elements exist in brewers yeast which can be volatilized by thermal
heating under vacuum distillation, and that the temperature dependence of
the quantitative distillation varies among the different trace elements.
47
-------
We have also shown that in a given biological material, some trace
elements exist in more than one chemical state. The potential exists for
use of this type of differential pyrolysis or fractional distillation in
studies of chemical speciation of these trace elements.
100
200
TEMPERATURE *C
300
FIGURE 6- IDEALIZED VACUUM DISTILLATION CURVE.
LITERATURE CITED
Jones, G. B., R. A. Buckley and C. S. Chandler. 1975. The volatil-
ity of chromium from brewers yeast during assay. Anal. Chem. Acta
80:389-392.
Koirtyohann, S. R. and C. A. Hopkins. 1976- Losses of trace metals
during the ashing of biological materials. Analyst 101:870-875-
Kjehner, E. C., R. Alvarez, P. J. Paulsen and T. J. Murphy. 1972.
Production and analysis of special high purity acids purified by
sub-boiling distillation. Anal. Chem. kk:2050-2056-
McClendon, L. T. Determination of chromium in biological matrices
using NAA application of SRM's. Radio Anal. Chem. 8:(Accepted for
publicat ion).
McClendon, L. T. 197't- Selective determination of chromium in
biological and environmental matrices. In: Tvaae Substances in
48
-------
Environmental Health - VIII, D. D. Hemphill, Ed., University of
Missouri, Columbia, pp. 255~257-
6. Parr, R. 197&- Problems of chromium analysis in biological
materials: An international perspective with special reference to
results for analytical quality control samples. In: PTOC. Int.
Conf. Modern Trends in Activation Analysis (to be published).
7. Wolf, W., W. Mertz and R. Masironi. 197^- Determination of chromium
in refined and unrefined sugar by oxygen plasma ashing by flameless
atomic absorption. J. Ag. Food Chem. 22:1037.
8. Wolf, W. 1976. Preparation of biological materials for chromium
analysis. In: Proa. 7fh Materials Research Synrp. , NBS Special
Publication ^22, 605-610, U.S. Government Printing Office, Washington.
DISCUSSION
Inquirer: W. R. Faulkner, Vanderbilt University, Nashville, TN
Q. Is it likely that vol Cr might be appreciable at temps greater than
305°C and less than charring temps?
A. Not likely because Cr" cone is beginning to plateau at temps less than
305°C. There was difficulty because of clogging of trap at temps
greater than 305°C.
Inquirer: Richard A. Peabody, V.A. Hospital, Albany, NY
0_. Has chromyl chloride been observed directly in your analytical system
for volatile components, presumably formed by interaction of HC1 with
Cr in digests?
A. No, not in our system. The procedure used in this study was vacuum
distillation of volatile materials directly from the brewers yeast.
No acid digestion, using HC1 or any other acids, was used. Also, the
neutron activation technique used to quantify the amounts of trace
elements volatilized would not detect chemical compounds, such as
chromyl chloride; it is a total elemental technique.
-------
Appendix III
Chemical Preparation of Biological Materials
for Accurate Chromium Determination by
Isotope Dilution Mass Spectrometry
Lura P. Dunstan and Ernest L. Garner
National Bureau of Standards
Washington, D. C.
ABSTRACT
The current interest in trace elements in biological
materials has created a need for accurate methods of analysis.
The source of discrepancies and variations in chromium concen-
tration determinations is often traceable to inadequate methods
of sample preparation. Any method of Cr analysis that requires
acid digestion of a biological matrix must take into considera-
tion the existence or formation of a volatile Cr component. In
addition, because Cr is often present at concentrations less
than 1 vg/g, the analytical blank becomes a potential source of
error.
Chemical procedures have been developed for the digestion
of the biological matrix and the separation of Cr without either
large analytical blanks or significant losses by volatilization.
These procedures have been used for the analysis of NBS Standard
Reference Material (SRM) 1569 Brewers Yeast; SRM 1577 Bovine
Liver; SRM 1570 Spinach and other biological materials including
human hair and nails. At this time, samples containing 1 ug of
Cr can be determined with an estimated accuracy of 2%.
INTRODUCTION
The accurate determination of chromium in biological materials
presents a considerable challenge to the analytical chemist. Because the
concentration of Cr is often less than 1 yg/g, it is imperative that the
analytical method be sensitive and non-contaminating. Isotope dilution
mass Spectrometry (IDMS) is capable of precise and accurate concentration
determinations on submicrogram amounts of an element, and as a result, a
major limitation of this method is the magnitude and variability of the
analytical blank. Through the use of clean room facilities, pure reagents
and apparatus fabricated from quartz and Teflon, the variability of the
Cr analytical blank can be maintained at a level of less than 10 ng.
Control of the analytical blank at this level permits the determination of
1 ng of Cr with an estimated accuracy of +2%.
For biological samples, accuracies of this magnitude are not only
dependent on controlling the analytical blank, but on preventing loss of
the volatile Cr compounds during the wet oxidation of the matrix. The
presence of volatile Cr compounds in various biological materials has been
discussed (I,1*) and it is suspected that many of these compounds may
volatilize at low temperatures. In addition, the formation of volatile
Cr compounds during the wet oxidation process must be avoided. For this
reason, perchloric acid should not be used since the formation of chromyl
chloride is likely to produce losses of Cr. The loss of other volatile
components can be controlled by performing the wet oxidation procedure in
a quartz reflux system.
Reprinted from Trace Substances In EnMronmental Heallh-XI 1977 A symposium. D D HemphiD. Ed .© University of
Missouri Columbia
50
-------
Once the digestion of the matrix is complete and the spike and
natural Cr have equilibrated, it is no longer necessary to maintain
complete recovery. However, recoveries greater than 80% are essential for
the accurate determination of ng amounts of Cr, since a few ng of blank
are sufficient to significantly bias the concentration measurement. Thus,
understanding and controlling the effects of the analytical blank and the
loss of volatile Cr compounds are the major factors in the accurate deter-
mination of Cr in biological materials by IOMS.
MATERIAL AND METHODS
The dissolution, separation and purification of all samples were
performed in a Class 100 clean air environment to minimize particulate
contamination. With the exception of the eerie sulfate solution, all
reagents used for these analyses were purified by sub-boiling distillation
(2). The eerie sulfate was prepared from high purity cerium (IV) oxide by
several evaporations with sulfuric acid.
The 50Cr spike, a 96% enriched isotope, was calibrated by comparison
with 2 different solutions of high purity natural Cr. Weighed aliquots
of the spike and natural solutions were mixed and equilibrated and the
concentration of the spike was determined by IDMS.
The quartz reflux system used for the wet oxidation of the organic
matrix consisted of a 125 ml Erlenmeyer flask fitted with a quartz conden-
ser. The refluxing action was aided by filling the condenser with quartz
chips and by running chilled water through the condenser jacket.
The chemical separation procedure is similar for all biological
materials discussed in this paper. The sample is placed in a quartz
Erlenmeyer flask and 50Cr spike is added to alter the natural s°Cr/52Cr
ratio. Sample wts ranging from 1 g for spinach and brewers yeast up to 5
g for blood serum were used. The flask is then fitted with a condenser
and chilled water is run through the jacket for approximately 1/2 hr. Ten
g of concentrated (70%) nitric acid are added through the top of the con-
denser and the temp is increased gradually to avoid excessive foaming and
frothing of the sample. After any initial reactions have ceased, 3 g of
concentrated (98%) sulfuric acid are added. This mixture is digested for
approximately 5 hr while gradually increasing the temp to ^250°C. The
condenser jackets are then drained and the refluxing action is continued
overnight. In cases where silica is present in the matrix (e.g. brewers
yeast and spinach), the contents of the flask are transferred to Teflon
beakers and treated with 1 g of concentrated (36%) hydrofluoric acid.
The Cr is then oxidized with eerie sulfate, separated by extraction
with methyl isobutyl ketone and back extracted into water. The resulting
solution is evaporated to dryness and the purified Cr is analyzed by
surface ionization mass spectrometry.
All isotopic ratio measurements were made using a 15~cm radius of
curvature, solid sample, single focusing mass spectrometer. The details
of the tungsten filament surface ionization procedure for Cr will be
published (1). All isotopic ratio data were corrected to the absolute
isotopic composition of a Cr isotopic reference standard (SRM 979) which
was analyzed under the same conditions as the samples.
RESULTS AND DISCUSSION
The results of Cr determinations on various biological materials are
shown in Tables l-lll. The effect of the blank varies with the concen-
tration level, and as the amount of Cr determined approaches the submicro-
gram range, the accuracy of the method becomes primarily dependent upon
the magnitude and variability of the blank. The uncertainty for the
determination of Cr in Brewers Yeast SRM 1569 (Table I) is the estimated
accuracy and includes allowances for the precision of the ratio measure-
51
-------
ment, the calibration of the 50Cr spike, the variability of the Cr analyt-
ical blank and sample inhomogeneity.
TABLE I. CHROMIUM CONCENTRATION IN
SRM 1569 BREWER'S YEAST
Bottle No.
5
127
128A
ISA
3^0
253A (!)
253A (2)
2.
2.
2.
2.
2,
2.
2.
Average 2
,g Cr/g
07
.10
.09
.08
.23
.13
. 12
.12 +_ 0.13a
a
The uncertainty (ts) is the esti-
mated accuracy and includes allow-
ances for the precision of the
ratio measurement, the calibration
of the 50Cr spike, the variability
of the analytical blank apd sample
inhomogene i ty.
TABLE II. CHROMIUM CONCENTRATION IN
SRM 1577 BOVINE LIVER
Bottle No.
1
2
3
k
No. of
Determinations
A
3
2
l»
Average
ng Cr/g
95.6
85.3
89.3
97.8
92. Oa
Analytical blanks averaged 13 ^ 7 ng
TABLE III. CHROMIUM CONCENTRATION IN
OTHER BIOLOGICAL MATERIALS
Matrix
SRM 1570, Spinach
Hair
ToenaiIs
Blood serum
52
-------
Since both SRM 1569 Brewers Yeast and SRM 1577 Bovine Liver were
suspected of containing volatile Cr compounds, an attempt was made to
check for possible losses. This process involved taking additional
samples of each SRM, which were spiked after the wet oxidation of the
matrix had been completed. Any losses of Cr up to this point in the
analysis should produce a measured ratio depleted in natural Cr and thus
yield lower concentration data. In both cases discussed above, the Cr
concentrations determined by spiking before and after dissolution were
within the estimated uncertainty of the analysis and as a result it would
appear that if a volatile constituent is being lost, the amount is proba-
bly very sma 1 1 .
The bovine liver data in Table II exemplify the analytical problems
encountered when analyzing ng levels of Cr. These measurements were part
of a preliminary experiment to evaluate the IDMS technique for Cr in the
bovine liver matrix and to test for sample inhomogeneity. Two-g samples
of the liver were analyzed so that the total amount of Cr was approxi-
mately 200 ng. However, because the variability of the analytical blank
was +7 ng, the accuracy was estimated to be +4% for each determination.
At this concentration level, the effect of even relatively small changes
in the blank made it difficult to ascertain the cause of sample variations.
Table Ml contains a summary of concentration determinations on other
biological materials. The blood serum represents the lowest Cr concen-
tration determined to date. Although 5~g aliquots of serum were taken for
analysis the magnitude of the analytical blank was 25% of the total Cr in
the sample.
CONCLUSION
By employing techniques that minimize the magnitude and variability
of the analytical blank as well as losses of volatile Cr compounds, the
concentration of Cr in biological materials can be accurately determined.
Although the magnitude and variability of the analytical blank becomes
increasingly significant as the amount of Cr approaches the low ng range,
uncertainties of 2% or less are possible for the determination of at least
one yg of Cr.
ACKNOWLEDGMENTS
We wish to express our appreciation to Joy J. Shoemaker and Karen A.
Brletic for manuscript preparation, to William A. Bowman,!I I for instru-
ment maintenance support and to our colleagues of the Analytical Spectro-
metry Section for their helpful discussions.
LITERATURE CITED
I. Dunstan, L. P. and E. L. Garner. 1977- The determination of
chromium by surface ionization mass spectrometry. (In preparation).
2. Kuehner, E. C., R. Alvarez, P. J. Paulsen and T. J. Murphy. 1972.
Production and analysis of special high-purity acids purified by sub-
boiling distillation. Anal. ne". 44:2050-2056.
3. Wolfe, W. R. and F. E. Greene. 1976. Preparation of biological
materials for chromium analysis. In' 4j:^, ~.^le ".^••dli--\c, i.m'.jsis. P. 0. LaFleur, Ed., NBs"spec.
Publ. 422, Vol. I, U.S. Government Printing Office, Washington,
pp. 605-610.
4. Wolfe, \l. R. , W. Mertz and R. Masironi. 1974. Determination of
chromium in refined and unrefined sugars by oxygen plasma ashing
flameless atomic absorption. •'. /.-. Pc:~ Che^, 22:1037-1042.
53
-------
Appendix IV
The Determination of Zinc, Cadmium and
Lead in Biological and Environmental Materials
by Isotope Dilution Mass Spectrometry
J. W. Gramlich, I. A. Machlan
T. J. Murphy and L. J. Moore
National Bureau of Standards
Washington, 0. C.
ABSTRACT
Techniques have been developed for the accurate and pre-
cise determination of the concentration of zinc, cadmium and
lead using thermal ionization mass spectrometry. These tech-
niques have been applied to the analysis of a variety of Biolog-
ical materials such as blood, hair and nails, and to environ-
mentally related materials such as water, air particulates and
fossi1 fuels.
Uncertainty in the accuracy of the isotope dilution method
is less than 0.5% for all 3 elements, and is nearly independent
of element concentration from the % level down to the ng/g (ppb)
range where the analytical blank becomes a significant contri-
bution to the uncertainty.
The mass spectrometric methodology and chemical separation
procedures are presented along with examples showing the effect
of sample impurities on the quality of the analytical data.
INTRODUCTION
The toxicity of cadmium and lead, and the associated hazards result-
ing from increased pollution of the environment by these elements, has
received considerable attention during the past decade. The serious patho-
logical effects of Cd (3,8) and the hematological disturbances associated
with lead poisoning (2) have been extensive!/ reviewed in the literature.
Zinc is an essential element for growth and maintenance of body functions;
it is a constituent of several enzymes and a cofactor for certain enzy-
matic reactions. There is ample evidence to suggest that the metabolisms
of Cd and Zn are closely related and that Cd has the ability to exchange
with Zn and thus produce changes in enzymatic activity (3). The close
biochemical and geochemical association of Zn and Cd make their collective
study in biological and environmental systems highly desirable.
As the concentration of these trace elements approaches the ng/g
(ppb) range, it becomes increasingly difficult to maintain accuracy and
precision in routine analytical measurements. Well characterized trace
element standard materials are essential to ensure quality control and to
develop or verify analytical methods. Although not suited for routine
analysis because of the time and expense involved, isotope dilution mass
spectrometry provides an extremely valuable tool for trace element anal-
ysij of standard materials, for refereeing discrepancies between other
analytical methods and for determining small sample inhomogeneities.
Samples to be analyzed are spiked with a known amount of a separated
isotope of the element of interest. After dissolution of the sample and
equilibration of sample and spike, quantitative recovery is no longer
necessary because only the altered isotopic ratio is of interest. Herein
Reprinted from Trace Suhstances In Emironmentdl Heallh-XI W A symposium. D D Hemphill. Ed . iQ Lniversn> of
Missouri. Columbia
54
-------
lies a major advantage of the isotope dilution technique over other
analytical methods. Because only a representative portion of the sample
is needed for analysis, rigorous chemical purification can be employed to
obtain a high purity, matrix-free sample that can be compared directly
with pure standards of known absolute or relative isotopic abundances.
The high degree of sensitivity afforded by the mass spectrometric
method allows the concentrations of Zn, Cd and Pb, as well as many other
elements, to be determined with both high accuracy and precision at con-
centrations below one mg/g. The practical lower limit of concentration
measurements is governed by the magnitude and variability of the analyt-
ical blank.
MATERIALS AND METHODS
Accurate mass spectrometric measurements require a high purity sample
for analysis. Trace amounts of impurities may result in reduced signal
intensity and alteration cf the filament fractionation pattern (k). In
addition, isobaric interferences from Ge, Ni, Pd, Sn and In may affect the
determination of Cd and Zn. Although the exact chemical separation and
purification procedures may vary somewhat, depending on the particular
IMCJ I i IA, IMC i u . luwiny piuccOui'c ib ycilti'ai i j app . ICdDit. n r^,iO^i\ ciiTtO^t. t
of sample is spiked with weighed aliquots of 6 Zn, Cd, and Pb
solutions. The sample is then digested with a mixture of hydrochloric,
nitric and perchloric acids. After dissolution and equilibration of the
sample and spike, the 3 cleh.er.ts are separated from each other and the
matrix by anion exchange chromatography. Further purification of Pb and
Cd is accomplished by electrodeposition: Pb by anodic deposition as Pb02
(1) and Cd by cathodic deposition from an ammoniaca1 solution. Further
purification of Zn is frequently necessary to remove traces of Fe which
interfere witn the mass spectrometric analysis. This is accomplished by
additional anion exchange chromatography.
Since the limiting factor for tre accuracy of trace element analysis
is often the magnitude and variability of the analytical blank, all chem-
ical preparation is conducted in a clean room facility (7) using acids and
water prepared in this laborator/ by sub-boiling distillation (5). A
class 100 c'ean air erv'"cr"-e-t is also utilized to dry the samples on the
mass spectrometer filaments.
All analyses reported here were performed on 30 cm radius, 90° mag-
netic sector mass spectrometers utilizing thin lens z-focusing sources and
Faraday cage collectors equipped with 50? transmission grids shadowing a
series of suppression grids (10). The suppression grids are designed to
provide cubic suppression of secondary particles between the transmission
grid and the collector-defining slit The remainder of the measurement
circuits consisted of a vibrating reed electrometer, voltage to frequency
converter, sealer and progranmable calculator. Timing, magnetic field
switching and data acquisition were controlled by the programmable calcu-
1 ator
Zinc, Cd and Pb were analyzed frof separate loadings onto single
filament rhenid-> sources employing tne silica gel-phosphor i c acid tech-
nique for lonization enhancement 0,?.'-
RESULTS AM jISCUSSION
The concentration of Pb at or celow the -g/g level has been routinely
deterr-.ined in this laboratory for several years. Samples relevant to this
discussio" include porcine, bovine and human olooa (0.03-1 i.g/g) , human
hair and nails (28 and 1 6 ,q/g respectively), botanical materials (1-10
>.g/g) , gasol ine i'l2-700 -q/g) and water (0.02-1 ~g/c) . The determination
of Zn and Cd concentrations D/ isotope dilution nas been recently under-
taken, based on results v.hicn indicated that suff c'ent signal intensities
c^uld be Generated using the s lica ge 1-pnosohoric acid technique (6,9).
55
-------
Cadmium produces signal intensities and stabilities comparable to
those obtained with Pb; however, the ionization appears to be easily
poisoned by trace impurities, especially Zn. Electrodeposition as a final
purification step has been found necessary to obtain maximum signal inten-
sity. The analytical blank for Cd is less than 1 ng, which is substan-
tially better than the typical 1-5 ng blanks for Pb. Measurement preci-
sion at present appears to be in the range of 0.1^-0.22 (see Table III).
Zinc produces a signal intensity at least an order of magnitude
lower than Cd or Pb. This fact, combined with relatively high and varia-
ble blanks, restricts the practical concentration limits to samples con-
taining more than 100 ng of Zn. Fortunately, Zn is abundant in nature and
samples below the u/g level are not prevalent. Further research is
expected to identify and reduce the sources of the blank.
The results obtained by isotope dilution mass spectrometry on 2 new
candidate NBS Standard Reference Materials are reported in Tables I and II.
These SRMs should be of particular interest to those involved in environ-
mental monitoring.
TABLE I. ZINC, CADMIUM AND LEAD IN URBAN PARTICULATES (SRM 1648)
Bottle No.
40
148
162
757
808
1636
Average
Std. Dev.
Range
Percent Zn
0.4769
0.4784
0.4730
0.4768
0.4741
0.4761
0.4759
0.0019
0.0054
ng cd/g
76.06
75.65
77-98
75-95
76.14
76.05
76.31
0.84
2.33
Percent Pb
0.6597
0.6590
0.6595
0.6598
0.6579
0.6611
0.6595
0.0010
0.0032
TABLE II. ZINC, CADMIUM AND LEAD IN COAL
(Cadidate NBS SRM 1632a)
Bottle No.
1
2
16
93
Average
Std. Dev.
Range
tig Zn/g
26.94
26.88
27-09
27-82
27- 18
0.43
0.94
ug Cd/g
0.1688
o. 1698
0.1692
0.1746
0.1706
0.0027
0.0058
pg Pb/g
12.21
12.35
12.10
12.33
12.25
0. 12
0.25
56
-------
TABLE III. REPETITIVE MASS SPECTROMETRIC ANALYSES OF ZINC,
CADMIUM AND LEAD IN COAL - BOTTLE NO. 1
(Candidate NBS SRM 1632a)
Average
Std. Dev.
Range
jjg Zn/g
26.932
26.938
26.973
26.919
26.914
26.935
0.023
0.059
ug Cd/g
0.16895
0. 1687!
0. 16896
0. 16869
0. 16893
0.16885
0.00014
0.00027
ug Pb/g
12.206
12.203
12.205
12.206
12.204
12.205
0.001
0.003
As mentioned previously, trie precision of the isotope dilution method
allows the identification of sample inhomogeneities which are often masked
by the imprecision of other analytical techniques. The data in Table III
represent 15 separate mass spectrometric analyses on a one g sample taken
from bottle No. I of SRM I632a. These data are typical of the precision
expected from the mass spectrometric measurements. Comparison of the data
in Tab'e III w'i_h those in Table II indicates that the variation between
bottles is the result of sample inhomogeneity rather than measurement
imprecision. Statements on the accuracy of the concentration measurements
must include not only the measurement precision and blank uncertainty con-
tribution, but also components to cover possible sources of systematic
errors such as the uncertainty in the concentration of the separated iso-
tope and the isotopic composition of the separated isotope and sample.
Thus, although the mass spectrometric precision for Pb analyses is gener-
ally less than O.I?, an accuracy statement of ^0.2% is generally assigned
to the data. At present the accuracy of the Zn and Cd analyses is esti-
mated to be +_ 0.5%. This value is expected to decrease as experience is
gained and sources of possible systematic errors are further investigated.
CONCLUSIONS
Isotope dilution mass spectronetry is an accurate and precise analy-
tical technique capable of determining Zn, Cd and Pb concentrations from
the % level into the ng/g concentration range where the analytical blank
becomes a limiting' factor in the accuracy. The technique is particularly
useful for the standardization of reference materials and for measuring
small sample inhomogeneities.
LITERATURE CITED
1. Barnes, I. L., T. J. Murphy, j. VI. Grarrlich and W. R. Shields. 1973.
Lead separation by anodic deposition and isotope ratio mass spectro-
metry of microgram and smaller samples. Anal. Che". 45:1881-1884.
2. de Bruin, A. 1971. Certain biological effects of lead upon the
animal organism. Arcr.. Er.-j. '-.It'".. 23:2'9~26).
3. Friberg, L. T., M. piscator, G F Nordberg and T. KjellstrSm. 1974.
,'j.^i.tr ir I.V-.& Snvi.Tcr,",en^, 2nd ed. CRC Press, Cleveland.
4. Gramlich, J. W. 1976. The effect of sample purity on mass spectro-
metric isotope ratio measurements. In: Pros. 24th 4nK. Ccmf. Afass
^."c •^rr'-et/v a".d Al lied Tjr iss, San Diego, pp. 4g8-499-
5. Kuehner, E. C., R. Alvarez, P J. Paulsen and T. J. Murphy. 1972.
57
-------
The production and analysis of acids purified by sub-boiling distil-
lation. Anal. Chem. kk.-2050-2056.
6. Moore, L. J., J. W. Gramlich and L. A. Machlan. 1975- Application
of isotope dilution to the high accuracy trace analysis of environ-
mental and health standards. In: Trace Substanaes in Environmental
Health - IX, D. D. Hemphill, Ed., University of Missouri, Columbia,
PP. 311-316.
7. Murphy, T. J. 1976. The role of the analytical blank in accurate
trace analysis. In: Accuracy in Trace Analysis: Sampling, Sample
Handling and Analysis, P. D. LaFleur, Ed., NBS Spec. Publ. 422, U.S.
Government Printing Office, Washington, pp. 509-539.
8. Nilsson, R. 1969. Aspects on the Toxicity of Cadmium and its
Compounds. Swedish Nat. Sci. Research Council, Ecological Research
Comm. Bull. 7, Stockholm.
9. Rosman, K. J. R. and J. R. Delaeter. 197^. The abundance of cadmium
and zinc in meteorites. Geochim. Cosmochim. Acta 38:1665-1677.
10. Shields, W. R., Ed. 1966. Analytical Mass Spectrometry Section:
Instrumentation and Procedures for Isotopic Analysis, NBS Tech. Note
277, U.S. Government Printing Office, Washington, pp. 1-99.
58
------- |