svEPA
                                United States
                                Environmental Protection
                                Agency
                                Chesapeake Bay
                                Program
                                Annapolis MD 21401
                                Research and Development
                                EPA-600/S3-82-085  Oct. 1982
Project Summary
                                The Characterization of the
                                Chesapeake Bay:  A Systematic
                                Analysis  of Toxic  Trace
                                Elements
                                Howard M. Kingston, Robert R. Greenberg, Ellyn S. Beary, Billy R. Hardas,
                                John R. Moody, Theodore C. Rains, and Walter S. Liggett
                                 As part of a multidisciplinary study
                               of Chesapeake Bay, the National
                               Bureau of Standards (NBS) was asked
                               to develop the techniques and proce-
                               dures necessary to measure trace and
                               toxic element concentrations within
                               the water column throughout the
                               length of Chesapeake Bay. The Inor-
                               ganic Analytical Research Division of
                               the Center for Analytical Chemistry at
                               NBS has completed  the analysis for
                               selected elements (Cd, Ce, Co, Cr, Cu,
                               Fe, Mn, Mo, Ni, Pb. Sc, Sn, Th, U, and
                               Zn), including some elements  at
                               concentrations consistently below
                               one picogram per milliliter (part per
                               trillion). The characterization of Ches-
                               apeake Bay can  be divided into five
                               major  phases. The first included the
                               development and construction of a
                               sampling system for the trace metallic
                               elements dissolved in water and a
                               filtration system for collecting the
                               particulate elemental component.
                                 The  second phase consisted of
                               sampling chemical stabilization by
                               acidification and storage of the sam-
                               ples in the field. The total complement
                               of 102 samples was obtained, filtered,
                               acidified  and stabilized.  There were
                               also 51  replicate bottom samples
                               obtained and frozen for archival use. A
                               series  of over 30 blanks was also
                               prepared and integrated with  the 102
                               water samples to be analyzed.
                                 The third major phase of activity
                                consisted of the chemical separation
                                and preparation of samples for the
                                analytical instrumental methods. The
                                chemical separation/sample prepara-
                                tion  stage of  this work has been
                                described in the literature for both
                                instrumental techniques.
                                 The fourth major phase consisted of
                                the instrumental analysis of the
                                samples for the trace elements. The
                                total number of elemental concentra-
                                tions resulting  from the analyses of
                                the contracted elements exceeded
                                3000 and involved several thousand
                                more unreported analyses totaling
                                over 5000 separate determinations.
                                 The fifth major phase involved data
                                reduction and evaluation of the statis-
                                tical  significance of the  blank. The
                                blanks were statistically modeled for
                                each element,  and the  blank and
                                uncertainty of the blanks were applied
                                to the data. The concentrations were
                                adjusted uniformly to at least the 95
                                percent confidence limit.
                                 This Project Summary was devel-
                               oped by EPA's Chesapeake Bay
                               Program, Annapolis, MD, to announce
                               key findings of the research project
                               that is fully documented in a separate
                               report of the same title (see Project
                               Report ordering information at back).

-------
Introduction
  This  report  describes the National
Bureau of Standards' (NBS) efforts in a
multidisciplinary study  of Chesapeake
Bay  coordinated by the U.S. EPA's
Chesapeake Bay Program. The  NBS
used the best  available technology to
determine the  trace and toxic element
concentrations in the water column. As
part of  this program,  the NBS has
collected and  analyzed both the dis-
solved  and suspended  participate
fractions of 102 water samples covering
the entire length of  Chesapeake Bay.
The elements of interest include Cd, Ce,
Co, Cr, Cu, Fe,  Mn, Mo, Ni, Pb, Sc, Sn,
Th, U,  and Zn. These  analyses were
accomplished  using specific chemical
preconcentration,  separations and
manipulations  to prepare  the samples
for analysis by Neutron Activation
Analysis (NAA) and Graphite Furnace
Atomic  Absorption Spectrometry
(GFAAS).
  Except for neutron activation analysis
and  anodic stripping voltammetry,  no
analytical techniques are currently
available  for  the untreated sample
determination  of trace elements  in
seawater at concentrations below five
fjg L~\ Usually it is necessary to pre-
concentrate the trace elements from a
large volume and separate the transition
elements  from the alkali  and alkaline
earth elements. In such sample prepa-
rations, the efficiency of concentration,
completeness  of separation, and total
analytical blank  become critical to the
final instrumental method.
  A more recent separation procedure
utilizing  Chelex resin  produced a
sample devoid  of alkali, alkaline earth,
and halogen elements, and left a dilute
nitric acid/ammonium  nitrate matrix
containing only the trace elements of
the seawater  sample. This procedure
was used in conjunction with GFAAS to
analyze Chesapeake Bay  estuarine
samples.

Procedure/Methodology
  The method  of preparation described
for solid  samples  from  100  mL  of
estuarine or seawater uses Chelex 100
resin, followed by the determination of
12 trace elements by NAA. This proce-
dure has  been used to analyze  NBS
SRM 1643a, as  well as high salinity
water samples collected near the mouth
of Chesapeake  Bay.
  The extremely low trace concentra-
tions in these estuarine waters caused
the procedural  blank to be of paramount
importance. The integrity of the sample
can be  compromised  by just a  brief
exposure to normal laboratory air or less
than exhaustively cleaned container
materials, etc In addition, the extremely
high concentrations of alkali, alkaline
earth, and  halogen elements  in  the
marine water .matrix  make  direct
analysis difficult or impossible for most
analytical techniques.
  To circumvent these problems, spe-
cial chemical and instrumental proce-
dures were developed and chemical
separation/preconcentration   proce-
dures based on the chelating resin
Chelex-100 were applied prior to NAA
and GFAAS analysis. The elimination of
the matrix elements allowed the deter-
mination of many elements that could
not otherwise  be analyzed and  en-
hanced the sensitivity of other elements
of interest.  The control of the blank in
this procedure has enabled its  contri-
bution to be low enough not to limit the
measurement of  most elements in
pristine  samples.
  To ensure sample  integrity and
accurate analytical blank determina-
tions, 30 dissolved and particulate
blanks were prepared during the sample
collection. The blanks were then carried
through all manipulations and analyses
as additional samples  interspersed
throughout  the  analyses, with a mini-
mum of three per set.
  Evaluations were made using com-
puter-assisted statistical comparisons
with data of known statistical reliability
The analysis, blank contribution, correc-
tions, and mathematical manipulation
of the data in this report have resulted in
58  data sets which are of known
statistical reliability. These data sets
contain  the sample numbers arranged
in a numerical sequence approximating
the geological arrangement of Chesa-
peake Bay, from Susquehanna River to,
and including, the Atlantic Ocean. The
concentrations are given as  a  best
value, and  a  maximum and minimum
value which represent at least  the 95
percent  confidence limit of the concen-
tration. The significant figures of each
concentration are  determined  by the
range of the  maximum and minimum
value.
  The potential  information  m  the
particulate elemental concentration
data is  even more  difficult to  under-
stand. Although it may appear initially to
be  uncertain in interpretive value,  a
technique long used  in the study of
atmospheric  particulate material is
applicable. The comparison of elemen-
tal ratios for different samples instead of
the absolute concentration is informa-
tive. By normalizing the concentration
of each element to  a crustal element,
such as Sc, problems caused by differ-
ing amounts of bottom sediment sus-
pended in water (loading effects) are
eliminated.
  Scandium was chosen for this pur-
pose  because  it has  relatively few
anthropogenic uses. Since it is not used
in a refined  form  in  industry  and is
refractory in nature, it is not expected to
be introduced into the environment in
an  enriched state  or in  significant
quantities.  When  these  ratios are
divided by ratios of average  crustal
material, a crustal  enrichment factor
(EF) results. This is done for conven-
ience and to allow a crude comparison
with naturally occurring material.
  I n these data the concentrations from
Wedepohls'1  compilation  for  crustal
elements has been used. Similar
though not identical results could  be
obtained using other compilations.
Additionally, the computation  of EFs
relative to  average  soils and average
sedimentary rocks would be of value to
see how the suspended sediments of
Chesapeake Bay  differ from those
natural materials.
  Ideally, the EFs for each element will
remain constant if the sources contribu-
ting to the suspended sediment  remain
the same. Although the concentration
of the  various elements may fluctuate
several  orders of magnitude  from
sampling to sampling, the EFs should be
constant if the sources are constant as
they are  not effected by mass loading.

Conclusions
  Uses of  these EFs to produce  an
interpretive model  for evaluating and
concluding  elemental relationship and
origins can be postulated. However,
actual  conclusions  cannot  be  drawn
until a rigorous scrutiny of the statistical
significance of  the  individual  sets of
enrichment factors has been com-
pleted. Because this technique  has not
been used for water particulates previ-
ously, many cross references between
elements and geological positioning, as
well as  within set limits,  must  be
evaluated.
  In this  report, the enrichment factors
normalized to the Wedepohl  crustal
numbers have been given without
interpretation to at least the 90 percent
confidence limit.
  These  data  are of sufficiently well
known reliability that statistical  com-

-------
parison can be performed resulting in
significant trends of known reliability.
This work has not been included in this
report and  is of a sufficiently complex
nature to comprise a separate effort,
which has been recently initiated.

References
  1. Wedepohl, K.H. 1966, "Origin and
Distribution of  the  Elements."  L.H.
Aherns, ed. Pergamon Press, London.
pp 999-1016.
Howard M. Kingston, Robert R. Green berg, EllynS. Beary, BillyR. Hardas. John
  R.  Moody, Theodore C.  Rains, and Walter S. Liggett are with the National
  Bureau of Standards,  Washington, DC 20234.
Duane Wilding was the EPA Project Officer (for information, see below).
The complete report, entitled "The Characterization of the Chesapeake Bay: A
  Systematic Analysis of Toxic Trace Elements," (Order No. PB 82-265 265;
  Cost: $18.00, subject to change) will be available only from:
       National Technical Information Service
       5285 Port Royal Road
       Springfield, VA 22161
        Telephone: 703-487-4650
For information contact David Flemer at:
       Chesapeake Bay Program
       U.S. Environmental Protection Agency
       2083 West Street. Suite 5G
       Annapolis, MD 21403
                                                                                      US GOVERNMENT PRINTING OFFICE 1SM-559-017/0851

-------
United States
Environmental Protection
Agency

Center for Environmental Research
Information
Cincinnati OH 45268

Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
F*
UtMML
Official Business
Penalty for Private Use $300
            55 lgff3S9

-------