Link to Bioassay Testing Report

Link to Bioassay Testing Report
Bioassay Testing of Baltimore Harbor Sediments
Spiked with Cr(VI)
Final Report from the Center for Contaminant Transport,
Fate, and Remediation
February 12, 2008
Report Authors:
Katherine Watlington1,2, Andrew Graham1, Edward Bouwer1*, William
Goodfellow2, and John Baummer2
1
Department of Geography and Environmental Engineering
Johns Hopkins University
3400 N. Charles Street
Baltimore, MD 21218
2
EA Engineering, Science and Technology, Inc.
15 Loveton Circle
Sparks, MD 21152
*
Corresponding Author email: b[email protected]
1
Table of Contents
Page
Objective And Executive Summary.............................................................................................4
Background....................................................................................................................................6
Introduction.........................................................................................................................6
Toxicity Evaluation Methods...............................................................................................7
Chromium Contamination of Sediments..............................................................................8
Experimental Approach..............................................................................................................10
Selection of Sample Sites...................................................................................................14
Selection of Bioassay Test Organism.................................................................................15
Selection of Cr(VI Spiking Concentrations........................................................................16
Selection of Test Type and Duration..................................................................................16
Experimental Procedures............................................................................................................17
Task 1: Sample Collection.................................................................................................17
Task 2: Sample Preparation and Analysis.........................................................................17
Task 3: Cr(VI) Spiking and Additional Characterization..................................................20
Task 4: Sediment Bioassays...............................................................................................21
Task 5: Final Characterization.........................................................................................24
Task 6: Cr(VI) Re-spiking and Characterization..............................................................24
Results...........................................................................................................................................26
Water Parameters at Sediment Collection Sites................................................................26
Initial Sediment and Water Parameters.............................................................................27
Initial Acute Toxicity Tests.................................................................................................31
Characterization of Spiked Sediments...............................................................................32
Acute Toxicity Tests...........................................................................................................33
Chronic Toxicity Tests.......................................................................................................34
Final Characterization.......................................................................................................37
Cr(VI) Re-spiking and Characterization...........................................................................41
Analysis of Results.......................................................................................................................44
Sediment Toxicity...............................................................................................................44
Cr(VI)/AVS Ratio..............................................................................................................47
Re-spiking and Characterization……………………………………………………………...48
Cr(VI)/AVS Ratio for Re-Spiked Sediment.......................................................................49
Cr Mass Balance Calculations for Re-Spiked Sediments................................................49
2
Table of Contents (Continued)
Page
Conclusions...................................................................................................................................50
References.....................................................................................................................................53
Appendix A: Previous Site Characterization Data...................................................................58
Appendix B: Sample Characterization Protocols.....................................................................61
Appendix C: Organics Analysis Results....................................................................................72
3
I. Objective and Executive Summary
The primary objective of this study was to determine if there is a relationship between
ecotoxicity and ingestion of Baltimore Harbor sediments containing chromium. This was
accomplished by spiking indigenous sediment samples with Cr(VI) and subsequently conducting
amphipod bioassay toxicity tests. Whole sediment samples were used in the bioassays.
Leptocheirus plumulosus, an indigenous amphipod, was used as the test organism. L.
plumulosus was selected due to the organism’s sensitivity to contaminants as well as its tendency
to reside in and ingest sediment.
Both acute (10 day) and chronic (28 day) toxicity studies were performed using L. plumulosus.
Survival was used as an endpoint for the acute toxicity studies, while survival, growth and
reproduction were used for the chronic study endpoints. Three different concentrations of
hexavalent chromium were spiked into sediments, from five locations, to determine if the
addition of Cr(VI) to sediment caused toxicity to the amphipods. The lowest Cr(VI) spike
concentrations were within the environmentally relevant concentration range. The middle and
highest Cr(VI) spike concentrations significantly exceeded levels of measured total Cr
concentrations in Baltimore Harbor sediments.
Based on the sediment chemistry in the Baltimore Harbor, it was hypothesized that Cr(VI) would
be rapidly reduced to Cr(III). Research conducted through an ongoing research project within
the Center for Contaminant Transport, Fate and Remediation (CTFR) at Johns Hopkins
University (JHU) has demonstrated rapid reduction rates of Cr(VI) in Baltimore Harbor
sediments (Graham et al., unpublished data). These rapid reduction rates were also found while
conducting these bioassay studies on Baltimore Harbor sediments. Results of both acute and
chronic toxicity tests demonstrated that spiking sediments with environmentally relevant
concentrations of Cr(VI), as well as, concentrations significantly exceeding chromium
concentrations found in the Harbor did not cause additional toxicity to L. plumulosus.*
*
Note: One sediment sample, spiked with concentrations (3210 µg/g) well above levels found in the Harbor, had 0% survival for both acute and
chronic toxicity in all test beakers. This was the only sediment sample where the added Cr(VI) greatly exceeded the available acid volatile
sulfides (AVS); and elevated Cr(VI) concentrations in sediments and overlying water were found..
Based on the results of this study, it can be concluded that the addition of environmentally
relevant levels of Cr(VI) does not cause additional toxicity in Baltimore Harbor sediments,
because Cr(VI) is rapidly reduced to nontoxic Cr(III). Moreover, since Cr(VI) addition did not
result in increased mortality to the bioassay test organism, the chromium currently present in
Baltimore Harbor sediments is unlikely to elicit a toxicological response from the ingestion
pathway.
5
II. Background
Introduction
In the past two decades, an awareness of sediment toxicity has emerged, raising concern
regarding the potential impacts to both ecological and human health. Contaminated sediments
threaten aquatic life through bioconcentration, ingestion and bioaccumulation in the food chain.
Sediments serve as both a sink and a reservoir for metals and hydrophobic organic contaminants
(Long et al., 1998). Evaluations of sediment quality have identified widespread toxicity across a
number of water bodies. National studies by the US Environmental Protection Agency (EPA)
have estimated that over 10% of the sediment in waterways across the country should be
classified as being contaminated (USEPA 1997).
Problems with sediment quality have become more visible in the face of improving surface water
quality. At a number of locations where water quality criteria (WQC) are being met, sediment
toxicity is still present. Establishing standards for sediment quality has proven more difficult
than establishing water criteria due to the complex nature of sediments and exposure routes. The
presence of metals and organics in sediment (normally expressed as dry weight concentration)
does not necessarily translate into a toxic response.
Physical and chemical properties of the sediment affect the bioavailability of pollutants, and thus
the toxicity of sediment at a given site. Total organic carbon, sediment grain size, and the
concentration of acid volatile sulfides (AVS) all influence bioavailability and thus toxicity.
Biological exposure routes also vary between organisms and sites. Benthic organisms can be
exposed to pollutants in the sediment through dissolved phase transport across organisms’
membranes (e.g., gills, skin) and/or ingestion of contaminated sediment (USEPA 2001, USEPA
2004a).
Data from past studies show that sediments collected from many locations around the Baltimore
Harbor exhibit toxicity and are classified as “impaired” (Klosterhaus et al., 2006, McGee et al.,
1999). At many sampling locations within the harbor, the WQC for measured constituents are
being met; however, sediments from the sites are still toxic to estuarine species. The nature of
6
sediment toxicity has made it extremely difficult to establish cause-and-effect relationships with
specific contaminants in the Harbor and toxicity. A recent toxicity identification evaluation
(TIE) carried out by the University of Maryland failed to find specific contaminants or classes of
contaminants responsible for toxicity in the Baltimore Harbor (Klosterhaus et al., 2006).
Toxicity Evaluation Methods
A number of methodologies have been used to determine the relationship of contaminants with
toxicity. Most recently, TIE methods have been applied to sediment samples. While TIE studies
have proven useful in establishing cause and effect relationships in water samples, sediment
samples have proven more difficult (Ho et al., 2002). In the past, TIE studies have been
performed on pore water as a means of examining sediment toxicity; however, pore water studies
do not factor in exposure through ingestion which is increasingly being viewed as a significant
source of contaminant exposure (Lee et al., 2000). Unfortunately, experimental constraints of
performing whole sediment TIE studies, namely procedures for sequestration of metals or
organics, have made them largely unsuccessful.
In looking for a better means for establishing cause and effect relationships in
contaminated sediment, CTFR recently reviewed the literature and prepared a literature review
and annotated bibliography for the Maryland Department of the Environment (Watlington et al,
2007). The review focuses on sediment evaluation methods. Based on the literature, CTFR
concluded that spiking studies with whole sediment bioassays provided the best method for
establishing causality with respect to toxicity. By spiking sediments with a single contaminant,
acute and chronic toxicity of the spiked samples can be compared with the original toxicity of the
sampling site (Murdoch et al., 1997).
No change in toxicity would indicate that increases in the concentration of the specific
contaminant do not cause increased toxicity. In this case, it would be unlikely that current
concentrations of a specific contaminant at the sampling site play a role in any observed toxicity
on site (e.g., no dose-response relationship). On the other hand, increases in toxicity would
strongly suggest that a specific contaminant would contribute to the toxicity of sediments at a
7
specific site. The nature of spiking allows this method to be used with a wide range of
sediments, from those that are relatively clean, to those that exhibit a moderate level of toxicity.
In addition, the use of whole sediment samples with test species that live in the sediment and
ingest the sediment allows for both dissolved phase transport and ingestion to be evaluated.
Chromium and Contaminated Sediments
While chromium is not an “emerging” contaminant, health concerns and public awareness of this
toxicant have risen in recent years. Due to the presence of chromium in Baltimore Harbor
sediments, this contaminant has been a concern to public and private organizations concerned
with the overall health of the Baltimore Harbor. In previous studies, chromium has been shown
to exist at elevated levels in Harbor sediments; however, chromium bioavailability and relative
toxicity has been called into question (Baker et al., 1997; MDE, 2005). Both trivalent chromium
(Cr(III)) and hexavalent chromium (Cr(VI)) can be found in aquatic systems; however, Cr(VI) is
much more soluble and thus more mobile than Cr(III). Cr(VI) is also highly toxic, while Cr(III)
is, to all intents and purposes, non-toxic at environmental concentrations.
It has been determined that chromium exists as Cr(III) in most anoxic sediments. A number of
studies have shown that sediments containing acid volatile sulfides (AVS) do not typically
contain Cr(VI), because any Cr(VI) entering the sediment is rapidly reduced to Cr(III). AVS, a
common component of anoxic sediments, is an operationally defined variable consisting of
hydrogen sulfide, polysulfides, mackinawite (FeS), greigite (Fe3S4), and metal sulfides (Rickard
and Morse, 2005). Mackinawite is believed to be the principal solid phase AVS component in
many anoxic sediments. Recent studies have shown synthetic mackinawite to be an efficient
reductant of hexavalent chromium (Boursiquot et al., 2002; Mullet et al., 2004).
Even in spiking studies with Cr(VI), reduction to Cr(III) in the presence of AVS occurs fairly
quickly. Studies have also shown, however, that when AVS is not present or when the Cr(VI)
concentration exceeds the AVS reducing capacity, Cr(VI) will persist (Berry et al., 2004; Besser
et al., 2004; Becker et al., 2006). When this situation occurs, toxicity is expected (Rifkin et al.,
2004). Other studies on chromium toxicity have demonstrated that Cr(III) located in sediments
is non-toxic to estuarine organisms. Both amphipods and polychaetes have been used in toxicity
8
tests with Cr(III) spiked sediments. Neither species demonstrated an observable reaction to
increased Cr(III) concentrations (Berry et al., 2002; Berry et al., 2004; Oshida et al.,1981).
Thus, it has been concluded that when sufficient AVS is present, nearly all of the chromium in
the sediment will be found in the trivalent form, and will be nontoxic to aquatic organisms.
Unfortunately, most documented studies of chromium concentrations in the Baltimore Harbor
sediments have only reported total chromium concentrations. Basing management and
regulatory decisions on total chromium concentrations is not a sound scientific approach as
discussed by Rifkin et al. (2004). The data for Baltimore Harbor show that AVS levels are
significantly higher than metal concentrations with a considerable level of excess sulfides.
Based on these findings, it has been concluded that the total sediment chromium concentration
would almost all be present in the Cr(III) state (MDE, 2005).
Although the literature and mass balance measurements available indicate a high probability that
chromium is not responsible for sediment toxicity in the Baltimore Harbor, efforts have been
unable to establish clear causality between any pollutant(s) and observed toxicity. Thus,
chromium is among the metals of potential concern in the Baltimore Harbor. Bioassay testing
was conducted to determine if a link exists between the ingestion of chromium and observed
toxicity in the Harbor. The results of the bioassay testing will also determine whether current
chromium levels are contributing to the present levels of sediment impairment.
9
Experimental Approach
The purpose of this study was to determine what effect, if any, the addition of Cr(VI) has on the
toxicity of selected Baltimore Harbor sediments via the ingestion pathway. CTFR investigated
sediment toxicity following Cr(VI) addition by spiking Baltimore Harbor sediment samples with
Cr(VI) and performing acute and chronic bioassays using the amphipod L. plumulosus. CTFR
was the lead institution and employed EA Engineering, Science and Technology (EA), Inc under
a subcontract arrangement. While CTFR conducted all of the spiking and analyses at the
Homewood campus, the bioassay tests were performed by EA under the supervision of CTFR.
EA’s bioassay facility, located 30 minutes north of the Homewood Campus, provided a
convenient location for testing.
Initially, this study was divided into five tasks.
Task 1. CTFR collected sediment samples, with EA’s assistance and equipment, from five
selected locations in the Baltimore Harbor and one reference location. Following collection,
sediment samples were transported to EA’s facility, by EA personnel, where a portion of the
sediments were sieved to remove any indigenous organisms.
Task 2. Sieved sediment samples were then transported to CTFR for analysis. Data on sediment
water content, total metals, acid volatile sulfides (AVS), total chromium concentration and
chromium species concentrations were obtained and analyzed. The chromium concentrations
obtained from these measurements are referred to as “baseline” concentrations throughout the
remainder of the report. “Baseline” concentrations should not be interpreted as concentrations in
sediments free of anthropogenic impact, but rather the native pre-spike concentrations of Cr(III)
and Cr(VI).
Task 3. The sieved samples were spiked with Cr(VI). Three different spike concentrations were
used for the sediment samples obtained from each test site. Spike A was designed to reflect
environmentally relevant concentrations of total chromium and ranged from 383 to 677 µg/g dry
weight. Spike B (ranged from 1250 to 1810 µg/g dry weight) and Spike C (ranged from 2000 to
10
4180 µg/g dry weight) contained concentrations well in excess of the levels of total chromium
measured in Baltimore Harbor sediments. The measured concentrations and calculated spike
concentration data are summarized in Table 1.
These spike concentrations were chosen to: 1) correspond with the range of total Cr
concentrations measured in the Baltimore Harbor sediments (e.g., environmentally relevant) in
the selected sites (Spike A); and 2) exceed measured total Cr concentrations (Spikes B and C).
Spikes B and C were used, in part, to assess the level of Cr(VI) necessary to cause toxicity, in the
event environmentally relevant concentrations showed no response. Spiked samples were then
transported back to the EA Bioassay Lab. Prior to determining toxicity with spiked sediment
samples, preliminary acute toxicity tests for each sample site were conducted to gauge the
toxicity of unspiked sediments.
Task 4. Following delivery of the spiked sediment samples to EA, both acute and chronic
toxicity studies were completed for each of the samples. A positive control sample spiked with a
high concentration of the PAH naphthalene was also tested, along with a negative control
reference sediment collected from the Wye River (a nearby site with similar sediment
characteristics and negligible toxicity).
Task 5. At the conclusion of the bioassay tests, final characterization studies were conducted on
each of the sediment samples. A flow chart summarizing the experimental approach is shown in
Figure 1.
11
Table 1. Measured total Cr concentrations for spiked and unspiked Baltimore Harbor sediments
used in bioassay.
Sample
Measured
Measured Calculated Measured Calculated Measured Calculated
Location
Baseline
Total Cr
Spike A
Total Cr
Spike B
Total Cr
Spike C
Total Cr
after
(µg/g)
after
(µg/g)
after
(µg/g)
(µg/g)
Spike A
Spike B
Spike C
(µg/g)
(µg/g)
(µg/g)
BSM-33
823
1500
677
2470
1650
2820
2000
BSM-38
271
654
383
1580
1310
3360
3090
BSM-45
148
548
400
1820
1670
4330
4180
BSM-54
126
535
408
1380
1250
3050
2920
BSM-68
354
964
610
2160
1810
3560
3210
Task 6. Cr(VI) concentrations ranging from 1,500 ug/g to 3,000 ug/g dry weight were spiked
into BSM-68 to determine the Cr(VI) spike concentration at which incomplete Cr(VI) reduction
occurs. The experiment allowed for a more refined determination of the threshold concentration
at which the Cr(VI) spike exceeded the reducing capacity of the sediment resulting in 100%
mortality to amphipods.
12
Collect sediment
samples from five
designated sites and
one nontoxic control
site.
Transport
sediment
to EA
Sieve 6 gallons of
sediment sample from
each site. Prepare
sieved sediment and
pore water samples
for analyses.
Perform initial acute
toxicity tests on sieved
sediment from each site
Note Results
Transport
samples
to JHU
JHUJH
U
Analyze sieved
sediment samples,
measuring pH, TOC,
AVS, etc., as well as
measuring metal and
organic concentrations.
Spike samples from each site with 3
different concentrations of Cr(VI).
Prepare a PAH spiked positive
control
with one sediment sample
Transport
samples
to EA
Perform acute and chronic bioassays on
the original sieved samples and spiked
samples, including the positive PAH spiked
sample, and a nontoxic control sediment.
(*Note: Chronic tests will be performed as
needed based on the results of the initial
acute tests)
Transport
samples
to JHU
Conduct final characterization studies on each
sediment, pore water, and overlying water sample.
Analyze data, interpret
results, prepare final report
Figure 1. Flow chart of experimental approach.
13
Selection of Sample Sites
Sample collection for the bioassay studies took place at five selected sites around the Baltimore
Harbor. The sites were chosen based on previous work detailed by McGee et al. (1999, 2004)
and a recent University of Maryland TIE DRAFT Report (Klosterhaus et al., 2006) that showed
these sites to have elevated total chromium concentrations. Characterization data (TOC, grain
size, etc) as well as contaminant analysis and toxicity data are detailed in the reports for each of
the sites, thus providing important background information for the analysis to be completed by
CTFR.
In selecting the sites, it was important that each site exhibit no more than 50% mortality of the
test organism in an acute study (10 days) so that potential changes in toxicity with increasing
Cr(VI) spike concentration would be observable. The selected sites are designated 33, 38, 45,
54, and 68. Figure 2 shows a map containing the chosen test sites along with other sites that
have been characterized in the Baltimore Harbor and surrounding tributaries (McGee et al.,
1999). Data from the past studies for each of the test sites including pollutant concentrations and
toxicity results are listed in Appendix A.
14
Reference Site (Wye
River)
Figure 2. Baltimore Harbor test sites (McGee et al., 1999). The sites shown with red dots
and the Wye River reference site (off map) were sampled for this study.
Selection of Bioassay Test Organism
The estuarine amphipod L. plumulosus was used as the test species for the entirety of the
bioassay experiments. This species was selected based on the organism’s sensitivity and its
presence in a wide range of sediment types and salinity ranges, as well as its tendency to reside
in and ingest sediment. Because previously observed concentrations of metal contaminants in
overlying water and pore water were quite low, ingestion of contaminated sediment was
expected to be the primary route of exposure for aquatic organisms in the Harbor. Therefore, it
was paramount that the chosen organisms ingest significant amounts of sediment as well as
exhibit sensitivity to contaminants. As compared to polychaetes, clams, and other amphipods, L.
plumulosus best matches the desired traits for the bioassay study (EPA, 2001). L. plumulosus is
also indigenous to Baltimore Harbor sediments and can be cultured in the laboratory. This was
advantageous since the cultured organisms could not have been previously exposed to
environmental contaminants.
15
Selection of Cr(VI) Spiking Concentrations
Three different concentrations of Cr(VI) were spiked into sediment from each test site, and are
noted as Spike A, Spike B, and Spike C throughout the report. Concentrations of chromium in
dry sediment samples were previously found to range from approximately 200 and 850 µg/g dry
weight at the five test sites (characterization data contained in Appendix A). These baseline
concentrations were confirmed in the initial sediment analysis conducted as part of this
experiment. These results are located in Table 1. Based on the observed chromium
concentrations present, Spike A (range of 383-677 µg/g dry weight) was chosen to span the
observed range of concentrations. Spike B (range of 1250-1810 µg/g dry weight) and Spike C
(range of 2000 to 4180 µg/g dry weight) were chosen to assess the level of Cr(VI) necessary to
cause acute and or chronic impacts to the bioassay organisms.
Selection of Test Type and Duration
Both acute (10 day) and chronic (28 day) bioassays were chosen for this study. In general, acute
tests are more commonly performed in the field due to their ability to provide rapid evaluation of
contaminated sediments. Chronic tests, however, have the ability to mimic conditions where
sublethal concentrations of contaminants are present (McGee et al., 2004) and could lead to
problems associated with long-term exposure. As can be seen from the site data in Appendix A,
the chosen sites exhibit varying levels of toxicity, with some sites demonstrating changes in
toxicity over the course of multiple tests. Thus, conducting both tests was intended to provide a
more complete understanding of any effects the addition of Cr(VI) has on toxicity.
16
Experimental Procedures
Task 1: Sample Collection
Sediment samples were collected from the five site locations on 20-21 February 2007 by EA and
CTFR personnel. A sediment sample from the non-toxic Wye River control site was also
collected. Equipment for water quality analyses, including the boat, GPS system and sampler
was provided by EA. The top 3 inches of sediment were collected from each station using a
stainless steel Ponar sampler. Multiple grab samples were collected to make up a total of 10
gallons of sediment per site.
Five gallons from each site were used for toxicity and physicochemical characterization. The
remaining five gallons of sediment were held in reserve at EA. Standard water quality
parameters, temperature, pH, dissolved oxygen and salinity were measured at the bottom of each
sampling station by EA personnel using a multimeter. The sediment samples were hand carried
to EA’s Ecotoxicology Laboratory in Sparks, Maryland on the day of collection, and placed in a
temperature controlled chamber at 4°C. When not being used for testing, sediment samples
remained stored in the dark at 4°C.
Task 2: Sample Preparation and Analysis
Once transferred to EA, a portion of the sediment samples were sieved and the remainder stored
(in the dark, at 4°C) for any further use. A total of 5-gallons of sediment were press sieved using
a 250 µm stainless steel wire mesh sieve, removing large particles and indigenous organisms.
After sieving, each sample was homogenized. The five gallon sieved samples were then divided
for characterization, spiking, and initial toxicity studies. Two gallons of the sieved sediment for
each sampling site were hand carried to CTFR for analysis and spiking.
The remaining portion of sieved sediment from each test site was kept at EA where initial acute
bioassay studies were performed to gauge the extent of mortality for the baseline mix of
contaminants in the sediments. Analysis of the sediment samples included a number of
parameters. Sediment water content was measured based on a protocol by Tan (1996). 10 g of
wet sample was weighed and then dried in an oven at 105 °C. The final weight was then taken
17
and used to calculate the water content (Tan 1996). Sieve analysis was conducted by wet sieving
sediment through 4.75 mm and 75 µm sieves. Particles retained on the 4.75 mm sieve were
classified as gravel, those retained on the 75 µm sieve as sand, and those passing the 75 µm sieve
as silt/clay. Total organic carbon content was determined by loss on ignition. Samples were
placed in a muffle furnace at 550 oC for eight hours after first pretreating with 5 M HCl to
remove carbonates.
The concentration of acid volatile sulfides (AVS) was also determined for each sieved sediment
sample. AVS concentrations affect the reducing capacity of the sediments, as well as the
bioavailability of metals; therefore, these values are important in examining causes of toxicity.
AVS concentrations were determined using a method developed by Boothman and Helmstetter
(1992).
Calibration standards were first prepared over a range of anticipated AVS concentrations. A
sulfide stock solution was prepared from washed Na2S⋅9H2O crystals in deaerated deionized
water. This sulfide solution was then standardized by iodimetric titration as described by
Boothman and Helmstetter (1992). Calibration standards were prepared in a sulfide antioxidant
buffer (SAOB) solution consisting of 2.0 M sodium hydroxide, 0.2 M EDTA, and 0.2 M ascorbic
acid. A sulfide-specific electrode was used to measure the electrochemical potential of each
standard, and a calibration curve was constructed. AVS was then recovered from the sediment
samples using the purge and trap method developed by Boothman and Helmstetter (1992).
Following extraction, the potential of each recovered solution was recorded and converted to
concentration.
Determining chromium concentrations in the sediment samples was accomplished using two
distinct analytical methods. Because chromium typically exists in aquatic environments in two
valence states, different analytical approaches were used to measure both total chromium
concentrations and concentrations of relevant individual chromium species, specifically Cr(VI).
The use of the two analytical techniques enabled a mass balance to be obtained for chromium.
18
Initially, total chromium (both Cr(VI) and Cr(III)) was measured for each sieved sample using
microwave assisted acid digestion coupled with inductively coupled plasma – mass spectrometry
(ICP-MS). Concentrated HNO3 was used for acid digestion of total metals following EPA
Method 3051A (US EPA 1994). A polyatomic spectral interference at m/z = 52 caused by ArC+
necessitated use of dynamic reaction cell (DRC) technology to remove the interference. The
DRC utilized an ammonia reaction gas at a flow rate of 0.4 mL/min and a rejection parameter q
of 0.4.
In addition to measurements of total chromium, the concentration of Cr(VI) in each sample was
measured. Cr(VI) was extracted from sediment samples using an alkaline digestion protocol
modified from EPA Method 3060A. Following alkaline digestion, samples were analyzed using
reverse phase ion-pair HPLC (high performance liquid chromatography)-ICP-MS based upon a
method developed by Chang and Jiang (2001). This method allowed for simultaneous
measurements of Cr(VI) and Cr(III) in the digested samples. Separation of Cr(VI) and Cr(III)
through HPLC-ICP-MS was achieved on a 3 mm i.d. by 3 cm C8 column using a mobile phase
flow rate of 1.5 mL/min and a mobile phase consisting of 2.0 mM tetrabutylammonium
hydroxide (TBAH), an ion pairing reagent, and 0.6 mM of EDTA, a chelating agent, with the pH
adjusted to 6.9 to 7.0.
Prior to analysis, alkaline digestates were diluted five-fold or greater into HPLC mobile phase
and the pH of each sample was adjusted to ~7.0 with concentrated ultrapure HNO3. Due to the
alkaline digestion procedure, most of the chromium extracted was present as Cr(VI); however,
the utilization of HPLC allowed for discrimination between Cr(VI) and any extracted Cr(III) at
the lowest possible detection limits. The alkaline digestion procedure has been shown to
minimize interconversion of Cr species during extraction (James et al., 1995). The concentration
of Cr(III) in the sediment samples could then be calculated as the difference between total Cr by
acid digestion and Cr(VI) determined by alkaline digestion.
Because of the multielement capability of ICP-MS, acid digestion coupled with ICP-MS, used to
measure total Cr concentrations, was also simultaneously used to determine additional total
metal concentrations. Cadmium, cobalt, copper, manganese, nickel, lead and zinc concentrations
19
were all analyzed in each sediment sample. Analysis of organics in the sediment samples,
including PAHs, PCBs, and chlorinated pesticides, was carried out by Martel Laboratories, Inc.,
a local analytical lab.
Similar parameters measured for sediment samples were also determined for sediment pore
water. Pore water was extracted from sediment samples using a centrifugation method.
Centrifugation at 4,000 xg and 4°C was conducted on each sediment sample for 30 minutes to
extract pore water (Bufflap and Allen, 1995; Winger et al., 1998). Following centrifugation, the
supernatant was filtered through a 0.2 µm nylon membrane filter to complete the pore water
isolation. Chromium concentrations were then measured in the pore water samples. Total Cr and
other metals were determined by ICP-MS following acidification of the pore water. Cr
speciation in pore water was determined by HPLC-ICP-MS following dilution of pore water
samples into the HPLC mobile phase. Measurements of ammonia concentrations for each
sediment sample were completed by EA using an Accumet Model 25 pH/ion meter with a gas
sensing ion selective electrode (Fisher Scientific). A more complete description of the analytical
methods is provided in Appendix B.
Task 3: Cr(VI) Spiking and Additional Characterization
Sieved sediment samples from each site location were spiked with three different concentrations
of Cr(VI). Spiking was accomplished by adding a standard solution of potassium dichromate to
each sediment sample to achieve the desired spike concentration. Sediment samples were then
mixed by hand in an attempt to achieve a uniform concentration. Analysis of samples following
spiking and mixing indicated that mixing was likely insufficient as total Cr concentrations were
not uniform throughout the spike container.
Based on previous studies performed by Berry et al. (2002), 24 to 48 hours of mixing was
needed to ensure equilibration between the spiking solution and the sediments. However, Berry
et al. (2002) did not measure the mixing time needed to achieve complete reduction of Cr(VI) in
sediments. Based on observed kinetics of Cr(VI) reduction in batch Baltimore Harbor sediment
suspensions conducted by the CTFR, a one-day equilibration period was determined to be
20
sufficient. After spiking was completed, sediment samples were refrigerated overnight before
being hand carried back to EA for the bioassay testing. The bioassay tests were set up on the day
of sediment delivery. A small aliquot from each spiked sediment sample was returned to CTFR
for analysis of total Cr and Cr speciation.
A positive control was also prepared using naphthalene as a model PAH and sieved sediment
from site BSM-38. The positive control was designed to be extremely toxic to the organisms,
thus eliciting a positive response. Site BSM-38 was chosen for the PAH positive control based
on data from the University of Maryland TIE Report (2006). Total PAHs were determined for
each sample site, and site 38 exhibited a mid-range total concentration of PAHs. These data are
located in Appendix A. A stock solution of naphthalene in acetone was prepared and then added
to the sediment sample to a concentration of 350 mg/kg dry weight naphthalene in sediment.
This value was chosen based on previous studies with similar PAHs (Verrhiest et al., 2001).
All spiking studies at CTFR were initiated within 30 days of sample collection in order to ensure
the anoxic nature of the sediments was preserved. In USEPA Method 3060A, “hexavalent
chromium has been shown to be quantitatively stable in field-moist soil samples for 30 days
from sample collection,” (USEPA 1996). Thus, all spiking was completed within this
timeframe. Due to the length of the chronic bioassays, final characterization was completed
more than 30 days after Cr(VI) spiking.
Task 4: Sediment Bioassays
As discussed in Task 1, each of the five sediment samples were initially screened for acute
toxicity (10 day) to ensure each sample exhibited a moderate level of survival for baseline
comparisons. The 10-day L. plumulosus acute toxicity testing was conducted by EA, in
accordance with US EPA (2001) guidance. The L. plumulosus tests were conducted in 1-L
beakers each containing 175 mL of sediment and 800 mL of overlying water. Artificial seawater
was used as the overlying water for the toxicity testing. The artificial seawater was prepared by
mixing Crystal Sea bioassay grade sea salts with dechlorinated tap water to a final salinity of 15
21
ppt. The test organisms were less than 48-hours old at test initiation. The sediment and
overlying water were added to the chambers 24 hours prior to introduction of the test organisms.
The beakers were left undisturbed overnight to allow any suspended sediment particles in the
water column to settle. Ten organisms were randomly introduced into each replicate beaker.
The introduction of the test organisms to the test chambers marked the initiation of the toxicity
test. The test was initiated 23 February 2007. Organisms were not fed during the initial 10-day
acute toxicity screenings. The test chambers were maintained at a target temperature of 25±1°C
with a 16-hour light/8-hour dark photoperiod. Temperature, pH, dissolved oxygen, and
conductivity measurements were recorded daily for the overlying water. The test solutions were
not renewed during the 10-day exposure period.
At the end of the 10-day exposure period, the surviving adult organisms from each replicate were
retrieved by screening through a 250 µm sieve. The number of surviving adult L. plumulosus
from each replicate was recorded.
Following the completion of sediment spiking and delivery of the sample to EA, both acute and
chronic tests were performed on each sample. The 10-day acute toxicity tests were performed
concurrently with the 28-day sediment toxicity test. The tests were conducted in 1-liter beakers
each containing 175 mL of sediment and 800 mL of overlying water. The tests were performed
using ten replicates per sediment sample, five for the acute test and five for the chronic series.
The sediment and overlying water were added to the chambers 24 hours prior to introduction of
the test organisms.
The beakers were left undisturbed overnight to allow any suspended sediment particles in the
water column to settle. Twenty organisms were randomly introduced into each replicate beaker.
The introduction of the test organisms to the test chambers marked the initiation of the toxicity
tests. The test chambers were placed in an environmental chamber and maintained at a target
temperature of 25±1°C with a 16-hour light/8-hour dark photoperiod. The overlying water was
gently aerated at a rate of 100 bubbles per minute throughout the 10- and 28-day exposure
22
period. During the 10-day exposure period, the L. plumulosus were fed three times a week with
1 mL/replicate of a 20 mg/mL slurry of finely ground Tetramin in deionized water.
The overlying water in the exposure chambers was renewed three times each week by siphoning
400 mL of the old overlying water from each test chamber, and then slowly siphoning fresh
replacement water into the chamber, taking care not to disturb the sediment. Temperature, pH,
dissolved oxygen, and conductivity in the overlying water were recorded daily for one replicate
of each sediment sample.
At the end of the 10-day exposure period, the five acute toxicity beakers for each spike sample
were taken down for analysis. The surviving adult organisms from each replicate were retrieved
by screening through a 250 µm sieve. The number of surviving adult L. plumulosus from each
replicate was recorded.
The 28-day chronic toxicity beakers were maintained for an additional eighteen days. As with
the acute toxicity beakers, the L. plumulosus for the chronic study were fed three times a week
with 1 mL/replicate of a 20 mg/mL slurry of finely ground Tetramin in deionized water during
the first two weeks of the exposure period. This feeding schedule was maintained during weeks
three and four, however the concentration of the slurry was increased to 40 mg/ml Tetramin, to
provide additional food for the older (larger) test organisms.
As was done throughout the acute toxicity test, the overlying water in the exposure chambers
was renewed three times per week. Water quality parameters, such as temperature, pH, dissolved
oxygen, and conductivity measurements were recorded daily for the overlying water in one
replicate of each sediment for the duration of the 28 day test. Ammonia measurements were
conducted on composite samples of overlying water from each sediment sample at test initiation.
At the end of the 28-day exposure period, the surviving adult organisms from each replicate were
retrieved by screening through a 500 µm sieve. The number of surviving adult L. plumulosus
from each replicate was recorded, and the surviving adults from each replicate were placed in a
dried, pre-weighed tin and placed in a drying oven overnight at 100°C. The tins were then
23
removed from the oven and placed in a desiccator to cool. Each pan was weighed to the nearest
0.01 mg to determine a mean dry weight per replicate, obtained by dividing the total organism
dry weight per replicate by the number of surviving organisms per replicate. Material that had
passed through the 500 µm sieve was then retained on a 250 µm sieve to retrieve the offspring.
Amphipods and residual sediments that were retained on the 250 µm sieve were rinsed with
freshwater to remove salts, and washed into a sample jar. The offspring were stained with a 1
g/L solution of rose bengal, and preserved with 70% alcohol. The offspring were counted, and
the reproduction endpoint was calculated as the number of offspring per surviving adult.
Task 5: Final Characterization
At the completion of the chronic studies by EA, further sediment characterization was performed
by CTFR. Portions of sediment from each test sample were returned to CTFR for final
characterization. Total Cr concentrations and Cr(VI) sediment concentrations were measured for
each sample using the protocols outlined in Task 2. Overlying water from the beakers was also
tested for total Cr and Cr(VI). In addition, pore water was isolated from each sediment sample
using centrifugation as described in Task 2. Total Cr concentrations and species concentrations
(when relevant) were measured in the pore water for each sample.
Task 6: Cr(VI) Re-spiking and Characterization
After completing the final characterization, it was determined that Cr speciation data were
needed for spiked sediment 24 hours after Cr(VI) addition. The additional measurements were
used to determine if the added chromium had been completely reduced to Cr(III) at the time the
bioassays were set up (24 hours after spiking). A range of chromium concentrations, from 1,5003,000 µg/g dry weight were spiked into BSM-68 to determine the threshold concentration where
Cr(VI) exceeds the reducing capacity of the sediment.
To better mimic bioassay testing conditions, 1-L beakers were set up, each containing 150 grams
(approximately 175 mL) of BSM-68 sediment. The sediments were then spiked with varying
concentrations of Cr(VI). The concentrations used were 1500 µg/g, 1750 µg/g, 2000 µg/g, 2250
24
µg/g, 2500 µg/g, 2750 µg/g, and 3000 µg/g. Following spiking, sediments were hand mixed
within the beakers. The beakers were then covered and refrigerated overnight. After a 24 hour
period, 800 mL of overlying water was added to the beakers. The beakers were then allowed to
sit for an additional 24 hours, for a total equilibration period of 48 hours. The overlying water in
each beaker was then observed to determine beakers with yellow tinted water, indicating Cr(VI)
in the water and incomplete reduction. Each sediment sample was then analyzed for total Cr and
Cr(VI).
25
Results
Water Quality Parameters at Sediment Collection Sites
Standard water quality parameters including temperature, pH, dissolved oxygen and conductivity
were measured at the surface, mid-point (half of total depth at each site), and bottom of each
sampling station by EA personnel, with the exception of BSM 45, BSM-2 and Wye River. Data
were not collected at these locations. These values are displayed in Table 2, below.
Temperature and pH were fairly consistent between the sites, measured at approximately 2 ºC
and pH 8, respectively. Conductivity, dissolved oxygen (DO), and turbidity showed more
variability from location to location. Dissolved oxygen was greatest in the near surface waters
and ranged from 1.52 to 4.41 mg/L for sampled stations. While DO was lower at depth for all
sampled stations, only site BSM-33 showed a significant difference (~1.9 mg/L) between surface
and bottom water DO concentrations. Conductivity was greatest in the bottom waters near the
sediment-water interface and ranged from 281.0 to 323.5 µS/m. Turbidity was also greatest for
bottom waters, ranging from 3.2 to 5.2 NTU for bottom waters.
Table 2. Water quality parameters measured at sampling locations.
BSM 33
21-Feb-2007 Surface
Middle
Bottom
Latitude
Longitude
Depth (ft)
Temperature
(ºC)
2.20
2.55
2.04
Surface
Middle
Bottom
Latitude
Longitude
Depth (ft)
DO
(mg/L)
4.41
3.53
2.54
pH
7.91
8.13
8.01
Turbidity
(NTU)
1.8
2.0
5.0
39.25354488 N
76.49044867 W
10
BSM 38
21-Feb-2007
Parameters
Conductivity
(!S/m)
230.4
253.3
281.0
Temperature
(ºC)
2.31
1.81
1.71
Parameters
Conductivity
DO
(!S/m)
(mg/L)
310.7
1.76
315.4
1.66
317.0
1.55
39.2564 N
76.5361 W
20
26
pH
8.16
8.18
8.05
Turbidity
(NTU)
3.5
3.1
3.4
BSM 54
20-Feb-2007
Surface
Middle
Bottom
Latitude
Longitude
Depth (ft)
Temperature
(ºC)
1.96
1.88
1.99
Surface
Middle
Bottom
Latitude
Longitude
Depth (ft)
DO
(mg/L)
1.52
1.45
1.34
pH
8.00
7.99
7.91
Turbidity
(NTU)
4.1
2.7
5.2
39.2583 N
76.5683 W
29
BSM 68
20-Feb-2007
Parameters
Conductivity
(!S/m)
318.8
319.9
323.5
Temperature
(ºC)
2.37
2.01
2.16
Parameters
Conductivity
DO
(!S/m)
(mg/L)
298.6
2.71
318.9
2.22
319.2
2.16
pH
7.83
7.84
7.82
Turbidity
(NTU)
2.1
2.1
3.2
39.2778 N
76.5833 W
12
Initial Sediment and Water Parameters
Results from measurements of sediment water content, grain-size, and TOC for the five test sites
are displayed in Table 3. Water content was fairly similar among the five test sites and averaged
about 49 weight percent. Grain-size analysis revealed that the sediments should be classified as
silty muds with roughly 82-97% of the total sediment weight in the silt/clay size fraction, with
the exception of site BSM-54 which was appreciably sandier. TOC content averaged about 67% and showed little site to site variation, with the exception of site BSM-54 (TOC = 4.9%)
27
Table 3. Percent water content, grain-size distribution, and total organic carbon content
for the five ecotoxicity test sites.
% Sand
(0.075-4.75
mm)
Water Content
(%)
% Gravel
(>4.75 mm)
% Silt/Clay
(<0.075 mm)
BSM-33
BSM-38
BSM-45
BSM-54
57%
43%
47%
51%
0%
0%
2.7%
12.2%
97.3%
87.8%
0%
0%
17.6%
39.8%
82.4%
60.2%
7.3%
6.1%
7.0%
4.9%
BSM-68
46%
0%
5.6%
94.4%
7.1%
% TOC
The concentration of acid volatile sulfides (AVS) was also determined for each sieved sediment
sample. The results of this analysis are displayed in Figure 3. As can be observed from the
graph, the AVS concentrations are similar across the sediment samples with the exception of
BSM-33. Measuring over 500 µmole/g dry weight, the AVS concentration in sediment from
BSM-33 is a factor of 10 greater than the concentrations in BSM-68 and BSM-54 sediments.
Figure 3. AVS content of sediments used in the bioassay. Error bars indicate 95%
confidence limits.
28
During pre-spike sediment characterization, total Cr was measured for each sample, as well as
the concentration of Cr(VI). The concentration of Cr(III) in the sediment samples was then
calculated by subtracting the Cr(VI) concentration from the total Cr concentration. The total Cr
concentration, Cr(VI) concentration and resulting balance of Cr(III) for each sediment is
displayed in Table 4. As can be seen from Table 4, total chromium concentrations for the five
test sites in Baltimore Harbor are elevated with concentrations ranging from 126 to 822 µg/g dry
weight. Cr(VI) concentrations in sediment, however, are quite low, ~4 orders of magnitude
lower than total Cr, meaning that the bulk of Cr in these sediments is present in the trivalent
form.
Table 4. Total Cr, Cr(VI) and Cr(III) balance for each unspiked sediment sample. All
concentrations are dry weight basis.
BSM 33
BSM 38
BSM 45
BSM 54
BSM 68
Total Cr
(ug/g)
Cr(VI)
(ug/g)
Cr(III) (Balance)
(ug/g)
823
271
148
126
354
0.066
0.055
0.055
0.050
0.078
823
271
148
126
354
Further analysis of sediment samples was performed to measure other contaminant
concentrations, including cadmium, cobalt, copper, manganese, nickel, lead and zinc. Table 5
displays the total metal results. Concentrations of Cu, Zn, and Pb were elevated for most of the
test sites with concentrations generally in the several hundred (Cu and Pb) to thousands (Zn) of
µg/g dry weight. Concentrations of other heavy metal contaminants (Co, Ni, Ag, and Cd) were
considerably lower with concentrations generally below 50 µg/g dry weight. A hot spot of Cd
contamination was evident for site BSM-38 in Colgate Creek that contained Cd levels nearly a
full order of magnitude above the other sampled stations. The data for total metals collected in
this study are consistent with previous measures of total metals for these sites shown in
Appendix A.
29
Table 5. Total metal concentrations in Baltimore Harbor sediments. All concentrations are
dry weight basis.
BSM-33
BSM-38
BSM-45
BSM-54
BSM-68
[Mn]
(ug/g)
[Co]
(ug/g)
[Ni]
(ug/g)
[Cu]
(ug/g)
[Zn]
(ug/g)
[Ag]
(ug/g)
[Cd]
(ug/g)
[Pb]
(ug/g)
144
373
411
309
376
11.4
12.7
8.7
10.9
13.3
16.5
35.5
27.7
26.6
30.2
97.4
379
193
223
250
2100
1380
281
1254
311
1.5
1.5
0.7
1.2
1.7
8.3
77.7
1.0
29.0
1.5
174
426
145
248
178
Sediment concentrations of organics, including PAHs, PCBs, and chlorinated pesticides, were
determined by the contracted analytical laboratory, Martel Laboratories, Inc. A large suite of
target analytes were analyzed and the full results for each sediment sample are included in
Appendix C. Organic analytes were generally below detection with few exceptions. Common
contaminants such as benzo[a]pyrene, 1,2 dichlorobenzene, 2,4, dichlorophenol,
hexachlorobenzene, naphthalene, nitrobenzene, phenanthrene, and phenol were not observed at
concentrations above the detection limit of 500 µg/g dry weight for any of the sediments sampled
in this study. Common PCB Aroclor mixtures were also generally below detection (0.05 µg/g
dry weight) with the exception of Aroclor PCB-1260, which was found at concentrations of 0.12
and 0.10 µg/g dry weight at sites BSM-38 and BSM-45, respectively. Chlorinated pesticides
were also generally observed at concentrations below detection (generally 5 µg/kg dry weight)
with the exception of dieldrin which was observed at 5.3 µg /kg dry weight at site BSM-38 and 9
µg/kg dry weight at site BSM-45.
In addition to sediment, analysis of the pore water was also conducted. Total Cr results for pore
water are displayed in Table 6 below. Because total Cr concentrations were well below ambient
water quality criteria for Cr(VI) (11 µg/L for freshwater and 50 µg/L for saltwater), Cr speciation
in unspiked samples was not determined. Previous work by the CTFR has shown pore water
Cr(VI) concentrations throughout Baltimore Harbor to be less than 1 µg/L.
30
Table 6. Total chromium concentrations in pore water isolated from unspiked Baltimore
Harbor sediments.
Total Cr
(ug/L)
BSM 33
BSM 38
BSM 45
BSM 54
BSM 68
3.0
3.7
1.2
1.0
4.8
Initial Acute Toxicity Tests
Table 7 presents the results of the initial 10-day acute toxicity sediment tests. At completion of
the test, survival in the BSM38, BMS68 and the reference sediment (52, 61 and 64 percent
survival respectively) was significantly different (p=0.05) from survival in the Wye River
control, which had 82 percent survival. Survival in the BSM33, BSM45 and BSM54 samples
was 65, 81 and 78 percent survival, respectively, and was not significantly different from control
survival. Although some sediment samples exhibited greater toxicity than the Wye River
control, the conclusion of the initial acute toxicity screens was that no sediment sample was
sufficiently toxic to exclude it from the chronic toxicity sediment spiking study. Temperature,
pH, dissolved oxygen, and conductivity measurements were recorded daily on the overlying
water in one replicate of each sediment sample, and ammonia was measured in sediment pore
water at test initiation. Temperature, pH, DO and salinity were determined to be fairly constant
between the samples. Ammonia concentrations, however, exhibited some variability, ranging
from 0.10 to 14.2 mg/L. Water quality measurements for the initial 10-day bioassay are
summarized in Table 8.
31
Table 7. 10-day % survival for initial acute toxicity tests.
10-Day Survival
(% )
BSM 33
BSM 38
BSM 45
BSM 54
BSM 68
BSM 2 Reference
Wye River Control
65
52(a)
81
78
61(a)
64(a)
82
(a) Significantly different from the Wye River control (p=0.05).
Table 8. Water quality measurements for initial acute toxicity tests.
Temperature
(°C)(b)
pH
Dissolved Oxygen
(mg/L)
Salinity
(ppt)
Ammonia
(mg/L)(c)
23.8-25.9
7.8-8.2
7.9 (±0.1)
4.9-6.9
6.2 (±0.6)
14.2-15.7
14.9 (±0.6)
14.2
25.0 (±0.6)
BSM38
24.0-25.7
7.8-8.1
5.3-6.7
14.5-16.6
6.85
BSM45
24.9 (±0.5)
24.3-25.9
7.9 (±0.1)
7.7-8.5
6.2 (±0.4)
5.7-6.8
15.2 (±0.6)
14.3-16.3
11
BSM54
25.2 (±0.5)
23.9-25.5
8.2 (±0.3)
7.9-8.6
6.2 (±0.4)
5.1-6.9
14.9 (±0.6)
14.9-16.8
7.77
BSM68
24.7 (±0.5)
24.2-25.9
8.2 (±0.2)
7.9-8.0
6.2 (±0.5)
5.4-7.0
15.6 (±0.5)
14.5-15.5
9.94
BSM2 REF
25.1 (±0.6)
24.1-25.9
8.0 (±0.1)
7.8-8.0
6.2 (±0.5)
5.9-7.2
14.8 (±0.3)
14.5-17.6
1.52
25.1 (±0.6)
23.7-25.3
7.9 (±0.1)
7.7-8.3
6.7 (±0.4)
5.4-6.9
15.4 (±0.1)
15.1-17.1
<0.10 - 6.61
24.7 (±0.5)
8.0 (±0.2)
6.3 (±0.5)
15.9 (±0.6)
BSM33
Wye River
Control
(a) Range and mean (± standard deviation).
(b) Temperatures of test solutions were within the acceptable 3°C deviation as defined by
USEPA (2002).
(c) Ammonia measured on overlying water at test initiation and termination.
Characterization of Spiked Sediments
Following spiking of sediment samples, samples were taken to EA for use in the bioassays.
Unused sediment was returned to CTFR for analysis for total chromium. These results are
presented in Table 1.
32
Acute Toxicity Tests
Results of the acute sediment toxicity tests for the various sediment samples are summarized in
Table 9. For the BSM 33 sediment samples, there was 93 percent survival for spike C (2000
µg/g Cr dry weight), 90 percent survival for spike B (1650 µg/g Cr dry weight) and 92 percent
survival for spike A (677 µg/g Cr dry weight) after 10 days. The unspiked baseline sediment had
90 percent survival. Sediment samples from BSM 38 displayed similar results.
After 10 days of exposure, there was 96 percent survival for spike C (3090 µg/g Cr dry weight),
90 percent survival for spike B (1310 µg/g Cr dry weight), 96 percent survival for spike A (383
µg/g Cr dry weight), and 91 percent survival in the unspiked baseline sediment. For BSM 45,
there was 100 percent survival for spike C (4180 µg/g Cr dry weight), 94 percent survival for
spike B (1670 µg/g Cr dry weight) and 98 percent survival for spike A (400 µg/g Cr dry weight).
The unspiked baseline sediment had 97 percent survival.
After 10 days of L. plumulosus exposure to the BSM 54 sediment samples, there was 88 percent
survival in spike C (2920 µg/g Cr dry weight), 89 percent survival in spike B (1250 µg/g Cr dry
weight) and 93 percent survival in spike A (408 µg/g Cr dry weight). The unspiked BSM-54
baseline sample had 92 percent survival. Unlike, the other four sediment samples BSM-68
displayed acute toxicity at the highest spike concentration (spike C). After 10 days of exposure,
there were no surviving organisms in the 3210 µg/g Cr(VI) spiked sediment, which was
significantly different from survival in the unspiked baseline sample (73 percent survival).
Survival in spikes B (1810 µg/g Cr dry weight) and A (610 µg/g Cr dry weight) was 75 and 81
percent survival, respectively, which was not significantly different from the baseline survival.
33
Table 9. Percent survival for each sediment sample at the completion of the 10 day acute
toxicity test. For specific values of the spike concentrations of spikes A to C, refer to the
text or Table 1. Error estimates indicate 95% confidence limits.
Baseline
Spike A
Spike B
Spike C
BSM-33
(%)
BSM-38
(%)
BSM-45
(%)
BSM-54
(%)
BSM-68 Wye River PAH Control
(%)
(%)
(%)
90.0
(±14.6)
92.0
(±5.6)
90.0
(±11.6)
93.0
(±8.3)
91.0
(±5.2)
96.0
(±2.8)
90.0
(±9.8)
96.0
(±11.9)
97.0
(±12.1)
98.0
(±3.4)
94.0
(±8.1)
100.0
(±8.8)
92.0
(±7.1)
93.0
(±7.1)
89.0
(±5.2)
88.0
(±11.3)
73.0
(±9.4)
81.0
(±9.2)
75.0
(±13.2)
0
(±0)
99.0(±2.8)
80.0(±28.1)
Chronic Toxicity Tests
Results of the chronic sediment toxicity tests for the various sediment samples are summarized
in Table 10. For the BSM-33 sediment samples, there was a substantial drop in survival from 10
days to 28 days; however, survival was similar among the Cr spiked sediments and not
significantly different from the unspiked baseline sediment. For BSM-33 spikes A and C (677
and 2000 µg/g Cr dry weight, respectively), 33 percent survival was observed. The spike B
sample had 47 percent survival at test termination and the baseline unspiked sediment had 32
percent survival. At the 95% confidence level, these values were not significantly different.
Mean biomass in the spiked sediment samples ranged from 0.37 to 0.53 mg/exposed organism,
which was not statistically significant from the unspiked baseline sediment (0.29 mg/organism).
Mean reproduction also appeared to be unaffected by additional chromium. Mean reproduction
for spikes A, B, and C was 3.6, 5.2 and 2.2 neonates per surviving organism, respectively. Mean
reproduction in the unspiked baseline sample was 2.4 neonates per surviving organism.
For sediment from site BSM-38, at 28-days, there was 37 percent survival in the baseline
sediment. Survival in the spiked sediments ranged from 50 to 60 percent survival, and was not
significantly different from the baseline sample at the 95% confidence level. Mean biomass in
the spiked sediment samples ranged from 0.40 to 0.71 mg/exposed organism which was not
significantly different from mean biomass in the unspiked baseline sample (0.46 mg/exposed
34
organism). Mean reproduction for spikes A, B, and C was 1.6, 0.7 and 1.8 neonates per
surviving organism, respectively. The unspiked BSM-38 sediment had a mean reproduction of
1.3 neonates per surviving organism.
In sediment samples from site BSM-45, there was 72 percent survival for spike A (400 µg/g Cr
dry weight), 87 percent survival for spike B (1670 µg/g Cr dry weight) and 91 percent survival
for spike C (4180 µg/g Cr dry weight). The baseline sediment sample had 85 percent survival.
Mean biomass in the spiked sediment samples ranged from 1.07 to 1.1 mg/exposed organism and
was not significantly different from the unspiked baseline sample (1.04 mg/exposed organism).
Mean reproduction in spikes A, B, and C was 1.9, 3.2 and 2.5 neonates per organism,
respectively. As with the biomass measurements, mean reproduction in spiked and unspiked
sediments were not significantly different. The unspiked mean reproduction was measured at 0.8
neonates per surviving organism.
At 28-days, survival in BSM-54 sediment ranged from 73-82% for spiked sediments and 76%
for the unspiked sediment. Compared to the unspiked baseline sediment, survival at 10-days
and 28-days was not significantly altered by spiking with Cr(VI). Mean biomass in the spiked
sediment samples ranged from 1.04 to 1.65 mg/exposed organism which was not significantly
different from mean biomass in the unspiked baseline control (0.99 mg/exposed organism).
Mean reproduction also was unaffected by Cr(VI) addition. Mean reproduction for spikes A, B,
and C was 2.9, 2.3 and 4.3 neonates per surviving organism, respectively, which was not
significantly different from mean reproduction in the unspiked baseline sample (2.5 neonates per
surviving organism).
As with the acute toxicity test, BSM-68 displayed chronic toxicity at the highest spike
concentration (3210 µg/g Cr dry weight). At the end of the 28-day testing period, there was 0%
survival at this spike level. Survival for the remaining spiked sediments (67% for spike A and
57% for spike B) was not significantly different from survival in the unspiked sediment (56
percent survival). Mean biomass for spikes A and B was 0.54 and 0.50 mg/exposed organism,
which was not significantly different from the unspiked baseline sample (0.39 mg/organism) at
the 95% confidence level. Mean reproduction for spikes A and B was 0.2 neonates per surviving
35
organism. The unspiked baseline had a mean reproduction of 0.60 neonates per surviving
organism.
Table 10. Percent survival, mean biomass, and mean reproduction for each sediment
sample at the completion of the 28 day chronic toxicity test. Error estimates indicate 95%
confidence limits.
28-Day Survival
Mean Biomass
(%)
(mg/organism exposed)
Mean Reproduction
(neonates per surviving
organism)
32.0 (±14.3)
33.0 (±12.1)
47.0 (±24.3)
34.0 (±16.1)
0.29 (±0.08)
0.37 (±0.21)
0.53 (±0.28)
0.40 (±0.19)
2.4 (±2.1)
3.6 (±2.8)
5.2 (±2.0)
2.2 (±1.2)
37.0 (±19.4)
50.0 (±9.8)
53.0 (±16.2)
60.0 (±15.2)
0.47 (±0.40)
0.68 (±0.33)
0.40 (±0.30)
0.72 (±0.35)
1.3 (±0.8)
1.6 (±0.5)
0.7 (±1.0)
1.8 (±1.7)
86.0 (±6.8)
72.0 (±26.2)
88.0 (±10.4)
93.0 (±8.3)
1.04 (±0.27)
1.07 (±0.0.63)
1.51 (±0.14)
1.44 (±0.48)
0.8 (±0.7)
1.9 (±1.1)
3.2 (±2.4)
2.5 (±1.6)
76.0 (±16.1)
78.0 (±9.4)
73.0 (±16.8)
82.0 (±12.1)
0.99 (±0.41)
1.38 (±0.43)
1.04 (±0.63)
1.65 (±0.45)
2.5 (±2.0)
2.9 (±1.8)
2.3 (±1.0)
4.3 (±1.1)
56.0 (±27.9)
67.0 (±12.1)
57.0 (±10.4)
--
0.39 (±0.26)
0.54 (±0.27)
0.50 (±0.28)
--
0.6 (±0.6)
0.2 (±0.4)
0.2 (±0.3)
--
BSM-33
Baseline
Spike A (677 ug/g)
Spike B (1650 ug/g)
Spike C (2000 ug/g)
BSM-38
Baseline
Spike A (383 ug/g)
Spike B (1310 ug/g)
Spike C (3090 ug/g)
BSM-45
Baseline
Spike A (400 ug/g)
Spike B (1670 ug/g)
Spike C (4180 ug/g)
BSM-54
Baseline
Spike A (408 ug/g)
Spike B (1250 ug/g)
Spike C (2920 ug/g)
BSM-68
Baseline
Spike A (610 ug/g)
Spike B (1810 ug/g)
Spike C (3210 ug/g)
36
Final Characterization
Following completion of the chronic bioassay studies, a representative sediment sample of each
site and spike concentration was transported to CTFR for final characterization. Total Cr and
Cr(VI) concentrations in sediment were measured and Cr(III) calculated by difference. These
data are displayed in Table 11. Cr(VI) concentrations in sediment in spiked sediments were
virtually the same as unspiked sediment with the exception of BSM 68 (spike C) which
contained Cr(VI) in sediment approximately 150 times above baseline concentrations. Despite
the increased Cr(VI) concentration for this sample, most of the Cr(VI) added was still reduced
for this sediment, with greater than 99 percent of the total mass of Cr in the sediment in the form
of Cr(III). From these speciation results, it is apparent that for all samples, nearly all of the
Cr(VI) added was reduced to the trivalent form.
37
Table 11. Total Cr, Cr(VI) and Cr(III) balance for each sediment sample following Cr(VI)
addition to the sediment. All concentrations are dry weight basis.
Total Cr
Cr(VI)
Cr(III) Balance
Site
(!g/g)
(!g/g)
(!g/g)
BSM-33
Baseline
Spike A
Spike B
Spike C
823
1325
2466
2820
0.05
0.00
0.00
0.02
823
1325
2466
2820
BSM-38
Baseline
Spike A
Spike B
Spike C
271
654
1580
3360
0.05
0.01
0.01
0.03
271
654
1580
3360
BSM-45
Baseline
Spike A
Spike B
Spike C
148
548
1820
4330
0.05
0.01
0.00
0.02
148
548
1820
4330
BSM-54
Baseline
Spike A
Spike B
Spike C
126
535
1380
3050
0.08
1.37
0.27
0.80
126
534
1380
3049
BSM-68
Baseline
Spike A
Spike B
Spike C
354
964
2160
3560
0.06
0.36
1.10
9.57
354
964
2159
3550
Overlying water concentrations were measured for each representative sediment sample of each
site and spike concentration. Total Cr and Cr(VI) were both measured and Cr(III) calculated by
difference for each sample. These values are reported in Table 12. Total Cr in the overlying
water was generally below 10 µg/L with only a few exceptions. The overlying water for the
baseline test reactor for site BSM-38 was anomalously high, averaging about 13 µg/L; however,
no Cr(VI) was detected in the overlying water for this sample. The only overlying water that
contained any quantifiable Cr(VI) was collected from the test beakers containing BSM 68
sediment spiked with 3210 µg/g dry weight Cr(VI). Total Cr in the overlying water for these test
38
conditions were 1455.9 µg/L, with Cr(VI) averaging 1054.1 µg/L and Cr(III) 401.8 µg/L. It
should be pointed out, however, that the overlying water was changed periodically throughout
the duration of the bioassay tests, and the chromium concentration in the overlying water was
diluted by 60% with every water change. For BSM-68 where incomplete Cr(VI) reduction
occurred, accumulation of Cr(VI) in the overlying water may have been appreciably higher than
the reported values in the initial stages of the bioassay.
Table 12. Total Cr, Cr(VI) and Cr(III) balance for overlying water taken from bioassay
test beakers.
Site
Total Cr
(ug/L)
Cr(VI)
(ug/L)
Cr(III) Balance
(ug/L)
BSM-33
Baseline
Spike A (677 ug/g)
Spike B (1650 ug/g)
Spike C (2000 ug/g)
0.05
ND
ND
ND
ND
ND
ND
ND
0.05
ND
ND
ND
BSM-38
Baseline
Spike A (383 ug/g)
Spike B (1310 ug/g)
Spike C (3090 ug/g)
13.4
ND
1.6
1.6
ND
ND
ND
ND
13.4
ND
1.6
1.6
BSM-45
Baseline
Spike A (400 ug/g)
Spike B (1670 ug/g)
Spike C (4180 ug/g)
0.1
0.3
1.5
1.3
ND
ND
ND
ND
0.1
0.3
1.5
1.3
BSM-54
Baseline
Spike A (408 ug/g)
Spike B (1250 ug/g)
Spike C (2920 ug/g)
0.5
ND
2.7
9.2
ND
ND
ND
ND
0.5
ND
2.7
9.2
BSM-68
Baseline
Spike A (610 ug/g)
Spike B (1810 ug/g)
Spike C (3210 ug/g)
0.8
1.7
3.1
1455.9
ND
ND
ND
1054.1
0.8
1.7
3.1
401.8
39
In addition, pore water was isolated from each sediment sample using centrifugation as described
in Task 2. Total Cr concentrations were measured in the pore water for each sample at CTFR.
The results are displayed in Table 13. Total Cr concentrations in pore water did not show much
variation except for site BSM-33 that showed total Cr concentrations ranging from 0.9 to 27.2
µg/L. Although we observed high Cr(VI) concentrations in the overlying water of test beakers
for BSM-68 spike C, we observed relatively low (10.3 µg/L) total Cr concentrations in the pore
water of this sample. This finding may also be related to the periodic water renewal throughout
the duration of the bioassay.
Table 13. Total Cr concentrations in pore water of spiked and unspiked Baltimore Harbor
sediments.
Site
Total Cr (ug/L)
BSM-33
Baseline
Spike A (677 ug/g)
Spike B (1650 ug/g)
Spike C (2000 ug/g)
3.0
13.8
27.2
1.0
BSM-38
Baseline
Spike A (383 ug/g)
Spike B (1310 ug/g)
Spike C (3090 ug/g)
3.7
2.8
14.3
5.2
BSM-45
Baseline
Spike A (400 ug/g)
Spike B (1670 ug/g)
Spike C (4180 ug/g)
1.2
1.5
1.2
0.9
BSM-54
Baseline
Spike A (408 ug/g)
Spike B (1250 ug/g)
Spike C (2920 ug/g)
1.0
4.0
1.1
1.9
BSM-68
Baseline
Spike A (610 ug/g)
Spike B (1810 ug/g)
Spike C (3210 ug/g)
4.8
6.1
4.8
10.3
40
Cr(VI) Re-spiking and Characterization
For site BSM-68 sediment, we observed toxicity to L. plumulosus at a Cr(VI) spike concentration
of 3210 µg/g dry weight but not a spike concentration of 1810 µg/g dry weight, implying that the
true lowest observed adverse effects level (LOAEL) for site BSM-68 lies somewhere between
1810 and 3210 µg/g dry weight of Cr(VI). In order to more accurately determine the threshold
spike concentration required to elicit a toxic response, a range of Cr(VI) concentrations, from
1,500-3,000 µg/g dry weight, were spiked into BSM-68 sediment. The resulting Cr(VI)
concentrations in the water column, pore water, and sediment are shown in Table 14. The results
are also depicted graphically. As can be seen, the threshold spiking concentration appears to be
approximately 2,250 µg/g. At this concentration, Cr(VI) is detected in pore water and the water
column concentration increases significantly. Relatively little Cr(VI) was found to be associated
with the solid phase (Table 14), indicating that little of the Cr(VI) remaining was adsorbed to the
sediment.
The water column and pore water data are also graphed against the corresponding spike
concentration in Figure 4. The data also show that a linear relationship exists between water
column and pore water and the spiking concentration once the threshold Cr(VI) spiking
concentration has been exceeded. Figure 5 shows a linear regression of the relationship between
water column and pore water concentrations with the spiking concentrations above the threshold
value. Also shown in Figure 5 is the AVS content of BSM-68 sediment. To first approximation,
the Cr(VI) concentration in the pore water and overlying water appears to increase linearly once
the Cr(VI) spike concentration exceeds the molar equivalent AVS concentration.
41
Table 14. Concentration of Cr(VI) in water column, pore water, and sediment samples for
re-spiking experiment with BSM-68 sediment.
BSM 68-1500 ug/g
BSM 68-1750 ug/g
BSM 68-2000 ug/g
BSM 68-2250 ug/g
BSM 68-2500 ug/g
BSM 68-2750 ug/g
BSM 68-3000 ug/g
Concentration of Cr(VI)
Water Column
Porewater
(ug/L)
(ug/L)
4
133
252
3400
56.8
26200
28300
39300
94704
61700
170000
Solid-associated Cr(VI)
(ug/g sediment)
0.3
1.3
2.5
Figure 4. Overlying water and pore water concentrations vs. Cr(VI) spike concentrations
for BSM-68 sediment.
42
Figure 5. Linear relationship between overlying water and pore water concentrations and
Cr(VI) spike concentrations for BSM-68 sediment. The dashed line indicates the AVS
concentration of the sediment in µmole/g dry weight.
43
Analysis of Results
Sediment Toxicity
The results of both the acute and chronic toxicity data indicate that none of the spiked sediment
samples displayed elevated toxicity when compared with baseline concentrations, with the
exception of BSM-68 spiked at 3210 µg/g dry weight. To illustrate these trends, bar graphs of
the toxicity data for each site are displayed below, with total Cr concentrations for the baseline
and three spike levels. The acute toxicity data are provided in Figure 6, followed by the chronic
toxicity data in Figure 7.
Looking at the graphs, it is clear that with the exception of BSM-68 spiked at 3210 µg/g, toxicity
values do not vary from that observed for the baseline concentrations. Similar to previous
findings (Berry et al, 2004; Besser et al., 2004), no correlation was observed between sediment
toxicity and total Cr concentrations. Interestingly, some pre-existing chronic toxicity is observed
for the baseline sediment samples; however, levels of chronic toxicity in the spiked samples are
not elevated with respect to the baseline samples, excluding BSM-68 spiked with Cr(VI) at 3210
µg/g dry weight.
44
Figure 6. Acute toxicity and total Cr concentrations of BSM-33, BSM-38, BSM-45, BSM54, and BSM-68 baseline and spiked sediments. Bars represent the mean acute 10 day
percent survival for each of the tests and the filled circles represent the dry weight total Cr
concentrations. Error bars indicate 95% confidence limits on toxicity data.
45
Figure 7. Chronic toxicity and total chromium concentrations of BSM-33, BSM-38, BSM45, BSM-54, and BSM-68 baseline and spiked sediments. Bars indicate chronic 28 day
percent survival and filled circles represent the total Cr concentration. Error bars indicate
the 95% confidence limits for toxicity data.
46
The addition of Cr(VI) to sediments at concentrations at or exceeding environmentally relevant
concentrations caused no changes in observed toxicity (e.g., no dose-response), with the
exception of BSM-68 spiked at 3210 µg/g dry weight. Moreover, conclusions can also be drawn
about the effects of residual chromium in the baseline sediment. Based on these findings, it can
be concluded that the chromium already present in baseline sediment samples does not
contribute to any observed toxicity. If chromium already present was contributing to acute or
chronic toxicity, increases in this toxicity would be expected with even slight additions of
chromium. Thus it can be concluded that chromium is not responsible for any of the observed
acute or chronic toxicity in the Baltimore Harbor sediments sampled in this study. The data also
lend further support to the hypothesis that ingestion of even high levels of trivalent chromium in
sediment does not result in toxicity to the indigenous, sensitive amphipods used in these
bioassays.
Cr(VI)/AVS Ratios as Predictors of Toxicity
The initial AVS data coupled with measured Cr concentrations for each sediment sample and
spike concentration were used to calculate the ratio of added Cr(VI) to AVS for each sample.
Numerous studies suggest that the metal/AVS ratio can be used as an indicator of toxicity
(DiToro et al., 1992, Berry et al., 1996). In the case of chromium, AVS constituents including
FeS(s) and H2S reduce Cr(VI) to Cr(III). As long as the total concentration of available sulfides
exceeds the total Cr(VI) added to the sediment, all Cr(VI) should be quickly reduced to Cr(III).
The results of these calculations for the sediment samples are displayed in the Table 15 below.
AVS concentrations exceed spiked Cr(VI) concentrations in all sediments except for BSM-54
(spike C) and BSM-68 (spike C). It should be noted that both of these samples had Cr(VI)
added/AVS molar ratios greater than unity but only the BSM-68 (spike C) sample showed acute
or chronic toxicity. Incremental spiking of BSM-68 sediment revealed that aqueous Cr(VI)
concentrations rapidly climbed above ambient water quality criteria once the Cr(VI) spike
addition exceeded the reducing capacity of the sediment.
47
When excess reducing capacity remains, presumably in the form of AVS, aqueous and solid
phase Cr(VI) concentrations are negligible. Because no toxicity was observed at a ratio slightly
above one for BSM-54 (spike C) it seems probable that BSM-54 contained non-AVS reducing
capacity (e.g. labile organic matter) that resulted in complete reduction of Cr(VI). It should also
be noted that stoichiometries for Cr(VI) reduction by AVS are largely speculative, as AVS is a
poorly defined operational quantity. Although a 1:1 stoichiometry for Cr(VI) reduction by FeS
appears possible (Mullet et al., 2004), it cannot be concluded with absolute certainty that Cr(VI)
will persist whenever Cr(VI)/AVS is greater than one. Rather, it appears that Cr(VI) is unlikely
to persist and elicit toxic responses when the Cr(VI)/AVS ratio is less than one.
Table 15. Molar ratio of Cr(VI) added to AVS present for the five test sites and three
different spike concentrations. For values of the various spike concentrations refer to the
text or Table 1.
Spike A
Spike B
Spike C
BSM-33
0.03
0.06
0.11
[Cr(VI)] added/AVS Molar Ratio
BSM-38
BSM-45
0.08
0.08
0.28
0.33
0.67
0.83
BSM-54
0.17
0.51
1.19
BSM-68
0.32
0.96
1.71
Re-spiking and Characterization
In a re-spiking experiment, a range of chromium concentrations, from 1,500-3,000 µg/g, were
spiked into BSM-68 sediment. The results allowed for determination of the threshold Cr(VI)
spike concentration at which the reducing capacity of BSM-68 sediment would be exceeded.
Based on the overlying water and pore water measurements of Cr(VI), the threshold
concentration was measured at 2,250 µg/g. This spike concentration represented the lowest
concentration where Cr(VI) was measured in the overlying water and pore water. One
interpretation of these results is that the 2,250 µg/g Cr(VI) spike concentration likely represents
the true lowest observable adverse effects level (LOAEL) for BSM-68. The bioassay data
constrained this threshold to between 1810 and 3210 µg/g dry weight, so the result found here is
48
consistent with the bioassay data. Interestingly, nearly all of the added Cr(VI) remaining was
present in the pore water and overlying water, with solid-associated Cr(VI) representing a
negligible contribution.
Cr(VI)/AVS Ratio for Re-spiked Sediments
In the re-spiking experiments, the Cr:AVS ratio was again calculated for each of the spikes. For
this calculation, the target spike concentrations were used to calculate the ratios, and AVS was
measured again for BSM-68 at the time of the re-spikes due to the potential for change in AVS
concentrations over time. The calculated ratios are displayed in Table 16. Looking at the table,
at the spike concentration of 2,250 µg/g, the Cr(VI) concentration just exceeds the available
AVS (with a ratio of 1.02). Because the Cr(VI) exceeds the AVS, it is possible that Cr(VI)
added will be incompletely reduced, leading to the observed accumulation of Cr(VI) in the
overlying water and pore water.
Table 16. Ratio of added Cr(VI) to AVS for re-spiked BSM-68 sediments.
BSM-68 Dry Weight
Spike (ug/g)
1500
1750
2000
2250
2500
2750
3000
Dry Weight Spike
(umole/g)
Dry weight AVS
(umole/g)
Total Cr :AVS
Ratio
28.8
33.7
38.5
43.3
48.1
52.9
57.7
42.3
42.3
42.3
42.3
42.3
42.3
42.3
0.68
0.80
0.91
1.02
1.14
1.25
1.36
Cr Calculations and Mass Balance for Re-spiked Sediments
Measurement of Cr(VI) in overlying water, pore water, and sediment permits mass balance on
the Cr(VI) spikes added to BSM-68, allowing for calculation of the total mass of Cr(VI) reduced
for the seven different spike concentrations. The overall percent reduction may be compared
with the expected Cr(VI) reduction based on a 1:1 stoichiometry for Cr(VI) reduction by AVS.
This comparison is plotted in Figure 8. We obtained relatively good agreement between
experimental Cr(VI) reduction and predicted reduction, but, because the slope of the line in
49
Figure 8 is slightly less than 1, a 1:1 stoichiometry for Cr(VI) reduction by AVS actually underpredicts the observed experimental Cr(VI) reduction by BSM-68 sediment. This underprediction may be due to less than 100% recovery of remaining Cr(VI) after spiking, the
presence of other non-AVS reductants such as natural organic matter, or due to an alternative
stoichiometry of Cr(VI) reduction by AVS.
Figure 8. Expected Cr(VI) mass reduction based on 1:1 Cr(VI)/AVS stoichiometry versus
observed experimental Cr(VI) mass reduction for BSM-68 sediments spiked at 1500, 1750,
2000, 2250, 2500, 2750, and 3000 µg/g.
Conclusions
The results of these experiments provide many insights into the nature of chromium in Baltimore
Harbor sediments. It can be concluded that the addition of Cr(VI) concentrations, significantly
above environmentally relevant levels, to Baltimore Harbor sediments does not cause additional
acute or chronic toxicity to the amphipod L. plumulosus. In these studies, no spiked sediments
demonstrated additional toxicity as a result of spiking with Cr(VI), with the exception of the
50
single sample spiked with 3210 µg/g dry weight of Cr(VI). Further, due to the lack of additional
toxicity observed in conjunction with spiking sediment with Cr(VI), it can also be concluded
that current chromium concentrations in Baltimore Harbor sediments are not causing the acute or
chronic toxicity observed in baseline samples.
This research also generated evidence that supports the theory that AVS is an important mediator
of chromium sediment toxicity. Looking at both the initial experiment, as well as the re-spiking
studies, there was a high correlation between Cr(VI) reduction and sediment AVS
concentrations. In the initial experiment, Cr(VI) was not completely reduced in only one
sediment sample- BSM-68 spiked with 3210 µg/g dry weight Cr(VI). This also represented the
only spike that exceeded a Cr(VI):AVS molar ratio of 1. Similarly, the experimentally
determined threshold concentration for incomplete Cr(VI) reduction roughly corresponded to the
predicted threshold concentration based on a 1:1 stoichiometry for Cr(VI) reduction by AVS.
The Cr(VI) spike of 2,250 µg/g was the lowest spike where Cr(VI) was observed in the overlying
water and pore water. It was also the lowest spike concentration that exceeded the AVS
concentration on a molar basis.
In conclusion, these experiments produced three important insights regarding chromium and
ecotoxicity of Baltimore sediments.
•
First, it was demonstrated that current chromium concentrations are not responsible for
observed toxicity to L. plumulosus.
•
Second, the addition of Cr(VI), at concentrations well above environmentally relevant
levels, to Baltimore Harbor sediments should not cause any additional toxicity.
•
Finally, these experiments support the hypothesis that AVS constituents are the major
contributors to Cr(VI) reduction in anoxic sediment. It also supports the hypothesis that
the stoichiometry of Cr(VI) reduction by AVS occurs on a 1:1 basis, and AVS
concentrations can be used as a predictor of potential toxicity due to Cr(VI) addition to
sediment.
51
Acknowledgements
The authors wish to express thanks to Honeywell International, Inc. for financial support of the
research described in this report.
52
References
American Public Health Association (APHA), American Waterworks Association, Water
Environment Federation. 1998. Standard Methods for Examination of Water and Wastewater,
20th edition or most recent version. APHA, Washington, D.C.
Baker, J., R. Mason, J. Cornwell, J. Ashley, J. Halka, and J. Hill. 1997. Spatial Mapping of
Sedimentary Contaminants in the Baltimore Harbor/Patapsco River/Back River System. Report
to the Maryland Department of Environment: Baltimore, Maryland. UMCES [CBL] 97-142
Becker, D. S., E. R. Long, D. M. Proctor, and T. C. Ginn. 2006. Evaluation of potential
toxicity and bioavailability of chromium in sediments with chromite ore processing residue.
Environ. Toxicol. Chem. 25: 2576-2583.
Berry, W.J., D. J. Hansen, J. D. Mahony, D. L. Robson, D. M. Di Toro, B. P. Shipley, B.
Rogers, J. M. Corbin, and W.S. Boothman. 1996. Predicting the toxicity of metal-spiked
laboratory sediments using acid-volatile sulfide and interstitial water normalizations. Environ.
Toxicol. Chem. 15: 2067-2079.
Berry, W. J., W. S. Boothman, J. R. Serbst, and P. A. Edwards. 2002. Effects of Chromium in
Sediment: 1. Toxicity Tests with Saltwater Field Sediments. U.S. Environmental Protection
Agency, National Health and Environmental Effects Research Laboratory , Atlantic Ecology
Division. Narragansett, RI. Presented at SETAC 23rd Annual Meeting Salt Lake City, Utah. 16 20 November 2002.
Berry, W. J., W. S. Boothman, J. R. Serbst, and P. A. Edwards. 2004. Predicting the toxicity
of chromium in sediments. Environ. Toxicol. Chem. 23: 2981-2992.
53
Besser, J. M., W. G. Brumbaugh, N. E. Kemble, T. W. May, and C. G. Ingersoll. 2004.
Effects of sediment characteristics on the toxicity of chromium(III) and chromium(VI) to the
amphipod, hyalella azteca. Environ. Sci. Technol. 38: 6210-6216.
Boothman, W. S. and A. Helmstetter. 1992. Vertical and seasonal variability of acid volatile
sulfides in marine sediments. U.S. EPA Report 600/X-93/036. Narragansett, RI.
Boursiquot, S., M. Mullet, and J.-J. Ehrhardt. 2002. XPS study of the reaction of
chromium(VI) with mackinawite (FeS). Surface and Interface Science 34: 293-297.
Bufflap, S. E. and H. E. Allen. 1995. Comparison of pore water sampling techniques for trace
metals. Water Research 29: 2051-2054.
Chang, Y. L. and S. J. Jiang. 2001. Determination of chromium species in water samples by
liquid chromatography-inductively coupled plasma-dynamic reaction cell-mass spectrometry. J.
Anal. At. Spectrom. 16: 858-862.
Di Toro, D. M., J. D. Mahony, D. J. Hansen, K. J. Scott, A. R. Carlson, and G. T. Ankley.
1992. Acid volatile sulfide predicts the acute toxicity of cadmium and nickel in sediments.
Environ. Sci. Technol. 26: 96-101.
Driscoll, S. K., G. A. Harkey, and P. F. Landrum. 1997. Accumulation and toxicokinetics of
flouranthene in sediment bioassays with freshwater amphipods. Environ. Toxicol. Chem. 16:
742-753.
Ho. K. T., R. M. Burgess, M. C. Pelletier, J. R. Serbst, S. A. Ryba, M. G. Cantwell, A.
Kuhn, and P. Raczelowski. 2002. An overview of toxicant identification in sediments and
dredged materials. Marine Pollution Bulletin 44: 286-293.
54
James, B. R., J. C. Petura, R. J. Vitale, and G. R. Mussoline. 1995. Hexavalent Chromium
Extraction from Soils: A Comparison of Five Methods. Environ. Sci. Technol. 29: 2377-2381
Klosterhaus, S., J. Baker, G. Zeigler, and D. Fisher. 2006. Toxicity Identification and
Evaluation and Long-Term Contaminant Trends in the Baltimore Harbor, DRAFT. Submitted to:
Technical and Regulatory Services Administration, Maryland Department of the Environment
Lee, B.-W., S. B. Griscom, J.-S. Lee, H. J. Choi, C.-H. Koh, S. N. Luoma, and N. S. Fisher.
2000. Influences of Dietary Uptake and Reactive Sulfides on Metal Bioavailability from Aquatic
Sediments. Science 287: 282-284.
Long, E. R., D. D. MacDonald, J. C. Cubbage, and C. G. Ingersoll. 1998. Predicting the
toxicity of sediment-associated trace metals with simultaneously extracted trace metal: acidvolatile sulfide concentrations and dry weight-normalized concentrations: a critical comparison.
Environ. Toxicol. Chem. 17: 972-974.
Maryland Department of the Environment. 2005. Water Quality Analysis of Chromium in
the Inner Harbor/Northwest Branch and Bear Creek Portions of Baltimore Harbor in Baltimore
City and Baltimore County, Maryland. Submitted to: Watershed Protection Division, USEPA.
McGee, B. L., D. J. Fisher, L. T. Yonkos, G. P. Ziegler, and S. Turley. 1999. Assessment of
sediment contamination, acute toxicity, and population viability of the estuarine amphipod
Leptocheirus plumulosus in Baltimore Harbor, Maryland, USA. Environ. Toxicol. Chem. 18:
2151-2160.
McGee, B. L., D. J. Fisher, D. A. Wright, L. T. Yonkos, G. P. Ziegler, S. D. Turley, J. D.
Farrar, D. W. Moore, and T. S. Bridges. 2004. A field test and comparison of acute and
chromic sediment toxicity tests with the estuarine amphipod Leptocheirus plumulosus in
Chesapeake Bay, USA. Environ. Toxicol. Chem. 23: 1751-1761.
55
Mullet, M., S. Boursiquot, and J.-J. Ehrhardt. 2004. Removal of hexavalent chromium from
solutions by mackinawite, tetragonal FeS. Colloids and Surfaces A: Physiochemical and
Engineering Aspects 244: 77-85.
Murdoch, M. H., P. M. Chapman, D. M. Norman, and V. M. Quintino. 1997. Spiking
sediment with organochlorines for toxicity testing. Environ. Toxicol. Chem. 16: 1504-1509.
Oshida, P.S., L. S. Word, and A. J. Mearns. 1981. Effects of Hexavalent and Trivalent
Chromium on the Reproduction of Neanthes arenaceodetata (Polychaeta). Mar. Env. Res. 5: 4149
Rickard, D. and J. W. Morse. 2005. Acid volatile sulfide (AVS). Marine Chemistry. 97:141197.
Rifkin, E., P. Gwinn, and E. Bouwer. 2004. Chromium and sediment toxicity. Environ. Sci.
Technol. 38: 267A-271A.
Tan, K. H. 1996. Soil Sampling, Preparation, and Analysis. Marcel Dekker, NY, 480 p.
U.S. Environmental Protection Agency. 1979. Methods for Chemical Analysis of Water and
Wastes. EPA/600/4-79/020. Environmental Monitoring and Support Laboratory, Cincinnati,
Ohio
U.S. Environmental Protection Agency. 1994. EPA Method 3051A- Microwave assisted acid
digestion of sediments, sludges, soils, and oils. SW-846.
U.S. Environmental Protection Agency. 1996. EPA Method 3060A- Alkaline Digestion for
Hexavalent Chromium. SW-846.
U.S. Environmental Protection Agency. 1997. The Incidence and Severity of Sediment
56
Contamination in Surface Waters of the United States: National Sediment Quality Survey,
Second Edition. EPA/823/R-97-006. Office of Science and Technology, Washington, D.C.
U.S. Environmental Protection Agency. 2001. Methods for Assessing the Chronic Toxicity of
Marine and Estuarine Sediment-associated Contaminants with the Amphipod Leptocheirus
plumulosus, First Edition. EPA/600/R-01020. Office of Research and Development, Washington,
D.C.
U.S. Environmental Protection Agency. 2004. The Incidence and Severity of Sediment
Contamination in Surface Waters of the United States: National Sediment Quality Survey,
Second EditionEPA/823/R-04-007. Office of Science and Technology, Washington, D.C.
U.S. Environmental Protection Agency. 2004b. EPA Method 9045D- Soil and Waste pH. SW846.
U.S. Environmental Protection Agency. 2004c. EPA Method 9060A- Total Organic Carbon.
Verrhiest, G., B. Clement, and G. Blake. 2001. Single and combined effects of sediment
associated PAHs on three species of freshwater macroinvertebrates. Ecotoxiciology. 10: 363372.
Watlington, K., A. Graham, and E. Bouwer. 2007. The Sediment Ingestion Pathway as a
Source of Toxicity in Baltimore Harbor, Literature Review Report. Submitted to Technical and
Regulatory Services Administration, Maryland Department of the Environment, Baltimore, MD.
Winger, P.V., P.J. Lasier, and B.P. Jackson. 1998. The influence of extraction procedure on
ion concentrations in sediment pore water. Archives Environ. Contam. Toxicol. 35: 8-13.
57
Appendix A: Previous site characterization data from the University of Maryland DRAFT TIE report (Klosterhaus et al.,
2006)
General Data for Sample Sites (University of Maryland TIE Report 2006)
Site
Name
Sediment Carbon
Pore Water DOC
Pore Water
Pore Water
Latitude
Longitude
(%)
(mg/L)
Ammonia (mg/L)
Sulfide (mg/L)
BSM33
Bear Creek #6
39.2536
76.4903
6
22.3
10.2
<0.18
BSM38
Colgate Creek
39.2564
76.5361
4.3
10.5
55.6
<0.18
BSM45
Curtis Bay
39.2175
76.5767
5.3
27
9.9
<0.18
BSM54
Lazaretto Pt.
39.2583
76.5683
3.9
13.4
11
<0.18
BSM68
Northwest Branch
39.2778
76.5833
5.7
14.5
11.6
<0.18
Sediment Grain Information for Sample Sites (University of Maryland TIE Report 2006)
Site Name
%H20
Bulk Density
% Gravel
% Sand
% Silt
%Clay
Shepard's Classification
BSM33
68.2
1.25
0
14.69
38.47
46.84
Silty-Clay
BSM38
71.06
1.22
0
14.15
34.18
50.87
Silty-Clay
BSM45
66.68
1.27
0
13.93
29.71
56.36
Silty-Clay
BSM54
67.8
1.26
1.1
20.53
26.48
51.89
Sand-Silt-Clay
BSM68
70.94
1.23
0
14.81
37.58
47.62
Silty-Clay
58
Metals in sediment (ug/g dry) (University of Maryland TIE Report 2006)
Site Name
Cr
Cu
Zn
Cd
Pb
As
Ag
BSM33
860
211
1640
7.97
265
41.2
2.44
BSM38
213
226
449
16
184
28.8
1.84
BSM45
265
269
487
1.42
270
46.5
1.98
BSM54
246
150
335
1.75
134
23.9
1.72
BSM68
472
356
529
2.95
335
30
3.07
Metals in Baltimore Harbor pore water (µg/L) (University of Maryland TIE Report 2006)
Site Name
Cr
Cu
Zn
Cd
Pb
As
Ag
BSM33
4.33
20.5
1.35
0.061
0.042
10.2
<MDL
BSM38
11.6
62.7
1.28
0.02
0.056
12.3
0.038
BSM45
1.88
24.3
2.22
0.023
0.026
25.1
<MDL
BSM54
3.24
31.7
1.95
0.021
0.045
23.9
0.023
BSM68
4.54
24.8
1.51
0.042
0.103
11
0.053
59
Organics in Baltimore Harbor Sediments (ng/g) (University of Maryland TIE Report 2006)
Site Name
Total PAHs
Total PCBs
Total BDEs
TBT
DBT
MBT
BSM33
6700
180
360
26
<DL
<DL
BSM38
9100
150
90
74
<DL
<DL
BSM45
11600
440
130
31
2
<DL
BSM54
6500
90
60
130
2
<DL
BSM68
15100
210
100
97
30
<DL
Organics in Baltimore Harbor Pore Water (ng/g) (University of Maryland TIE Report 2006)
Site Name
Total PAHs
Total PCBs
Total BDEs
TBT
DBT
MBT
BSM33
140
3.8
0.1
<DL
<DL
<DL
BSM38
300
2.5
<DL
<DL
<DL
<DL
BSM45
180
23.6
<DL
<DL
<DL
<DL
BSM54
110
2.2
<DL
<DL
<DL
<DL
BSM68
210
8.7
0.1
<DL
<DL
<DL
Toxicity of Baltimore Harbor Sediments, (McGee 19991, 20042)
Site Name
Acute Toxicity (% Survival)
BSM33
65.7
BSM38
55
BSM45
76.4 ; 46.7
BSM54
83.1
1
BSM68
53.0
1
2
Chronic Toxicity (% Survival)
2
2
2
2
39.0 ; 80.5
1
1
2
68.0 ; 99.6
60
Appendix B: Sample Characterization Protocols
Digestion for Cr(VI) – A Modified Version of EPA 3060A – Alkaline Digestion for
Hexavalent Chromium
I. Materials
•
•
•
•
•
•
•
•
Teflon microwave digestion vessels
CEM MARS X-press Microwave Digestion Unit
0.2 um nylon membrane syringe filters
10 mL polypropylene syringes
15 mL polypropylene centrifuge tubes
pH meter
Analytical balance
NIST approved temperature measurement device
II. Reagents
•
•
•
•
•
5.0 M HNO3
Anhydrous Na2CO3
NaOH pellets
Lead chromate (insoluble matrix spike) – 10-20 mg PbCrO4
Digestion Solution: 20.0 g NaOH and 30.0 g Na2 CO3 in 1 L flask and dilute to mark (IMPT: pH must be greater
than 11.5)
• K2Cr2O7 spiking solution (1000 mg/L Cr(VI)): dissolve 2.829 g of dried (105 oC) K2Cr2O7 in 1 L flask and dilute
to mark (certified primary standard may used instead)
• Matrix spiking solution (100 mg/L): add 10 mL of 1000 mg/L Cr(VI) solution to 100 mL flask and dilute to
volume
III. Procedure
1.
2.
3.
4.
5.
6.
7.
8.
9.
Weight ~1.0 g of field moist sediment into microwave digestion vessel.
Add 10 mL of alkaline digestion solution (described above)
Hand mix sediment and alkaline digestion solution
Microwave for 60 min at 95 0C
Allow vessels to cool, and pour contents into 15 mL centrifuge tubes
Centrifuge for 30 min at 4000xg
Filter supernatant through 0.2 um nylon membrane filter
Dilute filtered supernatant at least 5x into HPLC mobile phase (2 mM TBAH, 0.6 mM EDTA, at pH 6.9-7.0)
Adjust pH of sample to ~7 with concentrated HNO3 (generally need to add about 10 uL of concentrated nitric
to a 1 mL sample)
10. Transfer quantitatively contents of vessel to 100 mL flask and adjust volume to 100 mL with reagent water.
IV. QA/QC
1.
2.
Preparation blank prepared and analyzed with each digestion batch – detected Cr(VI) must be below
detection limit or 10% regulatory limit
Laboratory control sample (LCS) – utilize matrix spike solution or solid matrix spiking agent into 50 mL
of digestion solution
61
3.
4.
5.
6.
7.
8.
Separately prepared duplicate samples analyzed once per batch (Relative percent difference less than 20%
required)
Soluble and insoluble pre-digestion matrix spikes analyzed once per batch of 20 samples – Acceptable
spike recovery 75-125%
a. Soluble spike – 1.0 mL of spiking solution (40 mg/kg) or at twice sample concentration (greater of
two)
b. Insoluble spike – 10-20 mg PbCrO4 to sample
Compare LCS and matrix spike data. If LCS data is good, but matrix spike a failure, sediment may be
highly reducing and low recovery would be expected. Measure pH and oxidation reduction potential
(convert to Eh) of sediment and evaluate reducing or oxidizing properties of sediment.
If 0% matrix spike obtained – perform mass balance on total Cr analyzed for two samples:
a. Separate unspiked aliquot of sample used for spiking
b. Digested solids remaining after alkaline digestion and filtration – difference should equal to matrix
spike added.
If 0% matrix spike recovery obtained, perform a series of spike additions of increasing spike concentration
to exceed reducing capacity. Determine if response beyond reducing capacity is linear and quantitative.
Post-digestion Cr(VI) matrix spike once per batch (40 mg/kg or twice sample concentration, whichever is
greater) – 85-115% recovery guideline
V. References
James, B. R.; J. C. Petura; R. J. Vitale; and G. R. Mussoline. Hexavalent chromium extraction from soils: a
comparison of five methods. Environ. Sci. Technol. 1995, 29, 2377-2381.
Pettine, M. and S. Capri. Digestion treatments and risks of Cr(III)-Cr(VI) interconversions during Cr(VI)
determination in soils and sediments -- a review. Analytica Chimica Acta 2005, 540, 231-238.
U.S. Environmental Protection Agency. EPA Method 3060A- Alkaline Digestion for Hexavalent Chromium. SW846. Revised December, 1996.
Vitale, R. J.; G. R. Mussoline; K. A. Rinehimer; J. C. Petura; and B. R. James. Extraction of sparingly soluble
chromate from soils: evaluation of methods and Eh -- pH effects. Environ. Sci. Technol. 1997, 31, 390-394.
Digestion for Total Metals – EPA Method 3051 – Microwave Assisted Acid
Digestion of Sediments, Sludges, Soils, and Oils
I. Materials
•
•
•
•
•
•
Volumetric grad cylinder – 50 or 100 mL
Filter paper
Filter funnel
Analytical balance
NIST approved temperature measurement device
Fluorocarbon beaker
II. Reagents
62
•
•
•
High purity concentrated HNO3
Reagent water
2% v/v HNO3 solution prepared using concentrated nitric and DI water
III. Procedure
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Weigh the PFA or TFM vessel, cap assembly, and valve to nearest 0.001 g
Weigh well-mixed sample to nearest 0.001 g using no more than 0.500 g (recommend 0.250 g or 0.100 g if
carbonates or easily oxidized organics present in sample)
In fume hood, add 10 +/- 0.1 mL HNO3 (If reaction occurs, allow reaction to stop before capping)
Cap vessels and weigh to nearest 0.001 g
Place carousel in microwave and connect pressure vessels to central overflow vessel (IMPT – All sample
locations should be filled – fill extra vessels with 10 mL of HNO3 if necessary)
Irradiate for 10 min. Temperature should rise to 175 oC in less than 5.5 min and remain between 170-180
o
C for balance of 10 min period. Pressure should peak at ~ 6 atm.
Allow vessels to cool for at least 5 min before removing from microwave
Weigh each vessel assembly – if total weight has decreased by more than 10%, sample must be discarded
Uncap and vent each bottle in fume hood
Transfer sample to acid cleaned bottle
Centrifuge (2,000-3,000 rpm for 10 min) and settle
Dilute supernatant to 2% v/v HNO3 using DI water
Filter diluted sample through 0.2 um nylon membrane to remove any remaining particles
Dilute to known volume using 2% v/v HNO3 as diluent.
IV. QA/QC
•
•
•
•
Duplicate samples
Standard Reference Material every batch/20 samples – 85-115% recovery required. If standard not
met, SRM run again. If satisfactory result still not obtained, entire batch of samples must be run again
following identification of potential error.
Sample Spikes every batch/20 samples
Digestion with HF will permit determination of fraction of Cr associated with mineral phases not
digested using nitric acid
V. References
U.S. Environmental Protection Agency. EPA Method 3051- Microwave assisted acid digestion of sediments,
sludges, soils, and oils. SW-846. Revised September, 1994.
Cr Speciation Determination by HPLC-ICP-MS
I. Materials
•
•
Autosampler vials
Volumetric flasks for standard prep
II. Reagents
• 1.0 M solution of tetrabutylammonium hydroxide (TBAH)
• potassium salt of EDTA
63
• MeOH
• High purity concentrated HNO3
• Mixed calibration stds – dilute stock-standards to levels in linear range for instrument in 1% v/v HNO3 and
reagent water to obtain appropriate concentration for each analyte of interest
• Cr(VI) std solution – prepared from K2 Cr2O7
• Cr (III) std solution – prepared from CrCl3
• Cr(III) and Cr(VI) calibration standards in range of 0.5 to 100 ug/L diluted in HPLC mobile phase
• Calibration blank for ICP-MS (1% v/v HNO3 in reagent water)
• Cal. Blank for HPLC-ICP-MS
III. Procedure
Mobile Phase Prep
• 2 mM TBAH
• 0.6 mM EDTA
• pH adjusted to 6.9-7.0 with concentrated HNO3
Column Preparation
• Upon arrival column should be washed for 90 min with 100% MeOH at 1.0 mL/min
• Prior to daily use, run mobile phase through column for 30 min.
• Wash at end of day with 70/30 v/v MeOH/water
General ICP-MS Operating Procedure
1. Allow greater than 30 min for instrument equilibration
2. Verify instrument stability with tuning solution (at least 4 times with RSD less than 5%
a. Mass calibration should not differ more than 0.1 amu from true value; mass calibration should be
adjusted to correct value
b. Resolution should be less than 0.9 amu full width at 10% peak height
3. Calibrate for analytes of interest – blank and at least a single calibration std. Flush with rinse blank b/w
each std. Use average of more than 3 integrations for calibration and sample analysis
4. Flush with rinse blank (30 s) until signal returns to quantitation level. Nebulize sample until steady state
signal is achieved. Analyze calibration check standard and calibration blank at least once every 10 samples
HPLC-ICP-MS Calibration Procedure
a. Calibrate from 50 uL injections of 0.1, 0.5, 1, 5, and 10 ug/L of each Cr species in mobile phase
Table 1 – Operating Conditions for ICP-MS and DRC – from PE pamphlet and Chang and
Jiang, 2001
Plasma Conditions
rf power
Ar plasma flow
Auxiliary Ar flow
Nebulizer Ar flow
1050 – 1175 W
15 L/min
1.2 L/min
0.95 L/min
Mass spectrometer settings
Dwell time
300-1000 ms
64
Sweeps
Readings
Isotopes Monitored
DRC parameters
NH3 reaction gas flow
Quadropole rod offset
Cell path voltage
Cell rod offset
Rejection parameter q
Rejection parameter a
Autolens
1
250
52Cr (Chang and Jiang also monitored
53Cr)
0.40 mL/min to 0.65 mL/min
-8.5 V
-29 V
0V
0.4 to 0.70
0
On
Table 2 – HPLC conditions – modified from PE pamphlet and Chang and Jiang, 2001
Mobile Phase
2 mM TBAH+ (Chang and Jiang, 2001
used TBAP)
0.6 mM EDTA (potassium salt); pH 6.9
Flow Rate
1.0 mL/min to 1.5 mL/min
Column
3x3 CR C8
Sample volume
50 uL to 100 uL
Autosampler Flush Solvent
5% MeOH/95% water
IV. QA/QC
1.
2.
3.
4.
5.
6.
Instrument detection limit – calculated as std deviation of three runs on three non-consecutive days of blank
with seven consecutive measurements per day (DL must be determined every three months)
Dilution test – analysis of five fold dilution must agree with in 10% of original determination otherwise
interference suspected. Run dilution test once per 20 samples
Post-digestion spike addition – 75-125% recovery required
Laboratory control sample – same sample prep, analytical methods and QA/QC procedures as test samples.
Run once per 20 samples
Results of calibration blank must be within 3 x DL for each element
Check standards should be within 10% of expected values – run with each batch of samples.
V. References
Chang, Y-L and S-J Jiang. Determination of chromium in water and urine by reaction cell inductively coupled
plasma mass spectrometry. J. Anal. At. Spectrom. 2001, 16, 1434-1438.
Montes-Bayon, M.; K. DeNicola; and J. A. Caruso. Liquid chromatography-inductively coupled plasma mass
spectrometry. J. Chromatography A 2003, 1000, 457-476.
65
Neubauer, K.; W. Reuter; and P. Perrone. Chromium speciation in water by HPLC/ICP-MS. Perkin Elmer Bulletin.
2003.
Pantsar-Kallio, M. and P. K. G. Manninen. Speciation of chromium by coupled column HPLC-ICP-MS-the effects
of interfering ions. Fresenius J. Anal. Chem. 1996, 355, 716-718.
Powell, M. J.; D. W. Boomer; and D. R. Wiederin. Determination of chromium species in environmental samples
using high-pressure liquid chromatography direct injection nebulization and inductively coupled plasma mass
spectrometry. Anal. Chem. 1995, 67, 2474-2478.
Seby, F.; S. Charles; M. Gagean; H. Garraud; and O. F. X. Donard. Chromium speciation by hyphenation of highperformance liquid chromatography to inductively coupled plasma-mass spectrometry -- study of the influence of
interfering ions. J. Anal. At. Spectrom. 2003, 18, 1386-1390.
U.S. Environmental Protection Agency. EPA Method 6020- Inductively Coupled Plasma-Mass Spectroscopy. SW846. Revised September, 1994.
Pore water Extraction by Centrifugation
I. Materials
•
•
•
•
•
High speed centrifuge
250 mL polycarbonate centrifuge tubes
Analytical balance
0.2 um membrane filters
disposable plastic syringes with plastic tips
II. Reagents
•
LC mobile phase for dilution
III. Procedure
1.
2.
3.
4.
5.
Weigh centrifuge tube and cap
Weigh out desired sediment quantity (50 g) into centrifuge tube
Run for 20-30 min at 4 o C at 4000xg or 10000xg
Withdraw sample with syringe and filter supernatant through 0.2 um membrane filter into cleaned sample vial
Dilute to appropriate volume with LC mobile phase
IV. QA/QC
1.
2.
Mass balance to determine extraction efficiency – comparison of weight of pore water extracted vs. total pore
water weight determined by sample drying
Duplicate extractions
V. References
Ankley, G. T., and M.K. Schubauer-Berigan. 1994. Comparison of techniques for the isolation of sediment pore
water for toxicity testing. Archives Environ. Contam. Toxicol. 27, 507-512.
66
Annual Book of ASTM Standards, 2002. ASTM E 1391-94: Standard guide for collection, storage, characterization,
and manipulation of sediments for toxicological testing. ASTM Int.
Carignan, R., F. Rafin, and A. Tessier. 1985. Sediment pore water sampling for metal analysis: A comparison of
techniques. Geochim. Cosmo. Acta, 49, 2493-2497
Winger, P.V., P.J. Lasier, and B.P. Jackson. 1998. The Influence of extraction procedure on ion concentrations in
sediment pore water. Archives Environ. Contam. Toxicol. 35, 8-13.
Determination of Acid Volatile Sulfides – from Boothman and Helmstetter, 1993
I. Materials
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Impinger bottles
Round bottom flasks
N2 gas cylinder
Tygon tubing
Flow controller
O2 scrubber
pH meter
sulfide electrode
reference electrode
100 mL volumetric flasks
150 mL beakers
magnetic stirrer and stir bars
1 L volumetric flasks
Plastic syringes
Needles (Luer tip, 10 mL)
Analytical balance
125 mL Erlenmeyer flasks
glass pipettes – assorted
II. Reagents
•
•
•
•
Deaerated DI water (DDIW)
o Bubble N2 through 2.5 L of DI water for 1 hour
6 M HCl
SAOB solution (2 M NaOH, 0.2 M EDTA, 0.2 M ascorbic acid)
o Dissolve 80.00 g NaOH slowly in 700 mL DDIW
o When cool, add 74.45 g EDTA (disodium form) and stir
o Add 35.23 g ascorbic acid
o Pour solution into 1 L volumetric flask and dilute to mark
Primary Sulfide Standard
o Wash crystals of Na2S·9H2O with DDIW
o Weigh out 12 g Na2S·9H2O and dissolve in 900 mL DDIW
o Pour into 1 L volumetric flask and dilute to volume
67
•
•
•
•
•
SAOB diluent
o Mix 300 mL of SAOB with 300 mL DDIW
Working stock solution
o 50 mL of SAOB diluent in 100 mL volumetric
o Pipette in appropriate volume of primary standard and dilute to volume
Standard Iodine Solution (0.025 N)
o Dissolve 20-25 g KI in 100 mL DI water
o Weigh out 3.2 g I2 and dissolve in KI solution
o Pour solution into 1 L volumetric flask and dilute to mark
Thiosulfate titrant (0.025 N)
Starch indicator
o 1.0 g starch in 100 mL boiling DI water
III. Procedure
Standardization of Primary Stock Solution
1. Pipette 10.00 mL of standard iodine solution into each of two 125 mL Erlenmeyer flasks
2. Pipette 2.000 mL of sulfide primary stock solution into one flask, and 2.000 mL DDIW into the other
3. Add 5.00 mL of 6 M HCl into each flask, swirl, cover, and place in dark for 5 min
4. Titrate each with 0.025 N thiosulfate solution, adding starch indicator when yellow iodine color fades.
End point is reached when blue color disappears.
5. Sulfide concentration is calculated as:
sulfide (µ mol/mL) =
(Vtblank-Vtsample)[S2O32- ] 1 mole S2- 1000µ mole
x
x
Vsample
2 equiv S21mmol
where: Vtblank= volume of titrant added to blank
Vtsampe = volume of titrant added to sample
Vsample = volume of sample (2.000 mL)
[S2O32-] = concentration of S2O32- in mmol/mL
Calibration of Standards
1. Prepare 6 calibration standards from working stock solution that cover range of expected AVS
concentrations
2. Pour standard into 150 mL beaker, add magnetic stir bar and place on stirrer. Stir with minimum
agitation so as to minimize oxidation
3. Rinse electrodes with DDIW and blot dry. Immerse electrodes into sample
4. Allow 8-10 min for electrode response to stabilize and then record the reading in mV
5. Construct a calibration curve by plotting potential in mV vs. log concentration.
AVS recovery
1. Purge apparatus shown below with N2 for at least 30 min prior to sample introduction
2. Fill impinger bottles with 50 mL of SAOB and 30 mL of DDIW
3. Weigh out 1-2 g of wet sediment into 250 mL round bottom flask. Add 50 mL DDIW to cover
sediment. Add stir bar
4. Run purge gas flow at 100 mL/min for 10 min. Then reduce flow to 40 mL/min
5. Stop gas flow and slowly inject 10 mL of 6 M HCL with syringe through septum sidearm
6. Resume gas flow at 40 mL/min and stir well for 30 min.
7. Stop gas flow. Rinse impinger bottles with DDIW into 100 mL volumetric flask. Dilute to volume
8. Pour contents into 150 mL and immerse electrodes. Record measurement in mV.
68
IV. QA/QC
1.
2.
3.
4.
Duplicate analyses – calculated concentrations should differ by no more than 15% of the mean of the
two values.
Calibration blank
a. Add 25 mL SAOB and 15 mL of DDIW to 50 mL volumetric flask
b. Add 1-4 mL of secondary stock solution. Dilute to volume
c. Measured concentration should be within 15% of the expected value
Blank spike
a. Prepare apparatus as if sample, but without sediment.
b. Add 50 mL of DDIW and spike with 1-5 mL secondary stock solution
c. Calculate % recovery – should be between 85-115%. If not, examine apparatus for sources of
error. Correct problems and demonstrate successful blank spike recovery before analyzing
sediment samples
Sample Spike
a. Prepare sediment for analysis that has already been well characterized for AVS concentration
b. Add 1-5mL spike from secondary stock solution
c. Measure sulfide as before and calculate percent recovery – should fall between 85-115%.
Repeat spike if desired recovery not obtained. If recovery is still low, attempt spike with
lower sediment sample quantity to help reduce matrix interferences.
V. References
Allen H.E., Fu G., Deng B. Analysis of acid-volatile sulfide (AVS) and simultaneously extracted metals (SEM) for
the estimation of potential toxicity in aquatic sediments. Environ. Toxicol. Chem. 1993, 12:1–13.
Berry, W.J., W.S. Boothman, J.R. Serbst, and P.A. Edwards. Predicting the toxicity of chromium in sediments.
Environ. Toxicol. Chem. 2004, 23, 2981-2992.
Boothman, W.S. and A. Helmstetter. Determination of acid-volatile sulfide and simultaneously-extracted metals in
sediments using sulfide-specific detection. AVSSEM Standard Operating Protocol v. 2.0, 1993, US EPA
Environmental Research Laboratory, Narragansett, RI Internal Document.
69
Cornwell, J.C. and J.W. Morse. The characterization of iron sulfide minerals in marine sediments. Marine
Chemistry. 1987, 22, 193-206
Soil and Waste pH – EPA Method 9045 D
I. Materials
•
•
•
•
•
•
pH meter
glass electrode
reference electrode
50 mL beaker
thermometer or temperature sensor
analytical balance
II. Reagents
•
NIST primary buffer solutions
III. Procedure
1.
2.
3.
4.
5.
6.
Calibrate the pH meter using pH 2, 4, 7, and 10 NIST standard buffer solutions
Add 20 g of soil/sediment to 50 mL beaker and add 20 mL reagent water. Cover and stir for 5 min.
Let suspension stand for 1 hour to settle or centrifuge
Immerse glass electrode into clear supernatant solution
If sample temperature differs by more than 2 oC from buffer solution, apply temperature correction
Report temperature at which pH of soil in water was measured.
IV. QA/QC
1.
2.
Duplicate samples
Electrode thoroughly rinsed between samples
V. References
U.S. Environmental Protection Agency. EPA Method 9045D- Soil and Waste pH. SW-846. Revised November,
2004.
Gravimetric Method for Determining Sediment Water Content – Adapted from
Tan, 1996
70
I. Materials
•
•
•
•
Oven at 100 – 100 o C
Analytical Balance
Flasks with ground stoppered lid
Dessicator with magnesium perchlorate or calcium sulfate
II. Procedure
1.
2.
3.
4.
5.
6.
Weigh flask and stopper
Add 5-10 g of wet sample, stopper the flask, and weigh
Remove lid from flask and place in oven to dry for 24 hours at 105 o C.
Remove sample from oven and place in dessicator to cool
Place stopper on flask and record weight
wet mass % H2O = wet soil mass – dry soil mass /wet soil mass
III. QA/QC
1.
2.
Return sample and dry for several hours, cool in dessicator and determine weight. Repeat until no difference
in consecutive measurements.
Duplicate samples
IV. References
Tan, K.H. Soil Sampling, Preparation, and Analysis. 1996. Marcel Dekker, NY, 480 p.
71
Appendix C: Organics Analysis Results
72
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement