University of Wollongong Thesis Collections
University of Wollongong Thesis Collection
University of Wollongong
Year 
Distribution, speciation and
geochemistry of selenium in
contaminated marine sediments - Port
Kembla Harbour, NSW, Australia
Pattanan Tarin
University of Wollongong
Tarin, Pattanan, Distribution, speciation and geochemistry of selenium in contaminated
marine sediments - Port Kembla Harbour, NSW, Australia, PhD thesis, School of Earth
Environmental Sciences, University of Wollongong, 2006. http://ro.uow.edu.au/theses/714
This paper is posted at Research Online.
http://ro.uow.edu.au/theses/714
DISTRIBUTION, SPECIATION AND GEOCHEMISTRY OF
SELENIUM IN CONTAMINATED MARINE SEDIMENTS PORT KEMBLA HARBOUR, NSW, AUSTRALIA
A thesis submitted in fulfillment of the requirements for the
award of the degree
DOCTOR OF PHILOSOPHY
from
UNIVERSITY OF WOLLONGONG
by
PATTANAN TARIN
(BSc Hons)
SCHOOL OF EARTH AND ENVIRONMENTAL SCIENCES
- 2006 -
ii
Certification
I, Pattanan Tarin, declare that this thesis, submitted in fulfillment of the requirements for the
award of Doctor of Philosophy, in the School of Earth and Environmental Sciences,
University of Wollongong, is my own work unless otherwise referenced or acknowledged.
The thesis has not been submitted for a degree at this or any other academic institution.
….…………………………..
………….
Pattanan Tarin (Author)
(Date)
iii
Acknowledgements
I would like to express my gratitude to the following people who have helped and supported
me during the course of this research:
•
My supervisors: Prof John Morrison (Uni dad) and Dr Dianne Jolley (Uni sister) for
their guidance, encouragement and understanding, and all other support, especially with
the intense proofreading during the final stage.
•
Royal Thai Government for the scholarship support and Port Kembla Copper Ltd.
(PKC) for the first year project funding.
•
Drs Glennys O’Brien, Damris Muhammad and Bryan Chenhall for sharing knowledge
on sediment chemistry. Prof Bill Maher (UC) and Drs Stuart Simpson and Rob Jung
(CSIRO Lucas Heights) for comments and advice on selenium and sediment work.
•
Tonnes of thanks to Mark O’Donnell for all the technical help and a very hard work
during several core-sampling trips, and to Geoff Black for help with boat preparation.
•
Chris Chipeta (PKC) for prompt processing of numerous samples and support during the
first year harbour survey. Leigh Lemmon (PKC) and Louis Whant (Bluescope Steel) for
the labour work during collection of the grab samples. Captain Chris Haley and staff
(Port Kembla Port Corporation) for the grab-sampling boat service.
•
Atun Zawadski, Jennifer Harrison and Helen at ANSTO for help with the sediment
dating work. Thanks also to AINSE for providing the grant funding.
•
Sandra Quin, Marina, Jenny, Louisa and Wendy for administrative support, Pam
Morgan, Sue and Cathy, Heidi Brown, Peter Haines, Peter Sara, John Korth and the TO
team for technical and miscellaneous help. Special thanks are extended to Beth Peisley –
the Faculty Librarian, and Darien – a student service counselor.
•
Angel friends who have shared life outside Uni and kept me well during this time:
P΄Nang and Charles Pasfields (foster friends); Vivian, Ava, Perl and postgrad lady
support group; Jolley’s and Morrison’s past and present students; Sim Fui, Michi, Minh
Hue and many generations of flat mates; SGI and AusaidGang friends; my brother –
Suwat, my sister and nieces – Arunsri, little Yoa-Yoa and Yoke.
Finally, to god and my parents who provide me love and righteousness. Not having had
opportunity to be formally educated and not knowing any English, both parents will be
looking at this thesis like a work of an alien species but would be so over the moon to have
a daughter called ‘doctor’ which will be cool.
iv
Abstract
Selenium (Se) is an element of concern in Port Kembla Harbour as it was the only element
found in harbour fish tissues in the mid-1990s at concentrations that exceeded the Australia
New Zealand Food Authority Maximum Residue Limit. This thesis investigated the
distribution, speciation, binding phases and geochemical behaviour of selenium in Port
Kembla Harbour sediments, which potentially receive selenium pollution from local metal
processing and smelting and coal industries. Sedimentary selenium is a potential selenium
source for fish and organisms via benthic food chain transfer.
Grab surface sediments from 23 sites around the harbour and a total of 14 sediment cores
were collected from the contaminated Red Beach area (during 2003-2006) and analysed for
selenium concentrations by HG-AAS and also for sediment parameters including grain size
composition, pH, redox potential, other trace metals, porewater composition and sediment
macrocomponents. Two sequential extraction procedures were used to fractionate the solidphase selenium into soluble and adsorbed, carbonate, metal oxyhydroxide, organically
bound, elemental, organic matter and sulfide, and residual selenium fractions. The selenium
behaviour in the different geochemical phases was examined in association with the
measured sediment parameters.
The selenium concentrations in surface sediments from most harbour sites were low (below
3 µg/g) except those in sediments from the Red Beach area (up to 9.38 µg/g), which is in
close proximity to a local copper refinery. Selenium concentrations in the Red Beach
sediment cores ranged from 6 to 1735 µg/g, depending on depth and grain size, with peak
selenium concentrations observed at 6-10 cm and at 14-16 cm depths. The highest selenium
concentration (1735 µg/g), found in the <63 µm Red Beach sediments, was 100 times
higher than the highest sedimentary selenium concentration previously reported in
Australia. The sedimentary selenium was concentrated in fine (<63 µm) grains that are
easily mobile. Selenium was correlated mainly with Pb, Cu and Zn in the > 250 µm fraction
of the surface sediments and in the < 63 µm fraction of the sediment cores, indicating
association from both original ore sources and through post-depositional transformation.
The sediment 210Pb dating estimated the sedimentation rate of Red Beach cores to be 0.55 ±
v
0.03 cm/year. Sediment
210
Pb dating provided an indication that the deeper sediments were
not disturbed and high selenium concentrations in the sediment cores were a result of
historical selenium input potentially from a copper smelter.
The solid-phase selenium in the Red Beach sediment cores was present mainly as elemental
selenium. High proportions of the selenium were also bound to the organic matter in the
upper 10 cm region and associated with the residual fraction below 10 cm. Selenite was the
major selenium species found in the organically bound selenium fraction. Small proportions
of the solid-phase selenium were in soluble and adsorbed fractions, with peak
concentrations in the below 10 cm depth region. Minimal amounts of selenium were found
to associate with iron-manganese oxyhydroxides and carbonate minerals in the sediment.
The Red Beach sediment cores were oxic in the top 2 cm and anoxic below 2 cm depths.
The top 2 cm oxic sediment contained low solid-phase selenium concentrations and low
porewater selenium concentrations. The anoxic 2-10 cm core region contained the peak
solid-phase selenium concentrations but with low porewater selenium concentrations. This
layer was enriched with the organic matter, AVS, organically bound selenium, and
elemental selenium species, indicating a strong link between organic matter decay processes
and the reduction of sulfate and selenium. The below 10 cm-anoxic sediments contained
moderate solid-phase selenium, peak porewater selenium and high soluble and adsorbed
selenium concentrations, and stable pyritic sulfur species. Selenium was observed to
become associated with the residual fraction at the expense of the organically bound and the
elemental selenium in this region. This below 10 cm region contained lower proportions of
copper, lead and zinc in the residual fraction but significant amounts in the organically
bound fractions. The solid-phase selenium correlated with the solid-phase sulfur through the
association of their reduced forms: elemental selenium, pyrite and possibly as pyritic
selenium. Copper was the only major element that co-extracted with elemental selenium.
The reduced selenium forms (elemental and residual) correlated significantly with Cu, Pb
and Zn, suggesting possible formation of independent CuSe, PbSe and ZnSe minerals in the
sediment. Redox potential, sedimentation rate, organic matter components, sulfur and
transition elements are concluded to be the important factors affecting the selenium
geochemical behaviour in Red Beach cores.
vi
Table of Contents
Certification…...…………………………………………………………………….ii
Acknowledgements..……………………………………………………….……….iii
Abstract…………………………………………………………………….……….iv
Table of Contents……………………………………………………………….…..vi
List of Figures......……………………………………………………………….….xi
List of Tables…...…………………………………………………………………..xv
Abbreviations…...…………………………………………………………………xix
Chapter 1:
Introduction
1.1
General introduction………………………………………………………………….1
1.2
Port Kembla Harbour study site……………………………………………………...3
1.3
Objectives of this study………………………………………………………………5
1.4
Thesis outline………………………………………………………………………...6
Chapter 2:
Literature review of selenium in the aquatic environment
2.1
Introduction……………………………………………………………………….….7
2.2
Selenium……………………………………………………………………………...7
2.2.1
Properties………………………………………………………………….…7
2.2.2
Production and uses……………………………………………………….…8
2.3
Environmental sources and occurrence of selenium.……………………………….11
2.4
Selenium distribution in Australian aquatic environments………………………....13
2.4.1 Water………………………………………………………………….…….13
2.5
2.4.2
Sediment………………………………………………………………….....14
2.4.3
Organisms…………………………………………………………………..17
Biological uptake..………………………………………………………………….20
vii
2.6
2.7
2.8
2.5.1
Water-borne selenium pathway……………………………………………..20
2.5.2
Particulate and sedimentary selenium……………………………………....21
2.5.3
Dietary pathway……………………………………………………….…....22
Selenium toxicity……………………………………………………….….……….23
2.6.1
Aquatic life……………………………………………………….….……...23
2.6.2
Wildlife and animals………………………………………………….….....24
2.6.3
Human……………………………………………………….….…………..25
2.6.4
Toxicity mechanism……………………………………………...…….…...25
Selenium biogeochemical processes..…………………………………...…….……27
2.7.1
Speciation……………………………………………………….…………..28
2.7.2
Sorption and precipitation…………………………………….…………….30
2.7.3
Coupled redox processes…………………………………...…….………....31
2.7.4
Microbial activities……………………………………………...…….…….33
General conclusions……….……………………………………………...…….…..34
Chapter 3:
Evaluation and optimisation of a rapid method for total selenium determination
in marine sediments using microwave digestion and hydride generation-atomic
absorption spectrometry
3.1
Introduction……………………………………………………….….……………..35
3.2
Materials and methods……………………………………………...…….………...38
3.3
3.4
3.2.1
Reagents and glassware…………………………………….………………38
3.2.2
Microwave digestion procedures…………………………………….……..38
3.2.3
Sample pretreatment for HG-AAS analysis………………………………...40
3.2.4
Selenium determination by HG-AAS……………………………………....41
Results and discussion…………………………………….………………………..43
3.3.1
Evaluation of microwave digestion methods……………………………….43
3.3.2
Reduction of selenate to selenite…………………………………….……...45
3.3.3
Elimination of nitrogen oxide interferences………………………………...47
3.3.4
Analytical performance…………………………………….……………….48
Conclusions…………………………………….…………………………………...51
viii
Chapter 4:
Selenium speciation in marine sediment extracts using high performance liquid
chromatography and hydride generation-atomic absorption spectrometry
4.1
Introduction…………………………………….…………………………………...52
4.2
Materials and methods……………………………………………………………...56
4.3
4.2.1
Reagents and apparatus……………………………………………………..56
4.2.2
Test materials……………………………………………………………….58
4.2.3
Sediment extraction procedure…………………………………….………..59
4.2.4
HPLC separation and selenium detection…………………………………..60
Results and discussion……………………………………………………………....61
4.3.1
Sediment extraction………………………………………………………....61
4.3.1.1 Effects of extractant reagents on HG-AAS detection……………....61
4.3.1.2 Choice of extractants………………………………………………..63
4.3.1.3 Effects of extractant concentration and extraction time…………….65
4.4
4.3.2
Optimisation of the HPLC separation……………………………………....67
4.3.3
Application to sediment NaOH extracts…………………………………....71
Conclusions…………………………………….…………………………………...73
Chapter 5:
Selenium distribution in Port Kembla Harbour sediments
5.1
Introduction…………………………………….…………………………………...74
5.2
Materials and methods……………………………………………………………...75
5.3
5.2.1
Reagents and apparatus………………………………….………………….75
5.2.2
Collection of surface sediments and core samples………………………….75
5.2.3
Sample preparation and analysis for selenium……………………………...78
5.2.4
Lead-210 dating of Red Beach sediment cores……………………………..80
Results and discussion………………………………….…………………………...83
5.3.1
Selenium in surface sediments………………………………….…………..83
5.3.1.1 Sediment characteristics and grain size……………………………..83
5.3.1.2 Selenium spatial distribution………………………………….…….84
5.3.1.3 Selenium distribution in different grain sizes……………………....87
ix
5.3.1.4 Relationships with other trace elements…………………………….87
5.3.1.5 Preliminary hazard assessment………………………………….….92
5.3.2
Selenium in Red Beach sediment cores…………………………………….94
5.3.2.1 Sediment core characteristics and pH……………………………....94
5.3.2.2 Sediment 210Pb dating results………………………………….…....94
5.3.2.3 Selenium distribution in sediment cores…………………………....97
5.3.2.4 Relationships with other elements in core sediments……………..100
5.3.2.5 Factors affecting the selenium vertical distribution……………….104
5.4
Conclusions…………………………………….………………………………….106
Chapter 6:
Geochemistry of selenium in contaminated marine sediments – Red Beach, Port
Kembla Harbour
6.1
Introduction……………………………….……………………………………….107
6.2
Materials and methods………………………….…………………………………114
6.3
6.2.1
Reagents and apparatus………………………….………………………...114
6.2.2
Sample collection and analysis………………………….………………...115
6.2.3
Sequential extraction procedures………………………….………………120
Results and discussion………………………….………………………………….123
6.3.1
Sediment characteristics, redox potential and pH…………………………123
6.3.2
Sediment porewater compositions………………………….……………..125
6.3.2.1 Porewater sulfate and phosphate………………………….……….125
6.3.2.2 Porewater selenium………………………….…………………….127
6.3.3
Macrocomponent depth profiles………………………….……………….129
6.3.4
Forms and binding phases of selenium in Red Beach sediments………….136
6.3.4.1 SEP 1 fractionation………………………….…………………….136
6.3.4.2 SEP 2 fractionation………………………………………………..140
6.4
6.3.5
Selenium geochemical behaviour in Red Beach sediments…...…………..146
6.3.6
Implications for potential remobilization and bioavailability……………..150
Conclusions…………………………………….………………………………….151
x
Chapter 7:
Conclusions and recommendations
7.1
Introduction……………………………….……………………………………….152
7.2
Conclusions…………………………………….………………………………….152
7.3
Recommendations for future research………………………….…………………156
References…...…………………………………………………………………….152
Appendix A Surface sample data……………………………………………….….I
Appendix B Core sample data………………………………………………….VIII
xi
List of Figures
Figure 1.1
Port Kembla Harbour, NSW, showing major drains and surrounding
industrial environments………………………………………………………4
Figure 1.2
Selenium studies in Port Kembla Harbour sediments, highlighting the topics
covered in each of the main thesis chapters……………………………….....6
Figure 2.1
Selenium cycling in aquatic environment…………...……………………...27
Figure 2.2
A phase diagram for the Se-H2O system for pH and redox potential (Eh).
Between two solid lines is the stability region of water……………………28
Figure 2.3
A schematic of zones of organic matter degradation during diagenesis
processes in sediments...……………………………………………………32
Figure 3.1
In-house microwave rotators for multi-sample digestion, fitted general
50-mL centrifuge tubes……………………………………………………..39
Figure 3.2
Selenium analysis by HG-AAS using Varian VGA-76 vapour generator….42
Figure 3.3
Typical HG-AAS calibration curve for selenium…………………………..49
Figure 3.4
HG-AAS response of selenite standard added to 10 % nitric digested samples
(containing 40% HCl, 4% HNO3 and 0.16% urea), measured against 40%
HCl calibration standards…………………………………………………...50
Figure 4.1
Structures and pKa values of four selenium compounds studied…………..54
Figure 4.2
Effects of extractant matrix on HG-AAS signal (mean ± SE, n=3). HCl,
Ascorbic acid, H3PO4, H3PO4: Methanol, NH2OH.HCl and NaOH were 0.5
mol/L. KCl was 0.25 mol/L and phosphate (pH 8) was 0.1 mol/L…………62
Figure 4.3
Percentage (mean ± SE, n = 3) of selenium extracted from test sediments by
different extractant reagents: HCl, H3PO4 and NH2OH.HCl were 0.5 mol/L,
KCl was 0.25 mol/L, and NaOH and Phosphate (pH 8) were 0.1 mol/L…...63
Figure 4.4
Intense brown colour of NaOH extracts of SRM 2702, in comparison to (a)
other reagent extracts of SRM 2702: hydrochloric acid, phosphoric acid,
phosphoric: methanol and hydroxylamine hydrochloride; and (b) sodium
hydroxide extracts of other test sediments from Port Kembla Harbour: wet
anoxic, PKH-1 and Red Beach (19b) sediments…………………………...64
Figure 4.5
Percentage of selenium extracted in sodium hydroxide solutions with
different extraction time from (a) SRM 2702 and (b) wet anoxic sediment
……………………………………………………………………………...66
xii
Figure 4.6
HPLC of standard selenium compounds (0.1 µg Se) (a) in MilliQ water, and
(b) in 0.1 mol/L NaOH solutions. Hamilton PRP-X100 anion exchange
column, 40 mM /200 mM ammonium phosphate buffer, pH 6 mobile
phase………………………………………………………………………...68
Figure 4.7
HPLC of sediment NaOH extracts: (a) oxic Red Beach sediment (0.1 mol/L,
12 hour extraction) and (b) anoxic wet sediment (0.1 mol/L, 4 hour
extraction), Hamilton PRP-X100 anion exchange column 40 mM /200 mM
ammonium phosphate buffer, pH 6, mobile phase…………………………71
Figure 5.1
Locations of surface samples collected from 23 sites around Port Kembla
Harbour on 7th April 2003…………………………………………………..76
Figure 5.2
Laboratory set up for sediment sample processing. From right to left,
sediment core samples, nitrogen glove box and sediment core extruder…...77
Figure 5.3
Sample preparation and analysis flowchart for selenium spatial distribution
studies in Port Kembla Harbour sediments…………………………………79
Figure 5.4
Dominant grain size distribution in surface sediment samples….………….83
Figure 5.5
Spatial distribution of selenium (µg/g, d.w.) in surface sediments from Port
Kembla Harbour (a) whole sediments and (b) < 63 µm fractions………….85
Figure 5.6
Selenium concentrations (dry weight) in surface sediments from 23 sites of
Port Kembla Harbour (a) µg Se/g for each individual grain size fraction, data
points with error bars were means of oxic and anoxic results and those with
no error bars were of composite samples (b) µg Se in 1 gram of whole
sediment as a function of each grain size…………………………………...88
Figure 5.7
Correlations between selenium and several trace elements in whole surface
sediments of Port Kembla Harbour, excluding Sites 18, 19 and 20………..89
Figure 5.8
Selenium concentrations in whole surface sediments from Port Kembla
Harbour. The red line indicates the 4-µg/g biological effect threshold, as
suggested by the USA research guidelines…………………………………92
Figure 5.9
Pb-210 dating of Core C4 and Core D1 (top) plots of excess Pb-210 activity
(Bq/kg), normalized with < 63 µm grain size, against depth (bottom)
sediment age calculated from CIC model…………………………………..96
Figure 5.10
Grain size distribution and total selenium in three Red Beach sediment cores
………………………………………………………………………………98
Figure 5.11
Depth concentration profiles of selenium and other trace elements in Red
Beach (<63 µm) sediments, Cores A1-A3………………………………...101
xiii
Figure 5.12
Vertical profiles of selenium concentrations (mean±SE) in Red Beach
sediment cores (Cores B1-6 and C1-4 data are taken from Chapter 6). The
corresponding sediment age (Year) determined from 210Pb dating is plotted
against the annual refined copper production by ER&S/SCL…………….105
Figure 6.1
Sediment core sample preparation and analysis flowchart. ………………117
Figure 6.2
Sequential extraction procedures SEP 1 and SEP 2 used for selenium
fractionation in this study………………………………………………….122
Figure 6.3
Depth profiles of redox potential and pH (mean ± SE) of Red Beach
sediment cores collected in April 2004 (Cores B1-6, top row) and July 2005
(Cores C1-4, bottom row). ………………………………………………..124
Figure 6.4
Depth profiles of porewater sulfate (top row) and phosphate (bottom row) in
four individual Red Beach cores: C1-C4………………………………….126
Figure 6.5
Porewater selenium concentrations (mean ± SE) in Red Beach cores
collected in April 2004 (Cores B1-6) and July 2005 (Cores C1-4), in
comparison to the total solid-phase selenium in the corresponding cores...128
Figure 6.6
Concentrations (% d.w.) of Total Carbon, Total Organic Carbon, Total
Nitrogen, Total Sulfur, Acid Volatile Sulfides and Chromium Reducible
Sulfur (pyrites) in Red Beach whole sediment: Cores C1-C4…………….130
Figure 6.7
Cluster relationships between sediment macrocomponents, porewater
selenium, porewater sulfate and < 63 µm fraction in Cores C1-C4……….133
Figure 6.8
Macrocomponent ratios for Red Beach sediment cores (C1-C4, mean ± SE).
Top row: Ratios of TC to TOC, AVS, CrRS and TS. Bottom row: Ratios of
TOC to TN, AVS, CrRS and TS…………………………………………..135
Figure 6.9
Selenium concentrations (µg/g d.w.) in different sediment fractions of Cores
B1-6………………………………………………………………………..137
Figure 6.10
Selenium fractionation patterns (SEP 1) in Red Beach sediment cores (Cores
B1-6), as percentages of the total selenium extracted from sediments……138
Figure 6.11
Selenium concentrations (µg/g) in different sequential extracts (SEP 2) of
four Red Beach cores (whole sediments)………………………………….141
Figure 6.12
Fractionation patterns (SEP 2) of selenium and co-extracted trace elements in
Red Beach cores (mean of C1-C4)………………………………………...142
Figure 6.13
Comparison of selenium and sulfur Eh-pH diagrams. The stability region of
water is between the two solid lines……………………………………….148
xiv
Figure A.1
Dendrograms showing correlation patterns of selenium and common trace
metals in whole (left) and > 250 µm (right) fractions of surface sediments
from Port Kembla Harbour, excluding Site 20……………………………...VI
Figure A.2
Dendrograms (hierarchical clustering analysis) of selenium and common
trace metals in 63-250 (left) and <63 µm (right) fractions of Port Kembla
Harbour surface sediments………………………………………………....VII
Figure B.1
Grain size distribution in Cores B1-B6 collected in April 2005…………XXVI
Figure B.2
Grain size distribution in Cores C1-C4, collected in July 2005………...XXVII
Figure B.3
Porewater selenium and total selenium in individual Red Beach cores: B1B6, collected in April 2004……………………………………………..XXVIII
Figure B.4
Porewater selenium and total selenium in individual Red Beach cores: C1C4, collected in July 2005………………………………………………..XXIX
Figure B.5
Selenium concentrations (µg/g d.w.) in labile fractions of Cores B1-6….XXX
Figure B.6
Selenium fractionation patterns (SEP 1) in individual Red Beach cores (B1B6), Port Kembla Harbour……………………………………………….XXXI
Figure B.7
Selenium fractionation patterns (SEP 2) in individual Red Beach cores (C1C4), Port Kembla Harbour………………………………………………XXXII
xv
List of Tables
Table 2.1
Some selenium compounds and their uses…………………………………...9
Table 2.2
Common selenium minerals and their relative concentrations of selenium..12
Table 2.3
Concentrations of selenium in Australian waters…………………………..14
Table 2.4
Concentrations of selenium in Australian sediments……………………….15
Table 2.5
Concentrations of selenium in Australian marine organisms……………....18
Table 2.6
Biological effects of selenium in aquatic environments…………………....24
Table 2.7
Common selenium species found in the environment……………………...29
Table 3.1
Comparison of techniques for quantitative analysis of selenium in
environmental samples……………………………………………………...35
Table 3.2
HG-AAS operating conditions used in this study…………………………..42
Table 3.3
Acid extractable selenium from reference materials and test sediments using
microwave digestion at 90 ˚C for 20 minutes………………………………43
Table 3.4
Recoveries of acid extractable selenium from reference materials using two
digestion procedures: USEPA Method 3051 and Zhou et al. (1997)…….....44
Table 3.5
Efficiency of selenate reduction to selenite using HCl (microwave heating at
90 ˚C for 10 min, mean ± SE, n=3)…………………………………………46
Table 3.6
Comparison of selenite recoveries (mean ± SE) in aqua regia digestion with
and without the selenate reduction step…………………………………….47
Table 3.7
Recoveries of selenite spikes in nitrogen oxides-containing samples with
urea addition………………………………………………………………...48
Table 3.8
Analytical performance for analysis of total selenium in sediment extracts by
HG-AAS…………………………………………………………………….49
Table 4.1
Selenium speciation in soil/sediments by HG-AAS traditional method and
modern hyphenated techniques……………………………………………..53
Table 4.2
Selected hyphenated methods in recent literature for selenium speciation in
water samples……………………………………………………………….55
xvi
Table 4.3
Major constituents and selenium concentrations in oxic and anoxic reference
materials and test samples…………………………………………………..58
Table 4.4
Extractant reagents tested and sediment phases extracted………………….60
Table 4.5
Optimized HPLC conditions………………………………………………..61
Table 4.6
Analytical performance for selenium speciation by HPLC and HG-AAS…70
Table 5.1
Recoveries of aqua regia extractable metals from certified reference
materials analysed by ICP-OES…………………………………………….81
Table 5.2
Correlations (r) between selenium and common trace metals in different
grain size fractions of surface sediments from Port Kembla Harbour
sampling sites, including Red Beach area…………………………………..91
Table 5.3
Correlations (r) between selenium and other trace elements in Red Beach
sediment (<63 µm) Cores A1-A3………………………………………….103
Table 6.1
Common sequential extraction procedures employed in the literature to
extract selenium from soils/sediments…………………………………….109
Table 6.2
Summary of core samples collected for selenium fractionation studies…..116
Table 6.3
Correlations (r) between total selenium concentrations and measured
sediment parameters in Cores C1-C4……………………………………...132
Table A.1
GPS, pH and grain size data for surface sediment samples collected on 7
April 2003……………………………………………………………………II
Table A.2
Trace element concentrations (µg/g) in different grain size fractions (µm) of
surface sediment samples…………………………………………………....III
Table A.3
Correlations (r) between selenium and common trace metals in different
grain size fractions of surface sediments from Port Kembla Harbour sites,
excluding Red Beach area (Sites 18, 19 and 20)…………………………….V
Table B.1
Summary of all core samples………………………………………………..IX
Table B.2
pH values of core samples…………………………………………………...X
Table B.3
Redox potentials (mV) of core samples…………………………………….XI
Table B.4
Percentage grain size of <63 µm, 63-250 µm and >250 µm fractions in
sediment A, B and C cores……………………………………………….....XI
xvii
Table B.5
Porewater sulfate and phosphate concentrations (mg/L) in Cores C1-C4.
Total porewater volume extracted (mL) vs the sample solid wt. before the
porewater extraction (g) are included for information……………….…….XII
Table B.6
Porewater selenium concentrations (µg/L) in Cores B1-B6 and C1-C4…...XII
Table B.7
Sediment macrocomponent concentrations (% dry wt, whole sediment) in
Cores C1-C4……………………………………………………………….XIII
Table B.8
Total selenium concentrations (µg/g, d.w.) in different grain sizes of the
sediment A, B and C cores…………………………………………….…..XIV
Table B.9
Concentrations of selenium (µg/g, d.w.) in sequential fractions (SEP 1) of
Cores B1-B6 samples (< 63 µm sediment)………………………………...XV
Table B.10
Concentrations of selenium (µg/g, d.w.) in sequential fractions (SEP 2) of
Cores C1-C4 samples (whole sediment)…………………………………..XVI
Table B.11
Concentrations of chromium (µg/g, d.w.) co-extracted in the sequential
extracts (SEP 2) of Cores C1-C4 samples………………………………...XVII
Table B.12
Concentrations of copper (µg/g, d.w.) co-extracted in the sequential extracts
(SEP 2) of Cores C1-C4 samples………………………………………...XVIII
Table B.13
Concentrations of iron (µg/g, d.w.) co-extracted in the sequential extracts
(SEP 2) of Cores C1-C4 samples………………………………………….XIX
Table B.14
Concentrations of manganese (µg/g, d.w.) co-extracted in the sequential
extracts (SEP 2) of Cores C1-C4 samples………………………………….XX
Table B.15
Concentrations of nickel (µg/g, d.w.) co-extracted in the sequential extracts
(SEP 2) of Cores C1-C4 samples………………………………………….XXI
Table B.16
Concentrations of lead (µg/g, d.w.) co-extracted in the sequential extracts
(SEP 2) of Cores C1-C4 samples…………………………………………XXII
Table B.17
Concentrations of zinc (µg/g, d.w.) co-extracted in the sequential extracts
(SEP 2) of Cores C1-C4 samples………………………………………...XXIII
Table B.18
Concentrations of total trace metals (µg/g, d.w.) in < 63 µm fractions of
Cores A1-A3 samples………….………………………………………...XXIV
Table B.19
Concentrations of trace metals (µg/g, d.w.) co-extracted in the reactive iron
fraction of Cores C1-C4 samples………….……………………………...XXV
Table B.20
Sediment macrocomponent ratios (weight ratio, unless indicated) of Red
Beach cores (C1-C4)……………………………………………………..XXVI
xviii
Table B.21
Correlations (r) between sediment parameters and co-extracted elements in
the soluble and adsorbed fraction of Cores C1-C4. ……………………XXVIII
Table B.22
Correlations (r) between sediment parameters and co-extracted elements in
the organically bound selenium fraction of Cores C1-C4. ……………...XXIX
Table B.23
Correlations (r) between sediment parameters and co-extracted elements in
the elemental selenium fraction of Cores C1-C4. . ………………………XXX
Table B.24
Correlations (r) between sediment parameters and co-extracted elements in
the organic matter and sulfide fraction of Cores C1-C4. . ………………XXXI
Table B.25
Correlations (r) between sediment parameters and co-extracted elements in
the residual fraction of Cores C1-C4. …………………………………..XXXII
xix
Abbreviations
µg
Microgram
µL
Microlitre
ANSTO
Australian Nuclear Science and Technology Organization
AVS
Acid Volatile Sulfide
BCR
Community Bureau of Reference
BDL
Below Detection Limit
CrRS
Chromium Reducible Sulfide
dMedSe
Dimethyldiselenide
dMeSe
Dimethylselenide
Eh
Redox potential
GF-AAS
Graphite Furnace-Atomic Absorption Spectrometry
HG-AAS
Hydride Generation-Atomic Absorption Spectrometry
HPLC
High Performance Liquid Chromatography
hr
Hour
HS
Humic substance
ICP-MS
Inductively Coupled Plasma-Mass Spectrometry
ICP-OES
Inductively Coupled Plasma-Optical Emission Spectrometry
KOAc
Potassium Acetate
kPa
Kilopascal
Ksp
Solubility product
LOD
Limit of Detection
MeOH
Methanol
mV
Millivolt
NA
Not analysed
NaOAc
Sodium Acetate
ND
Not Detected
NIST
National Institute of Standards and Technology
NRCC
National Research Council of Canada
NS
No sample
pKa
Log acid dissociation constant
xx
PKC
Port Kembla Copper Limited
PKPC
Port Kembla Port Corporation
PW
Porewater
PZC
Point of Zero Charge
RNAA
Radiochemical Neutron Activation Analysis
RPC
Reversed Phase Chromatography
RSD
Relative Standard Deviation
SAX
Strong Anion Exchange
SE
Standard Error
SeCys
Selenocysteine
SeCys2
Selenocystine
SeCyst
Selenocystamine
SeEt
Selenoethionine
SeIV
Selenite
SeMet
Selenomethionine
SEP
Sequential Extraction Procedure
SeU
Selenourea
SeVI
Selenate
TC
Total Carbon
TFE
Tetrafluoroethylene
TMAH
Tetramethylammonium hydroxide
TMeSe
Trimethylselenonium
TN
Total Nitrogen
TOC
Total Organic Carbon
TS
Total Sulfur
UC
University of Canberra
XAS
X-Ray Absorption Spectrometry
XRD
X-Ray Diffraction
XRF
X-Ray Fluorescence Spectrometry
Chapter 1
Introduction
1.1
General introduction
Selenium has been a controversial element since the time of its discovery in 1817, when it
was first thought to be the element tellurium (Greenwood and Earnshaw, 1984; Carroll,
1999). Initially, selenium was considered a toxic element, and its toxicity in terrestrial
animals resulting from grazing on selenium-laden plants growing in seleniferous soils was
a well-known issue (Shamberger, 1983; Ewan, 1989; Magee, 1996). Then after it was
found to be an essential trace element in biological tissues in 1957 (Schwarz and Faltz,
1957), selenium biochemistry and its essential role in proteins have received wider
attention from research scientists. A selenium-containing molecule, selenocysteine, has
been recognized as the 21st essential amino acid (Rayman, 2000; Johansson et al., 2005)
and together with development of modern biomolecular sciences and advanced analytical
technology, selenium research has become one of the most interesting areas in many
integrated scientific fields, notably, in medicinal (Schweizer et al., 2004; Abdulah et al.,
2005; Brenneisen et al., 2005), nutritional (Hoenjet et al., 2005; Lyons et al., 2005) and
environmental fields (Sappington, 2002; Hamilton, 2004; Lemly, 2004).
Selenium in the environment is a significant research area as environmental selenium
provides a source for biological uptake. Selenium is generally widely distributed and is
cycled through environmental compartments via both natural and anthropogenic processes
(Fishbein, 1983). Selenium pollution becomes a problem when it is concentrated or
discharged from human activities, such as coal and oil combustion, metal mining and
smelting processes, and irrigation through seleniferous soils (Peters et al., 1999a; De
Gregori et al., 2002; Lemly, 2002; Lemly, 2004). Selenium contamination in the
environment has been known to cause serious (acute and chronic) biological effects due to
its unique toxicological properties (Lemly, 1999a; Ohlendorf, 2002; Spallholz and
Hoffman, 2002; Hamilton, 2004). Selenium has similar properties to sulfur and can be
incorporated into amino acids/proteins, accumulated in biological tissues and magnified
along food chains (Fan et al., 2002; Barwick and Maher, 2003; Bhattacharya et al., 2003) to
Chapter 1 – Introduction
2
a level that could potentially cause adverse biological effects. The difference between the
biologically safe and toxic selenium doses is narrow, covering only one order of magnitude
(Lemly, 1997a; Orr et al., 2006). In addition, toxicity of selenium is governed by its
chemical forms (Opresko, 1993; Hyne et al., 2002; Doyle et al., 2003). Environmental
selenium can occur in different oxidation states, and in both organic and inorganic forms
(Fishbein, 1983; Weres et al., 1989; Fox et al., 2003). Its biogeochemical behaviour is
influenced by various physical, chemical and biological conditions making it complex and
site-specific (Lemly, 1998; Hamilton and Lemly, 1999; Cutter and Cutter, 2004). Presently,
there is still much to understand about selenium’s paradoxical and complex behaviour in
the environment, in order to facilitate the management and control of selenium
contamination issues.
A review of literature for this research (presented in Chapter 2) found relatively limited
data on selenium concentrations and speciation in waters, sediments, and organisms in
Australia. The information on selenium behaviour and biotransformation pathways within
the aquatic environment is still speculative. The current Australian and New Zealand
Environment and Conservation Council and Agriculture and Resource Management
Council
of
Australia
and
New
Zealand
water
and
sediment
guidelines
(ANZECC/ARMCANZ, 2000) also indicate the absence of specific guideline values for
selenium (e.g., speciation and sediment quality guidelines) due to insufficient
local/overseas datasets. More intensive selenium studies are needed to support the
derivation of speciation criteria and sediment based criteria.
Selenium received little attention up to the 1990s due to a lack of awareness of its
significant environmental behaviour and a lack of suitable analytical techniques. Selenium
interest within Australia has focused on selenium deficiency due to geographically low
selenium concentrations (0.09 – 0.7 µg/g d.w.) in the soils that leads to low selenium
concentrations in local food, and hence dietary intake (Judson and Reuter, 1999;
McNaughton and Marks, 2002; Tinggi, 2003; Lyons et al., 2005). While many areas in
Australia are facing selenium deficiency due to low soil selenium concentrations, other
areas contain toxicologically high selenium concentrations particularly in some industrially
polluted coastal areas, including Port Kembla Harbour.
Chapter 1 – Introduction
1.2
3
Port Kembla Harbour study site
Port Kembla Harbour (Figure 1.1) is located 75 km south of Sydney (latitude 34°29′S and
longitude 150°54′E), NSW, Australia. The harbour body has been constructed into two
distinct areas, the Inner Harbour and Outer Harbour, divided by a 155 m wide channel
known as ‘the Cut’ (He and Morrison, 2001). The Inner Harbour was formed by dredging
of Tom Thumb Lagoon, occupying an area of 56.5 hectares with water depths from 9.2 to
16.3 metres. The Outer Harbour, occupying an area of 137.5 hectares with water depths
from 4 to 16 metres, was formed adjacent to a headland to the southeast by the construction
of two breakwaters (He and Morrison, 2001). Port Kembla Harbour is host to several heavy
industries that are potential sources of selenium contamination including a copper smelter
(Port Kembla Copper Limited), a steelworks (Bluescope Steel Limited), a coal export
terminal, and a sulfuric acid production plant (Pivot Limited). The harbour is ranked ninth
in Australian ports and contributes to approximately 1300 ships and 25 million tonnes of
cargo movements annually (He and Morrison, 2001) which will increase further with the
opening of the new jetty. The Inner Harbour area often experiences siltation due to fine
sediment and organic particle load from Allans Creek and Gurungaty Creek. The
navigation channel up to the main entrance channel requires regular maintenance dredging
(Delaney, 1996; Marine Science & Ecology, 1996).
Port Kembla Harbour has a long history of heavy metal pollution due to emissions and
discharges from the surrounding industries. Studies on metal contamination of Port Kembla
Harbour have focused mainly on other common trace metals with very limited research on
selenium (Moran, 1984; Marine Science & Ecology and Coastal Environmental
Consultants Pty. Ltd., 1992; Goodfellow, 1996; Low, 1998; Martley et al., 2004).
Concentrations of other metal pollutants in the harbour water and fish were reported to have
substantially decreased over the period 1970-1995 (He and Morrison, 2001). Selenium was,
however, reported to be the only element found at concentrations that exceeded the ANZFA
MRL in the tissue of the harbour fish during the 1990s (Environmental Protection
Authority, 1994; Marine Science & Ecology, 1996), despite the very low (< 1 ppb)
concentrations of selenium found in the harbour water (Phillips, 2002). Selenium is,
therefore, an element of concern for Port Kembla Harbour.
Chapter 1 – Introduction
4
Please see print copy for Figure 1.1
Figure 1.1 Port Kembla Harbour, NSW, showing major drains and surrounding industrial
environments, aerial photo from Hatch Engineering (2003).
Chapter 1 – Introduction
1.3
5
Objectives of this study
The studies presented in this thesis investigated the spatial distribution, speciation, binding
phases and the geochemical behaviour of selenium in a contaminated marine sediment
system, Port Kembla Harbour. Selenium research within our institution is in its infancy, so
initial assessment of appropriate sample preparation and analytical methods for the
determination of selenium and its species in environmental samples was also required to
support the field sample studies.
The general objectives of this research were to:
•
evaluate and optimise a rapid method for total selenium determination in sediment
samples employing microwave-assisted digestion and hydride generation-atomic
absorption spectrometry (HG-AAS);
•
develop and optimise a selenium speciation method for determination of organic
and inorganic selenium compounds in sediments based on HPLC separation and
HG-AAS detection;
•
investigate the spatial distribution of selenium in surface sediments from Port
Kembla Harbour and in cores from the more contaminated Red Beach area, in
relation to sediment grain sizes and concentrations of other trace elements (As, Cd,
Cr, Cu, Fe, Mn, Ni, Pb, Sb and Zn);
•
determine sedimentation rate and sediment age for Red Beach sediment cores using
210
Pb radiodating technique and to subsequently identify selenium contamination
history;
•
determine solid-phase speciation and binding phases of the selenium in Red Beach
sediment cores using sequential extraction procedures; and,
•
investigate the geochemical and early diagenetic behaviour of selenium in Red
Beach sediment cores, in relation to the sediment pH, redox potential, pore water
anion composition, sediment macrocomponents, and common trace elements.
The selenium contamination and behaviour in the sediments are also discussed with the
implications to its relative mobility and potential availability to aquatic organisms.
Chapter 1 – Introduction
1.4
6
Thesis outline
Following this introductory chapter, Chapter 2 presents a literature review on selenium in
the environment with an emphasis on aquatic ecosystems. Then, there will be four main
chapters representing two selenium-method studies and two field-sample studies as outlined
in Figure 1.2. Initial development and optimisation of methods for determination of total
selenium and its species in sediment samples based on HG-AAS techniques are reported in
Chapter 3 and Chapter 4, and the harbour survey and selenium geochemical studies are
presented in Chapter 5 and Chapter 6. The final chapter (Chapter 7) provides the
summary of the key findings and recommendations for further research.
Selenium studies in Port Kembla
Harbour sediments
Chapter 2
Literature review of
selenium in the environment
Chapter 3
Chapter 4
Determination of total selenium
in sediments using microwave
digestion and HG-AAS
Selenium speciation in
sediment extracts using HPLC
separation and HG-AAS
Chapter 5
Chapter 6
Spatial distribution of selenium
in the harbour surface and core
sediments. Its relationships with
sediment grain sizes, depth
profiles and sedimentation rate,
and trace metals
Studies of selenium
geochemistry and diagenesis in
the harbour sediment cores.
Selenium species and binding
phases determined using
sequential extraction procedures.
Figure 1.2 Selenium studies in Port Kembla Harbour sediments, highlighting the topics
covered in each of the main thesis chapters.
Chapter 2
Literature review of selenium in the aquatic environment
2.1
Introduction
This chapter reviews the research literature on selenium in the environment, its
biogeochemistry, toxicity and environmental contamination issues. The emphasis of the
review was on the aquatic environment.
2.2
Selenium
2.2.1
Properties
Selenium (Se) is a metalloid with an atomic number of 34 and an atomic weight of 78.96,
positioned between sulfur and tellurium in Group 16 of the Periodic Table. Elemental
selenium has a melting point of 220 ˚C, a boiling point of 685 ˚C and a density of 4.81
g/cm3 (Miessler, 1999). In nature, selenium exists as six stable isotopes: Se-74, 76, 77, 78,
80 and 82, with Se-80 and 78 being the most common, accounting for 49.8 and 23.5 %,
respectively (Nazarenko, 1972). Selenium’s physical structure can be either crystalline or
amorphous. Three allotropic forms of selenium are generally recognized, including two
crystalline monoclinic forms and a single crystalline hexagonal form. Crystalline hexagonal
selenium, the most stable form, is metallic grey, whereas crystalline monoclinic selenium is
deep red. The amorphous selenium is either red (in powder form) or black (in vitreous
form) (Nazarenko, 1972).
Chemical properties of selenium are similar to those of sulfur (biochemical) and tellurium
(metallurgical and industrial). Selenium has six valence electrons as in sulfur but it is a
larger and softer atom than sulfur, i.e., its electron cloud is large, diffuse and easily
distorted. This means that each selenium atom can readily spread its electrons over many
neighbours, and also donate electrons to other selenium atoms or to other non-metals to
which it is bound. Because selenium is comfortable with carrying a negative charge, it can
also accept electrons from metals or hydrogen (Rayman, 2002). The ability of selenium to
Chapter 2 – Literature review
8
donate and accept electrons reflects the dual nature of the element as a metalloid. When
selenium gives electrons, it is expressing its metallic character such as in SeO2. When
receiving electrons, selenium is expressing its non-metallic character such as in H2Se.
Hydrogen atoms in H2Se can leave as protons (H+…HSe-) making H2Se acidic (H2Se is
more acidic than H2S) (Greenwood and Earnshaw, 1984; Rayman, 2002).
2.2.1
Production and uses
Selenium is mainly obtained as a by-product of copper refining processes, since naturally
occurring selenium minerals such as eucairite (CuAgSe), crooksite (CuThSe) and
clausthalite (PbSe) are too rare to provide sufficient sources for commercial selenium
(Fishbein, 1983). Most of the world's selenium, i.e., more than 90% of the USA and more
than 80% of the world’s production, is derived from anode slimes generated in the
electrolytic production of copper (Brown, 2000). Secondary sources and recovery of
selenium include factory scraps generated during manufacture of selenium rectifiers,
burned-out rectifiers, spent catalysts, and used photocopying cylinders (Fishbein, 1983).
Approximately 250 metric tonnes of secondary selenium are produced annually worldwide
(Brown, 2000). A minor quantity of selenium is obtained from the accumulated residue of
selenium from sulfuric acid manufacture (Nazarenko, 1972).
The industrial isolation process of selenium is dependent on the types of other compounds
or elements present (Fishbein, 1983). All commercial processes for the production of
selenium may be considered as modifications or combinations of three fundamental steps:
smelting with soda ash, roasting with soda ash and roasting with sulfuric acid (Fishbein,
1983). Generally, the first step involves an oxidation in the presence of sodium carbonate
(soda ash) as exemplified in Equation 2.1:
Cu2Se + Na2CO3 + 2O2
2CuO + Na2SeO3 + CO2
…………………(Equation 2.1)
Then, the selenite Na2SeO3 is acidified with sulfuric acid. Any tellurites precipitate out
leaving selenous acid (H2SeO3) in solution. Selenium is liberated from selenous acid by
SO2 as per Equation 2.2:
Chapter 2 – Literature review
H2SeO3 + 2SO2 + H2O
9
Se + 2H2SO4
……………………(Equation 2.2)
The world refinery production (excluding USA) of selenium in 2000 was 1,460 metric tons
with Japan being the largest producer, followed by Canada, Belgium, and Germany
(Brown, 2000).
Selenium exhibits photovoltaic, and photoconductive properties. In solid elemental
selenium and in solid metal selenides, the small gap of electron energy levels between the
ground state and the excited states allows electrons to be easily excited, e.g., by light
energy, and to readily pass from one atom to nearby atoms. Selenium and many metal
selenides are therefore good semiconductors and photoconductors and strongly coloured
(Rayman, 2002). Various applications of selenium are based on these properties as detailed
in Table 2.1.
Table 2.1
Some selenium compounds and their uses (Fishbein, 1983).
Please see print copy for Table 2.1
Chapter 2 – Literature review
10
Glass manufacturing accounted for about 25% of the selenium usage in 2000 (Brown,
2000). Selenium is used principally as a decolorant in container glass and other soda-lime
silica glasses. Under weak oxidizing conditions, the addition of selenium adds a pink colour
to the glass that combines with the green colour imparted by ferrous ions to create a neutral
grey colour that has low perceptibility to the human eye. Selenium is also used to reduce
solar heat transmission in architectural plate glass and to add red colour to glass, such as
that used in traffic lights (Brown, 2000).
Metallurgical uses comprised an estimated 24% of the selenium market in 2000. It is
estimated that more than half of the metallurgical selenium is used as an additive to steel,
copper, and lead alloys to improve machinability. Several USA producers of rolled steel
bars produce selenium-bearing free-machining rods. Selenium-containing free-cutting
steels, however, are generally cost competitive only when used with high-speed automatic
machine tools. Recently, there has been an increase in selenium uses (with bismuth) in
pluming applications in place of lead, as they provide the same free machining properties as
lead without its negative environmental effects, especially for the installation and repair of
facilities providing water for human consumption. In addition, a smaller amount of
metallurgical selenium is used as an additive to low-antimony lead alloys forming the
support grids of lead-acid storage batteries. The addition of 0.02% selenium by weight as a
grain refiner, improves the casting and mechanical properties of the alloy. Hybrid batteries
have been gaining in usage, thus increasing the demand for selenium (Brown, 2000).
Electronics uses are decreasing in recent years, accounting for 10% of selenium use in 2000
(Brown, 2000). High-purity selenium compounds were used principally as photoreceptors
on the drums of plain-paper copiers. Photoreceptors had been the largest single application
for selenium during the 1970s and 1980s. Selenium compounds, however, are being
replaced by organic photoreceptor compounds, which reportedly offer better performance,
lower cost and are free of the environmental concerns that are associated with the disposal
of selenium compounds. Other electronics uses of selenium included rectifiers, devices
which convert alternating current electricity into direct current electricity, and photoelectric
applications (Brown, 2000).
Chapter 2 – Literature review
11
Agricultural uses accounts for about 19% of the selenium usage. Dietary supplements for
livestock are the largest agricultural use. Selenium may also be added to fertilizers used in
growing animal feed (Brown, 2000).
Chemical uses of selenium, including industrial and pharmaceutical applications, accounted
for about 14% of usage. Selenium is gaining greater recognition as a nutrient essential for
human health; small quantities of selenium are used as human dietary supplements. As
ongoing research verifies the apparent cancer-preventive properties of selenium, this
application is also increasing, but the low dosage requirement precludes it from becoming
significant in terms of quantity consumed. The principal pharmaceutical use of selenium is
in antidandruff hair shampoos. Miscellaneous industrial chemical uses include lubricants,
rubber compounds, and catalysts (Brown, 2000).
In pigment applications, selenium is used to produce colour changes in cadmium-sulfidebased pigments. Yellow cadmium pigments become redder as the selenium-to-sulfur ratio
increases. Sulfoselenide red pigments have good heat stability and are used in ceramics and
plastics, as well as in paints, inks, and enamels. Because of the relatively high cost and the
toxicity of cadmium based pigments, their use is generally restricted to applications
requiring long life, brilliance, high thermal stability, and chemical resistance. Pigments
accounted for about 8% of the selenium usage in 2000 (Brown, 2000).
2.3
Environmental sources and occurrence of selenium
Selenium is the 68th most abundant element of the Earth’s crust and widely dispersed in the
environment depending on local geological structures (Nagpal and Howell, 2001).
Selenium usually occurs in association with various sulfide minerals/ metallic ores in which
it replaces sulfur atoms and only forms minerals with elements having a comparatively high
atomic number, e.g., Pb, Hg, Bi, Ag, Cu, Co, Fe, Tl, Ni, Zn and Cd (Nazarenko, 1972).
Particularly, selenium has strong affinity for copper and the accumulation of copper in ores
is usually accompanied by concentration of selenium (Nazarenko, 1972). Some common
selenium minerals are listed in Table 2.2.
Chapter 2 – Literature review
Table 2.2
12
Common selenium minerals and their relative concentrations of selenium
(Nazarenko, 1972; Louderback, 1976).
Please see print copy for Table 2.2
Rare native selenium rocks were recently discovered in Yutangba, Enshi City, Hubei
Province China (where selenium poisoning occurred among the villagers in 1963).
Selenium occurrence in these rocks was by isomorphous substitution of selenium into the
pyrite lattice with the maximum selenium content of 6.68%. Some of the selenium was
found as eskebornite (CuFeSe2) and both forms (pyritic selenium and eskebornite) were
found to account for 33.9% of the total selenium (Zhu et al., 2004). Native selenium
sources are divided into three categories: the primary native selenium occurring in
carbonaceous-siliceous rocks and tiny selenium crystals formed in cracks of rocks during
tectonic activities; micro-selenium crystals formed in the weathering processes of Se-rich
rocks, and; larger selenium crystals derived from natural burning of stone coal in the
subsurface of abandoned stone coal spoils (Zhu et al., 2004).
Through natural processes, selenium is randomly dispersed in the environment depending
on the geological conditions (Nagpal and Howell, 2001). Naturally high selenium
concentrations can be found in some types of soils, called seleniferous soils, which are
found in the arid and semi-arid areas of the world, including the western areas of Canada
and the USA (Nagpal and Howell, 2001). Seleniferous soils in arid and semi-arid regions of
San Joaquin Valley, USA, are known to be the major selenium sources responsible for fish
and wildlife poisoning in Kesterson Reservoir, through irrigation of water for crop
production (Lemly, 1997a).
Chapter 2 – Literature review
13
Anthropogenic activities can greatly contribute to selenium dispersion in the environment.
The recognized anthropogenic sources of selenium contamination include the combustion
of fossil fuels (coal and oil), primary and secondary non-ferrous metal industries (e.g., Pb,
Cu-Ni and Zn-Cd), selenium-containing waste disposal and incineration (e.g., rubber tyres
and papers), manufacturing processes, metal mining and refining processes, and fossil fuelor coal-fired power generation (Nobbs et al., 1997; Nagpal and Howell, 2001; Lemly,
2004). Additionally, sulfuric acid production may release some selenium into the
environment via atmospheric SeO2 emission during roasting of selenium-containing pyrite.
The sludges from the sulfuric acid industry contain 0.9 – 63.7 % selenium (usually in the
elemental state) (Nazarenko, 1972). Coal mining is also a potential selenium source in
Australia due to the oxidation of selenium-bearing pyrite exposed to surface oxic
conditions. Coals from Australia are reported to contain 0.2 – 1.6 µg/g of selenium
(Swaine, 1990). In NSW, coal selenium concentrations range from 0.25-2.5 µg/g in the
north, 0.21-0.63 µg/g in the south to 0.9 – 2.2 µg/g in the west of the state (Swaine, 1990).
2.4
Selenium distribution in Australian aquatic environments
2.4.1
Water
Data on selenium concentrations in Australian waters are limited as summarized in Table
2.3. Selenium concentrations in water are reported to be low and vary between sites. Apte
et al. (1998) conducted a baseline study in NSW coastal waters off Eden, Port Macquarie,
Terrigal, Ulladulla and Yamba and found selenium concentrations to be below 0.073 µg/L
(HG-AAS detection limit). Similarly low selenium concentrations have been reported in
ocean waters: 0.079 µg Se/L for the North Pacific Ocean (Cutter and Bruland, 1984), and
0.045 µg Se/L for the Atlantic Ocean (Cutter and Cutter, 1995). Water selenium
concentrations from relatively more contaminated sites, such as Lake Macquarie, have been
reported to be higher, up to 4.7 µg/L (Carroll, 1999). Unusually high concentrations (up to
480 µg Se/L) of water selenium were reported for Peel Inlet and Harvey Estuary, WA
(Summers and Pech, 1997). They were obtained during February and August samplings
and believed to result from flushes of rainfall that washed down selenium and trace
elements from the catchment area (Summers and Pech, 1997) .
Chapter 2 – Literature review
Table 2.3
14
Concentrations of selenium in Australian waters.
Please see print copy for Table 2.3
Selenium speciation data for water are limited, possibly due to the very low concentrations
of total selenium, making those of individual species below the detection limits of most
analytical methods. Selenate was reported to be a major selenium species in both fresh
water (Derwent River Estuary) and marine water (Maria Island) (Wake et al., 2004).
2.4.2
Sediment
Concentrations of selenium reported in Australian sediments are shown in Table 2.4. The
most extensive data are available for Lake Macquarie, NSW, where selenium
contamination was reported. Sedimentary selenium concentrations for other sites such as
Peel Inlet and Harvey Estuary in Western Australia, Sydney’s continental shelf sediment
and North Head, Bondi and Malabar, NSW were below or near 1 µg/g dry wt.
Chapter 2 – Literature review
Table 2.4
15
Concentrations of selenium in Australian sediments.
Please see print copy for Table 2.4
Chapter 2 – Literature review
16
A selenium concentration of 14 µg/g was reported for a sediment core from Cockle Bay at
the northern end of Lake Macquarie by an early study (Batley, 1987), indicating selenium
contamination issues which triggered later studies. Selenium concentrations in surface
sediments were found to be up to 1.94 µg/g at Bennet Park (northeast) and up to 1.8 µg/g
from the southern perimeter of Cockle Bay and adjacent to the power station at Vales Point
(Carroll et al., 1996). The mean selenium concentrations in surface sediment from the
southern basin of Lake Macquarie ranged from 0.9±0.2 µg/g at Nord’s Wharf
(undisturbed), to 5.6 ± 3.1 µg/g at Chain Valley Bay (proximate to the coal-fired power
station) (Peters et al., 1999b; Kirby et al., 2001a). These concentrations were reportedly 319 times the background concentration (~0.3 µg/g). The widespread selenium
contamination in the southern basin of Lake Macquarie was believed to result from the
atmospheric deposition from power generation activities, or dispersion of dissolved and
sediment transport of selenium from enriched sources such as fly ash, urban runoff, or
sewage (Kirby et al., 2001a).
Selenium concentrations in sediment cores from Lake Macquarie were later reported to be
up to 17.2 µg/g at 30-40 mm depth from Mannering Bay (Peters et al., 1999b) and those
from Wyee Creek near the Vales Point ash dam were reported to be up to two orders of
magnitude higher than previously reported by a Batley (1987) study (Nobbs et al., 1997).
Selenium porewater concentrations in the top 25 mm of sediments from Mannering Bay
(where high sedimentary selenium concentrations were found) ranged from 0.3 to 5.0 µg/L
with a progressive decrease with sediment depth (Peters et al., 1999b).
Selenium concentrations in whole sediments of Port Kembla Harbour have been reported in
few studies including Goodfellow (1996), Hoai (2001) and White (2001). Selenium
concentrations in sediment samples from the Dolphin and at the storm water channel outfall
(see Figure 1.1 for map) were found to be approximately 2 and 4 µg/g (d.w.), respectively
(White, 2001). Goodfellow (1996) and Hoai (2001) reported selenium concentrations of 1057 µg/g (d.w.) in some harbour sediment samples collected from areas near the mouth of
the Darcy Road Drain. The sedimentary selenium concentrations in Port Kembla Harbour
were very high compared to other Australian data, indicating some selenium contamination
issues, which will be investigated more closely in Chapter 5.
Chapter 2 – Literature review
17
2.4.3 Organisms
There have been relatively more studies done on selenium in biological organisms,
compared to water and sediment studies within Australia. Selenium concentrations
(commonly reported as µg/g, wet wt basis) in Australian aquatic organisms are summarized
in Table 2.5. The majority of studies were carried out on the South Eastern coastline and
have investigated whole organism concentrations, individual tissues, and factors affecting
selenium accumulation. The most common organisms studied were commercial fish,
bivalves and mud crabs.
Selenium concentrations in organisms varied between sites, species of organisms, types of
tissues, and other factors such as size, age and feeding behaviour. In general, the organisms
from pristine sites (Maroochy and Pine River, QLD and Jervis Bay, NSW) (Baldwin et al.,
1996; Baldwin and Maher, 1997; Mortimer, 2000) contain lower selenium concentrations
in their muscle tissue than the organisms from the relatively more contaminated sites
(listed). Higher selenium concentrations are generally found in digestive and liver tissue,
compared to muscle tissue. Note that liver tissue may not be a good indicator for a
contaminated surrounding environment as it can accumulate high concentrations of
selenium even at pristine sites such as Jervis Bay, NSW (Baldwin et al., 1996; Baldwin and
Maher, 1997). The tissue distribution of selenium in the bivalve A. trapezia was found to be
in the decreasing order of gill > intestine > adductor muscle > mantle > foot, indicating the
pattern of selenium uptake via food/ direct ingestion of water- or sediment-borne selenium
(Maher et al., 1997). In sea mullet M. cephalus (a benthic feeding fish), the tissue
distribution of selenium was liver>stomach >heart>muscle>kidney (Maher et al., 1997).
More than 70% of the selenium recovered from the bivalve and fish tissues was associated
with proteins, particularly as selenocysteine and selenomethionine (Maher et al., 1997;
Peters et al., 1999a).
No significant correlation was reported between selenium and other elements (e.g.,
mercury, cadmium and arsenic), except the weak correlation between selenium and
mercury (r2 = 0.505) and selenium with cadmium in Black Marlin (M. indica Cuvier) livers
reported by Mackay and coworkers in 1975 (Maher and Batley, 1990; Maher et al., 1992).
Chapter 2 – Literature review
Table 2.5
18
Concentrations of selenium in Australian marine organisms*
Please see print copy for Table 2.5
Chapter 2 – Literature review
19
In sea turtles, partial correlations were believed to exist between selenium and cadmium in
both liver (r = 0.535; p < 0.05) and kidney (r = 0.539; p < 0.05), between selenium and zinc
in both liver (r = 0.621; p < 0.05) and kidney (r = 0.571; p < 0.05). It was commented that
turtle size or age alone could not explain the association between metal concentrations in
those tissues (Gordon et al., 1998).
Many selenium studies in Australian organisms aimed primarily to protect human
consumers from consumption of selenium contaminated food items from marine/aquatic
sources. The selenium results were commonly compared with the Australia New Zealand
Food Authority Maximum Residue Limit (ANZFA MRL) of 1 µg/g-wet wt (ANZFA,
1992). From Table 2.5, edible tissues of fish collected from several locations contained
selenium concentrations that exceeded the ANZFA MRL, notably those collected from
Lake Macquarie, Port Kembla Harbour, Allans Creek, and Ninety Mile Beach.
Selenium concentrations in organisms from Allans Creek and Port Kembla Harbour were
relatively high compared to other data from NSW and other parts of Australia (except in
Lake Macquarie, where fish from Wyee Creek and Vales Point were reported to contain
from 5 to 14 times the ANZFA MRL (Nobbs et al., 1997)). Two fish species (blackfish
Girella tricuspidata and sea mullet Mugil cephalus) from Port Kembla Harbour were found
to contain selenium concentrations (1.40±0.21 and 1.33±0.08 µg/g wet wt, respectively)
that were higher than the ANZFA MRL (Environmental Protection Authority, 1994). Also,
five species of fish (blackfish Girella tricuspidata, bream Acanthopagrus australis, sea
mullet Mugil cephalus, sand mullet Myxus elongatus and whiting Sillago ciliata) and one
species of crab (Scylla serrata) from Allans Creek were found to contain elevated selenium
concentrations (up to 3.7 µg/g wet wt.) in fish muscle tissue with the highest values
recorded in sea mullet and bream (Marine Science & Ecology, 1996). Up to 25 µg/g wet
wt. of selenium was detected in the composite liver tissue sample of sea mullet caught at
approximately 1 km upstream in Allans Creek. Approximately two thirds of the fish and all
six composite samples of crab tissues from Allans Creek (the 1995 study) contained
selenium concentrations that exceeded the ANZFA MRL (Marine Science & Ecology,
1996). From the Allans Creek study in 1995, it was concluded that selenium was a
contaminant of concern in Port Kembla Harbour.
Chapter 2 – Literature review
2.5
20
Biological uptake
Impacts of selenium contamination on biological systems in the aquatic environment may
be classified into two areas: the health of aquatic organisms (e.g., affecting growth,
reproduction and survival), and the public health concern with respect to human
consumption of selenium contaminated food items (Kirby et al., 2001a; Kirby et al.,
2001b). Three major pathways of selenium accumulation by organisms have been reported
including: (1) uptake from solution (water-borne selenium); (2) ingestion of seleniumenriched sediment and suspended particles; and (3) accumulation through diets (John and
Leventhal, 1995; Fan et al., 2002; Barwick and Maher, 2003; Jolley et al., 2004).
2.5.1
Water-borne selenium pathway
Waterborne selenium can be accumulated by organisms, but has been reported to be not
very toxic to fish and wildlife (NIWQP, 1998). When water is the only exposure route,
toxic thresholds for selenium are generally > 1,000 µg/L for adult fish (NIWQP, 1998).
However, the speciation of waterborne selenium can substantially affect the potential for
bioaccumulation in fish and wildlife issues. For example, waterborne selenite (from coal
fly-ash effluent and oil refinery wastewater) is more readily bioaccumulated than
waterborne selenate (from irrigation wastewater) (NIWQP, 1998). Much lower
concentrations of selenium (1-3 µg/L) were reported as thresholds for aquatic ecosystem
toxicity for both selenate and selenite dominated waters (Lemly, 1997a; Lemly, 1999b).
Biological uptakes of water-borne selenium to a level of observable toxic effects also
depended on species of organism. Selenium concentrations in water at which toxicity has
been observed for algae, invertebrates, and vertebrates were 10-80,000; 70-200,000; and
90-82,000 µg/L, respectively (Conde and Alaejos, 1997).
The acute toxicity of four chemical species of selenium to juvenile amphipods (Corophium
sp.) was assessed in water-only test by Hyne et al. (2002). The selenoamino acids, selenoL-methionine and seleno-DL-cystine were found to be more toxic (96-h LC50 values of 1.5
and 12.7 µg Se/L) than the inorganic selenite and selenate (96-h NOEC values of 58 and
116 µg/L). It was also found that life stages were highly sensitive to seleno-L-methionine
Chapter 2 – Literature review
21
spiked sediment. The juveniles were approximately five times more sensitive with a 10-day
LC50 of 1.6 µg/g (d.w.) compared to 7.6 µg/g (d.w.) for the adults (Hyne et al., 2002).
However, poor relationships have been observed between waterborne selenium
concentrations and biological impacts. Waterborne selenium concentrations, which
exceeded the current USEPA chronic criterion of 5 µg/L and often exceeded the acute
criterion of 20 µg/L, were found to have no impacts on biological systems (Canton and
Vanderveer, 1997). It has been argued that selenium uptake mechanism is not directly
through waterborne selenium exposure but as a result of selenium accumulation from
sediment, movement into the food chain, and resulting dietary uptake (Canton and
Vanderveer, 1997).
2.5.2
Particulate and sedimentary selenium
Sediment serves as a sink or reservoir for metals and metalloids and therefore a potential
source of the pollutants to the water column and biological organisms (John and Leventhal,
1995). Sedimentary selenium is known to represent an important link and exposure source
to the benthic-driven food webs with further transfer to higher trophic feeders such as fish
(Peters et al., 1999b; Sappington, 2002). The elevated concentrations of selenium in
sediments of Lake Macquarie were found to correlate with high selenium concentrations in
the fish tissues that exceeded the ANZFA MRL of 1 µg/g (w.w.). A significant relationship
between the mean concentration of selenium in the muscle tissue and sediments was found
for the benthos-feeding fish, mullet (r2 = 0.740, p = 0.05), flathead (r2 = 0.562, p = 0.05)
and bream (r2 = 0.398, p = 0.01) (Peters et al., 1999b). Kirby et al. (2001a) also found a
significant correlation between selenium concentrations in tissues from M. cephalus, a
benthic feeding fish, and the sediment, confirming sediment contamination as a selenium
source for the aquatic food chain, identifying a possible human exposure route.
Particulate or sediment-based criteria have been proposed for selenium as a more reliable
predictor of adverse biological effects, than waterborne selenium criteria (Canton and
Vanderveer, 1997). It has been stated that a preliminary toxic threshold for sedimentary
selenium existed at about 2.5 µg/g (d.w.), and adverse biological effects were always
Chapter 2 – Literature review
22
observed at selenium sediment concentrations greater than 4.0 µg/g (d.w.) (Canton and
Vanderveer, 1997; Van Derveer and Canton, 1997). The risk of selenium toxicity through a
detrital food pathway is argued to continue if water-borne selenium has been depleted but
the underlying sediment is contaminated (Lemly and Smith, 1987). However, this issue has
not been completely resolved to date due to insufficient evidence of a relationship between
sedimentary selenium and chronic toxicity. This has led to important research needs on
sedimentary selenium in relation to food web accumulation being discussed by leading
selenium experts during a peer consultation workshop on the selenium toxicity and
bioaccumulation organized by the USEPA in 1998 and emphasized by Sappington (2002).
2.5.3
Dietary pathway
Selenium bioaccumulation through the diet is usually greater than the direct uptake from
water, particularly when selenium occurs in natural dietary ingredients as compared to
inorganic selenite or selenate (Kennish, 1997; Garcia-Hernandez et al., 2000). Dietary
uptake has been reported as one of the dominant pathways for selenium bioaccumulation by
benthic invertebrates such as bivalves (Schlekat et al., 2000). Lemly and Smith (1987)
reported 2 to 6 times bioconcentration of selenium from producers (algae and plants) to
lower consumers such as invertebrates an forage fish. In the San Francisco Bay-Delta, an
invasion of exotic bivalves (Potamocorbula amurensis) has increased selenium
concentrations in higher trophic organisms (sturgeons and diving ducks) due to the
bivalve’s ability to become enriched in selenium (6-20 µg Se/g d.w.) through filter feeding
(Linville et al., 2002). Concentrations of selenium in some predatory fish from the Bay area
were high although selenium concentrations in water and sediments were low (< 1 µg/L in
water) (Stewart et al., 2004).
Biotransformation of selenium into proteinaceous forms (such as selenomethionine) is
known to be an important factor leading to selenium accumulation in organisms, which can
be transferred to higher trophic organisms (Fan et al., 2002). Formation of organoselenium
compounds and subsequent uptake and transfer to higher fish tissue was reported to be
higher in lentic (slow-flowing) habitats than in lotic (fast-flowing) habitats of a western
Canadian watershed (Orr et al., 2006). A slow rate of selenium excretion from the bodies of
Chapter 2 – Literature review
23
some aquatic species, such as the bivalve Potamocorbula amurensis, was reported to
account for selenium accumulation even in a relatively low selenium contaminated area
(Stewart et al., 2004).
Adverse biological effects at much lower concentrations of environmental selenium are
believed to be predicted more accurately when food web transfer is considered than when
only water-borne selenium is considered (Luoma et al., 1992; Fan et al., 2002). Protective
criteria based on tissue-based selenium and food web transfer have been recommended
(Hamilton, 2002; Hamilton, 2003).
2.6
Selenium toxicity
2.6.1
Aquatic life
Selenium toxicity has been described as an ‘insidious time bomb’ as selenium is
accumulated and stored in the eggs of adult fish, which may survive and appear healthy.
However, the accumulated selenium can be transferred to offspring after hatching and can
affect larvae development and survival (Lemly, 1999a). Fish populations can, therefore,
decline or disappear over the course of several years for no apparent reasons (Lemly,
1999a). Chronic and sublethal toxicity was reported in fish of Belews Lake, North Carolina,
which received selenium contamination from a coal-fired power plant wastewater during
the mid-1970s (Lemly, 1993; Lemly, 1997a; Lemly, 2002). The detailed symptoms
included developmental abnormalities (e.g., swelling of gill lamellae, elevated
lymphocytes), anemia, pathological alterations in liver, kidney, heart and ovary and
possible mortality in young fish. A close relationship was found between selenium
concentrations in eggs, incidence of teratogenic deformities in larvae, and magnitude of
reproductive failure (Lemly, 2002).
Another well-known case of selenium poisoning in fish and birds occurred at Kesterson
Wildlife Refuge in the San Joaquin Valley, California, in the early 1980s (Saiki and Ogle,
1995; Ohlendorf, 2002). It was caused by selenium leaching from seleniferous soils, carried
by irrigated water, which massively collected in the refuge areas, e.g., reservoirs and
Chapter 2 – Literature review
24
wetlands. Several types of fish in the reservoir died off and the remaining ones were found
to contain high concentrations of selenium. In addition, the birds nesting in the area were
observed to experience high levels of mortality and abnormal embryos and chicks. The
birds, and especially their eggs, were found to contain elevated concentrations of selenium
(Ohlendorf, 2002).
Some criteria for assessment of selenium biological effects in the aquatic environment have
been derived in the USA and are given in Table 2.6. Additionally, the toxicity threshold
(dry wt.) for the health and reproductive success of freshwater fish and fish that swim into
rivers from the sea to spawn have been suggested to be 4 µg/g in whole-body, 8 µg/g in
skeletal muscle, 12 µg/g in liver, and 10 µg/g in ovaries and eggs (Lemly, 1997).
Table 2.6
Biological effects of selenium in aquatic environments (NIWQP, 1998).
Please see print copy for Table 2.6
2.6.2
Wildlife and animals
Acute selenium poisoning was reported to occur in animals that grazed on indicator plants
(or plants that accumulate high amount of selenium) with selenium concentrations up to
10,000 µg/g. The symptoms include abnormal movement and posture, anorexia, watery
diarrhea, fever, fatigue, nausea, labour breathing and death due to respiratory failure
(Shamberger, 1983). Chronic toxicity could also occur when animals consume plants with
moderate selenium concentrations (100-10,000 µg/g) for a long period of time. The
condition is known as ‘blind staggers’, with symptoms including stumbling, impaired
Chapter 2 – Literature review
25
vision, loss of appetite, weak legs, paralysis, respiratory failure and eventual death
(Shamberger, 1983). Another common form of chronic selenium poisoning in animals is
the alkaline disease, which results from continuous ingestion of food containing 5- 40 µg/g
of selenium. The symptoms are rough hair coat and hair loss, malformation and sloughing
off hooves, loss of appetite, weight loss, liver cirrhosis, atrophy of heart and anemia
(Shamberger, 1983).
2.6.3
Human
Selenium toxicity in humans has occurred in Southwest China as a result of the use of high
selenium coals. The major symptoms were hair and nail losses, along with various
symptoms of the nervous system (Zheng et al., 1999). The source of the selenium was coal
deposits from which the selenium leached to surrounding agricultural soils. The practice of
liming these soils resulted in the selenium being readily available to plants, which, in turn,
led to high concentrations of selenium accumulating in the plants. Other chronic symptoms
were thickened and brittle nails, ‘garlicky odour’ of breath, sweat and urine, hair and nail
loss, mottled teeth and skin lesions (Opresko, 1993).
Acute selenium toxicity may occur due to occupational exposure of workers in copper
smelters and selenium rectifier plants, e.g., via airborne selenium and SeO2 aerosols, which
hydrate to selenous acid on contact with the skin and mucous membranes. Symptoms of
over-exposure to selenium include immediate irritation to the mucus membranes of the
upper respiratory tract, followed by headache, nausea, fatigue, vomiting, dizziness, a bitter
taste in the mouth and garlic breath; pulmonary edema may also result (Opresko, 1993).
2.6.4
Toxicity mechanism
The toxicity of selenium depends on the valence state, chemical forms and water solubility
of the compound (Opresko, 1993). Selenite and selenate are considered highly toxic as
being the most mobile forms, therefore can be readily assimilated or exposed by organisms.
A study of the toxicity of selenium species present in plants on the insect herbivore
(Spodoptera exigua) found that sodium selenite was the most toxic form with an LC50 of
Chapter 2 – Literature review
26
9.14 µg/g (wet wt), selenocystine being moderately toxic (LC50 of 15.21 µg/g wet wt) and
selenomethionine being the least toxic form (Trumble et al., 1998). However, selenoamino
acids are toxic if present at very high concentrations in the diet. Elemental selenium is
insoluble and not readily available to aquatic organisms, therefore is the least toxic form
(Canton and Vanderveer, 1997). It may be assimilated by organisms via sediment ingestion
(Luoma et al., 1992; Schlekat et al., 2000). In general, the toxicity of different selenium
forms, in decreasing order of magnitude is: hydrogen selenide ≈ selenomethionine (in diet)
> selenite ≈ selenomethionine (in water) > selenate > elemental selenium ≈ metal selenides
≈ methylated selenium compounds (ANZECC/ARMCANZ, 2000).
Selenium toxic effects are believed to be a result of excess selenium analogs of sulfur
containing enzymes and structural proteins (Spallholz and Hoffman, 2002). Selenium has
similar biochemical properties and ionic radii to those of sulfur and is, therefore, able to
substitute for sulfur atoms in biological molecules (Lemly, 1997b). However, there are
some differences in the chemical behaviours of Se and S in vivo. For example, selenium
tends to undergo reduction whereas sulfur tends to undergo oxidation, leading to
differences in their metabolic pathways. The relative acidic strength of H2Se is a much
stronger acid than H2S, so the selenohydryl group of selenocysteine (pKa 5.24) dissociates
at physiological pH while cysteine (pKa 8.25) exists in a protonated form (Shamberger,
1983). Excessive substitution of selenium for sulfur could disrupt the normal functioning of
biological molecules. Excess selenium, as selenocysteine, was reported to inhibit selenium
methylation metabolism, leading to the accumulation of a toxic intermediate metabolite,
hydrogen selenide, in animals, which in turn causes hepatotoxicity and other seleniumrelated adverse effects (Spallholz and Hoffman, 2002). In the case of proteins, S–S bonds
are necessary in order for protein molecules to coil into the tertiary structure that is required
for their proper functioning. Excess selenium alters the chemical bonding, resulting in
improperly formed and dysfunctional proteins or enzymes (Lemly, 1997b).
Another mechanism of selenium toxicity in aquatic birds involves the formation of CH3Se-,
which either enters a redox cycle and generates superoxide and oxidative stress, or forms
free radicals that bind to and inhibit important enzymes and proteins (Spallholz and
Hoffman, 2002).
Chapter 2 – Literature review
2.7
27
Selenium biogeochemical processes
Selenium biogeochemical cycling in aquatic environment involve several interdependent
processes as represented in Figure 2.1.
Please see print copy for Figure 2.1
Figure 2.1
Selenium cycling in aquatic environment (Lemly and Smith, 1987).
Selenium is immobilized in sediment through reduction, adsorption, coprecipitation, and
complexation with sediment components. It can be mobilized and made available for
biological uptake through processes including: oxidation and methylation of inorganic and
organic selenium (by plant root and microorganisms); bio-mixing and associated oxidation
of sediments resulting from burrowing of benthic and vertebrates and feeding activities of
fish and wildlife; and physical perturbation and chemical oxidation associated with water
circulation and mixing (e.g., current, wind, stratification, precipitation and upwelling)
(Lemly and Smith, 1987; Lemly, 1997a).
The following sections summarise further important selenium processes that occur within
water and sediment systems including speciation, sorption and precipitation, coupled redox
processes, and microbial activities.
Chapter 2 – Literature review
2.7.1
28
Speciation
Selenium is known to exist in both inorganic and organic forms and in four oxidation states
(0, -2, +4 and +6): elemental selenium, selenide, selenite and selenate, respectively (Weres
et al., 1989; Masscheleyn and Patrick, 1993; Fox et al., 2003). Changes of selenium
oxidation state and speciation are greatly dependent on redox potential (Eh) and pH
(McNeal and Balistrieri, 1989; Masscheleyn et al., 1990; Peters et al., 1997; Seby et al.,
2001). Figure 2.2 shows a thermodynamic stability diagram of selenium species under a
range of environmental pH and redox conditions. In general, selenium is in the oxidised
selenate and selenite forms under alkaline and high redox potential conditions. Reduced
selenium forms (elemental selenium and selenides) are favoured under acidic and anoxic
conditions.
Please see print copy for Figure 2.2
Figure 2.2
A phase diagram for the Se-H2O system for pH and redox
potential (Eh). Between two solid lines is the stability region
of water (McNeal and Balistrieri, 1989).
Chapter 2 – Literature review
29
In the environment, selenium speciation may also be affected by other physical, chemical
and biological properties and processes such as solubility, coupled redox reaction and
biological interactions (discussed in the following sections) (Shrift, 1964; Fishbein, 1983;
Conde and Alaejos, 1997).
Common selenium species known to be present in the environment are listed in Table 2.7.
Selenium in water is present mainly as inorganic selenite and selenate species (AbdelMoati, 1998; Cutter and Cutter, 2004; Wake et al., 2004). Elemental selenium and some
organoselenium compounds have also been found in solution as soluble forms or adsorbed
onto suspended particles (Pyrzynska, 1998; Cutter and Cutter, 2004; Zhang et al., 2004b).
In complex biological samples, the major selenium species reported were organic selenium
compounds with direct Se-C bonds, including methylated compounds (such as dimethyl
selenide, dimethyldiselenide, and trimethylselenonium), selenoamino acids, selenoproteins
and their derivatives (Maher et al., 1997; Potin-Gautier et al., 1997; Pyrzynska, 1998;
Moreno et al., 2004). In soils and sediment, selenite and selenate have been reported to be
present in oxic sediment (Gao et al., 2000). In anoxic sediment, selenium is present as
reduced elemental selenium or selenides, which can be in the forms of volatile selenium
compounds, organic species or associated with other heavy metals (Masscheleyn et al.,
1991; Velinsky and Cutter, 1991; Dungan et al., 2000).
Table 2.7
Common selenium species found in the environment (Greenwood and
Earnshaw, 1984; Velinsky and Cutter, 1991; Maher et al., 1997).
Please see print copy for Table 2.7
Chapter 2 – Literature review
2.7.2
30
Sorption and precipitation
Selenium may be incorporated into the solid phase of sediment through precipitation,
adsorption onto sediment surfaces, or absorption into minerals or organic matter (Fox et al.,
2003). Precipitation of selenium species is governed by their solubility properties. Metal
selenides and elemental selenium are insoluble in sediment and may precipitate in sediment
under reducing conditions (Masscheleyn et al., 1991; Canton and Vanderveer, 1997).
Selenium salts such as metal-selenates and metal-selenites are highly soluble so are
unlikely to persist in soils/sediments, especially at alkaline pH (Greenwood and Earnshaw,
1984; Elrashidi et al., 1987). The fate of selenate and selenite anions, therefore, can largely
be affected by adsorption and complexation processes, which depend on soil/sediment
characteristics including types of soil/sediment minerals, pH, salinity and ligand exchange.
Selenium anions are known to strongly adsorb onto sediment mineral components such as
iron/manganese oxyhydroxides, aluminium oxyhydroxides, aluminosilicate clays and
organic matter (Balistrieri and Chao, 1990; Dhillon and Dhillon, 1999; Schulthess and Hu,
2001; Blackmore, 2002; Wang and Chen, 2003). Selenium adsorption by iron/ manganese
oxyhydroxides is greater than by aluminium oxyhydroxides and manganese dioxide
(Balistrieri and Chao, 1990). Aluminosilicate clay particles are also reported to strongly
adsorb selenium (Wang and Chen, 2003). Selenite adsorption onto iron oxyhydroxide may
also be affected by organic matter. Tam et al. (1995) reported an increase in selenite
immobilization in the presence of organic matter. However, Masset et al. (2000) found that
organic matter (humic acid) decreased the sorption of selenite onto goethite due to
competition for the adsorption sites from humate ions. This issue needs further
investigation.
Between the two selenium anions, in general, selenite is more strongly adsorbed to
sediment surfaces than selenate. Selenite adsorption onto iron oxyhydroxides occurs via
inner-sphere complexes (Equation 2.3), which are not affected by ionic strength of the
solution (Pezzarossa and Petruzzelli, 2001).
Surface-OH + SeO32- + H+
Î
Surface-SeO32- + H2O…………...(Equation 2.3)
Chapter 2 – Literature review
31
Selenate forms weaker outer-sphere complexes with iron oxyhydroxides in sediment
(Equation 2.4). The complexes are less stable and selenate sorption decreases with an
increase of solution ionic strength (Pezzarossa and Petruzzelli, 2001).
Surface-OH
+ SeO42- + H+
Î
SeO42- + Surface-H2O…………...(Equation 2.4)
Adsorption of selenium anions is highly dependent on pH. In general, high sorption ability
occurs at low pH, especially for sediment minerals such as iron/manganese oxyhydroxides
and clays (Dhillon and Dhillon, 1999; Goh and Lim, 2004). However, selenite sorption to
calcite and hydroxyapatite is low in the acidic region and peaks at pH 8 (Fox et al., 2003) .
Salinity can affect selenium adsorption and complexation behaviour in water. Total
dissolved selenium and organic selenium concentrations were reported to decrease when
salinity increased (Conde and Alaejos, 1997). The decrease in the concentration of
dissolved and suspended particulate selenium and other trace metals in the water column
was believed to result from inorganic complexation between positively charged salt ions
and negatively charged clay particles (that transport trace metals) forming salt complexes
which then precipitate and become associated with sediment (Jolley, 1999; Hoch et al.,
2002). The presence of other anions in solution may also affect the adsorption of selenite
and selenate on the surfaces of clay minerals and oxyhydroxides. Phosphate, hydroxide,
arsenate, molybdate and silicate (with fluoride and sulfate to a lesser extent) were reported
to decrease selenite and selenate adsorption by competing for adsorption sites, thus
increasing the mobilization of selenium (Balistrieri and Chao, 1990; Jackson and Miller,
2000; Blackmore, 2002; Goh and Lim, 2004; Zhang et al., 2005).
2.7.3
Coupled redox processes
As a redox sensitive element, selenium behaviour can be influenced by other redox species
within the sediment system (Rue et al., 1997). Selenium oxyanions can also be electron
acceptors for bacterial degradation of organic matter during diagenesis in sediments
(Oremland, 1994; Lovley, 1995). Figure 2.3 shows a schematic zone of selenium reduction
in comparison to other redox reaction sequences occurring during diagenetic processes.
Chapter 2 – Literature review
32
Please see print copy for Figure 2.3
Figure 2.3
A schematic of zones of organic matter degradation during diagenesis
processes in sediments, adapted from Wakeham (2002).
The coupled redox process is important in determining selenium mobility in sediment.
From Figure 2.3, if oxygen, nitrate (Wright, 1999), iron and manganese oxyhydroxides
were the dominant species in sediments (such as in oxic surface sediment), selenium would
undergo oxidation and be present in mobile selenite or selenate forms. If the sediment is
high in organic matter and sulfate, and is depleted in oxygen and nitrate, selenium would
under reduction and be present as insoluble elemental selenium or selenide forms. The
availability of trace elements in sediment may enhance selenium precipitation after
diagenetic reduction processes as metal selenide species (such as AgSe, Ag2Se, FeSe, and
HgSe) (Mercone et al., 1999; Crusius and Thomson, 2003; Herbel et al., 2003). The
complex redox behaviour of selenium in relation to sediment diagenesis will be
investigated for Port Kembla Harbour in Chapter 6.
Chapter 2 – Literature review
2.7.4
33
Microbial activities
Microbial activities play an important role in selenium biogeochemical cycling in aquatic
systems by either facilitating a loss of selenium from the system via volatilization processes
(Chau et al., 1976; Flury et al., 1997; Dungan et al., 2000) or reducing selenium to
immobilized elemental or selenide forms (Garbisu et al., 1996; Oremland and Stolz, 2000;
Herbel et al., 2003; Siddique et al., 2005). Volatilization of selenium by microorganisms
and aquatic plants is reported to be a significant mechanism for selenium loss from a
wetland system (Masscheleyn and Patrick, 1993; Zhang and Moore, 1997; Hansen et al.,
1998). Selenium is generally converted from oxidised selenite and selenate species into
volatile compounds: dimethyl selenide or dimethyldiselenide (Masscheleyn and Patrick,
1993; Pilon-Smits et al., 1999). Selenium volatilization has been reported as favoured in the
surface soil with high moisture, preferably with the amendment of the protein casein (Flury
et al., 1997; Zhang and FrankenbergerJr., 2000). In Australia, the study of microorganisms’
roles in selenium cycling in Lake Macquarie sediment near a coal-ash dam reported the
ability of four bacterial species, Bacillus mycoides, Shewanella putrefaciens, Cellulomonas
biazotea or Bacillus sp., and Pseuodomonas sp., to produce nonvolatile organocompounds,
and two organisms, Bacillus brevis and 30-8-5-A, were reported to produce methylated,
volatile compounds as a result of the reduction processes (Nobbs et al., 1997).
Certain microorganisms have the ability to reduce selenium anions to insoluble elemental
selenium forms (Zhang et al., 2004a). Six bacterial species from Lake Macquarie sediments
(Bacillus brevis, Bacillus sphaericus, Bacillus mycoides, Shewanella putrefaciens,
Cellulomonas biazotea or Bacillus sp., and Pseuodomonas sp.) were reported to reduce
selenite to elemental selenium (Nobbs et al., 1997; Carroll, 1999). Garbisu et al. (1996)
reported the ability of Pseuodomonas fluorescens and Bacillus subtilis to reduce selenite
and selenate as part of detoxification mechanisms. The elemental selenium was reported to
deposit as granules throughout the Pseuodomonas fluorescens cells, but between the cell
wall and the plasma of Bacillus subtilis (Garbisu et al., 1996). Herbel et al. (2003) observed
reduction of elemental selenium to HSe- species in aqueous media and to FeSe in estuarine
sediment slurries by a selenite-respiring bacterium (Bacillus selenitireducens), indicating
the possible formation of metal selenide in sedimentary systems.
Chapter 2 – Literature review
34
The bacterial reduction of selenium anions to elemental or selenide forms can occur via
dissimilatory and assimilatory mechanisms. In the dissimilatory reduction, selenium
respiring bacteria utilize selenate or selenite as electron acceptors in the mineralisation of
organic matter (Oremland, 1994; Siddique et al., 2005). Bacterial dissimilatory reduction
constitutes a major mechanism for selenium immobilization in anoxic sediments
(Masscheleyn and Patrick, 1993; Stolz and Oremland, 1999). In the assimilatory reduction,
selenium oxyanions are incorporated into bacterial cells while they are reduced to selenides
and subsequently transformed into proteinaceous forms (Frankenberger and Karlson, 1994;
Fan et al., 2002). The reductive assimilation and biotransformation is important for
selenium bioaccumulation and transfer through the food chain, which is a major
biogeochemical pathway in aquatic ecosystems.
2.8
General conclusions
This literature review has provided background information about the metalloid selenium,
its environmental significance and its currently known behaviour in the aquatic
environment. Relatively less selenium research is being done, compared to other trace
elements (such as As, Cd, Cu, Fe, Mn, Ni, Pb and Zn), especially within Australia and more
selenium research is warranted. For Port Kembla Harbour, the limited literature on
selenium in water, sediment and fish has highlighted the selenium contamination issue.
Water selenium concentrations are being monitored as part of the harbour water qualitymonitoring program and are known to be very low (Green, 2003). Therefore, the research
direction for selenium in Port Kembla Harbour is toward selenium in sediment and
biological organisms. The research direction for this thesis is focused on sedimentary
selenium, since this is a potential selenium source for the harbour water and for the
organisms via benthic food chain transfer. The selenium geochemical behaviour in relation
to sediment diagenetic processes is unknown, and this also is a focus on the research
reported in this thesis.
Chapter 3
Evaluation and optimisation of a rapid method for total selenium
determination in marine sediments using microwave digestion and
hydride generation-atomic absorption spectrometry (HG-AAS)
3.1
Introduction
Several analytical techniques are available to measure selenium in environmental samples,
including fluorimetry, voltammetry, radiochemical neutron activation analysis (RNAA),
hydride generation-atomic absorption spectroscopy (HG-AAS) and graphite furnace-atomic
absorption spectroscopy (GF-AAS), inductively coupled plasma - atomic emission (ICPAES) and mass spectrometric (ICP-MS) methods (Nazarenko, 1972; Haygarth et al., 1993;
Olivas et al., 1994; Pyrzynska, 1998). The technique of choice depends on the sample
matrix, sample concentration, and type of information required (e.g., total selenium,
isotopes or speciation). Haygarth et al. (1993) provided a good review and comparison of
widely employed instrumental methods for selenium determination in environmental
samples. Some common techniques are compared in Table 3.1.
Table 3.1
Comparison of techniques for quantitative analysis of selenium in
environmental samples (Shamberger, 1983; Haygarth et al., 1993; Borella et
al., 1998).
Please see print copy for Table 3.1
Chapter 3 – Total selenium determination
36
Hydride generation - atomic absorption spectrometry (HG-AAS) is a traditional but still
widely used technique for selenium analysis in environmental samples. The technique
offers good accuracy and is reliable, rapid, and relatively inexpensive. During HG-AAS
analysis, selenite (the only reactive selenium species) in an aqueous sample reacts with a
reducing agent, sodium borohydride (NaBH4), in the presence of hydrochloric acid to
generate gaseous selenium hydride (H2Se) (Equation 3.1). The H2Se is stripped by N2 in a
gas-liquid separator, passing through a drying tube into a quartz tube furnace mounted in
the light-path of an AAS running a selenium hollow cathode lamp. The H2Se is thermally
decomposed into selenium atoms, which absorb light at 196.0 nm.
4H2SeO3 + 3BH4- + 3H+
→
3H3BO3 + 3H2O +
4H2Se ……(Equation 3.1)
Selenium analysis by HG-AAS requires digestion of solid samples to release selenium into
a solution. Traditionally, wet acid digestion has been used which involves digestion/heating
in open vessels with strong acids to destroy the organic matter and dissolve the metal
analytes. However, this technique has several disadvantages in Se analysis due to possible
loss of Se volatile compounds formed during digestion at high temperature. In addition, the
method is time-consuming and requires continuous attention from operators. Samples could
be potentially exposed to contamination from reagents plus the environment (multiple
contamination if several reagents are used (Wang et al., 2001)), causing significant errors in
the final determination. These procedures often involve the use of perchloric (HClO4) and
hydrofluoric acids (HF), which are potentially extremely hazardous.
Microwave heating is currently one of the most widely employed methods for sample
digestion in analytical laboratories (Agazzi and Pirola, 2000). Pressurized acid digestion in
closed vessels with microwave heating speeds up the dissolution of various solid samples
and (sediment) sample dissolution can often be achieved with the use of nitric acid alone
(Wang et al., 2001). Closed vessel digestion effectively prevents sample contamination
from the environment and less contamination from reagents if only nitric acid is used
(Wang et al., 2001). Advantages of microwave digestion over traditional techniques include
the ability to strictly control heating power and the length of time that the heat is applied,
and all the processes can be automated with real time graphics and data acquisition of
Chapter 3 – Total selenium determination
37
temperature and pressure parameters (Ducros et al., 1994). Other advantages include: a
shorter acid digestion time; no losses of volatile elements; lower contamination levels;
minimal volumes of reagents; more reproducible procedures; and a safer working
environment (Agazzi and Pirola, 2000). In addition, microwaves only heat the liquid phase
because the vapours do not absorb microwave energy.
As a component of the selenium studies in Port Kembla Harbour, this initial work aimed to
evaluate and optimise a rapid method for total selenium determination in sediment samples,
combining a microwave digestion technique with selenium detection and quantification by
hydride generation-atomic absorption spectrometry (HG-AAS). HG-AAS was the
technique of choice employed in this work due to the equipment availability in our
laboratory. The method aimed to preclude any use of hazardous acids, especially HF and
HClO4, despite their strong power in decomposing silicate materials and releasing selenium
and other metals into solution (Zhou et al., 1997; Radojevic and Bashkin, 1999; van Staden
et al., 2000). Hydrogen peroxide (H2O2), in combination with nitric and sulfuric acids, has
been reported to successfully digest selenium in soil samples (Kos et al., 1998). However,
in our laboratory, H2O2 was previously found to interfere with the reduction of selenite by
borohydride during HG-AAS analysis of selenium in biological samples (Jolley,
unpublished results). Sulfuric acid (H2SO4) was also not chosen for this work as it naturally
contains traces of selenium impurities.
This study assessed the selenium digestion efficiencies of three digestion procedures: (a)
Kirby et al. (2001a); (b) the USEPA Standard Method 3051 (1994); and (c) Zhou et al.
(1997), detailed below in Section 3.2.2. These procedures employed mainly nitric acid
(HNO3) and hydrochloric acid (HCl) as digestion matrices. Other specific issues
investigated and encountered in this work were the optimisation of selenate reduction to
selenite for the required HG-AAS detectability, the elimination of nitrogen oxide
interferences, and foaming problems in samples with high organic content. Test sediments
used in the method development process were certified reference materials (NIST Estuarine
Sediment 1646a, NIST Marine Sediment SRM 2702, NRCC Estuarine Sediment MESS-3,
and BCR Estuarine Sediment CRM 277), an in-house reference material (PKH-1) with high
selenium concentration, and other marine sediments collected from Port Kembla Harbour.
Chapter 3 – Total selenium determination
3.2
Materials and methods
3.2.1
Reagents and glassware
38
All glassware and plastic containers were cleaned by soaking in 10% (v/v) HCl (UNIVAR,
32%) for at least 24 hours, followed by rinsing with MilliQ water (Millipore Australia) and
dried in the laboratory under ambient conditions in the inverted position to prevent
contamination. Small vials and syringes were dried in an oven at 50˚C.
Chemicals and reagents were of analytical reagent grade or better. Selenite stock standard
(1000 mg/L) was either from a commercial AAS standard stock solution (AAS
SPECTROSOL, Crown Scientific, Cat. No. 2594) or prepared in the laboratory by
dissolving of selenous acid (Sigma Aldrich) (0.8167 g) in 500 mL MilliQ water, acidified
with 5 mL conc HCl. Intermediate standard solutions (1 mg/L and 10 mg/L) were prepared
fresh or weekly and calibration standard solutions (0 – 50 µg/L) were prepared fresh in 4
mol/L HCl. Selenate stock standard solutions (1000 mg/L) were prepared from sodium
selenate anhydrous (Na2SeO4) (SIGMA® Sigma Ultra, Cat. No. S-8295) by dissolving
0.5982 g in MilliQ water containing 2.5 mL conc HCl and diluting to 250 mL. Selenium
standard stock solutions were stored below 4 ˚C.
Sodium borohydride reagent (0.3% NaBH4 w/v) was prepared fresh by adding 0.5 g NaOH
and 0.6 g NaBH4 (ALDRICH® VenPure® AF granules, 98+%, ALDRICH Chemical
Company Inc. USA, Cat No. 452173) in 200 mL MilliQ water. Concentrated nitric acid
(70%, UNIVAR) and concentrated hydrochloric acid (32 %, UNIVAR) were purchased
from Crown Scientific Australia. Urea (20% w/v) was prepared by dissolving urea
(ALDRICH® U2709, 99+%) (20 g) in 100 mL MilliQ water.
3.2.2
Microwave digestion procedures
Sample digestion was carried out in a microwave oven (MILESTONE ETHOS SEL),
equipped with a carousel-rotator containing 10 x 80 mL-Teflon vessels. The program
control was operated on MLS easyWAVE 3.2 software (MLS GmbH Germany).
Chapter 3 – Total selenium determination
39
The digestion procedure (a) was carried out using in-house rotators (Figure 3.1) made to fit
the same microwave oven.
Please see print copy for Figure 3.1
Figure 3.1
In-house microwave rotators for multi-sample digestion,
fitted general 50-mL centrifuge tubes.
Three strong acid digestion methods evaluated in this work were:
(a)
Kirby et al. (2001a) procedure: samples were digested at 90˚C in 50-mL
polypropylene centrifuge tubes. Dry sediment sample (0.3-0.5 g) was digested in 5 mL
conc. nitric acid (70%, UNIVAR) in a microwave at 90 °C for 20 min. For high organic
samples, the acid-sample mixture was allowed to digest at room conditions for 30 min
before heating to reduce pressure from organic reaction in the centrifuge tubes. The sample
was filtered through an acid-resistant filter paper (Whatman No. 541) and diluted to 50 mL
in a volumetric flask. The digest solutions were stored in a refrigerator below 4˚C until
further analysis.
(b)
The USEPA Method 3051 (1994): the method employed the regular 80 mL-Teflon
vessels for digestion. Dry sediment sample (0.25-0.3 g) was digested in 10 mL conc. HNO3
acid (70%, UNIVAR) (solid sample on the vessel internal surface rinsed down into the
solution when adding acid), in the microwave oven at 180 °C for 10 min (Program: 10 min
Chapter 3 – Total selenium determination
40
ramp to 180 °C, maintaining at 180 °C for 10 min). The vessels were allowed to vent and
cool for 60 min to below 40 °C before opening. The sample was transferred into a 100 mLvolumetric flask by filtering through Whatman No. 541 filter paper, diluted to the mark
with MilliQ water and stored below 4˚C. Dilution to 100 mL was required to dilute the
HNO3 to a concentration that would not interfere with the HG-AAS analysis.
(c)
Zhou et al. (1997) procedure: this procedure was also carried out using the regular
80 mL-Teflon vessels. Dry sediment samples (0.25-0.3 g, up to 1 g for samples containing
very low selenium concentrations) were accurately weighed into a Teflon vessel. 2.5 mL of
conc. HNO3 (70%, UNIVAR) and 7.5 mL of conc. HCl (32%, UNIVAR) was added
(rinsing down any sediment on the internal surface of the vessel and swirling to suspend
dry sediment into the solution). The samples were digested at 200˚C for 20 min (10 min to
200˚C, maintaining at 200 °C for 20 min). The vessels were allowed to cool for 60 min to
below 40 °C before opening. The digested samples were filtered, diluted to 25 mL for low
selenium samples and 50 mL for high selenium concentration samples and stored in a
refrigerator until further analysis. The final digest solutions contained 5%-10% HNO3 and
15%-30% HCl.
3.2.3
Sample pretreatment for HG-AAS analysis
Digested samples were pretreated to convert all selenium species to selenite to generate a
response in the HG-AAS analysis. Nitric acid digests (5 mL-aliquot) were treated by
acidifying with 5 mL of 10 mol/L HCl, heating at 90 ˚C for 20 min and allowing to cool.
Then, 250 µL of 20% urea was added to the mixture, mixed and allowed to stand for 10-20
min to degas N2 (see also Section 3.3.3). The sample was diluted to a final volume of 12.5
mL before HG-AAS analysis. The heating step (i.e., selenate reduction step) can be done in
a microwave at 90˚C for 20 min (5 min ramping period plus 15 min at 90˚C).
Aqua regia digests were treated by transferring 5 mL aliquot into a 15-mL polypropylene
centrifuge tube (this sample contained 0.25-0.5 mL HNO3 and 0.75-1.5 mL HCl, for the 50
mL and 25 mL previous digest dilution, respectively). 250 µL of 20 %(w/v) urea was added
and the mixture was allowed to degas N2 as mentioned above. HCl and MilliQ water were
Chapter 3 – Total selenium determination
41
then added to give a final volume of 12.5 mL containing 40% HCl (v/v). For high selenium
concentration samples, a lower volume of the digest was treated with an equivalent volume
of urea and HCl.
3.2.4 Selenium determination by HG-AAS
The pretreated samples were analysed using the atomic absorption spectrometer (Varian
SpectrAA220, equipped with VGA-76 vapour generation unit, Varian Australia Pty Ltd.).
The selenium analysis diagram using continuous flow sample introduction by hydride
generation is shown in Figure 3.2. The HG-AAS operating conditions were based on the
manufacturer recommended procedure (Elrick and Horowitz, 1986), as shown in Table 3.2.
Sodium borohydride (NaBH4) concentration is important and affects the sensitivity of
selenium hydride-formation (Welz and Schubert-Jacobs, 1991). The maximum AAS
selenium signal has been reported with NaBH4 concentrations of 0.2-1% and 0.3-0.65% by
Welz and Schubert-Jacobs (1991) and Schloske et al. (2002), respectively. In this work, the
concentration of NaBH4 reductant was decreased from the recommended 0.6 to 0.3 % to
reduce the reaction of the borohydride with other elements that could cause interference
with the analysis (Vanclay, 2003). The concentration of NaOH in the borohydride reagent
was only sufficient to stabilize the borohydride ions and needed to be low enough to be
neutralized by the sample acidity, because acidic conditions are required for the hydride
formation.
The carrier HCl concentration was 10 mol/L and the sample solution contained 40% v/v
HCl (~4 mol/L). HCl concentrations between 0.5 and 5.0 mol/L have been reported to have
no effect on the selenium signal, neither in peak absorbance nor in peak area (precision)
(Welz and Schubert-Jacobs, 1991). A high (4 mol/L) concentration of HCl was used in this
work to stabilize selenite species and to minimize interferences from transition metals (Kos
et al., 1998; Vanclay, 2003). Other parameters were set as per manufacturer
recommendations.
42
Table 3.2
HG-AAS operating conditions used in this study.
Parameters
Wavelength
Slit width
Background correction
Lamp current
Fuel
Delay time
Replicates/ measurement time
Reductant (% NaBH4: % NaOH)
Acid concentration
Acid-channel flow rate (mL/min)
NaBH4 Channel flow rate (mL/min)
Sample Channel flow rate (mL/min)
Carrier gas
Carrier gas pressure
Operating conditions
196.0
1.0 nm
Deuterium, On
10 mA
Air-acetylene
40 s
3 replicates /5 seconds
0.3: 0.25
10 M
1
1
7.5-8.0
Nitrogen
> 200 kPa
Please see print copy for Figure 3.2
Figure 3.2
Selenium analysis by HG-AAS using Varian
VGA-76 vapour generator (Voth-Beach and
Shrader, 1985).
43
3.3
Results and discussion
3.3.1
Evaluation of microwave digestion methods
Three digestion procedures: (a) Kirby et al. (2001a); (b) the USEPA Method 3051 (1994);
and (c) Zhou et al. (1997), were evaluated for digestion of dry sediment samples and
subsequent selenium determination by HG-AAS.
The Kirby (2001a) method (a) using nitric acid digestion was the most rapid and
convenient. Samples were digested in 50-mL centrifuge tubes. No pressurized vessels were
required and as many as 39 samples could be digested at the same time on the in-house
microwave rotators which were specifically made for this purpose. This digestion method
was reported as successful in the literature but in this study, gave low Se recoveries (Table
3.3). An increase in digestion time from 20 min to 40 min (data not shown) provided a
better recovery, with an average of 74 % for MESS-3 (c.f. 62 % from a 20-min digestion).
A 60-min digestion was tried but more incidences of sample explosion through
polypropylene caps were observed and led to an unsafe procedure and unreliable results.
The use of Teflon tubes or a better heating regulation of the temperature probe might be
helpful, but the method was not optimized further in this study.
Table 3.3
Acid extractable selenium from reference materials and test sediments using
microwave digestion at 90 ˚C for 20 minutes (Kirby et al., 2001a).
Please see print copy for Table 3.3
Chapter 3 – Total selenium determination
44
Nitric acid digestion of sediment based on the USEPA Method 3051 (b) used 0.25 g dry
sediment, 10 mL conc. nitric acid, 10 min ramp to 180 ˚C and hold at 180 ˚C for 10 min
(U.S. Environmental Protection Agency, 1994). In conjunction with HG-AAS detection,
this method provided satisfactory selenium recoveries for marine sediment reference
materials (Table 3.4). Nitric acid, as nitrogen oxide species (NOx) in digested samples, was
initially found to interfere with HG-AAS. This was overcome by addition of urea as
discussed in Section 3.3.3. However, samples still required a considerable dilution to
reduce the interfering effect of nitric acid, leading to higher detection limits.
Aqua regia (3HCl: 1HNO3) digestion, modified from the method described in Zhou et al.
(1997) (c), in conjunction with HG-AAS analysis provided satisfactory recoveries of Se
from marine sediment reference materials (Table 3.4). Aqua regia is an effective acid
composition for sediment digestion for selenium analysis as numerous sulfides (such as,
those of As, Se, Te, Bi, Fe, Mo), arsenides, selenides, telurides, sulfosalts, and native Au,
Pt, and Pd and oxyhydroxides minerals (e.g., Fe-Mn) are effectively decomposed by hot
aqua regia (Hall, 1997b).
Table 3.4
Recoveries of acid extractable selenium from reference materials using two
digestion procedures: USEPA Method 3051 and Zhou et al. (1997).
Percentage recoveries are numbers in brackets.
Please see print copy for Table 3.4
Chapter 3 – Total selenium determination
45
The mixture of 3 parts HCl to 1 part HNO3 has a strong oxidizing power due to the
formation of nascent chlorine and nitrosyl chloride, and thus the organic component of
sediment is efficiently wet ashed (Equations 3.2 and 3.3, Hall, 1997b):
HNO3 + 3HCl
Æ
NOCl + 2H2O + 2(Cl)
……...(Equation 3.2)
NOCl
Æ
NO + (Cl)
……...(Equation 3.3)
The use of aqua regia was considered to be unsuitable in sample digestion for selenium
analysis when open vessel digestion is used, due to formation of volatile selenium
compounds (such as SeCl4) or selenium oxochlorides (such as SeOCl2) (Nazarenko, 1972)
which can be lost and lead to underestimation of selenium concentrations. However, this
loss was eliminated for this work, as digestion was carried out in closed microwave vessels.
Both aqua regia and nitric acid provided comparable selenium recoveries in the reference
materials tested. Aqua regia was chosen for digestion of sediment samples in further work,
as all selenium is in the selenite form (Hall, 1997b) after digestion by aqua regia and hence
no pre-reduction step was required to convert selenate to selenite as was needed when using
HNO3 alone. In addition, lower detection limits could be achieved, as less sample dilution
was required to dilute nitric acid effects.
3.3.2
Reduction of selenate to selenite
Measurement of selenium by HG-AAS requires all selenium to be present as selenite
(Borella et al., 1998). Selenium in strong sediment digests, such as nitric acid, may be
present in other oxidation states, such as selenate. Therefore, conversion of other forms of
selenium into selenite is necessary. The common procedure involves reduction of selenate
into selenite by treating the sample solution with 4-6 mol/L HCl at 80-100˚C for 10-50
minutes (Zhou et al., 1997; Apte et al., 1998; Zhang et al., 1999a; Zhang et al., 1999b;
Schloske et al., 2002) . Effective and reproducible reduction of selenate to selenite using
microwave heating in similar acid media has also been reported (Brunori et al., 1998; Li et
al., 1998; Olivas and Donard, 1998).
Chapter 3 – Total selenium determination
46
Concentrated HCl solutions are needed as the Cl-ion is an important component of the
reducing solution (Brimmer et al., 1987) (Equation 3.4).
SeO42- +
Æ
2HCl
SeO32- + H2O + Cl2
………..Equation 3.4
If reducing selenium species are expected to be present in the sample, oxidation of all
selenium into selenate is required before the reduction of selenate into selenite prior to HGAAS analysis.
In this work, selenium species in the nitric acid digested samples, which should be in
oxidised selenate form, were converted into selenite by heating sample solutions at 90 ˚C
for 10-20 minutes in 5-6 mol/L HCl, using either a hot water bath or a microwave. The
efficiency of the reduction step was assessed by spiking selenate standard (in triplicates)
and the recovery results are shown in Table 3.5. For comparison, a minimum 4.4 mol/L
HCl medium was found to give good recoveries at selenate concentrations up to 20 µg/L
but recoveries decreased slightly at higher selenate concentrations. The selenate recoveries
after reduction to selenite in both standard matrices and sediment digests containing up to
20 % nitric acid were between 98-106 % with the relative standard deviation (RSD) of
below 3%.
Table 3.5
Efficiency of selenate reduction to selenite using HCl (microwave heating at
90 ˚C for 10 min, mean ± SE, n=3).
Selenate added
(µg/L)
% Recovery
4.4 M
Standard matrix
Standard matrix
Sediment digest
HCl
10 % nitric
20 % nitric
20% nitric
4
96 ± 2
104 ± 0.1
102 ± 3
98 ± 3
8
-
105 ± 2
100 ± 2
102 ± 3
12
105 ± 2
104 ± 0.7
103 ± 3
103 ± 2
20
105 ± 1
102 ± 2
101 ± 1
102 ± 0.3
28
97 ± 6
103 ± 0.7
101 ± 0.5
103 ± 1
36
92 ± 7
106 ± 0.9
101 ± 0.5
-
Chapter 3 – Total selenium determination
Table 3.6
47
Comparison of selenite recoveries (mean ± SE) in aqua regia digestion with
and without the selenate reduction step.
Selenite added before digestion
% Recovery
(µg/L)
No reduction step
With extra reduction step
16 (n=5)
96 ± 1
101 ± 2
32 (n=4)
98 ± 3
99 ± 2
The digestion of sediments using aqua regia (as discussed above in Section 3.3.1) yielded
all selenium species as selenite due to the high content of HCl in the digesting media and
sufficient Cl- to stabilize selenite (Brimmer et al., 1987; Hall, 1997b). This was confirmed
by measuring selenite in the digested solution with and without prior heating. The
differences between selenium recoveries in the two steps were considered negligible (Table
3.6). Therefore, for total Se determination in sediment samples after aqua regia digestion,
no reduction step was performed.
3.3.3
Elimination of nitrogen oxide interferences
In HG-AAS analysis, it was found that a trace amount of nitric acid in the sample or from
the pre-nitric acid-cleaned containers caused severe suppression to the AAS signal.
Observable symptoms included low and unstable liquid levels in the gas liquid separator,
non-reproducible absorbance, complete absence of signals and a lasting memory effect.
Nitrogen oxide interferences occurred as a result of HNO3 used in the sediment digestion,
forming observable brown NOx fumes. The problem was severe for a closed-vessel
microwave digestion in this study, as NOx were prevented from escaping from the sample
matrix, unlike in the open-vessel procedure. A similar problem was encountered by Li et al.
(1998) and Schloske et al. (2002). Nitrogen oxide interferences were believed to cause
signal suppression due to oxidative potential against H2Se (Voth-Beach and Shrader, 1985;
Schloske et al., 2002). The interferences have been successfully minimized by treating the
sample with amidosulfuric acid or sulfanilamide (Schloske et al., 2002) and addition of
urea (Li et al., 1998). In this work, the addition of urea was attempted and found to be
successful as shown in Table 3.7.
Chapter 3 – Total selenium determination
Table 3.7
48
Recoveries of selenite spikes in nitrogen oxides-containing samples with
urea addition. Sample volume: 25 mL, containing 1 mL heated nitric acid.
Urea concentration
% Recovery of spiked selenite
(% w/w)
16 µg/L
32 µg/L
0.04 (400 mg/L)
99
96
0.08 (800 mg/L)
95
101
0.16 (1600 mg/L)
98
102
0.24 (2400 mg/L)
95
101
The interference from 1 mL of conc. nitric acid in the digest sample can be eliminated by
adding 50 µL of 20 % urea solution. An excess of urea was found to have no affect on the
AAS response. The urea addition was found to be only effective with heated nitric acid or
nitrogen oxide species. Bubbles of gas (N2) were formed and visible during the urea
treatment. A possible reaction is in Equation 3.5.
H+
2NO2-
+
NH2CONH2
→
2N2
+
HCO3- +
2H2O
……… (Equation 3.5)
The resultant N2 from the urea reaction with NOx was observed to give a slightly high
background signal in selenium analysis by HG-AAS. The blank signal (and hence the
detection limit) was lower when the N2 gas resulting from urea addition was removed from
the solution by degassing (allowing to stand with occasional shaking for approximately 10
min before final sample dilution for the AAS analysis).
3.3.4
Analytical performance
The analytical performance of the HG-AAS technique for the determination of total
selenium after strong acid digestion is summarized in Table 3.8. The optimum calibration
range (curve fit) was between 0–50 µg/L. A typical HG-AAS calibration curve for
selenium is shown in Figure 3.3. The linear correlation was always greater than 0.99, but a
new rational curve fit setting in the Varian software provided better results. The instrument
detection limit determined from the mean of 10 blank measurements plus three times the
standard deviation was 0.2 µg/L. The method blank was lower when the N2 gas resulting
49
Absorbance ( 196. 0 nm)
0.6
Table 3.8
y = 0.0103x + 0.0228
R2 = 0.9914
0.5
Analytical performance for analysis of total
selenium in sediment extracts by HG-AAS.
Performance
0.4
Calibration range, µg/L
0.3
Correlation coefficient, r2
0.2
0.1
0
10
20
30
40
50
60
Se concentration (µg/L)
Figure 3.3
Typical HG-AAS calibration curve for selenium
as selenite in 4 mol/L HCl.
0-50
> 0.99
Instrument detection limit, µg/L
0.2
Method detection limit*, µg/g
0.01
% RSD (> 5 µg/L)
0
Value
*1 g sample digested.
<5%
Chapter 3 – Total selenium determination
50
from urea addition was completely removed from the solution by degassing. The method
detection limit from a sample size of 1 g dry sediment digest was 0.01 µg Se/g.
The reproducibility of the instrument was excellent with RSD well below 5% for selenium
concentrations greater than 5 µg/L. The method reproducibility from sample digestion,
reduction step and HG-AAS analysis was generally within 10%. Deviation of the results of
replicate samples mainly arose from the reduction step. Microwave reduction provided a
better RSD (<<5%) than hot plate heating. Satisfactory reproducibility for hot plate heating
procedures was also obtained with increasing experience and laboratory skill of the
operator.
0.5
Absorbance (196.0 nm)
y = 0.0097x + 0.1224
R2 = 0.9997
0.4
0.3
0.2
0.1
0
0
5
10
15
20
25
30
Selenite added (µg/L)
Figure 3.4
HG-AAS response of selenite standard added to 10 % nitric digested
samples (containing 40% HCl, 4% HNO3 and 0.16% urea), measured against
40% HCl calibration standards. % RSD of triplicate samples ranged from
1.3-2.2%.
Final treated samples for HG-AAS analysis, which contained 5% v/v HNO3, 40% v/v conc.
HCl (~4 mol/L) and 0.08-0.24% urea, were analysed against the acidified standards (0-50
µg/L in 40% HCl). The recoveries of selenite standard in this matrix against the acidified
standards were > 96 %. Matrix matched standards were not required. Standard addition
tests provided a good linear response as shown in Figure 3.4.
Chapter 3 – Total selenium determination
51
Samples containing high organic matter (such as those encountered in field sample studies
in Chapter 5) were found to foam during hydride generation, which caused the total AAS
signal suppression and destabilized the system. This was overcome by adding a trace
amount of Antifoam B (Sigma-Aldrich, A-5757) (25 µL antifoam solution per 10 mL
sample solution) (Chipeta C., 2003, Port Kembla Copper Ltd, per comm.). Analysis of
samples containing antifoam solution required matrix-matched standards. Also, it was
found that approximately 10% signal suppression occurred over time during a long analysis
period. Rinsing between samples (~30s) and frequent recalibration was found to provide
good quality control of the analysis.
3.4
Conclusions
Evaluation and optimisation of a rapid method for total selenium determination in sediment
samples using a microwave assisted digestion and hydride generation-atomic absorption
spectrometry found strong aqua regia (3HCl: 1HNO3) digestion to be the best method. It
provided good selenium extractability and no extra selenate reduction step was required for
the subsequent HG-AAS analysis. A nitrogen oxide interference was overcome by addition
of urea and a foaming problem found with high organic content samples was overcome by
addition of an antifoam solution. The method detection limit of 0.01 µg/g dry sediment was
achieved with > 95 % instrumental and > 90 % method confidence (precision and accuracy)
levels.
Chapter 4
Selenium speciation in marine sediment extracts using high
performance liquid chromatography and hydride generation-atomic
absorption spectrometry
4.1
Introduction
Selenium toxicity and bioavailability are governed by its chemical forms (Opresko, 1993;
Hyne et al., 2002; Doyle et al., 2003). Information on selenium species will assist in
environmental risk assessments of selenium-contaminated systems. The information is also
helpful in understanding the environmental transformation of selenium (Szpunar and
Lobinski, 1999). For these reasons, an effective method for the determination of selenium
species in sediments is required for accurate quantification of individual selenium species
in Port Kembla Harbour sediments.
The literature on selenium speciation in soils and sediments revealed a common and
traditional HG-AAS speciation method that was used by many researchers as represented in
Table 4.1 (first three rows of the ‘pretreatment’ column). The traditional procedures
involve a selective measurement of three selenium fractions after appropriate chemical
treatments: (1) selenite fraction, which is measured directly after acidification with 4 mol/L
HCl (no heating); (2) selenate fraction, measured after a reduction step by heating with 4-6
mol/L HCl; and (3) total selenium fraction, measured after oxidative digestion followed by
the reduction step. The organic plus elemental selenium (Se (-II, 0)) fraction is obtained by
the difference between the total selenium and the sum of selenite and selenate fractions
(Seby et al., 1997; Zhang et al., 1999a; Zhang et al., 1999b; Bujdos et al., 2000). Initial
assessment of the traditional HG-AAS method to measure selenite, selenate and elemental
and organic fractions in this study found up to 20 % of an organic selenium compound
(selenocystamine) to be oxidised during a typical selenate reduction step (6 mol/L HCl at
90 ˚C for 20 min), which can cause an overestimation to the selenate fraction. A similar
finding has also been reported elsewhere (Martens and Suarez, 1997). The traditional HGAAS method also contains limitations that the selenium species measured were a result of
chemical interferences rather than a direct measure of individual selenium compounds.
53
Table 4.1
Selenium speciation in soil/sediments by the HG-AAS traditional method and modern hyphenated techniques.
Please see print copy for Table 4.1
Chapter 4 – Selenium speciation
54
There has been an analytical trend for selenium speciation in soil/sediments toward the area
of hyphenated techniques, which couple chromatographic separation procedures to an
element-specific detector such as AAS, AFS or ICP-MS (Guerin et al., 1999; Uden, 2002;
Capelo et al., 2006). The hyphenated techniques allow direct quantification of individual
selenium species and, with a sensitive detector, are capable of achieving the low detection
limits necessary for speciation analysis at environmentally relevant concentrations (Jackson
and Miller, 1998; Szpunar and Lobinski, 1999; Uden, 2002). Modern hyphenated
techniques (such as HPLC-ICP-MS) have been widely adopted for selenium speciation in
biological samples, such as yeasts (Casiot et al., 1999; Kotrebai et al., 1999), cooked cod
(Crews et al., 1996), human urine (Cao et al., 2001), garlic (Kotrebai et al., 1999) and
selenium-containing proteins in human and mouse plasma (Koyama et al., 1999). There has
been increasing application of hyphenated techniques for water samples but research on the
more complex matrix of sediments has been limited. Table 4.2 summarises hyphenated
methods for selenium speciation in water samples, which can be applied to sediment
extracts.
CH3
Selenite
Selenate
pKa1 = 2.46
pKa2 = 7.31
pKa2 = 1.92
Selenocystine
Figure 4.1
Selenomethionine
pKa1 = 2.19
pKa2 = 9.05
pKa1 = 1.68
pKa2 = 2.15
pKa3 = 8.07
pKa4 = 8.94
Structures and pKa values of four selenium compounds studied.
This chapter reports the development and optimisation of a selenium speciation method for
selected inorganic and organic selenium compounds: selenite, selenate, selenomethionine
and selenocystine (Figure 4.1) in marine sediments based on HPLC separation and HGAAS detection.
55
Table 4.2
Selected hyphenated methods in the recent literature for selenium speciation in water samples.
Please see print copy for Table 4.2
Chapter 4 – Selenium speciation
56
The specific aims of the study were to:
•
investigate the appropriate sediment extraction procedure (reagent type,
concentration and extraction time), with the purposes of (a) obtaining the best
possible selenium recovery, (b) to preserve original selenium species, and (c)
the extractant reagent being compatible with the HG-AAS detection;
•
optimise the HPLC separation of four standard selenium compounds; and
•
evaluate the application of the optimised speciation method on Port Kembla
Harbour sediment samples.
Both oxic and anoxic sediment materials were studied and compared.
4.2
Materials and methods
4.2.1 Reagents and apparatus
All glassware and plastic containers were acid-cleaned before use. Chemicals and reagents
were of analytical reagent grade or better. Selenite and selenate standard solutions were
prepared according to the procedure described previously in Section 3.2.1.
Organoselenium compounds were purchased from Sigma-Aldrich: seleno-DL-cystine
(SIGMA®, S1650), selenocystamine dihydrochloride (SIGMA®, S0520), seleno-DLmethionine (SIGMA®, S3875). Stock standard solutions (100 mg Se/L) were prepared by
dissolving the entire contents of a 25-mg bottle in the appropriate volume of MilliQ water
in a fume hood (for example, 25 mg of seleno-DL-methionine contained 10.07 mg Se,
requiring 100.7 mL of MilliQ water to make 100 mg Se/L stock standard). This approach
was found to be effective and convenient, eliminated the inaccuracy of weighing small
amounts of the compounds, and prevented personal exposure to the toxic compounds
during weighing. The organic selenium stock solutions were standardized against the
commercial selenite standard for best accuracy, and were covered with aluminium foil to
prevent photodegradation while stored at below 4˚C.
Chapter 4 – Selenium speciation
57
Ascorbic acid (0.5 mol/L) was prepared by dissolving 44.0325 g of L-Ascorbic acid (BDH
AnalaR®) in less than 500 mL of slightly warm MilliQ water, stirring until all the solid had
dissolved and allowing it to cool to room temperature. The solution was transferred to a
500-mL volumetric flask and made up to volume with MilliQ water. HCl (0.5 mol/L) was
prepared by dispensing 50 mL conc. HCl (32%, UNIVAR) into a 1-L volumetric flask and
making up to volume with MilliQ water. H3PO4 (0.5 mol/L) was prepared by transferring
33 mL conc. H3PO4 (85%, UNIVAR) into a 1-L volumetric flask and diluting to the mark
with MilliQ water. H3PO4 (0.5 mol/L): CH3OH (1:1 v/v) was prepared by adding 33 mL
conc. H3PO4 (85%, UNIVAR) into a 1-L volumetric flask and diluting to the mark with 1:1
methanol: MilliQ water. NaOH (0.5 mol/L) was prepared by dissolving 20.00 g of NaOH in
MilliQ water in a 1-L volumetric flask. NH2OH.HCl (0.5 mol/L) was prepared by
dissolving 34.745 g of NH2OH.HCl (UNIVAR) in MilliQ water in a 1-L volumetric flask.
Ascorbic acid was stored below 4 ˚C and other reagents were stored at room temperature in
glass regent bottles (Schott, Q Stores).
Potassium chloride and phosphate solutions were taken from the sequential extraction
procedure studies in Chapter 6. Potassium chloride (0.25 mol/L) was prepared by
dissolving 18.638 g of KCl in 1 L of MilliQ water. Phosphate solution (0.1 mol/L, pH 8)
was prepared by dissolving 22.820 g of K2HPO4.3H2O in 1 L MilliQ water and the pH was
adjusted using dilute HCl. Sea water used was collected from Port Kembla Harbour.
Ammonium phosphate mobile phase (40 mmol/L; buffer pH 6) was prepared by adding 800
mL of 40 mmol/L (NH4)2HPO4, di-ammonium hydrogen orthophosphate (BDH AnalaR®)
(5.2824 g in 1 L MilliQ water, initial pH 8) to 200 mL of 40 mmol/L monobasic
(NH4)H2PO4, (SIGMA®) (4.6012 g in 1 L MilliQ water, initial pH 4.5). The pH of the
buffer mixture was usually close to 6 and precisely adjusted with dilute phosphoric acid or
dilute ammonia. Ammonium phosphate (200 mmol/L; buffer pH 6) was prepared in a
similar way from 200-mmol/L solutions of each salt (26.412 g and 23.006 g in 1 L MilliQ
water, respectively). K2S2O8 (0.2 mol/L) was prepared by dissolving 5.4066 g and 1 pellet
NaOH (~ 0.1 g, for stabilization) in MilliQ water in 100 mL volumetric flask.
Chapter 4 – Selenium speciation
4.2.2
58
Test materials
Test sediment samples were certified reference materials and marine sediments collected
from Port Kembla Harbour as summarized in Table 4.3.
Table 4.3
Major constituents and selenium concentrations in oxic and anoxic reference
materials and test samples*.
Marine sediment
SRM 2702
Anoxic sediment
(XRF, n= 5)†
PKH-1
(RNAA, n = 5)†
Red Beach
(ICP-OES)†
52 ± 2‡
Whole sediment
66.8 ± 0.4
182‡
<250 µm
-
<63 µm
-
-
-
Se (µg/g)
4.95 ± 0.46
Grain size
Al (%)
<70 µm
8.41
C (%)
Ca (%)
3.36
0.343
Cl (%)
-
0.87 ± 0.02
3±0
-
K2O (%)
-
1.00 ± 0.01
1.0 ± 0.1
-
Na (%)
0.681
1.85 ± 0.03
1±0
-
Fe (%)
7.91
5.11 ± 0.07
5.6
P (%)
0.1552
0.412 ± 0.005
10 ± 0
-
S (%)
1.5
0.25 ± 0.02
-
-
Si (%)
-
63.8 ± 0.4
-
-
TiO2 (%)
-
0.56 ± 0.01
-
-
µg/g
Ag
0.622
-
20 ± 4
-
As
45.3
68 ± 7
243 ± 3
340
Ba
-
406 ± 41
-
Cd
0.817
110 ± 15
380 ± 40
-
Cr
352
234 ± 9
Cu
177.7
Hg
Mn
0.438
1757
2140 ± 132
-
Mo
10.8
244 ± 5
-
Ni
75.4
Pb
Zn
8.7 ± 0.1
3.2 ± 0.2
-
-
29
277 ± 6
-
14156
-
385
-
168 ± 2
8±1
-
362
132.8
946 ± 13
-
4377
485.3
1238 ± 80
2806 ± 50
4993
180
* Selenium and metal concentrations are in µg/g dry weight basis and as mean ± SE, where data are available.
†
XRF: X-Ray Fluorescence (Blue Scope Steel); ICP-OES: Inductively Coupled Plasma-Optical Emission
Spectrometry (Port Kembla Copper); RNAA: Radiochemical Neutron Activation Analysis (ANSTO).
‡
Analysed by the HG-AAS method in Chapter 3.
Chapter 4 – Selenium speciation
59
Commercially available reference materials of marine/estuarine sediments typically contain
low selenium concentrations (NRCC MESS-3 and PACS-2 contain 0.72±0.05 and
0.92±0.22 µg/g, respectively, and NIST SRM 2702 and 1646a contain 4.95±0.46 and
0.193±0.028 µg/g, respectively). These selenium concentrations were detectable in total
selenium analysis but not in the speciation work. The concentrations of selenium in
aqueous and acidic extracts of those certified reference materials were below the detection
limit of the HG-AAS procedures. An in-house reference material (PKH-1) and test
sediment samples (oxic and anoxic) from Port Kembla Harbour with sufficiently high
selenium concentrations were therefore used for optimization of the extraction procedure.
Anoxic sediments were prepared from two sediment cores collected from Red Beach, Port
Kembla Harbour in June 2003 (GPS: 0307855/6183222 and 0307847/6183237). The
sediment cores were extruded, the oxic layer removed and the remaining anoxic sediment
homogenized under N2 atmosphere in a glove box. Several sub-samples were prepared and
stored at 4 ˚C for short periods or in a freezer for long-term storage. Five separate subsamples were dried to obtain moisture content. The dry samples (n=5) were finely ground
and metal compositions analysed by X-Ray Fluorescence (XRF) (Whant, 2003).
4.2.3
Sediment extraction procedure
Sediment extraction for the speciation work requires extractant reagents and conditions that
can recover a quantifiable amount of metals as well as retain their original species. The
reagents known to extract metals from particular sediment phases (see also Table 4.1) were
tested for their suitability to extract selenium from sediments. The tested reagents included
MilliQ water, seawater, potassium chloride (KCl), potassium phosphate (K2HPO4, pH 8),
hydrochloric acid (HCl), phosphoric acid (H3PO4), phosphoric acid: methanol (H3PO4:
MeOH, 1:1), ascorbic acid, hydroxylamine hydrochloride (NH2OH.HCl), and sodium
hydroxide (NaOH). The sediment phases, which each reagent will extract, are given in
Table 4.4.
In a typical procedure, ~ 0.2-0.5 g (or equivalent dry weight) of test sediments were
extracted with 10 mL of reagents on a mixing wheel (end-over-end mixing) for a minimum
Chapter 4 – Selenium speciation
60
of 1 hour. Samples were centrifuged at 2400 rpm for 20 min, decanted and filtered through
0.45 µm membrane filters. The extracts were stored below 4 ˚C for further analysis. The
HPLC separation was performed on the same or next day. Acid preservation was not
carried out, as it has been reported to promote selenite binding to dissolved organic matter
co-extracted in the sediment solutions (Zhang et al., 1999a), also discussed in Section
4.3.3.
Table 4.4
Extractant reagents tested (Tokunaga et al., 1991; Seby et al., 1997; Zhang et
al., 1999a; Ellwood and Maher, 2003) and sediment phases extracted.
Please see print copy for Table 4.4
4.2.4
HPLC separation and selenium detection
Standard selenium compounds in MilliQ water, NaOH solutions, and sediment extracts
were separated using HPLC protocols optimised from the procedure by Orero Iserte et al.
(2004) and summarised in Table 4.5.
The eluate was collected every one minute in the early stages and if it was found that some
overlapping of peaks occurred then the fractions were collected every 30 seconds. Postcolumn samples were digested according to the procedure by Zhang et al. (1999a) and its
application was assessed (based on recoveries) in the laboratory to be valid. Typically, 0.5
mL of 0.2 M K2S2O8 was added to 0.5-1 mL fractions in a 15-mL polypropylene tube and
heated in a water bath at 90 ˚C for 30 min; then 2 mL conc. HCl was added and heating
Chapter 4 – Selenium speciation
61
continued for 20 min. Samples were allowed to cool, made up to 5 mL with MilliQ water
and analysed by HG-AAS as per the procedure in Section 3.2.4. Total selenium in the
extract was determined from digestion of 0.5-1 mL aliquot of the extract and then analysed
by HG-AAS.
Table 4.5
Optimized HPLC conditions.
Parameters
Operating condition
HPLC system
Shimadzu Liquid Chromatograph LC-10 AT VP and Class-VP software
Column
Hamilton PRP-X100 anion exchange column (Phenomenex), 250 x 4.6
mm, 10 µm PEEK, 100 A˚ pore size/ pH dependent, Poly(styrenedivinyl)benzene polymers-Trimethyl ammonium exchange resin
Regeneration
50 mL of 1% 6 M HNO3 (AnalaR®) in methanol
Mobile Phase
A: 40 mM ammonium phosphate buffer pH 6
B: 200 mM ammonium phosphate buffer pH 6, (ammonia or phosphoric
acid)
Injection volume
100 µL
Temperature
Ambient
Flow rate
1.5 mL/min
Detection
HG-AAS (operating conditions as in Table 3.2)
Gradient program
1.00-4.00 min 100% A, 4.01-9.00 min 100%B, 9.01-15.0 min 100%A.
4.3
Results and discussion
4.3.1
Sediment extraction
4.3.1.1
Effects of extractant reagents on HG-AAS detection
Suitable reagents for sediment extraction must be compatible with the HG-AAS detection.
The compatibility of extractant reagents with the HG-AAS was first tested by spiking
different reagents (at 0.5 mol/L) with selenite standards (4–20 µg/L) and analysed by HGAAS against typical 40% HCl matrix standards. The results are shown in Figure 4.2.
Chapter 4 – Selenium speciation
62
120
% Selenite recovered
100
80
60
40
20
Figure 4.2
e
Ph
os
p
ha
t
HC
l
H
NH
2O
H
Na
O
O4
:M
eO
H
O4
H3
P
H3
P
cid
or
bi
ca
As
c
KC
l
HC
l
0
Effects of extractant matrix on HG-AAS signal (mean ± SE, n=3). HCl,
Ascorbic acid, H3PO4, H3PO4: Methanol, NH2OH.HCl and NaOH were 0.5
mol/L. KCl was 0.25 mol/L and phosphate (pH 8) was 0.1 mol/L.
Good recoveries of selenite were obtained in the 0.5 mol/L HCl, NH2OH.HCl and NaOH
matrices with average recoveries of 93, 95 and 104 %, respectively. Potassium chloride
(KCl) and phosphate solutions also provided good selenite recoveries and did not interfere
with the HG-AAS analysis. Poorer recoveries were obtained for the phosphoric acid and
phosphoric acid: methanol (1:1 v/v) matrices (75 % and 27 %, respectively). The methanolcontaining matrices also decreased the selenium AAS signal of the calibration standards
during a subsequent analysis. The interfering mechanism of methanol in hydride generation
or atomic absorption spectrometry is unknown. Ascorbic acid gave no recovery, possibly
because the ascorbic acid solution reduced selenite to elemental selenium under acidic
condition such as in this HG matrix (Schlekat et al., 2000). Elemental selenium does not
react with the hydride generation reagent, so provided no AAS signal.
The reagents chosen for further sediment extraction tests were HCl, KCl, NaOH,
NH2OH.HCl, and phosphate solution. Phosphoric acid was also tested for its extraction
efficiency of selenium from sediments, despite its relatively low AAS signal recovery, as
phosphoric acid has been reported to successfully extract and preserve arsenic species in
marine sediment samples (Ellwood and Maher, 2003; Orero Iserte et al., 2004).
Chapter 4 – Selenium speciation
4.3.1.2
63
Choice of extractants
The efficiencies of different reagents to extract different selenium species from the
reference material SRM 2602, oxic and anoxic test materials are shown in Figure 4.3.
Sodium hydroxide was the most effective reagent, which extracted 21.5-47.3% of selenium
from the test sediments. Phosphate solution and hydrochloric acid were the next best
reagents extracting 3.6-8.3% and 1.3-5.9%, respectively. Other reagents tested (phosphoric
acid, potassium chloride and hydroxylamine hydrochloride) were able to extract less than 3
% of the selenium. Overall, the reference material SRM 2702 provided higher selenium
recoveries for all extractants used, while the wet anoxic sediment and the dry Red Beach
sediment contained a slightly lower percentage of extractable selenium.
60
% Se extracted
50
40
30
20
10
SRM 2702
Figure 4.3
Anoxic Sediment
PKH-1
Ph
os
p
ha
t
e
HC
l
H
NH
2O
H
Na
O
KC
l
O4
H3
P
Cl
H
er
aw
at
Se
M
ill
i
Q
0
Red Beach
Percentage (mean ± SE, n = 3) of selenium extracted from test sediments by
different extractant reagents: HCl, H3PO4 and NH2OH.HCl were 0.5 mol/L,
KCl was 0.25 mol/L, and NaOH and Phosphate (pH 8) were 0.1 mol/L.
The results showed that selenium extraction was more efficient in alkaline reagents and in
oxic sediments. Significantly less selenium was extracted by hydrochloric acid, phosphoric
acid, and hydroxylamine hydrochloride indicating that either, selenium was not associated
with acid-soluble components such as carbonates or iron-manganese oxyhydroxides in the
Chapter 4 – Selenium speciation
64
sediment (more discussion on sediment geochemical phases in Chapter 6), or the acidity of
the reagents protonated the sediment surface making the adsorption sites available for readsorption of selenium oxyanions, and so less selenium was found extracted in the acid
solutions (Gruebel et al., 1988; Goh and Lim, 2004). It is reasonable for anoxic sediment to
contain less extractable selenium due to its reducing conditions, but that does not explain
the low selenium recovery from the dry oxic Red Beach sample. The similar sediment
compositions shared by the two Port Kembla Harbour test samples (anoxic sediment and
Red Beach) could be a factor contributing to their lower extractability, in comparison to the
less contaminated SRM 2702.
Please see print copy for Figure 4.4
Figure 4.4 Intense brown colour of NaOH extracts of SRM 2702, in
comparison to (a) other reagent extracts of SRM 2702: hydrochloric acid,
phosphoric acid, phosphoric:methanol and hydroxylamine hydrochloride;
and (b) sodium hydroxide extracts of other test sediments from Port Kembla
Harbour: wet anoxic, PKH-1 and Red Beach (19b) sediments.
The high organic matter content could be the major factor contributing to high percentage
of selenium released in NaOH extracts of SRM 2702, which were observed to have an
intense brown colour, relative to the colour of other reagent extracts (Figure 4.4 (a)).
Sodium hydroxide extracts of other test sediments from Port Kembla Harbour had a much
lighter brown colour (Figure 4.4 (b)) indicating that less organic matter was released in the
extracts and hence a lower percentage of selenium was extracted than from the SRM 2702.
Organic carbon contents of the test sediments were not specifically known. The average
total organic carbon measured for Red Beach sediment cores (Section 6.3.2, n=36) was
2.68 ± 0.15 %. The total carbon (not all as organic) reported for SRM 2702 was 3.36 %.
Chapter 4 – Selenium speciation
65
A basic stability test of organic and inorganic selenium compounds in the NaOH reagent
was carried out by spiking NaOH solutions (1 mol/L) separately with selenite, selenate,
seleno-DL-methionine and selenocystine (at 1 mg/L) and allowing them to stand for 2
weeks below 4 ˚C. Direct measurements of selenite by HG-AAS after diluting a small
aliquot to 10 µg Se/L found 98.4, 0.4, 1.2 and 4.5 % of selenite recovered from selenite,
selenate, seleno-DL-methionine and selenocystine, respectively. This indicated that selenite
was certainly stable in the tested 1 mol/L NaOH reagent. Selenate and selenomethionine
also showed their high stability with negligible amounts converted to selenite.
Selenocystine was less stable with approximately 5 % converted into selenite under tested
conditions. However, this simple test did not verify directly that selenate, seleno-DLmethionine and selenocystine were not converted into other selenium species. Providing the
oxic and alkaline conditions, the three compounds were unlikely to be reduced to elemental
selenium (see also selenium Eh-pH diagram, Figure 2.2). Further oxidation of seleno-DLmethionine and selenocystine to selenate should not occur, as the selenite tested was not
found to be oxidised to selenate (within a range of analytical errors). Sodium hydroxide
was used for all selenium extractions in further speciation work.
4.3.1.3
Effects of extractant concentration and extraction time
Extractant concentrations and extraction time might affect the amount of selenium
extracted from soils/sediment. The efficiencies of different sodium hydroxide
concentrations (0.1 M to 1 M) in extracting selenium from test sediments: SRM 2702 and
the wet anoxic sediment, over a time period of 1 to 24 hours were tested and the results are
presented in Figure 4.5. This indicated that as the extraction time increased the percentage
of selenium extracted from both sediments increased. More than 80% of selenium was
extractable from SRM 2702 (containing total selenium of 4.95 µg/g) over a 24-hour
extraction period. The anoxic sediment released a lower percentage of selenium in the
NaOH solutions than SRM 2702 over the same extraction period. This was possibly
because the anoxic sediment was in a reducing state with less alkaline-extractable selenium
(or having less selenium associated with organic matter) as discussed previously. In
addition, the 10 times higher total selenium (52 µg/g) in the anoxic sediment than in SRM
2702 might affect the equilibrium between the solution and solid phases during extraction.
Chapter 4 – Selenium speciation
66
More selenium extracted during a longer extraction time was believed to result from solidstate selenium such as selenides or elemental selenium being slowly oxidised into selenite
and selenate and released into the NaOH solutions (Seby et al., 1997).
100
(a)
% Se extracted
80
60
40
0.1 M
0.2 M
0.5 M
20
1M
0
0
4
8
12
16
20
24
28
Extraction time (hour)
60
(b)
% Se extracted
48
36
24
0.1 M
0.2 M
0.5 M
12
1M
0
0
4
8
12
16
20
24
28
Extraction time (hour)
Figure 4.5
Percentage of selenium extracted in sodium hydroxide solutions with
different extraction time from (a) SRM 2702 and (b) wet anoxic sediment –
data point for 12 hr extraction of 0.2 mol/L NaOH (39%) was anomalous
therefore not included in the graph.
The sodium hydroxide concentration did not significantly affect the selenium extraction
efficiencies. No significant differences (P > 0.05) between NaOH concentrations were
detected for SRM 2702 (Figure 4.5 (a)). The higher concentrations of NaOH provided
Chapter 4 – Selenium speciation
67
slightly better extraction efficiencies for anoxic sediment results (Figure 4.5 (b)), however,
0.1 mol/L concentration was chosen primarily for further work as it was found to have less
effect on the subsequent HPLC separation as discussed in Section 4.3.2.
The effect of multiple extractions on removing selenium was also examined by conducting
two consecutive extractions of 1 hour using 0.1 mol/L NaOH. The second extraction was
found to increase the extracted selenium from the first extraction by up to 50 %. However,
the additional extraction was considered less beneficial as the additional extractant volume
significantly diluted the first extractant selenium concentrations and decreased selenium
detectability. Only a single extract was used for further speciation analysis.
4.3.2
Optimisation of the HPLC separation
HPLC protocols including the type of mobile phase, concentration, pH and flow rate were
investigated and optimised for the separation of four selenium standards in acidified MilliQ
water and NaOH (0.1 mol/L) solutions and in sediment extracts.
The compatibility of an ammonium phosphate mobile phase with HG-AAS analysis was
first examined by spiking selenite and selenate standards (0.25µg corresponding to 10 µg/L
AAS reading in 25 mL) in triplicate and subjecting the solutions to the digestion procedure
described above in Section 4.2.4. The recovery was 92 ± 4 % for selenite and 95 ± 3 % for
selenate. Freeze-drying the matrix and re-dissolving the solid in HCl before digestion gave
slightly better recoveries of 97± 2 % for selenite and 99 ± 2 % for selenate, but the freezedrying step was omitted due to the significant time (overnight) required.
Four selenium standards in MilliQ water and in 0.1 mol/L NaOH solution were well
separated by chromatography using the conditions in Table 4.5. The chromatographic
results of both matrices (Figure 4.6) showed an elution sequence of selenocystine, selenite,
selenomethionine and selenate, respectively. The retention times for the four standards
were 2, 4, 6 and 10 min in MilliQ water matrix while NaOH matrix was found to retain the
selenium compounds in the column for approximately 1 min longer, shifting the elution
chromatograms to the right.
Chapter 4 – Selenium speciation
68
This elution sequence was considered reasonable since the anion exchange column
separated the selenium compounds on a basis of their ionic charges (Vassileva et al., 2001).
According to their pKa values (Figure 4.1), all four selenium compounds are present as
anionic species in 0.1 mol/L NaOH extracts (pH ~ 13), with -2, -2, -2 and -1 charges for
selenite, selenate, selenocystine and selenomethionine, respectively. During separation in
the mobile phase buffer at pH 6, selenocystine was neutral or zwitterionic therefore eluted
first. Selenite containing –1 charge (HSeO3-) and the zwitterionic selenomethionine were
eluted in between. Selenate remained –2 anionic and was eluted last.
0.1
µg Se
(a)
Selenate
Selenite
0.08
Se-Cys2
0.06
Se-Met
0.04
0.02
15
14
13
12
10
9
8
7
6
5
4
3
2
1
0
0
Time (min)
0.1
µg Se
(b)
Selenite
0.08
Selenate
Se-Cys2
0.06
Se-Met
0.04
0.02
15
14
13
12
10
9
8
7
6
5
4
3
2
1
0
0
Time (min)
Figure 4.6
HPLC of standard selenium compounds (0.1 µg Se) (a) in MilliQ water, and
(b) in 0.1 mol/L NaOH solutions. Hamilton PRP-X100 anion exchange
column, 40 mM /200 mM ammonium phosphate buffer, pH 6 mobile phase,
according to Table 4.5.
Chapter 4 – Selenium speciation
69
Other observations and findings included:
•
Lower eluent concentrations (10 mM and 20 mM ammonium phosphate) were
examined to assess implications for later ICP-MS detection, as samples containing
lower salt concentrations are preferred. The lower eluent concentrations provided a
better separation, but caused significant peak broadening at the baseline.
•
A larger injection volume (200 µL) was attempted to give more selenium for the
separation and to enhance the HG-AAS detection limit. This was found to cause
peak broadening and eventual overlapping.
•
Different gradient programs were tried, including a gradual increase in eluent
concentrations, but this also caused either peak overlapping or broadening. The
gradient program given in Table 4.5 was optimal for samples containing 0.1 mol/L
NaOH. High eluent concentrations (40 mM ammonium phosphate) provided good
sharp peaks and a step up to 200 mM was necessary to elute selenate.
•
A slight initial pH change, especially with a NaOH extract, could alter the
separation of early peaks (e.g., 100 µL of 0.1 mol/L NaOH shifted Se-Cys2 to a
longer elution time in Figure 4.6 (b)). Sodium hydroxide at 1 mol/L concentration
caused overlapping between the Se-Cys2 and selenite peaks. Therefore, a lower
NaOH concentration is preferred in sediment extract samples in order to obtain a
good chromatographic separation.
The HPLC protocol and elution time were used to confirm the stability of the four selenium
species in the sodium hydroxide solution, mentioned previously in Section 4.3.1.2. The
selenium species tested were again found to be stable in 0.1 mol/L NaOH extractant
reagents after 2 week-storage below 4 ˚C.
Calibration curves were constructed for each selenium standard for the range of 0 – 1,000
µg/L, corresponding to 0 – 20 µg/L AAS readings, based on the required digestion method
with the final volume of 5 mL before HG-AAS analysis. The analytical performance of the
selenium speciation by the HPLC and HG-AAS method is given in Table 4.6.
Chapter 4 – Selenium speciation
Table 4.6
70
Analytical performance for selenium speciation by HPLC and HG-AAS.
Selenocystine
Selenite
Selenomethionine
Selenate
0 - 1000
0 - 1000
0 - 1000
0 - 1000
> 0.99
> 0.99
> 0.99
> 0.99
Retention time (min)
3
4
6
11
Method detection limit, µg/L
50
10
10
10
% RSD (1 mg/L, n = 3)
3
1
6
2
Linear range, µg/L
Correlation coefficient, r2 (n=7)
One of the drawbacks in this method was that the post column selenium detection by the
HG-AAS method required the digestion and up to 10-times dilution of the eluate fraction
(e.g., 0.5 mL to 5 mL digest), leading to any minor selenium species present at initial low
concentrations falling below the detection limit.
Online detection of HPLC-post column selenium has been reported to provide better
detection limits with minimal sample pretreatment steps (see also Table 4.1) (CoboFernandez et al., 1995; Chatterjee et al., 2001; Goldberg et al., 2006). The AAS available
for use in this study contained software that allowed the continuous detection of signal for
the maximum of only 5 minutes (300 seconds). Using the exact same instrumentation and
software, Muhammad (2003) was able to continuously detect four arsenic standards:
(arsenite, dimethylarsinate, monomethylarsonate and arsenate) from direct coupling to a
HPLC column. In this study, the optimum separation of selenium species was achieved in
12 min, therefore the selenium analysis was possible through a fraction collection and postcolumn digestion steps.
At the time of writing, a new ICP-MS has been acquired recently by the School of Earth
and Environmental Sciences at the University of Wollongong. Any future direct coupling
of this HPLC separation protocol with the ICP-MS detector is feasible, which would
shorten the analysis time significantly by eliminating the lengthy digestion steps,
subsequently improving method detection limits.
Chapter 4 – Selenium speciation
4.3.3
71
Application to sediment NaOH extracts
The optimised HPLC with HG-AAS method was used to analyse sodium hydroxide (0.1
mol/L) extracts of oxic (Red Beach test sediment) and the anoxic sediment. The HPLCs of
selenium species in the sediment NaOH extracts are shown in Figure 4.7.
0.50
(a)
µg Se
0.40
Selenite
0.30
0.20
?
0.10
?
?
15
14
13
12
10
9
8
7
6
5
4
3
2
1
0
0.00
Time (min)
0.028
(b)
Selenite
µg Se
0.021
0.014
0.007
Selenate
15
14
13
12
10
9
8
7
6
5
4
3
2
1
0
0
Time (min)
Figure 4.7
HPLC of sediment NaOH extracts: (a) oxic Red Beach sediment (0.1 mol/L,
12 hour extraction) and (b) anoxic wet sediment (0.1 mol/L, 4 hour
extraction), Hamilton PRP-X100 anion exchange column 40 mM /200 mM
ammonium phosphate buffer, pH 6, mobile phase (as per Table 4.5).
The chromatographic results of the test oxic sediment (dry Red Beach) (Figure 4.7 (a))
showed a strong peak between 4-5 min, corresponding to the standard chromatogram peak
for selenite. There were also other unidentifiable peaks that were higher than the baseline at
2 min and between 6.5-10 min. These peaks were not matched to the standards used in this
study. The early-unknown peak might be a selenium species (potentially organic) with
neutral or zwitterionic nature and so not retained by the anion exchange column under the
Chapter 4 – Selenium speciation
72
HPLC running conditions. The later unknown peaks eluted near selenomethionine retention
time possibly have a similar ionic property. The recovery of the selenium from the HPLC
column was 71% for the Red Beach oxic sediment extract.
In the anoxic sediments (Figure 4.7 (b)), two peaks were observed and identified, as
selenite and selenate, with selenite being the major species. No other selenium peaks were
observed over the 15-min run, indicating that no other (unknown) selenium species were
present in the anoxic sample extracts (at concentrations above the detection limit). Only 53
% of the selenium in the anoxic sediment extract solution was recovered after the HPLC
column (100 µL injection).
The stability of selenium species in the real sediment extracts was examined by spiking the
alkaline sediment extracts with a mixture of four selenium standards (500 µg/L) and
passing the spiked sample through the HPLC column. Full recoveries of selenomethionine
and selenate were obtained; however, none of selenocystine and selenite was recovered. It
is not known why selenocystine was not recovered from the NaOH extract. However, it was
suspected that the added selenite (originally from a mildly acidic standard solution) might
be adsorbed or bound to dissolved organic matter that is co-extracted in the NaOH solutions
(Seby et al., 1997; Ferri and Sangiorgio, 1999), and this might alter the selenite ionic
structure that was required for the chromatographic separation. Selenite has also been
frequently reported to bind strongly to iron-manganese oxyhydroxide species (Balistrieri
and Chao, 1990; Dhillon and Dhillon, 1999; Duc et al., 2006; Martinez et al., 2006).
However, this hypothesis can be ruled out, as negligible amounts of iron and manganese
were co-extracted (Section 6.3.4.2) (NB: copper and chromium were only two elements
found in significant quantities in the NaOH extracts).
To test for any possible association of the selenium compounds with humic acid or fulvic
acid fractions, the NaOH extracts were acidified to below pH 2 and samples were
centrifuged to separate the humic precipitate residue from the fulvic acid soluble fraction
(Seby et al., 1997). This method was only found to be successful for high organic content
samples. Selenium concentrations were determined for both humic and fulvic fractions of
successful samples (SRM 2702) and revealed that 90-95% of the selenium in the acidified
Chapter 4 – Selenium speciation
73
NaOH extracts remained in the soluble fulvic acid fraction. However, this test did not
verify that the selenium was not associated with dissolved organic matter in the extracts.
A use of XAD resins to remove dissolved (hydrophobic and neutral) organic matter from
soil/sediment extracts before selenium speciation has been reported in the literature (Fio
and Fujii, 1990; Ornemark and Olin, 1994; Martens and Suarez, 1997). Some researchers
have noted a significant loss of selenite from XAD treatment steps due to complexation of
selenite with humic substances during the pre-column acidified treatment leading to
removal of the complex by the resin (Pyrzynska, 1995; Zhang et al., 1999a; Zhang et al.,
1999b). An application of XAD resins to remove organic matter interferences was
considered, but a separate study may be required due to the complex nature of this issue.
4.4
Conclusions
A sediment extraction procedure and a HPLC & HG-AAS method for measurement of
labile selenium compounds in marine sediments was evaluated and optimised. Extraction of
labile selenium compounds from sediments using water, salt, acid and alkaline solutions
found an alkaline sodium hydroxide to be the most effective reagent in extracting selenium
from the reference material and Port Kembla Harbour test sediments. An anion exchange
column was used successfully to separate four selenium compounds (selenite, selenate,
selenomethionine and selenocystine) in a sodium hydroxide (0.1 mol/L) solution in 12 min
with gradient ammonium phosphate elution. Combined with HG-AAS detection, the
method operated with high detection limits but is feasible for measurement of individual
selenium compounds in contaminated samples. Using the optimised HPLC & HG-AAS
method, selenite and selenate were identified in NaOH extracts of both oxic and anoxic
sediments. At this stage, due to interferences from a complex sediment extract matrix, the
method could not be used with full confidence to accurately quantify individual selenium
compounds in the NaOH extracts. Further development on the quantification technique is
required.
Chapter 5
Selenium distribution in Port Kembla Harbour sediments
5.1
Introduction
Limited studies have been completed on selenium in sediments of Port Kembla Harbour
(Figure 1.1), including Goodfellow (1996), Hoai (2001) and White (2001) as summarised
in Section 2.4.2. Therefore, there is a need for a new full survey of selenium contamination
in the harbour sediments in this project. The sedimentary selenium data were mainly
obtained as a supplement to other metal studies (Goodfellow, 1996; White, 2001). The
analytical techniques, such as RNAA, XRF or ICP-OES, used for the trace metal
determination in local studies had high detection limits (RNAA: 5 µg/g (O’Donnell M.,
2003, University of Wollongong, per comm.), XRF: 10 µg/g (Whant, 2003) and ICP-OES:
5 µg/g (Chipeta C., 2003, Port Kembla Copper Ltd, per comm.), which are not sensitive
enough to accurately determine selenium concentrations that are present at relatively low
levels. The conventional sample preparation for trace element analysis involves drying
sediment samples at 110˚C, which could potentially underestimate selenium concentrations
due to selenium loss via volatile compounds. The study employed the HG-AAS technique
(optimised in Chapter 3) that is specific to selenium analysis and can accurately quantify
selenium concentrations in sediments at low background levels.
The specific aims of the harbour survey work were to:
•
determine the total concentrations and the spatial distribution of selenium in the
surface sediments from 23 sites throughout Port Kembla Harbour and three core
samples from the Red Beach;
•
identify any possible differences in selenium concentrations between top oxic
sediments and lower anoxic layers;
•
determine any relationship between selenium concentrations and sediment particle
size (<63, 63-250 and >250 µm fractions);
Chapter 5 - Selenium distribution
•
75
determine any relationship between selenium concentrations and other trace
elements (As, Cd, Cr, Cu, Fe, Mn, Ni, Pb, Sb and Zn) in the sediments; and
•
identify the likely selenium sources and pathways into the harbour.
The results are discussed in comparison with selenium in Lake Macquarie, NSW, Australia,
as well as in reference to an aquatic hazard assessment guidelines developed from the
extensive selenium research in the United States of America (USA). Sediment 210Pb dating
work to determine sedimentation rate in Red Beach sediment cores is also carried out and
reported in this chapter.
5.2
Materials and methods
5.2.1
Reagents and apparatus
All glassware and plastic containers were acid-cleaned. Chemicals and reagents were of
analytical reagent grade or better. Preparations of reagents and standards are as described in
Section 3.2.1 and Section 4.2.1.
5.2.2
Collection of surface sediments and core samples
Surface sediment samples were collected using a grab sampler from a boat on 7th April
2003 from 23 harbour sites as shown in Figure 5.1. The sampling sites were chosen to
cover all possible contaminated areas of the harbour. Defined-oxic and -anoxic sediments
(identified by colour) were collected from grab samples that appeared to retain their
original shape and layers. Oxic samples were collected by careful scooping from the top ~5
cm sediment using plastic spoons. Anoxic sediments were collected in a similar way after
the entire top 5 cm layer was removed. Samples that appeared to be sandy or their
representative oxic/anoxic layers were not clearly obtained, were collected and labeled as a
composite sample.
Samples were stored in acid-washed polycarbonate containers below 4 ˚C for the shortterm, and –20 ˚C for long-term storage. Well-separated oxic and anoxic grab sediments
were obtained for 15 sites (i.e., Sites 1, 2, 3, 5, 8, 9, 10, 11, 12, 13, 14, 17,18, 21 and 21).
Chapter 5 - Selenium distribution
Please see print copy for Figure 5.1
Figure 5.1
Locations of surface samples collected from 23 sites around Port Kembla
Harbour on 7th April 2003. GPS coordinates are given in Table A.1.
76
Chapter 5 - Selenium distribution
77
Two grab samples from Site 19 had distinctly different physical characteristics so were
analysed separately as 19a and 19b. Samples from the remaining 7 sites (i.e., Sites 4, 6, 7,
15, 16, 22 and 23) were analysed as a composite sample. A total of 39 initial samples were
obtained, homogenized and sub-sampled for (1) pH measurement (samples discarded
afterward); (2) un-sieved samples (whole sediment); and (3) wet sieving for particle size
separation. Details of grab samples including the GPS coordinates (Global Positioning
System) are listed in Appendix A (Table A.1).
Please see print copy for Figure 5.2
Figure 5.2
Laboratory set up for sediment sample processing. From right to left,
sediment core samples, nitrogen glove box and sediment core extruder.
Three sediment cores (Cores A1-A3) were collected using a hand-held sediment corer fitted
with acid-washed 30-cm polycarbonate corers (method described in Muhammad, 2003)
from Red Beach area near the mouth of the Darcy Road Drain (Sites 19 and 20 in Figure
5.1). Core A1 was collected on 25 February 2003 and Cores A2 and A3 on 25 June 2003.
Details of these cores including the GPS coordinates are given in Appendix B (Table B.1).
The sediment cores were stored in a polystyrene box in an upright position, and transported
to the laboratory where they were immediately sectioned or stored on ice until sectioning
could be completed (generally within 2 days of sampling).
Chapter 5 - Selenium distribution
78
Sediment cores were maintained vertically, extruded using a locally manufactured device
(Figure 5.2) and cut into 2-cm sections. Each section was homogenized and divided into
three sub-samples as carried out for the surface samples above. Core A1 sample processing
was carried out in a N2 glove box. Since only total selenium concentrations were to be
measured, use of the nitrogen glove box was considered not an important factor. Therefore,
Cores A2 and A3 processing was completed outside the glove box.
5.2.3
Sample preparation and analysis for selenium
Sub-samples from both surface and core samples described above were wet-sieved through
250 µm and 63 µm screens by spraying seawater from a spray bottle over sediment until
clear through water was obtained. Three particle size fractions < 63 µm, 63-250 µm and >
250 µm were obtained for each sample. For < 63 µm samples, suspended sieved-samples
were centrifuged at 2400 rpm for 15 min to settle out the fine sediment and remove
excessive water. Larger particle samples were transferred into a 600-mL beaker and
allowed to stand for 20-30 minutes to allow sediment settling from the sieving water.
Samples (both sieved and un-sieved) were freeze-dried or dried in an oven at 40 °C (for 4-6
days depending on particle sizes, sample quantities and surface area of the containers used).
Freeze-drying required much longer than oven drying with limited space for multiple
samples. There was no significant difference in selenium concentrations from using either
freeze-drying or oven drying at 40 ˚C, so the more convenient oven drying method was
used. Dried sediments were manually ground using a porcelain mortar and pestle, and
stored in acid-washed polycarbonate containers at room temperature.
All dry sediment samples were microwave-digested (MILESTONE ETHOS SEL) in aqua
regia as per the method optimized in Chapter 3. The digested samples were stored below
4˚ C until analysis and pretreated for HG-AAS analysis as described in Section 3.2.3. Total
selenium concentrations were determined using HG-AAS (Varian SpectrAA220, equipped
with VGA-76 vapour generation system) according to the method described in Section
3.2.4. Diluted AAS samples were analysed within 1-2 days. Sub-samples of the digests
were analysed for other major metals (As, Cd, Cr, Cu, Fe, Mn, Ni, Pb, Sb and Zn) by ICP-
Chapter 5 - Selenium distribution
79
OES at Port Kembla Copper Limited (in 2003) using the company’s standard procedure.
The sample preparation and analysis flowchart is shown in Figure 5.3.
Surface sediment
Sediment core
Cut to 2-cm sections
Homogenised
pH
Un-sieved samples
/ whole sediment
To-be-sieved
samples
Wet sieved in to > 250, 63-250
and < 63 µm fractions
Freeze or oven dried at 40°C, then finely ground
Microwave aqua regia digestion
Total Se by HG-AAS
Figure 5.3
Total metals by ICP-OES
Sample preparation and analysis flowchart for selenium spatial distribution
studies in Port Kembla Harbour sediments.
Accuracy and reproducibility of the HG-AAS selenium analysis in the field samples was
assured by using freshly prepared calibration standards. Standards and blanks were checked
regularly (every 10 samples or when selenium concentrations were found to be distinctly
high or low) for instrumental drifting. A MilliQ water rinse was used after concentrated Se
samples. Triplicate analyses of the same solutions usually gave excellent precision (RSD <
1%) under normal analysis conditions. Over-range samples were diluted with 40 % HCl
(v/v) to give a selenium concentration that fell within the linear calibration range (0-50 µg
Se/L).
Chapter 5 - Selenium distribution
80
Certified reference materials MESS-3, CRM 277 and SRM 2702 for low selenium
concentration range, in-house reference material (PKH-1) for high selenium concentration
range, and blanks were analysed as samples for quality control. Consistently good
recoveries were obtained for selenium, similar to Table 3.6. Relatively low recoveries were
obtained for other acid-extractable elements by ICP-OES analysis (Table 5.1), as HF was
not used for sample digestion in this work (an OH&S restriction). Low recoveries of
simultaneously extracted As and Sb might not be caused by the digestion method but the
ICP-OES instrumental method, which may not be the best analysis method for those
hydride-forming elements (Chipeta C., 2003, Port Kembla Copper Ltd, per comm.).
However, the metal recoveries were reasonably consistent for each element and for the
reference materials analysed.
The selenium and trace element concentrations are reported on a dry weight basis and are
not corrected for the method recovery. As, Cd, and Sb concentrations in the surface
sediment samples were below or near ICP-OES detection limits (0.050 mg/L for As and
0.010 mg/L for Cd and Sb), and, therefore, not included in this report. The oxic and anoxic
data sets for pH, grain size, selenium and trace metals, were not significantly different;
therefore they were averaged for subsequent data interpretation and analysis (for all
parameters).
5.2.4
Lead-210 dating of Red Beach sediment cores
Sediment dating was carried out in January 2006 using two sediment cores collected from
the Red Beach area. The first sediment core was Core C4 (24 cm), with 2-cm interval
samples from the sequential extraction work, collected in July 2005 (see Chapter 6). The
second sediment core (Core D1, 36 cm) was collected fresh in January 2006 using the same
core-sampling technique but with a longer and sharper aluminium tube (as compared to the
polycarbonate tubes) that made penetration and sampling of deeper sediment possible (NB:
an aluminium tube was not used in sediment core sampling for other studies in this thesis
due to potential sample contamination from metal compositions of the tube). Core D1 was
sectioned at 1-cm intervals, which was preferred for the Pb-210 dating work. The samples
81
Table 5.1
Recoveries of aqua regia extractable metals from certified reference materials analysed by ICP-OES*.
Reference materials
MESS-3
CRM 277
As
Cd
Cr
Cu
Fe (%)
Mn
Ni
Pb
Sb
Zn
Certified (µg/g)
-
-
105
33.9
4.34
324
46.9
21.1
-
159
Measured (n= 5)
-
-
65 ± 16
33 ± 3
2.29 ± 0.29
237 ± 7
35 ± 1
19±3
-
121± 6
% Recovery
-
-
62
97
53
73
75
90
-
76
Certified (µg/g)
47.3±1.6
11.9±0.4
192 ± 7
101.7 ± 1.6
41.7
1615
43.4± 1.6
146±3
-
547 ± 12
Measured (n= 4)
35±16
9±1
129 ± 5
93 ± 4
23.5 ± 0.2
1146 ± 35
33 ± 3
117 ± 4
-
413 ± 1
74
76
67
91
56
71
76
80
-
76
Certified (µg/g)
-
-
352 ± 22
177.7±5.6
7.91± 0.24
1757±58
75.4± 1.5
132.8± 1.1
5.6±0.24
485 ± 4
Measured (n=10)
-
-
260 ± 6
101± 5
4.8± 0.0
1390± 38
58± 1
106 ± 3
5±1
388 ± 7
% Recovery
-
-
74
57
61
79
77
80
86
80
% Recovery
SRM 2702
†
* Certification is for total metals, not aqua regia extractable. †Not certified.
Chapter 5 - Selenium distribution
82
were homogenised and sub-sampled for grain size analysis. The remaining samples were
dried (40˚C oven), finely ground and stored in a plastic bag ready for further
210
Pb-dating
analysis.
The further sample processing and analyses for the radiodating work were carried out at the
Radiochemical Laboratory, Australian Nuclear Science and Technology Organization
(ANSTO), Lucas Heights, NSW (supported by AINSE Grant (No. 06089), in collaboration
with Atun Zawadski and Jennifer Harrison, Institute for Nuclear Geophysiology). The
sediment samples were subject to acid-digestion and isolation of
210
Po and
226
Ra, which
were collected as the alpha sources. Polonium-209 and Ba-133 yield tracers were used to
determine the recoveries of 210Po and 226Ra, respectively, and to correct for their activities.
The
the
210
226
Po sources were counted to determine their activity on an alpha spectrometer, and
Ra/133Ba sources were counted on a gamma spectrometer for the
133
Ba activity and
then on an alpha spectrometer to measure the 226Ra activity. The preparation and analysis of
210
Po and
226
Ra sources were carried out under good quality control according to the
ANSTO standard methods (ENV-I-044-031: Sedimentation rate determination; ENV-I044-006: Bulk iron removal by ether extraction, ENV-I-044-016: Manganese dioxide coprecipitation; ENV-I-044-023: Polonium analysis; and ENV-I-044-027: Radium-226
analysis).
The
210
Po sources prepared from Red Beach sediment core samples were initially found to
contain interferences, possibly from trace metals that co-deposited onto the sources’
surface. These interferences prevented the spectrometers from detecting
210
Po alpha
emission, and hence from accurately quantifying its activities. A different method (ENV-I044-015: Lead-210 analysis) using anion exchange chromatography was required for
purification and separation of
210
Pb and
210
Po radionuclides from our samples for the
preparation of the alpha sources. Satisfactory results were achieved using this method. The
sediment dating data were processed and analysed using ANSTO databases and software.
Chapter 5 - Selenium distribution
5.3
Results and discussion
5.3.1
Selenium in surface sediments
5.3.1.1
83
Sediment characteristics and grain size
The pHs of the grab samples were close to neutral ranging from 7.19-7.82. Full pH data are
given in Appendix A (Table A.1). The pH differences between defined-oxic and anoxic
samples from all sites were not greater than 0.12 units. The pHs of samples from Site 6
(mouth of Allans Creek) and Site 16 (near the storm water channel inflow) were slightly
higher at 7.82 while the pHs of Site 4 and Site 5 (near the Bluescope Steel Company’s iron
ore pile) were slightly lower than other survey sites.
Figure 5.4
Dominant grain size distribution in surface sediment samples.
Chapter 5 - Selenium distribution
84
The spatial distribution of the surface sediment grain sizes is shown in Figure 5.4. Sites 2,
3, 4, 5, 7, 9 and 11 contained more than 80 % of < 63 µm sized particles, with a
predominantly black colour and silty texture. Site 19a, near the Darcy Road Drain inflow,
and Sites 15 and 16, near No. 6 Jetty, contained more than 80% sand particles (> 250 µm).
Sediments from Site 6, near the mouth of Allans Creek, and Site 8, near the Gurungaty
Creek inflow, however, contained less sand particles, possibly due to relatively low flow
rates allowing large and heavy particles to settle out before reaching the harbour. Large
particles found in these two sites were comprised of detritus and plant materials. Overall,
grab sediments from Inner Harbour sites contained mainly clay and silt particles while
those from the Outer Harbour were mainly silt and sand particles. This is not unusual for
such a harbour system. Fine sediment input from the catchment travels further into the
Outer Harbour but has longer residence time in the water column so has the opportunity to
return to deposit in Inner Harbour region by tidal action.
Other foreign materials observed in the surface samples included small worms, which were
only found in samples from Sites 11 and 12, and visible black granules, clearly coal
materials, abundant at Sites 9 and 10 and in lesser amounts at Sites 4 and 5. Yellow-red
granules and stones, clearly iron ore particulates, which broke up when lightly crushed by
fingers, were abundant at Site 4 and in a lesser quantity at Site 9. Site 20 and Site 7 samples
had a strong hydrocarbon smell. Sandy grains at Site 19a samples had a distinct red colour
indicating high iron content.
5.3.1.2
Selenium spatial distribution
The spatial distribution of selenium in whole surface sediments and < 63 µm fractions from
the 23 survey sites is shown in Figure 5.5. In whole surface sediments (Figure 5.5 (a)),
distinctly high selenium concentrations were found in samples from the Red Beach area
(Sites 18, 19 and 20) near the mouth of the Darcy Road Drain and progressively decreased
with distance away from this drain. This indicated that the copper smelter was a selenium
source via the drainage. However, whole surface sediments from other Outer Harbour sites
were found to contain less than 1 µg Se/g. Whole surface sediments from the Inner Harbour
sites contained slightly elevated selenium concentrations, especially at Site 3 and
85
(a)
Figure 5.5
(b)
Spatial distribution of selenium (µg/g, d.w.) in surface sediments from Port Kembla Harbour (a) whole sediments and
(b) < 63 µm fractions. Data are given in Appendix A (Table A.2).
Chapter 5 - Selenium distribution
86
Site 4 (see also Figure 5.6 (a)), which are in close proximity to the inflow of the iron
making east drain (NB: Iron ores can contain up to 3.0 % of selenium (Nazarenko, 1972)).
Figure 5.5 (b) shows that the < 63 µm sediments contained higher selenium concentrations
than whole sediments in general. Site 18 was found as a hot spot, indicating the ability of
the fine-grained sediments to travel further away from the Darcy Road drain mouth (Site
19). Particularly high selenium concentrations in < 63 µm sediments from Sites 12 and 13
were considered as possibly contaminated from the Port Kembla Coal Terminal, and at Site
14 from Bluescope Steel’s coal storage area. Illawarra regional coals have been shown to
contain 0.21-0.63 µg Se/g (Swaine, 1990).
Sediments of all grain sizes from several sites in Port Kembla Harbour had selenium
concentrations less than 1 µg/g (Site 15, 22 and 23). These are similar concentrations to
those found in pristine sites in Lake Macquarie (e.g., Nord’s Wharf, Croudace Bay,
Belmount Bay and Kilaben Bay) and those reported in Sydney continental shelf sediment,
North Head, Bondi and Malabar surface sediments, NSW and Peel Inlet and Harvey
Estuary sediments, WA (Table 2.4 and references therein). However, selenium
concentrations (up to 6.93 µg/g) in the fine grains (< 63 and 63-250 µm) from Red Beach
surface sediments are slightly lower than those reported for Chain Valley Bay and
Mannering Bay surface sediments, sites considered polluted from coal-fired power plants of
Lake Macquarie, which contained up to 10 and 12 µg Se/g, respectively (Kirby et al.,
2001a; Peters et al., 1999b). This is possibly because selenium input is currently continuing
in Lake Macquarie but has ceased in Port Kembla Harbour as a result of the closure of the
copper smelter (1994-2001, also from 2003).
The background selenium concentration has not been formally studied for Port Kembla
Harbour or for the Illawarra region. Considering our three lowest selenium concentrations
found in whole sediments of Site 15 inside the harbour and Sites 22 and 23 immediately
outside the harbour (0.10, 0.07 and 0.14 µg Se/g, respectively), a mean background
selenium concentration for the harbour area is estimated to be 0.1 µg/g. Concentrations of
selenium in most harbour sites are higher than this background value, indicating some level
of contamination.
Chapter 5 - Selenium distribution
5.3.1.3
87
Selenium distribution in different grain sizes
Selenium distributions in different grain sizes of the surface sediments are shown in Figure
5.6. The highest selenium concentrations (red colour bars) were found in < 63 µm fractions
from 18 sites out of the 23 survey sites. Figure 5.6 (b) clearly shows that < 63 µm fractions
contributed to selenium accumulation in most harbour sites per gram of whole surface
sediment, with the exception of Sites 19 and 20.
Larger particle size fractions (63-250 µm and > 250 µm), overall, contained lower but still
significant selenium concentrations. Interestingly, > 250 µm fractions from Site 4 (adjacent
to the iron ore pile) and Site 20 (near the copper smelter drain) contained higher selenium
concentrations (the latter site with large variation between oxic and anoxic samples) than
the <63 µm and 63-250 µm fractions. Original copper or iron ore materials abundant at the
two sites were retained in the large particle size fractions from sieving so they are believed
to contribute to the unusually high selenium concentrations. The large grain samples, by
nature, were inhomogeneous so this contributed to the high variation of results for > 250
µm samples.
5.3.1.4
Relationships with other trace elements
Concentrations of major acid-extractable trace elements (Cr, Cu, Fe, Mn, Ni, Pb and Zn) in
the surface sediments are given in Appendix A (Table A.2). Arsenic (As) and Antimony
(Sb) data were not included for the surface sediment study as the majority of the
concentrations were below the detection limits of the ICP-OES method. The extent of other
metal contamination in Port Kembla Harbour sediments and other harbour compartments
have been discussed in detail elsewhere (Goodfellow, 1996; Low, 1998; He and Morrison,
2001; Martley and Gulson, 2003; Beavington et al., 2004).
Relationships between selenium concentrations and acid-extractable trace metal
concentrations in the surface sediments were examined. The selenium data sets for the
contaminated Red Beach sites (18, 19 and 20) were outliers compared to other harbour
sites. However, they were considered important for this study so were not discarded. Two
separate correlations were therefore performed, with and without the Red Beach data sets.
Chapter 5 - Selenium distribution
88
8.00
(a)
10.88
7.00
Se ( µg /g )
6.00
5.00
4.00
3.00
2.00
1.00
0.00
1
2
3
4
5
6
7
8
9
10
11
12 13
14
15 16
17
18 19a 19b 20 21
22
23
Site
2.50
(b)
Se ( µg /g )
2.00
1.50
1.00
0.50
0.00
1
2
3
4
5
6
7
8
9
10
11
12 13
14
15
16 17
18 19a 19b 20
21
22
23
Site
Figure 5.6
Selenium concentrations (dry weight) in surface sediments from 23 sites of
Port Kembla Harbour (a) µg Se/g for each individual grain size fraction,
data points with error bars were means of oxic and anoxic results and those
with no error bars were of composite samples (b) µg Se in 1 gram of whole
sediment as a function of each grain size.
( whole sediment,
< 63µm,
63-250 µm, > 250 µm).
Chapter 5 - Selenium distribution
2.00
89
2.00
R2 = 0.6615
1.80
1.60
1.60
1.40
1.20
Se ( µg/g)
Se ( µg/g)
1.40
1.00
0.80
1.20
1.00
0.80
0.60
0.60
0.40
0.40
0.20
0.20
0.00
0.00
0
50
100
150
200
0
Cr (µg/g)
2.00
250
2.00
R2 = 0. 4636
1.80
1.80
1.60
1.60
1.40
1.40
1.20
1.00
1000
30
40
1500
2000
R2 = 0.5203
1.00
0.80
0.60
0.60
0.40
0.40
0.20
0.20
0.00
0.00
0
250
500
750
0
1000
10
2.00
2.00
R2 = 0.7901
1.80
20
Ni (µg/g)
Mn (µg/g)
R2 = 0. 7354
1.80
1.60
1.60
1.40
1.40
1.20
1.20
Se ( µg/g)
Se ( µg/g)
750
1.20
0.80
1.00
0.80
1.00
0.80
0.60
0.60
0.40
0.40
0.20
0.20
0.00
0.00
0
50
100
150
Pb (µg/g)
Figure 5.7
500
Cu (µg/g)
Se ( µg/g)
Se ( µg/g)
R2 = 0.1628
1.80
200
250
0
500
1000
Zn (µg/g)
Correlations between selenium and several trace elements in whole surface
sediments of Port Kembla Harbour, excluding Sites 18, 19 and 20.
Chapter 5 - Selenium distribution
90
Figure 5.7 shows correlations between selenium and other trace metals in whole surface
sediments from Port Kembla Harbour sites, excluding the Red Beach area (Sites 18, 19 and
20). Significant correlations (P < 0.0001) were found, in decreasing order, between Se-Pb
(r2 = 0.790), Se-Zn (r2 = 0.735) and Se-Cr (r2 = 0.662). Correlations of selenium and trace
elements in individual grain size fractions (> 250, 63-250 and < 63 µm) for the same
harbour sites found significant correlations between Se-Cu (r2 = 0.826), Se-Pb (r2 = 0.764),
Se-Zn (r2 = 643) in > 250 µm sediments (Appendix A, Table A.3). No correlations were
found in < 63 and 63-250 µm fractions. Copper, lead and zinc are chalcophilic and the
elements of high atomic number with which selenium prefers to bind (Nazarenko, 1972;
Beavington et al., 2004). The strong correlations in > 250 µm sediments indicated that the
selenium was associated with Cu, Pb and Zn in the selenium sources arriving from original
ore materials (also discussed above in Section 5.3.1.3). The correlations in the > 250 µm
fraction may also be a main contributor to selenium correlations in the whole sediments.
In a separate correlation analysis, the inclusion of the Red Beach data sets (Sites 18, 19 and
20) showed similarly significant correlations between Se-Cu (r2 = 0.774) and Se-Pb (r2 =
0.672) in whole sediments and between Se-Pb (r2 = 0.904), Se-Cu (r2 = 0.895) and Se-Ni (r2
= 0.819) in > 250 µm sediments (see Table 5.2). Again no correlations were found in < 63
and 63-250 µm fractions in this case (NB: Zn appeared to correlate with Cu, Ni, and Pb in
these fine grain fractions but not in > 250 µm fraction).
Overall, the selenium in the surface sediments was associated strongly with Pb, Cu, Zn,
which might be from the original pollution sources such as iron ores and discharges from
the copper smelter. Selenium was also associated with Ni (Red Beach area included) and
with Cr (excluding Red Beach). No selenium was associated with Fe and Mn in all cases.
No selenium correlations in < 63 µm fractions might indicate that the selenium, which
highly accumulated in the < 63 µm fraction through sediment processes, has different postdepositional behaviour to and not associated with the trace elements examined in this
section. However, it should be noted that the correlations were for the total selenium and
total acid-extractable metals only. The relationships between selenium–metals in different
geochemical phases might provide a better understanding of selenium post-depositional
behaviour in the sediment, which will be examined in Chapter 6.
91
Table 5.2
Correlations (r)* between selenium and common trace metals in different grain size fractions of surface sediments from
Port Kembla Harbour sampling sites, including Red Beach area.
> 250 µm (n = 24)
Whole surface sediment (n = 23)†
Se
Cr
0.187
Cr
Cu
0.880
Fe
0.096
Mn
-0.025
Ni
0.448
Pb
0.820
Zn
0.377
-0.023
0.855
0.786
0.910
0.568
0.761
-0.023
-0.169
0.239
0.722
0.171
0.880
0.753
0.513
0.717
0.663
0.368
0.727
0.691
0.691
Cu
Fe
Mn
Ni
Cu
0.946
Fe
0.159
Mn
-0.034
Ni
0.905
Pb
0.951
Zn
0.491
0.289
0.678
0.667
0.327
0.357
0.613
0.085
-0.088
0.984
0.991
0.421
0.821
0.141
0.155
0.541
-0.057
-0.005
0.549
0.970
0.370
0.707
Pb
Table 5.2
Cr
0.361
0.533
(continued).
63-250 µm (n = 22)‡
Se
Cr
Cu
Fe
Mn
Ni
Cr
0.066
< 63 µm (n = 22)‡
Cu
0.625
Fe
-0.114
Mn
-0.264
Ni
0.480
Pb
0.572
Zn
0.577
0.245
0.578
0.632
0.452
0.345
0.600
-0.029
-0.152
0.943
0.983
0.874
0.890
0.120
0.053
0.296
0.038
-0.071
0.198
0.968
0.894
Pb
* Values highlighted in bold are significant at P < 0.0001 level.
†
Excluding Site 20 outlier. ‡ No samples/data for Sites 15 and 22.
0.920
Cr
0.496
Cu
0.106
Fe
0.071
Mn
-0.374
Ni
-0.130
Pb
0.044
Zn
0.114
0.302
-0.297
-0.493
0.254
0.354
0.477
0.238
-0.237
0.862
0.925
0.880
0.562
0.032
0.096
0.179
-0.199
-0.250
-0.243
0.971
0.881
0.934
Chapter 5 - Selenium distribution
5.3.1.5
92
Preliminary hazard assessment
There are currently no standard guidelines to assess biological effects of sedimentary
selenium in Australia (ANZECC/ARMCANZ, 2000). The guideline protocol recommended
by the extensive selenium research from the USA was used in a preliminary assessment of
the potential selenium hazard found in Port Kembla Harbour surface sediments as
illustrated in Figure 5.8.
Please see print copy for Figure 5.8
Figure 5.8
Selenium concentrations in whole surface sediments from Port Kembla
Harbour. The red line indicates the 4-µg/g biological effect threshold, as
suggested by the USA research guidelines (Lemly, 1997a; Engberg et al.,
1998; NIWQP, 1998).
The guideline is applicable for interpretation of the biological effects of selenium in whole
surface sediments. The protocol indicates no hazard of selenium accumulation into the
benthic food chain from sediments containing < 1 µg Se/g (d.w.). However, concentrations
from 1-2 µg Se/g constitute a minimal hazard, 2-3 µg/g a low hazard, 3-4 µg/g a moderate
hazard, and > 4 µg/g a high hazard (Lemly, 1995; Lemly, 1997a; Engberg et al., 1998).
In Port Kembla Harbour, selenium concentrations in sediments from most sites (except 18,
19 and 20) were below 3 µg/g, indicating a low hazard. Sites 15, 22, 23 with selenium
concentrations of less than 1 µg/g, indicated no hazard for bioaccumulation of the element
Chapter 5 - Selenium distribution
93
into the benthic food chain. However, Site 19 contained a selenium concentration above the
biological effect threshold, and thus warrants further investigation.
There have been limited studies on biological effects of selenium within the harbour
environment as discussed in Chapter 2. Selenium was reported to be present at high
concentrations that exceeded the National Food Authority Maximum Residue Limit (1 µg/g
wet wt.) in fish tissue from Port Kembla Harbour and Allans Creek. However, the selenium
studies in the harbour organisms were limited to the two studies (Environmental Protection
Authority, 1994; Marine Science & Ecology, 1996) and a relationship between the
selenium contaminated sediments and high selenium concentrations in those organisms can
not be extrapolated at this stage.
During the course of this work, conversations with some local fishermen who have fished
in the harbour for the past 30 years revealed observations of severe physical deformities in
some fish species (such as bream) caught from the harbour during the 1970s and 1980s.
The symptoms could have been a possible biological effect of selenium contamination in
fish. These fish deformities were observed approximately 10-20 years after the first peak of
the refined copper production during the post World War II industrial boom (1955-1968,
see Figure 5.12) (Eklund and Murray, 2000). This was a time when pollution issues were
not fully in the public arena, and minimal action was taken to limit industrial discharges
until after 1970s (He and Morrison, 2001). Selenium could be one of the pollutants
responsible for the depletion of fish and other organisms in the harbour during 1970s.
However, a systematic scientific study on selenium biological effects, in comparison to
bioavailability and toxicity of other harbour pollutants is required for Port Kembla Harbour.
Chapter 5 - Selenium distribution
94
5.3.2 Selenium in Red Beach sediment cores
5.3.2.1
Sediment core characteristics and pH
The pH values of the sediment core samples were near neutral, ranging from 7.34 to 8.22
and found to increase slightly with depth. The full pH data of Cores A1-A3 are given in
Appendix B (Table B.2). Redox potentials of the sediment cores from Red Beach were
measured in the fractionation study and are discussed in Section 6.3.1.
Sediment cores collected from the Red Beach area near the inflow of the Darcy Road Drain
had a strong hydrocarbon smell. The surface layers (1-3 cm) appeared brown-grey as
compared to black colour of the lower sediment. In general, core sections with dominant
coarse textures appeared reddish in colour while sections with fine textures appeared black
and silty. Distinctly, Core A2 at 11, 13 and 15 cm depth contained heterogeneous bands
with hard yellow clay lumps, while the sediment above and below those sections appeared
the normal black and silty (measured grain size distributions of three Red Beach sediment
cores (A1-A3) are shown in Figure 5.10).
5.3.2.2
Sediment 210Pb dating results
There have been no previous studies on the sedimentation rate in the Red Beach area of
Port Kembla Harbour. In this study, to identify historical selenium input and accumulation
in the area, two Red Beach sediment cores (Core C4 and Core D1) were dated using the
210
Pb method, which is based on a measurement of a naturally occurring 210Pb radionuclide
formed as a product of the 238U decay series (Appleby and Oldfield, 1992). The total 210Pb
activity measured in sediments is a sum of the supported
210
Pb activity and excess
210
Pb
activity. The supported 210Pb activity results from the in situ decay of 238U radionuclides in
the sediment. The excess
222
210
Pb activity is produced by the decay of parent radionuclide
Rn that entered the sediment through sedimentation process from atmospheric
deposition, drains and catchment wash-in (Oldfield and Appleby, 1984; Spencer et al.,
2003). In practice, the total
210
Pb activity can be determined from the activity of 210Po and
the supported 210Pb activity can be determined from the activity of 226Ra. The excess 210Pb
activity can then be calculated from the difference between the two.
Chapter 5 - Selenium distribution
The depth profiles of excess
210
95
Pb activity in Core C4 and Core D1 from Red Beach area,
Port Kembla Harbour, are shown in Figure 5.9. Good linear relationships between the
excess 210Pb activity and most sediment depths were obtained for both cores, with r2 values
of 0.972 and 0.993 for 4-22 cm of Core C4 and 0-24 cm of Core D1, respectively. The
excess
210
Pb activity presented was normalised with <63 µm grain size as
210
Pb has been
reported to adsorb and accumulate in the fine grain size (He and Walling, 1996) (a good
linear relationship (plots not shown) was also obtained without the grain size normalization
with r2 values of 0.980 and 0.943 for 0-14 cm of Core C4 and 0-30 cm of Core D1,
respectively).
The results indicated that the deeper sediments from the study site were not disturbed or
mixed, thus supporting valid use of the
210
Pb dating technique to determine the
sedimentation rate and hence the sediment age. The sedimentation rate for both sediment
cores was calculated employing the Constant Initial Concentration (CIC) model with an
assumption of a constant input of the excess
210
Pb to the sediment or a constant
sedimentation rate (Brugam, 1978; Jha et al., 1999). The plots of the sediment age with
depth are shown in Figure 5.9. Using the <63 µm normalised excess
210
Pb data, the
sedimentation rate was determined to be 0.30 ± 0.03 cm/year for Core C4 and 0.55 ± 0.03
cm/year for Core D1. When using the whole sediment excess 210Pb data, the sedimentation
rates were 0.46 ± 0.05 cm/year and 0.53 ± 0.07 cm/year, respectively. The sedimentation
rates are higher than the rates calculated for two sites of Lake Macquarie: 0.13 ± 0.02
cm/year for Mannering Bay and 0.57 ± 0.09 cm/year in 0-7 cm and 0.15 ± 0.04 cm/year in
8-18 cm for Nords Wharf (Peters et al., 1999a). The higher sedimentation rate in Red Beach
area is considered to be typical of Port Kembla Harbour, which receives substantial
sediment inputs from a range of local creeks and drains (He and Morrison, 2001). They
were comparable to the sedimentation rates estimated for two fluvial bay-head deltas (0.33
cm/ year for Mullet Creek delta and 0.67 cm/year for Hooka Creek bay-head delta) of Lake
Illawarra, an estuarine lake located near Port Kembla Harbour (Sloss et al., 2004).
The
210
Pb dating results obtained in this study showed considerable confidence with good
linearity of the excess 210Pb activity with depth. However, the sedimentation rate estimated
using the CIC model might contain some uncertainty, as the model did not take into
96
Excess Pb-210 Activity (Bq/kg)
Excess Pb-210 Activity (Bq/kg)
1
10
100
1
1000
5
R2 (4-22 cm) = 0.9716
5
100
1000
R2 (0-24 cm) = 0.9928
10
Depth (cm)
Depth (cm)
10
0
0
10
15
15
20
25
30
20
35
25
40
Age (years)
0
20
40
Age (years)
60
80
0
100
20
40
60
80
0
0
5
Depth (cm)
Depth (cm)
5
10
15
10
15
20
25
30
20
35
25
40
Core C4
Figure 5.9
Core D1
Pb-210 dating of Core C4 and Core D1 (top) plots of excess Pb-210 activity (Bq/kg), normalized with < 63 µm grain
size, against depth (bottom) sediment age calculated from CIC model.
Chapter 5 - Selenium distribution
97
account real sedimentation dynamics, for example, vertical mixing and bioturbation
(Appleby and Oldfield, 1992; Matthai et al., 2001; Marques et al., 2006). In the local
region, there had been two major flood events, in 1984 (Huang and Nanson, 1997) and
1998 (Reinfelds and Nanson, 2004), that could have potentially caused a large influx of
sediment into Port Kembla Harbour and causing variation in the actual sedimentation rate.
In addition, Core C4 results obtained from 2-cm interval samples might contain larger
errors, as compared to Core D1 results obtained from 1-cm interval samples. The
sedimentation rate of Core D1 (0.55 ± 0.03 cm/year) is preferred for use in further
interpretation of other results in this thesis.
5.3.2.3
Selenium distribution in sediment cores
The depth profiles of selenium concentrations in three-grain size fractions (> 250, 63-250
and < 63 µm) of three Red Beach sediment cores (A1-A3) are shown in Figure 5.10. Total
selenium concentrations for the sediment cores are given in Appendix B (Table B.8).
Selenium concentrations in three sediment cores (Cores A1 to A3) varied, ranging from 6
µg/g to 1735 µg/g (d.w.) depending on depths and different particle sizes. Peak selenium
concentrations were found at 6-10 cm and at 14-16 cm depth for Cores A1/A2 and Core
A3, respectively. Similar to the surface sediments, selenium was found to accumulate in
<63 µm fractions in comparison to larger grain size fractions. The accumulation pattern of
selenium in fine grain fractions is more clearly observed in deeper sediment cores than in
surface sediments. The two highest selenium concentrations were found in <63 µm
fractions of Core A1 to be 1735 µg/g at 6-8 cm depth and 1620 µg/g at 8-10 cm depth. Core
A1 was closer to the Drain inflow than Cores A2 and A3, and was found to contain higher
selenium concentrations.
The selenium depth concentrations indicated clearly that the Red Beach sediments are
extremely contaminated, compared to Lake Macquarie sediment which has been reported to
receive selenium from the Pb-Zn smelter and coal-fired power stations (Peters et al.,
1999b). The highest selenium concentration found in the Core A1 (1735 µg/g) was 100
times higher than the highest sedimentary selenium concentration reported for Lake
Chapter 5 - Selenium distribution
98
% grain size
0%
50%
Se ( µg/g)
100%
0
1
Depth (cm)
Depth (cm)
7
9
2000
6
8
10
12
13
63-250 um
< 63 um
14
15
16
17
18
C ore A1
50%
C ore A1
0
100%
50
100
150
200
250
0
1
2
3
4
Depth (cm)
5
Depth (cm)
1500
4
5
11
7
9
11
6
8
10
Who le
63-250 um
< 63 um
12
13
14
15
16
17
18
19
20
C ore A2
0%
50%
80
160
240
320
400
0
2
3
4
5
Depth (cm)
7
9
11
13
Who le
63-250 um
< 63 um
6
8
10
12
14
15
16
17
18
19
20
21
22
C ore A3
Figure 5.10
C ore A2
0
100%
1
Depth (cm)
1000
2
3
0%
500
0
C ore A3
Grain size distribution and total selenium in three Red Beach sediment cores
< 63µm,
63-250 µm,
> 250 µm).
(Grain sizes:
Chapter 5 - Selenium distribution
99
Macquarie: 17.2 µg/g at 3-4 cm depth from Mannering Bay sediment (<100µm) (Peters et
al., 1999b). Red Beach sediment cores also contained higher selenium concentrations than
most reported for overseas soil/sediment depth profiles (Takayanagi and Belzile, 1988;
Tokunaga et al., 1991; Belzile et al., 2000; Wang and Chen, 2003). For example, selenium
concentrations reported for sediment cores from the lower St. Lawrence Estuary, Canada,
were very low (~0.75 µg/g d.w., whole) and constant along 30-cm sediment depth
(Takayanagi and Belzile, 1988). Selenium concentrations in freshwater sediment cores (15
cm) ranged from 0.16 to 11.8 µg/g (d.w., whole) in Clearwater Lake and McFarlane Lake
in Sudbury area (Belzile et al., 2000). Selenium depth concentrations of below 6 µg/g were
reported for a 170-cm soil profile of a vegetated site at Kesterson Reservoir (Tokunaga et
al., 1991) and a maximum of 16 µg/g concentrations were reported for wetland sediment
cores (30 cm) of Benton Lake, Montana (Zhang and Moore, 1996). In general, selenium
depth profiles peak at deeper depth for aquatic sediments (Zhang and Moore, 1996;
Tokunaga et al., 1997; Belzile et al., 2000) in comparison to peak concentrations being
found at surface layers of terrestrial soils (Tokunaga et al., 1991; Wang and Chen, 2003),
possibly due to loss of selenium into water column from the surface of aquatic sediments.
High selenium concentrations at depth strongly indicated a historical input, potentially from
the copper smelter, which was operated from 1908-1995 and 2000-2003, and in the early
years, there was no control of pollutants discharged into the harbour. Selenium was a
certain component released by the copper smelter (Cleland, 1995). It is also known that the
anode sludge formed during electrolytic refining of crude copper at copper smelting plants
contains approximately 3-14 % selenium (Nazarenko, 1972). The selenium might be
released from the copper smelter into the harbour via wastewater discharge (the Darcy Rd
Drain) and via stack emission. Southern Copper Ltd. was permitted, under a NSW EPA
license, to discharge a primary treated wastewater containing up to 1 mg Se /L (Cleland,
1995). A historical selenium input from the copper smelter is examined in detail, in
conjunction with the sediment dating information below, in Section 5.3.2.5.
Compared to the surface sediment results in Section 5.3.1.3, selenium concentrations in the
Red Beach deeper sediments were distinctly higher than those in the surface sediments; it is
therefore possible that sediment cores from other harbour sites, e.g., Sites 3, 4, 12, 13 and
Chapter 5 - Selenium distribution
100
14, may also contain selenium concentrations comparable to those in the Red Beach cores.
In this study, core sampling was limited to the Red Beach area, as significant selenium
concentrations were found in the grab samples there and the sediments are relatively
undisturbed, compared to sediments from other harbour areas that have been dredged from
time to time for breakwater/berth construction or for ship navigation. Sediment cores from
this area could be sampled using a hand-held corer at low tide (with the water depth of
approximately 3-5 m). Deeper water in other harbour areas required core sampling through
scuba diving, which was not possible at the time due to port security and safety issues.
5.3.2.4
Relationships with other elements in core sediments
In the surface sediment study (Section 5.3.1.4), selenium was found to correlate with Pb,
Cu and Zn in the whole surface sediments and mainly in > 250 µm fraction. The
relationships between selenium and simultaneously aqua regia-extractable metals in the
core samples were also examined. For the core sediments, the concentrations of As, Cd and
Sb were sufficiently high to be included in this assessment. The data sets for 63-250 µm
fraction and whole sediment were incomplete, as some samples were below detection
limits. Therefore, only < 63 µm data sets are examined.
Depth profiles of selenium concentrations in Cores A1-A3, in comparison to those of other
elements are shown in Figure 5.11 (data given in Appendix B, Table B.18). The trace
element correlation matrices for Cores A1-A3 batch and for individual cores are listed in
Table 5.3. Similar to the surface sediment results in general, significant correlations were
found between Se-Cu (r2 = 0.787), Se-Zn (r2 = 0.671) and Se-Pb (r2 = 0.627) in < 63 µm
sediments of Cores A1-A3 (n = 30), indicating possible association of the selenium with the
three elements through sediment deposition processes (including coming from the same
pollution sources and at similar deposition time). Their relationships were not observed in <
63 µm fraction of the surface sediments, possibly due to the sediment age and redox
conditions of the sediment. It is possible that in the sediment cores, selenium and trace
metals were co-deposited for a sufficiently long period under reducing conditions to allow
geochemical interaction to occur. Such interaction might not occur in oxic and recently
deposited surface sediments.
101
As (µg/g)
Se (µg/g)
0
100
200
300
0
400
600
Cd (µg/g)
800
1000
0
0
2
2
2
4
4
6
6
x5
Depth (cm)
6
8
10
12
Depth (cm)
0
8
10
12
16
16
16
18
18
20
20
22
22
22
Cu (mg/g)
Cr (µg/g)
0
50
100
150
200
250
0
300
10
20
40
3. 0
2
2
2
4
4
4
6
6
6
8
12
10
12
Depth (cm)
0
Depth (cm)
0
10
100
4. 5
6. 0
7. 5
9. 0
8
10
12
14
14
14
16
16
16
18
18
18
20
20
20
22
22
22
Figure 5.11
80
Fe (%)
30
0
8
60
12
14
20
40
10
14
Co re A1
Co re A2
Co re A3
20
8
14
18
Depth (cm)
400
0
4
Depth (cm)
200
Depth concentration profiles of selenium and other trace elements in Red Beach (<63 µm) sediments, Cores A1A3 (cont.).
102
Se (µg/g)
100
200
Mn (µg/g)
300
400
0
150
450
Ni (µg/g)
600
750
900
0
0
2
2
2
4
4
4
6
6
x5
8
10
12
16
18
Co re A1
Co re A2
Co re A3
20
22
8
10
12
0. 0
3. 0
6. 0
9. 0
16
18
18
20
20
22
22
0
2
4
8
10
12
Depth (cm)
6
50
100
150
0. 0
200
0
0
2
2
4
4
6
6
8
10
12
500
600
2. 0
4. 0
6. 0
8. 0
10.0
8
10
12
14
14
14
16
16
16
18
18
18
20
20
20
22
22
22
Figure 5.11
400
Zn (mg/g)
Sb (µg/g)
0
300
12
14
12.0
200
10
16
Pb (mg/g)
100
8
14
Depth (cm)
6
Depth (cm)
0
14
Depth (cm)
300
0
Depth (cm)
Depth (cm)
0
Depth concentration profiles of selenium and other trace elements in Red Beach (<63 µm) sediments, Cores A1-A3.
103
Table 5.3
Correlations (r)* between selenium and other trace elements in Red Beach sediment (<63 µm) Cores A1-A3.
Cores A1-3 (n = 30)
Se†
As
0.550
As
Cd
0.755
0.678
Cd
Cr
0.133
Cu
0.887
Fe
-0.059
0.129
0.621
0.430
0.089
0.596
0.909
0.111
0.776
0.070
-0.245
0.499
0.837
0.193
0.381
0.197
0.114
0.231
0.145
-0.129
-0.320
0.840
0.779
0.584
0.893
0.815
-0.230
0.301
0.405
0.148
Se-Sb, P < 0.001
-0.333
-0.053
0.000
-0.160
Se-Cd, P = 0.002
0.598
0.390
0.695
0.925
0.895
Cr
Cu
Fe
Mn
-0.250
Core A1 (n = 9)
Mn
Ni
0.611
Ni
Pb
0.792
Pb
Sb
0.580
Zn
0.819
As
Cd
Cr
Cu
0.872 0.873 0.473 0.963
Fe
0.025
0.899
0.725
0.693 0.445 0.953
-0.060
0.837
0.811
0.367 0.820
0.003
0.456
0.541
-0.749
-0.954
-0.056
-0.482
0.783
0.980
0.896
0.976
0.833
-0.339
-0.052
0.121
-0.134
-0.548
-0.466
-0.281
-0.551
0.730
0.503
0.712
0.942
0.985
Sb
Cd
Sb
0.927
Zn
0.966
-0.412
0.903
0.912
0.790
0.896
-0.349
0.417
0.890
0.967
0.894
0.509
0.488
0.328
0.592
0.937
(continued).
As
0.232
Core A3 (n = 11)
Cd
0.282
Cr
0.493
Cu
0.736
Fe
0.069
Mn
0.318
Ni
0.542
Pb
0.309
Sb
0.169
Zn
0.846
0.819
0.072
0.227
0.799
0.642
0.271
0.974
0.958
0.556
0.267
0.322
0.756
0.540
0.107
0.845
0.865
0.640
0.407
0.695
0.160
0.245
0.120
0.670
-0.071
0.078
0.871
0.190
0.051
0.765
0.835
-0.111
0.867
0.873
0.476
0.096
0.767
0.690
0.637
Se-Ni, P = 0.001
0.179
0.017
0.589
Se-Zn, P = 0.007
0.323
Cr
Cu
Fe
Mn
Se-Zn, P = 0.002
Ni
Se-Cu, P = 0.015
Pb
Pb
0.982
Se-As, P = 0.002
Core A2 (n = 10)
As
Ni
0.670
0.759
Table 5.4
Se†
Mn
-0.452
0.977
Sb
* Values highlighted in bold are significant at P < 0.0001 level.
†
For information, r-values in italics contain significance at P levels listed on the tables.
0.629
0.497
As
0.544
Cd
-0.002
Cr
0.527
Cu
0.564
Fe
0.552
Mn
0.442
Ni
0.836
Pb
0.630
Sb
0.629
Zn
0.754
-0.089
0.313
0.463
0.829
0.891
0.673
0.970
0.932
0.716
0.525
0.370
0.264
0.187
0.128
0.031
0.149
0.306
0.950
0.704
0.464
0.771
0.454
0.623
0.855
0.802
0.606
0.861
0.596
0.709
0.905
0.930
0.779
0.871
0.921
0.900
0.641
0.886
0.880
0.778
0.802
0.821
0.937
0.960
0.818
0.891
Chapter 5 - Selenium distribution
104
For Core A1, additional correlations were observed between Se-Sb (r2 = 0.859), Se-Cd (r2 =
0.762), and Se-As (r2 = 0.760). No such correlations were found in Cores A2 and A3. Core
A1 also showed very strong relationships between Se-Pb. Se-Zn and Se-Cu, while selenium
correlations were not observed with those three elements in Cores A2 and A3. The possible
explanation for the differences in the correlation results between Core A1 (collected in
February 2003) and Cores A2/A3 (collected in June 2003) was the use of a N2 glove box
for sample processing (extruding, cutting and wet-sieving). Core A1 was processed inside a
glove box, and Cores A2 and A3 were processed outside the glove box. Although all
samples were treated similarly in further analysis (e.g., oven-dried, ground and aciddigested under ambient conditions), a rapid change of the sediment anoxic conditions to
oxic conditions during sample processing might be a crucial factor, leading to a change in
selenium-metal geochemical association in Cores A2 and A3.
5.3.2.5
Factors affecting the selenium vertical distribution
The deeper sediments of Red Beach cores were found to contain much higher selenium
concentrations than the top 5 cm layers and in the grab surface samples (Sites 19 and 20).
This possibly resulted from historical selenium input from nearly 100 years of operation of
the copper smelter. Using the sedimentation rate of 0.55 cm/year (Core D1) from the
sediment
210
Pb dating results, the age of Red Beach sediment cores could be estimated.
Figure 5.12 compares the sediment age with the selenium vertical profiles and the annual
refined copper production by the copper smelter (Electrolytic Refining and Smelting
Company of Australia Limited (ER&S)/Southern Copper Limited (SCL)).
A strong peak of selenium concentrations was found in sediment deposited during 19871994 (6-10 cm), corresponding considerably with the peak copper production period during
1987-1994. The second selenium peak during 1976-1980 (14-16 cm) did not correspond
directly with the high copper production period and might possibly be due to other factors
such as vertical remobilization of selenium from deeper sediments (potentially receiving
pollution input from high copper production during 1955-1975 period). In addition, there
might be some uncertainty in the sediment age estimated from the 210Pb dating technique as
discussed in Section 5.3.2.2.
Chapter 5 - Selenium distribution
105
Please see print copy for Figure 5.12
Figure 5.12
Vertical profiles of selenium concentrations
sediment cores (Cores B1-6 and C1-4 data are
corresponding sediment age (Year) determined
against the annual refined copper production
Eklund and Murray, 2000).
(mean±SE) in Red Beach
taken from Chapter 6). The
from 210Pb dating is plotted
by ER&S/SCL (data from
During the 1987 to 1994 period, the copper smelter was operated under the management of
Southern Copper Limited. Pollution reduction programs were implemented as part of the
NSWEPA license requirements to decrease the release of contaminants including selenium
into surrounding environment (Cleland, 1995). However, a significant increase in the
copper production capacity during 1992-1994 might have nullified the effects of the
pollution reduction programs and led to high selenium input into the harbour.
The decommissioning of the copper smelter in 1995 led to minimal input of new selenium
from the plant into the harbour sediment. The selenium concentrations found in the top 5
cm sediment could mainly be as a result of selenium mobilization from the deeper
sediment. According to the selenium core profiles in Figure 5.12, the < 63 µm sediments
(Cores B1-6) contained peak selenium concentrations in the upper core region. This
indicated possible upward mobilization of selenium through an upward movement of < 63
µm sediments (NB: horizontal movement of < 63 µm sediments was observed in surface
Chapter 5 - Selenium distribution
106
sediments in Section 5.3.1.2). The relatively low selenium concentration found in the
surface layer might be due to loss of selenium into the overlying water column because of
the Se solubility and the high concentration gradient across the sediment-water interface.
In addition to the historical selenium input and sediment grain size factors, several other
physical, chemical and biological processes might also affect selenium distribution in
sediments. Further understanding of the selenium vertical distribution and depositional
transformation in Red Beach sediments is necessary. In addition, with the highly
contaminated selenium deposited not far below the sub-oxic layer, there is potential that the
selenium can be mobilized from the deeper sediments when environmental conditions
become favourable (such as introduction of oxygen via sediment dredging). A further
geochemical study of selenium in the sediment cores was carried out and presented
following in Chapter 6.
5.4
Conclusions
The investigation of selenium contamination in Port Kembla harbour surface sediments and
Red Beach sediment cores found selenium concentrations in surface sediments from most
harbour sites to be below 3 µg/g except those in sediments from the Red Beach area (up to
9.38 µg/g), which is in close proximity to the copper refinery. Selenium concentrations in
Red Beach sediment cores ranged from 6 to 1735 µg/g, depending on depth and grain size,
with peak selenium concentrations observed at 6-10 cm (Cores A1/A2) and at 14-16 cm
depth (Core A3). Pb-210 dating of the sediment cores indicated a likely historical selenium
input potentially from the copper smelter. Overall, selenium was concentrated in fine grain
size (<63 µm) sediments with high mobilization potential horizontally in surface sediments
and vertically upward in sediment cores. Selenium was correlated mainly with Pb, Cu and
Zn, in the > 250 µm fraction of the surface sediments and in the < 63 µm fraction of the
sediment cores, indicating possible association from both original ore sources and through
post-depositional transformation. The study provided good quality datasets on sedimentary
selenium in Port Kembla Harbour, especially for the Red Beach (hot spot) area, and
provided an interesting overview of selenium behaviour in this contaminated sediment
system.
Chapter 6
Geochemistry of selenium in contaminated marine sediments – Red
Beach, Port Kembla Harbour
6.1 Introduction
The investigation of selenium contamination in Port Kembla Harbour sediments reported in
Chapter 5 indicated that the deeper sediments from the Red Beach area were highly
contaminated with selenium that had a potential to mobilize upward into the overlying
water column. The study reported in this chapter investigated the solid-phase speciation and
the binding phases of selenium in Red Beach sediment cores in an attempt to understand its
geochemical transformations after deposition in the sediments. The information obtained
was used to predict the potential mobility and bioavailability of selenium.
A sequential extraction procedure (SEP), i.e., selective chemical extraction or fractionation,
is commonly used to characterize solid-phase speciation and the geochemical phase
distribution of trace metals in soils/sediments (Tessier et al., 1979; Cutter, 1985; Batley,
1987; Tokunaga et al., 1991; Martens and Suarez, 1997; Gao et al., 2000; Bujdos et al.,
2005). The technique is based on the use of a series of selective reagents (with increasing
chemical strength) to successively solubilise the different mineralogical fractions
responsible for binding trace elements (Kersten and Forstner, 1989; Gleyzes et al., 2002).
Different metal fractions associated with specific sediment phases are defined by their
target extractants under specific operational/extraction conditions, such as reagent
concentrations, shaking mechanism, temperature, and extraction time, depending on the
sequential extraction scheme used. Common sequential extract fractions are: soluble metal
fraction; exchangeable surface-adsorbed fraction; carbonate-bound elements; elements
associated with iron-manganese oxyhydroxides; elements bound to organic matter and
sulfide minerals; and residual fraction or elements retained within mineral silicate and
crystalline structure of the soils/sediments (Tessier et al., 1979; Batley, 1987; Tokunaga et
al., 1991). Specifically for selenium, a labile organic fraction (sodium hydroxide
extraction) and elemental selenium fraction have been incorporated into a conventional
Chapter 6 – Selenium geochemistry
108
sequential extraction scheme in the recent literature (Velinsky and Cutter, 1990; Zhang and
Moore, 1996; Wright et al., 2003; Zawislanski et al., 2003), as selenium has different
chemical properties to common cationic elements, which requires a different extraction
chemistry. Common sequential extraction schemes employed in the recent literature to
extract selenium from soil and sediment phases are summarized in Table 6.1.
It should be noted that the sequential extraction techniques may have some limitations
regarding their operationally-defined nature with: poor precision and reproducibility; a lack
of available standardization and reference materials; and possible re-adsorption and
redistribution of metals between phases during extraction (Sheppard and Stephenson, 1995;
Gleyzes et al., 2002). Nevertheless, they are useful tools for metal studies in
soils/sediments. The information on metal fractions may be used to assess their
geochemical behaviour as well as their relative mobility (Velinsky and Cutter, 1991;
Lussier et al., 2003; Zawislanski et al., 2003). The relative mobility of elements in
soils/sediments decreases as the strength of the extractant reagents increases along the
conventional extraction sequence, i.e., the early fractions are considered more mobile than
the later fractions, with the final residual fraction considered as immobilized forms of
elements.
In the literature, a major flaw of many selenium speciation studies in soil/sediments was the
sample preparation step, for example, use of dried-ground samples (Velinsky and Cutter,
1991; Bujdos et al., 2000; Wang and Chen, 2003; Zhang and Frankenberger, 2003). Sample
drying before a sequential extraction has been reported to increase the soluble and adsorbed
selenium and decrease selenium concentrations in organic fractions (Zhang and Moore,
1996). Therefore, in this study, special care was taken during the sample collection and
preparation steps to preserve sample integrity, which assisted in validating the overall study
objectives.
Soluble selenium fraction
Soluble selenium is considered mobile and readily available to organisms upon exposure.
To extract soluble selenium from soils/sediments for quantification, several authors have
109
Table 6.1
Common sequential extraction procedures employed in the literature to extract selenium from soils/sediments.
Please see print copy for Table 6.1
Chapter 6 – Selenium geochemistry
110
utilized water (Martens and Suarez, 1997; Lussier et al., 2003; Zhang and Frankenberger,
2003) or salt solutions, such as potassium chloride (KCl) (Tokunaga et al., 1991; Gao et al.,
2000; Wright et al., 2003; Zawislanski et al., 2003) and magnesium chloride (MgCl2)
(Nobbs et al., 1997; Peters et al., 1997). The salt solutions help facilitate dissolution of
evaporite salts such as halite, gypsum and anhydrite (MacGregor, 1997). The original
selenium species extracted by water and these salt reagents are chemically preserved and
can be individually speciated. The major selenium species reported for the soluble fraction
were selenate and selenite with some reduced (0, -II) species present in lesser quantities
(Zhang and Frankenberger, 2003).
Exchangeable or weakly adsorbed selenium
This fraction includes selenium adsorbed or bound to the surfaces of sediment particles; the
relative mobility and availability is dependent on the change in water ionic composition
(Tessier et al., 1979; Batley, 1987). A phosphate buffer (pH 7-8, 0.1-1 mol/L) is commonly
used to extract adsorbed selenium from soils/sediments (see Table 6.1). Phosphate ions are
known to competitively displace selenium anions at the sediment adsorption sites releasing
selenium into the solution (Jackson and Miller, 2000; Pezzarossa and Petruzzelli, 2001;
Goh and Lim, 2004). Buffering the solution at pH 7-8 is suitable to prevent selenium
leaching from carbonate phases and pH 8 is preferred as it is greater than the Point of Zero
Charge (PZC) of iron-manganese oxyhydroxide surfaces making their surface charge
negative so repelling selenium anions (Stumm, 1992; Blackmore, 2002). Higher pHs are
considered unsuitable due to possible dissolution of organic matter (Lipton, 1991;
MacGregor, 1997).
Selenium associated with carbonate minerals
The extraction procedure for trace elements associated with carbonate minerals was derived
from a conventional selective extraction procedure such as Tessier et al. (1979). Sodium
acetate (NaOAc) solution adjusted to pH 5 with acetic acid has also been used to extract
selenium from soils/sediments (Lipton, 1991; Tokunaga et al., 1991; MacGregor, 1997;
Gao et al., 2000). The acetic acid attacks the carbonate minerals in the sediment. The
Chapter 6 – Selenium geochemistry
111
NaOAc provides buffering conditions while the carbonate is being dissolved by the acetic
acid (Lipton, 1991; MacGregor, 1997).
Selenium adsorbed or coated onto iron-manganese oxyhydroxide particles
As with the carbonate fraction, the extraction procedure for selenium associated with
amorphous (or short-range order) iron-manganese oxyhydroxides has been derived from
conventional selective extraction procedures for common trace metals, which used a
reducing agent (hydroxylamine hydrochloride (NH2OH.HCl)) to reduce amorphous iron
and manganese oxyhydroxides and release the selenium into the solution (Tessier et al.,
1979; Batley, 1987; John and Leventhal, 1995). Selenium, released by such reductive
dissolution and present as oxyanions, was reported to readsorb onto other mineral phases
with positive charged surfaces that result from the low pH nature (pH 2) of the extractant
(Gruebel et al., 1988; Lipton, 1991). A subsequent extraction by a phosphate buffer (pH 8)
has later been employed to extract readsorbed selenium, which is then combined into the
NH2OH.HCl extract to rectify the problem (Lipton, 1991; Tokunaga et al., 1991; Gao et al.,
2000).
Hydrochloric acid at 4 mol/L has been used to extract selenium associated with crystalline
oxyhydroxides (Lipton, 1991; Tokunaga et al., 1991; Sharmasarkar and Vance, 1995;
Zhang and Moore, 1996). The crystalline oxyhydroxide fraction is not common in recent
sequential extractions as only small quantities of selenium were found and reported by the
above authors to be associated with this fraction.
Elemental selenium
The extraction of elemental selenium was first developed by Velinsky and Cutter (1990);
this involved the use of sodium sulfite (Na2SO3) solution at 1 mol/L (pH 7) to leach
elemental selenium after labile selenate and selenite species have been removed. The
procedure has been incorporated into the selective chemical extraction scheme for selenium
by several researchers (Zhang and Moore, 1996; Wright et al., 2003; Zawislanski et al.,
2003). Wright et al. (2003) have evaluated a Na2SO3 extraction and reported it to be an
Chapter 6 – Selenium geochemistry
112
acceptable procedure for extraction of elemental selenium, although it could co-extract up
to 17% of metal selenide from their model sediments.
Selenium associated with organic matter and sulfide minerals
The selenium and metals associated with organic matter and sulfide minerals may be
remobilized when the two sediment components become oxidised. To solubilise the
selenium in the organic/sulfide phase, a strong oxidizing agent such as sodium hypochlorite
(NaOCl) is frequently used (see Table 6.1). Other strong oxidizing agents used included
hydrogen peroxide (H2O2) (Nobbs et al., 1997; Peters et al., 1997); potassium persulfate
(K2S2O8) (Martens and Suarez, 1997; Bujdos et al., 2005); and potassium chlorate (KClO3)
(Sharmasarkar and Vance, 1995; Lussier et al., 2003). It should be noted that H2O2 was not
suitable for the work in this thesis as it was previously found to interfere with selenium
analysis by HG-AAS. Due to oxidative dissolution of the sediment organic and sulfide
phases, selenium released in this extract would be in an oxidised form, mainly selenate
(Gruebel et al., 1988).
Sodium hydroxide (NaOH), which is used independently in selenium extraction for
speciation studies (Seby et al., 1997; Sharmasarkar and Vance, 1997; Zhang et al., 1999a),
has been incorporated into sequential extraction procedures (generally after adsorbed
selenium fraction) to extract selenium in the labile organic matter fraction (Cutter, 1985;
Velinsky and Cutter, 1991; Wright et al., 2003; Zawislanski et al., 2003). Sodium
hydroxide extractions preserve selenium species and further speciation of individual
compounds in this extract has been reported (Seby et al., 1997; Zhang et al., 1999a).
Selenium associated with the crystalline structure of minerals
The selenium associated with mineral silicates and other crystalline components of
sediments is defined as the residual fraction and is obtained by digesting samples using a
strong acid solution as in a total selenium determination procedure in general (see Table
6.1).
Chapter 6 – Selenium geochemistry
113
This study employed the selective chemical extraction techniques as a tool to investigate
the geochemistry and early diagenetic behaviour of selenium in Red Beach sediment cores.
Two methods (SEP 1 and SEP 2, Figure 6.2) modified from the most suitable selenium
extraction procedures, Tokunaga et al. (1991) and Wright et al. (2003), respectively, were
chosen for the selenium fractionation. The specific aims of the study were to:
•
investigate solid-phase speciation and binding phases of the selenium in Red Beach
sediment cores using sequential extraction procedures (detailed below in Section 6.2.3)
and HG-AAS analysis. The operationally defined selenium phases were (1)
exchangeable and adsorbed; (2) acid soluble fraction; (3) organically bound selenium;
(4) reducible fraction; (5) elemental selenium; (6) oxidisable fraction; and (7) refractory
or residual fraction;
•
measure the concentrations of other trace elements (Cr, Cu, Fe, Mn, Ni, Pb and Zn) coextracted in the above sequential fractions using ICP-OES and examine the relationship
between selenium and those trace elements co-extracted in the sequential extracts;
•
measure selenium concentrations, anion species and trace metals present in the Red
Beach sediment pore waters;
•
measure sediment parameters including pH, redox potential, grain size compositions
(<63, 63-250 and >250 µm fractions), and macro-components (total sulfur, acid volatile
sulfides, pyrites, total carbon, total organic carbon, and total nitrogen); and
•
examine the relationships between these sediment parameters and selenium in different
sequential fractions.
The data analysis (cluster and correlation analysis) was performed using the JMP 5.1
statistics program (SAS Institute, Inc). The information obtained was used to assess
selenium geochemical and diagenetic behaviour, and to predict its relative mobility and
biological availability in contaminated sediments.
Chapter 6 – Selenium geochemistry
6.2
Materials and methods
6.2.1
Reagents and apparatus
114
All glassware and plastic containers were acid-cleaned as described previously in Section
3.2.1. Chemicals and reagents were of analytical reagent grade or better.
For sequential extraction work, the phosphate buffer (0.1 mol/L, pH 8) was prepared by
dissolving 22.82 g of K2HPO4.3H2O in 1 L MilliQ water. Initial pH 9 was adjusted to pH 8
using dilute HCl. The pH 10-phosphate buffer was achieved using KOH. Sodium acetate (1
mol/L, pH 5) was prepared by dissolving 82.03 g in 1 L MilliQ water, and the initial pH 7.8
was adjusted to pH 5 using acetic acid. NaOCl (5%) solution was purchased from
Australian Chemical Reagents or prepared by diluting 50 mL of NaOCl (SIGMAALDRICH®, Cat No. 239305) in 1 L MilliQ water, and the initial pH 11.5 was adjusted to
pH 9 with dilute HCl. NH2OH.HCl (0.25 mol/L) was prepared by dissolving 17.3725 g of
NH2OH.HCl (UNIVAR, APS) in MilliQ water in a 1-L volumetric flask. Sodium sulfite
solution (1 M, pH 7) was prepared by dissolving 31.51 g of Na2SO3 (SIGMA®, Sigma
Ultra, > 98 %, S4672) in 250 mL MilliQ water and the pH was adjusted using dilute HCl.
NH2OH.HCl solution (1 mol/L) for total reactive iron extraction was prepared by
dissolving 69.49 g in 1 L MilliQ water containing 25% acetic acid. Extractant reagents
were stored at room temperature in glass reagent bottles (Schott, Q Stores). Deoxygenated
seawater and reagents were obtained by bubbling N2 gas through solutions in a glove box
for approximately 1 hour before use.
For anion analysis by ion chromatography, a Na2CO3/NaHCO3 mobile phase (8 mM/1 mM)
was prepared by dissolving 16.96 g of Na2CO3 and 1.68 g of NaHCO3 in 1 L to obtain a
concentrate 0.16 M Na2CO3/ 0.02 M NaHCO3. The concentrated solution (44 mL) was
diluted to 1 L with MilliQ water and filtered through a 0.45 µm membrane filter before use.
H2SO4 regenerant (0.05 mol/L) was prepared by adding 55 mL of conc. H2SO4 to 1 L water
to obtain an intermediate 1 mol/L H2SO4 solution, and then 50 mL of the intermediate
H2SO4 solution was diluted to 1 L. Anion standard stock solutions (1000 mg/L) were
prepared by dissolving appropriate weights of dry salts: 1.4997 g NaNO2, 1.3708 g NaNO3,
Chapter 6 – Selenium geochemistry
115
1.4179 g KH2PO4, and 1.4786 g Na2SO4 in 1 L MilliQ water. Preparations of other reagents
and standards were as described in Section 3.2.1 and Section 4.2.1.
6.2.2
Sample collection and analysis
Ten sediment cores were collected from the Red Beach area near the mouth of Darcy Road
Drain as described in Chapter 5 (Section 5.2.2). The sediment core details including GPS
positions and sample treatment and analysis are given in Table 6.2. The sediment cores
were extruded and cut into 2-cm sections. Each section was homogenized and divided into
four sub-samples for: (1) pH and redox potential measurement with the samples discarded
afterward; (2) un-sieved samples; (3) macro-component TC, TOC, TS and AVS analysis
(frozen below 0˚C); and, (4) the remaining samples were segregated for porewater
extraction, wet-sieving and subsequent fractionation work. Sample processing steps were
carried out in a nitrogen-purged glove box to preserve the anoxic conditions of the samples.
The core sample preparation flowchart is shown in Figure 6.1.
Sediment pHs were measured using a portable Rex pH meter (Shanghai Rex Instrument
Factory, Model pHB-4), calibrated with pH 7 and pH 10 buffer solutions prior to use.
Redox potentials of the core samples were measured using an Orion redox meter during
sediment core sectioning in a nitrogen glove box. The redox probe was calibrated with
quinhydrone in buffer pH 4 and pH 7 solutions according to the manufacturer instructions.
Porewater separation was carried out by transferring sediments into a 50-mL polypropylene
centrifuge tube in a N2 glove box, capping tightly and centrifuging at 2400 rpm (maximum
speed allowed) for 20 min (Bufflap and Allen, 1995). The supernatant was decanted and
filtered through a 0.45 µm PTFE-membrane unit (Minisart SRP25, Sartorius AG
Germany). Porewater samples were preserved in HCl and stored under 4˚C until analysis.
Samples awaiting porewater extraction were stored in a refrigerator below 4˚C (Jung and
Batley, 2004). Porewater samples were analysed for total selenium concentrations by HGAAS and anions (no HCl preservation) using ion chromatography (Dionex ICS90 Ion
Chromatograph) equipped with suppressed electrical conductivity detector (DIONEX®
DS5 Detection Stabilizer Model DS5) and DIONEX IonPac®, AS14A-5µm 3x150mm
116
Table 6.2
Sampling
Date
16/4/04
29/7/05
Summary of core samples collected for selenium fractionation studies.
Core
GPS
UTM
(E/N)
Length
(cm)
pH
20
~ Distance
from the
Darcy Rd
Drain (m)
90
B1
030 7890
618 3200
B2
Sample treatment and analysis
030 7875
618 3183
14
B3
030 7812
618 3212
B4
Sieved
Porewater
Total Se
SEP 1
SEP 2
x
Redox
potential
x
x
x
x
x
-
Macrocomponents
-
65
x
x
x
x
x
x
-
-
14
80 West
x
x
x
x
x
x
-
-
030 7891
618 3220
16
110
x
x
x
x
x
x
-
-
B5
030 7904
618 3193
14
100 East
x
x
x
x
x
x
-
-
B6
030 7832
618 3214
20
70
x
x
x
x
x
x
-
-
C1
030 7791
618 3252
18
120 West
x
x
-
x
x
-
x
x
C2
030 7784
618 3251
14
120 West
x
x
-
x
x
-
x
x
C3
030 7971
618 3192
16
150 East
x
x
-
x
x
-
x
x
C4
030 7925
618 3186
24
90 East
x
x
-
x
x
-
x
x
117
Sediment core
Macro-component
analysis: TC,
TOC, TN, TS,
AVS, CrRS
Un-sieved
samples
Cut to 2-cm sections and homogenised
pH and redox
potential (Eh)
Porewater extraction by
centrifugation at 2400 rpm for 20 min
Porewater
Sediment
Homogenised or wet sieved to
collect <63 µm sediment then homogenised
Anions
by IC
Total Se by
HG-AAS
Figure 6.1
Metals by
ICP-OES
Sequential extraction
· Adsorbed
· Carbonate
· Organically bound
· Fe/Mn oxides
· Seo
· SOM and sulfides
· Residual
Oven dried at 40°C
Microwave aqua regia digestion
Total Se by HG-AAS
Moisture
Sediment core sample preparation and analysis flowchart. Black text: samples were processed in an N2 purged
environment; red colour region: outside a glove box. SOM: Sediment Organic Matter.
Chapter 6 – Selenium geochemistry
118
analytical column, according to the in-house standard procedure (Mobile phase: 8 mM
Na2CO3/1 mM NaHCO3; Regenerant: 0.1 N H2SO4). Only phosphate and sulfate were
detected and quantifiable with the retention times of 6.70 min and 8.86 min, respectively.
The percentage relative standard deviation of triplicate analyses of selected samples was
equal to or less than 5% and the recovery of the spiked standard (50 mg/L) was 88 % for
both anions.
After porewater extraction, the solid sediments of Cores B1-B6 were wet-sieved through a
63-µm nylon screen to collect < 63 µm sediments using seawater in a N2 glove box. The
seawater was deoxygenated by bubbling with high purity N2 gas in a glove box for at least
1 hour prior to use. The suspended fine sediment was settled out by centrifugation and
decantation as in Section 5.2.3. The wet sediments were homogenized and sub-sampled for
(1) moisture determination and total selenium determination, and (2) sequential extraction
(as per flowchart in Figure 6.1). The solid sediments of Cores C1-C4 were homogenized
and sub-sampled directly after the porewater separation without the sieving step.
The sub-samples (1) were dried (oven at 40 °C), finely ground, microwave-digested in aqua
regia and total selenium concentrations determined according to the procedure described
previously in Chapter 3. Total metal (Cd, Cr, Cu, Fe, Mn, Ni, Pb and Zn) concentrations in
the digests were also analysed using ICP-OES (Liberty AX (Axial) Sequential ICP-OES,
Varian Australia, Pty. Ltd) according to the procedure optimised from Johnson (1996).
The sub-samples (2) above were subject to sequential extraction procedures as described in
Section 6.2.3. The B cores (< 63 µm wet sediments) were subject to the SEP 1 procedure
and the C cores (whole wet sediments) were subject to the SEP 2 procedure as the
flowcharts in Figure 6.2. Fractions 1 and 2 reagents in 1-L containers were de-oxygenated
before being added to samples. Decantation of the extracts in the glove box was
troublesome and time-consuming so it was carried out outside the glove box in as fast as
possible to minimise sample oxidation. Triplicate extractions were carried out on selected
samples for each core to check for the method consistency and reproducibility. The use of
fine < 63 µm sediments for sequential extraction in the B cores’ study (2004) was to
enhance sample homogeneity and extraction reproducibility. However, whole wet
Chapter 6 – Selenium geochemistry
119
sediments were used for the C cores’ study (2005) to further minimise disruption of sample
anoxia from the wet-sieving step. The use of un-sieved samples for the 2005 study allowed
correlation with macro-component results, which were obtained from whole sediment subsamples.
The sequential extracts were analysed for total selenium by HG-AAS after appropriate
sample digestion. For phosphate, NaOH, NaOAc, NH2OH.HCl and NaOCl extracts, 0.5-3
mL aliquots (depending on selenium concentrations), were digested with 0.5-1 mL of 0.2
M K2S2O8 in a hot water bath at 90 ˚C for 30-50 min, the higher the brown colour organic
content, the longer the digestion time (Zhang et al., 1999a). The selenium was reduced to
selenite species for HG-AAS analysis in 5 mol/L HCl medium in a 90 ˚C hot water bath for
20 min. Sodium sulfite matrix interfered with HG-AAS, therefore, was eliminated by
digesting with nitric acid as recommended by Velinsky and Cutter (1991) before the
selenium reduction step. The residual extract was treated with urea and analysed for
selenium as described for the total selenium determination in Chapter 3.
Initially, the individual selenium compounds in the labile sequential Fraction 2.1 (K2HPO4
extract) and Fraction 2.2 (NaOH extract) were going to be measured using chromatography.
However, separating the individual selenium species in the two sequential extracts was not
performed as there was no confident method to accurately quantify the individual
compounds in those extracts (due to problems with organic interferences as noted in
Chapter 4). Original selenium species were not retained upon extraction using the strong
extractant reagents for Fractions 3-5 of both SEP 1 and SEP 2 (Figure 6.2). Therefore, in
this fractionation study, selenium in all fractions was determined as total concentrations,
which still provided useful information for the selenium geochemistry study. Sub-samples
of all sequential extracts were also analysed for co-extracted trace metals using ICP-OES as
for the total metal determination method above.
The sums of all sequential fractions (SEP 2), compared to the total selenium measured were
± 20% for the majority of the samples analysed (32 samples out of total 36 samples). The
RSDs of the sequential extractions (n =3) were less than 10% for phosphate and NaOH
extracts, 13% for sodium sulfite extracts, and up to 30% for NaOCl and residual fractions.
Chapter 6 – Selenium geochemistry
120
Total reactive iron was determined for C cores samples (whole wet sediments) using the
procedure modified from Hall et al. (1996). The samples (0.5 g) were extracted using 1
mol/L hydroxylamine hydrochloride in 25% acetic acid (25 mL) at 90 ˚C for 3 hours
(vertex mixing every 30 min). The extraction was repeated for 1 hour and the final residue
rinsed with MilliQ water. The two extracts and the washed solution were combined, filtered
through a 0.45 µm filter, preserved with 1% HCl and stored below 4˚C. The samples were
analysed for iron and co-extracted elements using the ICP-OES as above.
Analyses of sediment macro-components were completed on wet bulk samples of Cores
C1-C4 at the NATA accredited Environmental Analytical Laboratory (EAL), Norsearch
Ltd, Lismore, Australia. Total carbon (TC), total sulfur (TS) and total nitrogen (TN) were
determined using a LECO CNS 2000 analyser. Total organic carbon (TOC) was determined
using the Walkley Black method. Acid volatile sulfides and chromium reducible sulfides
were determined subsequently on the same subsample by the EAL methods (Sav – Method
22A and Scr – Method 22B, respectively) (Stone et al., 1998).
6.2.3
Sequential extraction procedures
The extractants used in the two sequential extraction procedures are shown in Figure 6.2.
The soluble fraction was omitted for both procedures due to minimal selenium detected in
this fraction from a preliminary extraction assessment.
Sequential extraction procedure (SEP 1): was modified from the procedure used by
Tokunaga et al. (1991) and Lipton (1991). In a typical extraction, approximately 2.5 g wet
sediment was transferred into a pre-weighed 50-ml polypropylene centrifuge tube in a N2
glove box, capped and weighed.
Fraction 1.1 Soluble and adsorbed: 25 mL of 0.1 mol/L K2HPO4 (adjusted to pH 8
using KOH) was added to the sample. The sample was shaken for 20 hours on a mixing
wheel at room temperature, centrifuged at 2400 rpm for 20 min and the solution decanted.
The solid was washed with 2-3 mL MilliQ water to remove any remaining soluble
selenium. The solution was decanted and combined with the phosphate extract. The
Chapter 6 – Selenium geochemistry
121
combined solution was filtered through a 0.45 µm membrane filter, preserved in 1% HCl
and stored below 4 ˚C until analysis.
Fraction 1.2 Carbonate: To the residue from Fraction 1.1, 25 mL of 1 mol/L NaOAc
(adjusted to pH 5 using acetic acid) was added. The sample was shaken for 5 hours,
centrifuged and decanted. 20 ml of 0.1 mol/L K2HPO4 (pH 8) was added and the sample
was extracted for 20 hours at room temperature. The sample was centrifuged, decanted and
washed with 2-3 mL MilliQ water. The solutions were combined, filtered through a 0.45
µm membrane filter, preserved in 1% HCl and stored below 4 ˚C until analysis.
Fraction 1.3 Reducible Fe/Mn oxyhydroxides: To the residue from Fraction 1.2, 25 mL
of 0.25 mol/L NH2OH.HCl (pH 2.5) was added. The sample was heated at 50 ˚C with
occasional shaking for 30 minutes, centrifuged and decanted. 20 mL of 0.1 mol/L K2HPO4
(pH 10) was added and the sample, with average final pH 8±0.5, was extracted for 20 hours
at room temperature. K2HPO4 (0.1 mol/L, pH 10) was used instead of the KOH employed
in original procedure by Tokunaga et al. (1991) as strongly alkaline KOH could remobilize
organic matter and potentially over-estimate selenium in this fraction. All extracts and
washing waters were combined, filtered and preserved in 1% HCl, and stored below 4 ˚C
until analysis.
Fraction 1.4 Organic matter: 10 mL of 5 % NaOCl adjusted to pH 9.5 using HCl was
added to the residue from Fraction 1.3. The sample was placed in a water bath at 90˚C for
30 min with occasional shaking. The sample was allowed to cool, centrifuged and the
solution decanted. The extraction was repeated once, then the residue was washed with
MilliQ water. All the extracting solutions were combined, filtered and preserved in 1% HCl
and stored below 4 ˚C until analysis.
Fraction 1.5 Residual: The solid residue from Fraction 1.4 was digested in aqua regia
(3HCl:1HNO3, 10 mL) at 200˚C for 30 min in a microwave and processed the same way as
for total selenium determination (Chapter 3).
122
SEP 1
SEP 2
F 1.1: 0.1 M K2HPO4
(pH 8) 10:1 solution:
solid, 20 hr at 25 ˚C.
F 2.1: 0.1 M K2HPO4
(pH 8) 10:1 solution:
solid, 20 hr at 25 ˚C.
Soluble and
adsorbed Se
F 1.2: 1 M NaOAc (pH5),
5 hr, followed by 0.1 M
K2HPO4 (pH 8, 20hr)
Soluble and
adsorbed Se
F 2.2: 1 M NaOH
(1:10 ratio, 4 hr)
Acid soluble
(e.g., carbonate)
Organically-bound
F 2.3: 1 M Na2SO3
(pH 7, 1:10 ratio, 8 hr,
rinsed with water)
F 1.3: 0.25 M NH2OH.HCl
(0.5 hr @50 ˚C), followed by
0.1 M K2HPO4, pH 10 (20 hr)
Reducible fraction
(Fe/Mn oxyhydroxides)
F 1.4: 5% NaOCl
(pH 9.5) 4:1 ratio, 30 min
at 90˚C; repeated once.
Elemental selenium
F 2.4: 5% NaOCl
(pH 9.5) 4:1 ratio, 30 min
at 90˚C; repeated once.
Oxidisable fraction
(Organically-bound
and sulfides)
F 1.5:
Remaining
F 2.5: Aqua regia
microwave digestion
Residual (unreactive
oxides and silicates)
Figure 6.2
Oxidisable fraction
(Organically-bound
and sulfides)
Residual (unreactive
oxides and silicates)
Sequential extraction procedures SEP 1 and SEP 2 used for selenium fractionation in this study
Chapter 6 – Selenium geochemistry
123
Sequential extraction procedure (SEP 2): was modified from the procedure used by
Velinsky and Cutter (1991) and Wright et al. (2003). Fraction 2.1, Fraction 2.4 and
Fraction 2.5 retained the same procedures as in SEP 1. Fraction 2.2, organically-bound
selenium, and Fraction 2.3, elemental selenium were extracted by sodium hydroxide and
sodium sulfite solutions, respectively, as detailed below:
Fraction 2.2 Organically bound selenium: To the residue from Fraction 2.1, 25 mL of 1
mol/L sodium hydroxide was added. The sample was shaken for 4 hours, centrifuged and
decanted. The sample was washed with 2-3 mL MilliQ water, centrifuged and decanted.
The solutions were combined, filtered through a 0.45 µm membrane filter, preserved in
HCl and stored below 4 ˚C until analysis.
Fraction 2.3 Elemental selenium: To the residue from Fraction 2.2, 25 mL of 1 mol/L
sodium sulfite solution (pH 7) was added. The sample was shaken for 8 hours at room
temperature, centrifuged and decanted. The solid residue was washed with 2-3 mL MilliQ
water, centrifuged and decanted. The solutions were combined, filtered through a 0.45 µm
membrane filter, preserved in HCl and stored below 4 ˚C until analysis.
6.3
Results and discussion
6.3.1
Sediment characteristics, redox potential and pH
Depth profiles of mean redox potentials and pH in the sediment cores collected from the
Red Beach area, Port Kembla Harbour, in April 2004 (Cores B1-B6) and July 2005 (Cores
C1-C6) are shown in Figure 6.3. Raw pH and redox potential data are given in Appendix
B (Table B.2 and Table B.3).
The average redox potentials of the sediment cores ranged from –269 mV to –395 mV for
B cores and from +231 mV to –388 mV in C cores, indicating a very reducing condition of
the sediment apart from the top surface layer. Sediments below 2 cm depth were highly
anoxic with redox potentials of less than –300 mV, similar to the sediment redox conditions
found in the previous studies (Hoai, 2001; Muhammad, 2003). The measured redox
Chapter 6 – Selenium geochemistry
Redox Potential (mV)
-400 -350 -300 -250
pH
7.4
-200
0
0
3
3
6
6
Depth (cm)
Depth (cm)
-450
124
9
12
7.6
7.8
8.0
8.2
8.0
8.2
9
12
15
15
18
18
21
21
24
24
Cores B1-6, April 2004
pH
7.4
400
0
0
3
3
6
6
9
12
Depth (cm)
Depth (cm)
-600
Redox Potential (mV)
-400 -200
0
200
7.6
7.8
9
12
15
15
18
18
21
21
24
24
Cores C1-4, July 2005
Figure 6.3
Depth profiles of redox potential and pH (mean ± SE) of Red Beach
sediment cores collected in April 2004 (Cores B1-6, top row) and July 2005
(Cores C1-4, bottom row).
Chapter 6 – Selenium geochemistry
125
potentials corresponded to the observable light grey colour in the approximately 1-3 cm
region, as compared to black colour of the deeper sediments. Based on these results, the
redox boundary is defined in this study at 2 cm. The top 2 cm is defined as the oxic layer
and below 2 cm is defined as anoxic sediment.
Mean pH values of the sediment core samples were near neutral, ranging from 7.6 to 7.9
and found to increase slightly from the oxic layer (1-2 cm) to the anoxic layer (3-4 cm and
below). A reverse trend between redox potential and pH profiles was observed in these
cores. Slightly more acidic conditions in the oxic layers are believed to be normal due to
oxidation and hydrolysis of iron species which releases protons as in Equations 6.1 and 6.2
(Blowes and Jambor, 1990). Another possible pH influence might be due to sulfate
reducing bacteria producing bicarbonate alkalinity.
FeS2
2+
2Fe
+
+
7/2O2 +
1/2O2 +
H2O
5H2O
Æ
Æ
Fe2+ +
2Fe(OH)3
2SO42- +
+
+
4H
2H+
…. ...(Equation 6.1)
……. (Equation 6.2)
The general sediment core texture and characteristics were similar to those observed and
described previously in Section 5.3.2.1. The grain size distribution in Cores B1-B6 and
Cores C1-C4 are given in Appendix B (Table B.4, Figure B.1 and Figure B.2).
6.3.2
Sediment porewater compositions
Pore waters were extracted during both sampling programs in April 2004 (B cores) and July
2005 (C cores). Only total selenium concentrations were measured for the B cores samples.
Additional concentrations of two anions (sulfate and phosphate) were determined for C
cores samples and are presented below. Concentrations of other trace metals were low and
below/near the detection limit of the ICP-OES method (0.010 mg/L), therefore, are not
discussed further.
6.3.2.1
Porewater sulfate and phosphate
Depth profiles of sulfate and phosphate concentrations in the sediment porewaters are
shown in Figure 6.4. The sulfate and phosphate data are given in Appendix B (Table B.5).
126
Porewater Sulfate, mg/L
Porewater Sulfate, mg/L
1000
2000
0
3000
2000
3000
4000
0
1000
2000
3000
4000
0
0
3
3
3
3
6
6
6
6
9
12
9
12
Depth (cm)
0
Depth (cm)
0
9
12
15
15
18
18
18
18
21
21
21
21
24
20
40
0
60
20
40
24
C3
Porewater P hosphate, mg/L
Porewater P hosphate, mg/L
0
24
C2
P orewater Phosphate, mg/L
60
0
20
40
0
3
3
3
3
6
6
6
6
12
12
Depth (cm)
0
Depth (cm)
0
Depth (cm)
0
9
9
12
15
15
18
18
18
18
21
21
21
21
Figure 6.4
24
C2
24
C3
20
40
60
80
12
15
C1
C4
9
15
24
3000
P orewater Phosphate, mg/L
60
0
9
2000
12
15
C1
1000
9
15
24
Depth (cm)
1000
Porewater Sulfate, mg/L
0
Depth (cm)
Depth (cm)
0
Porewater Sulfate, mg/L
24
C4
Depth profiles of porewater sulfate (top row) and phosphate (bottom row) in four individual Red Beach cores: C1-C4.
Dot line (Core C4): no data due to insufficient porewater sample volume (coarse grain region).
Chapter 6 – Selenium geochemistry
127
The porewater sulfate concentrations varied between cores ranging from 406 to 2940 mg/L,
and generally decreased with depth (r = - 0.776, P < 0.0001 for Cores C1-C4 samples, n =
33). High sulfate concentrations at the upper core layers might be derived largely from the
dissolved sulfate in the seawater (Gerritse, 1999). Analysis of sulfate in the overlying
seawater samples of Cores C1-C4 revealed similar sulfate concentrations as in the
porewater samples of the upper core region (2762 ± 27 mg/L, n = 4).
The porewater phosphate concentrations ranged from below the detection limit to 59 mg/L
and increased with the sediment depth. No phosphate was detected in the porewaters above
4 cm for all four cores and this is possibly because the phosphate was trapped in the solid
phase by iron oxyhydroxides in the oxic region. The phosphate in porewaters of deeper
sediments will be released from the iron oxyhydroxides as they become reduced and
solubilised under anoxic conditions in the deeper sediment (Cha et al., 2005).
6.3.2.2
Porewater selenium
Depth profiles of porewater selenium concentrations (mean ± SE), in comparison to the
total solid-phase selenium profiles in Red Beach cores (Cores B1-6 and Cores C1-4) are
shown in Figure 6.5. Porewater selenium data are given in Appendix B (Table B.6) and
porewater profiles of individual cores are shown in Figures B.3 and B.4. The average
selenium concentrations in the porewaters varied with depth ranging from 13.3 to 48.8 µg/L
in B cores (individual samples from 4.0 to 99.5 µg/L), and from 1.1 to 43.3 µg/L in C
cores, and peaked at depth.
The overall profiles show that the dissolved-phase selenium concentrations did not reflect
the total selenium concentrations in the sediments (i.e., solid-phase selenium). In most
cores, the dissolved selenium concentrations were found to peak at a lower depth than the
maximum peak of the solid-phase selenium concentrations. The patterns are clear in Cores
C1-4, where the solid phase selenium peaked at above 12 cm but the dissolved phase
selenium concentrations peaked below 12 cm.
Chapter 6 – Selenium geochemistry
128
Se conc
50
100
150
200
250
0
300
0
0
3
3
6
6
Depth (cm)
Depth (cm)
0
Se conc
9
12
18
18
21
T otal Se, <63 um, ug/g
24
120
150
Porewater Se, ug/L
T otal Se, whole, ug/g
24
Cores B 1-6
Figure 6.5
90
12
15
P orewater Se, ug/L
60
9
15
21
30
Cores C1-4
Porewater selenium concentrations (mean ± SE) in Red Beach cores
collected in April 2004 (Cores B1-6) and July 2005 (Cores C1-4), in
comparison to the total solid-phase selenium in the corresponding cores.
The profiles of porewater selenium are similar to those of porewater phosphate, both
increasing with depth. However, an inverse-trend was observed between porewater
selenium and porewater sulfate in Cores C1-C4. While the porewater selenium increased
with depth, the porewater sulfate decreased with depth.
The possible explanations for the porewater selenium results might be that in the upper core
region, selenium (some possibly being present as oxyanion oxidised forms) was bound with
the solid-phase such as Fe/Mn oxyhydroxides and organic matter, which are abundant in
the oxic sediments. This corresponds with the results in Section 6.3.4.2, which found large
percentages of solid-phase selenium in the organically bound fraction (see also Figure
6.12, Selenium) in the top 12 cm of the sediment cores. High concentrations of dissolved
selenium in the deeper core region may result from a direct release from Fe/Mn
oxyhydroxides, which become reduced and solubilised under reducing conditions: Fe (III)
Æ Fe (II). However, this is considered a minor pathway, as only small amount of solidphase selenium was found to associate with Fe/Mn oxyhydroxides (see Section 6.3.4.1). A
Chapter 6 – Selenium geochemistry
129
simple reductive solubilisation of selenium (SeIV Æ Se0 Æ Se-II) from the solid phase into
the dissolved phase at deeper depth was unlikely, as the anoxic condition was observed
from below 2 cm but the dissolved-phase selenium peak appeared at much lower depth (1219 cm) than the redox boundary (2 cm). This suggested that there should be other factors
(additional to the redox conditions) that controlled the peak porewater selenium
concentrations below 12 cm region (possibly such as coupled redox reactions: selenium
oxidative solubilisation (Se0 Æ SeIV) from the solid phase during a reductive burial of other
elements).
Another potential pathway is that the formerly solid-phase selenium in the surface layer
was desorbed by the high concentrations of dissolved phosphate (PO43-) at the lower core
region (see Figure 6.4). Phosphate ions are known to have stronger binding ability to
sediment particles than selenium (Jackson and Miller, 2000; Goh and Lim, 2004). There
was a slight decrease in the porewater phosphate concentrations below 13 cm in Cores C1
and C4, which might suggest that some of the dissolved phosphate were transferred back to
the solid-phase (corresponding to the high porewater selenium in this region). However, the
limited data points below 13 cm prevent the evidence from being conclusive.
It should also be noted that high percentages of coarse grains in core sections below 15 cm
of Core C4 (see Figure B.2 and Figure B.4) might be a contributing factor responsible for
the peak porewater selenium in this region due to a lower adsorption ability of coarse grains
to selenium, so more selenium was remained in a dissolved phase. The correlation
coefficient (r), between >250 µm fraction and the porewater selenium concentration in
Core C4, was 0.864 (P = 0.0006, n = 11). The coarse grain features in Core C4 may also be
contributing to the different chemical features observed in the deeper sections of this core.
6.3.3
Macrocomponent depth profiles
Depth profiles of the macrocomponent concentrations (% dry wt.) in the Red Beach cores
(C1-C4, whole sediment) are shown in Figure 6.6. Detailed data are given in Appendix B
(Table B.7).
130
T otal Carbon (% d.w.)
10
20
30
40
0. 0
4. 0
6. 0
0. 0
8. 0
0
3
3
3
6
6
6
9
12
Depth (cm)
0
9
12
15
18
18
18
21
21
21
24
24
24
Acid volatile sulfides (% d.w.)
1. 5
3. 0
4. 5
0. 0
6. 0
0. 2
0. 4
0. 6
0. 0
3
3
6
6
6
12
12
Depth (cm)
3
Depth (cm)
0
0
9
0. 8
1. 0
0. 6
1. 2
1. 8
2. 4
3. 0
9
12
15
15
15
18
18
18
21
21
21
24
24
24
Figure 6.6
0. 6
Pyrites (% d.w.)
0. 8
0
9
0. 4
12
15
0. 0
0. 2
9
15
T otal Sulfur (% d.w.)
Depth (cm)
2. 0
0
Depth (cm)
Depth (cm)
0
T otal Nitrogen (% d.w.)
T otal Organic Carbon (% d.w.)
Concentrations (% d.w.) of Total Carbon, Total Organic Carbon, Total Nitrogen, Total Sulfur, Acid Volatile Sulfides and
Chromium Reducible Sulfur (pyrites) in Red Beach whole sediment: Cores C1-C4 (
).
Chapter 6 – Selenium geochemistry
131
The concentrations of total carbon (TC), total organic carbon (TOC) and total nitrogen
(TN) varied between cores, ranging from 1.28 – 37.6; 0.58 – 6.40; and 0.09 – 0.69% for
TC, TOC and TN, respectively. Cores C1 and C2, collected from the west side of the Darcy
Road Drain, shared similar TC, TOC and TN profiles (top row, Figure 6.6), decreasing
with depth to approximately 8-12 cm then increasing with further depth. Cores C3 and C4,
collected from the east side of the Drain, also shared similar TC, TOC and TN profiles,
which appeared opposite-trended to the profiles of Cores C1 and C2.
The concentrations of total sulfur (TS), acid volatile sulfides (AVS), and chromium
reducible sulfur (CrRS) or pyrites showed much greater variation between cores and depths
than the TC, TOC and TN results, and ranged from 0.60 – 4.70; 0.016 – 0.67; and 0.38 –
2.67%, respectively. In general, the TS, AVS and CrRS profiles appeared to increase with
depth (bottom row, Figure 6.6). A sharp decrease of those three profiles below 15 cm of
Core C4 was mainly attributed from the high percentages of large grain sizes in this region,
subsequently decreasing the binding surface area.
The correlations of the sediment macrocomponents with total solid-phase selenium and
porewater selenium and porewater sulfate in Cores C1-C4 are shown in Table 6.3.
Significant correlations (P < 0.01) were found between TOC-TN in Cores C2 (r = 0.905)
and C4 (r = 0.788), between TOC-TC in Core C4 (r = 0.864). The relationships between
TC, TOC and TN are also evident in a cluster analysis shown in Figure 6.7, which
indicates their association through the organic matter components of the sediment.
Total sulfur, AVS and CrRS were inversely correlated with the porewater sulfate as clearly
found in Cores C3 and C4, indicating that the accumulation of the solid-phase sulfur was at
the expense of the dissolved sulfate species. Sulfate could undergo reduction and deposition
as sulfides and pyrite during degradation of organic matter under anoxic conditions (such as
in the Red Beach sediment cores) (Canfield, 1993; Gagnon et al., 1996). Relationships
between AVS and TOC were observed in Cores C3 (r = 0.870) and C4 (r = 0.838), while no
such relationship was observed between CrRS and TOC. These results correspond with the
fact that acid volatile sulfides are earlier products of the sulfate reduction than pyrites, and
so more active in the organic matter decay processes (Gagnon et al., 1996). AVS was also
132
Table 6.3
Correlations (r)* between total selenium concentrations and measured sediment parameters in Cores C1-C4.
Core C1 (n = 9)
PW Se PW SO4 <63 µm
AVS
CrRS
-0.952
0.607
0.576
0.514 -0.664 0.387
0.420 -0.541
-0.247
-0.611
0.462
-0.025 -0.142 -0.684 0.228 -0.158 -0.724
-0.432
-0.424
-0.671 -0.571 0.589 -0.613 -0.498 0.408
0.254
0.489
0.277 -0.435 -0.184 0.267 -0.321
0.103
AVS
0.748 -0.096 0.449
0.743
0.158
CrRS
0.019
0.896
0.233
-0.018 0.066
Depth
0.733
PW Se
PW SO4
<63 µm
TC
Core C2 (n = 7)
TC
TOC
0.224
TOC
TS
TN
Total Se
PW Se PW SO4 <63 µm AVS CrRS
0.544
-0.042
-0.218 0.228 -0.409 -0.270 -0.608 -0.229 -0.454
-0.589
0.570
0.527
0.651 0.784
0.004 0.297 0.710 -0.023
0.738
0.176
0.489 -0.532 -0.227 0.546 -0.274
0.317
0.349
0.368 0.708 0.963 0.535
0.924
0.674
0.701
0.859 0.317 0.824
0.328
0.225 -0.075
-0.051
0.622 0.905
0.686
0.463
0.455
0.489
0.877
0.434
(continued).
PW Se PW SO4 %<63 um AVS
AVS
CrRS
TC
TOC
TS
TN
0.602
Total Se
-0.369
Core C3 (n = 8)
<63 µm
TN
0.504
Table 6.3
PW Se
TS
0.720 0.657 0.580 0.483
TS
PW SO4
TOC
-0.122 0.580 -0.340 -0.903 -0.843 -0.186 -0.713
0.024 0.644
TN
Depth
TC
-0.666
-0.937
-0.627
0.931
CrRS
TC
Core C4 (n = 12)
TOC
TS
TN
0.986 0.954 0.480 0.877 0.959 0.472
Total Se
0.823
PW Se PW SO4 <63 µm AVS
0.630
CrRS
TC
TOC
TS
TN
Total Se
-0.762
-0.622 -0.234 0.399 -0.040 -0.179 0.376 -0.309
-0.238
-0.504
-0.691 -0.333 -0.005 -0.200 -0.105 -0.024 -0.476
-0.108
0.172
-0.185 -0.817 -0.396 -0.368 -0.814 -0.027
-0.117
0.403
0.614 0.447 0.592 0.804 0.510 0.767
0.462
-0.917
-0.957 -0.928 -0.270 -0.801 -0.965 -0.280
-0.735
0.907 0.931 0.168 0.766 0.931 0.192
0.773
0.728 0.236 0.449 0.599 0.310 0.667
0.491
0.940 0.441 0.870 0.961 0.428
0.757
0.528 0.776 0.838 0.524 0.775
0.610
0.398 0.743 0.992 0.300
0.889
0.675 0.639 0.987 0.394
0.387
0.568 0.369 0.910
0.590
0.864 0.653 0.873
0.658
0.766 0.711
0.697
0.651 0.788
0.823
0.295
0.850
0.418
0.435
0.491
* Values highlighted in bold are significant at P < 0.01 level. PW Se = Porewater selenium; PW SO4 = Porewater sulfate.
0.770
Chapter 6 – Selenium geochemistry
133
found to correlate with < 63 µm fraction in a cluster analysis shown in Figure 6.7, which
indicated the favoured formation of AVS in the fine grain sediment. The relationships
between CrRS-TS were strong in all sediment cores (also in the cluster analysis),
corresponding to the CrRS and TS concentration results that the CrRS (or pyrite) is a major
form of solid-phase sulfur in the Red Beach cores.
Figure 6.7
Cluster relationships between sediment macrocomponents, porewater
selenium, porewater sulfate and < 63 µm fraction in Cores C1-C4.
Sulfate is the final electron acceptor in the sequence of organic matter oxidation (Figure
2.3). Reduction of considerable amounts of sulfate to AVS and pyrite might suggest that
other redox species (including selenium), positioned earlier in the sequence could have
been converted to their reduced forms in the Red Beach cores. However, depending on the
microbial diversity, bacterial degradation of organic matter in the sediment may prefer
utilizing sulfate, which is readily available in the dissolved porewater phase, to other
electron accepters, which are bound tightly to the solid sediment phase (Muhammad, 2003).
A binding of selenium with organic matter or involvement of selenium in organic matter
degradation processes was found to exist with significant correlation between total solidphase selenium concentration and TOC, observed in Core C4 (r = 0.823). Strong
correlations between the total solid-phase selenium with CrRS (and hence TS) in Core C2
(r = 0.924) and Core C3 (r = 0.889) also suggest association of selenium with pyrite, a
stable and reduced form of the solid-phase sulfur, in the sediment potentially due to having
similar geochemical behaviour.
Chapter 6 – Selenium geochemistry
134
The ratios of the sediment macrocomponents might provide better representation of the
sediment core characteristics by minimizing large variation of macrocomponent
concentrations in different sediment cores. Depth profiles of macrocomponent ratios (mean
± SE) in Cores C1-C4 are shown in Figure 6.8 (data are given in Table B.20). The profiles
of TC/TOC and TOC/TN ratio showed no clear patterns with depth. All the profiles of TC
and TOC ratios to the sulfur species (AVS, CrRS and TS) showed a decreasing trend with
depth, indicating the accumulation of the solid-phase sulfur with depth. The TC/AVS and
TOC/AVS profiles show large variation above 5 cm and below 14 cm. The upper core
variation may indicate the dynamics of AVS species as the sediment transformed from oxic
conditions to anoxic conditions. Large variation in the core region below 14 cm was due to
less numbers of longer core data being averaged. And the integrity of sample anoxic
conditions in this core end region may be compromised due to the nature of field sampling.
However, both cases indicated the sensitivity of AVS species to the redox conditions.
A sharp decrease in the TC/CrRS, TC/TS, TOC/CrRS and TOC/TS ratios was observed
between the top 0-2 cm layer and the 2-4 cm layer, corresponding well with the redox
conditions of the Red Beach cores, which were oxic at the top 2 cm and highly anoxic
below 2 cm. The reducing condition below 2 cm led to a sharp increase in pyrite formation
in the sediment layer. Some large variation was observed at the upper core region, also
indicating the dynamics of pyrite formation in this upper core region (similar to the AVS
results). However, the TC/CrRS and TOC/CrRS profiles were more stable below 6 cm. The
TC/TS and TOC/TS profiles were also stable at deeper sediment, as pyrite was a major
species of the solid-phase sulfur as discussed above.
It may be concluded that Red Beach cores showed a high dynamics of diagenetic processes
and organic matter mineralisation, especially in the region around the sediment redox
boundary. The processes might be facilitated by the enrichment of reactive iron (Appendix
B, Figure B.19), dissolved sulfate from the seawater and organic matter input as well as
other redox elements in the sediment. High sedimentation rate of the Red Beach cores
(reported in Section 5.3.3.2) might also be a factor that facilitated the anoxic condition of
any new deposited sediment, leading to a fast rate of the organic matter degradation, and
hence a fast rate of the sulfate reduction (Canfield, 1993; Gerritse, 1999; Muller, 2002).
135
TC /AVS
TC /TO C
2
4
6
8
150
450
TC /TS
TC /C rRS
600
750
0
5
10 15
20 25
0
30 35
0
0
2
2
2
2
4
4
4
4
8
10
6
Depth (cm)
6
8
10
Depth (cm)
0
6
8
10
12
14
14
14
14
16
16
16
16
18
18
18
18
10
TO C /AVS
15
20
0
30
60
90
TO C /C rRS
120
150
0
2
6
0
2
2
2
2
4
4
4
4
8
10
8
10
6
8
10
Depth (cm)
0
Depth (cm)
0
Depth (cm)
0
6
25
1
2
3
4
5
6
8
10
12
12
12
12
14
14
14
14
16
16
16
16
18
18
18
18
Figure 6.8
20
TO C /TS
4
0
6
15
10
12
5
10
8
12
0
5
6
12
TO C /TN
Depth (cm)
300
0
Depth (cm)
Depth (cm)
0
0
Macrocomponent ratios for Red Beach sediment cores (C1-C4, mean ± SE). Top row: Ratios of TC to TOC, AVS, CrRS
and TS. Bottom row: Ratios of TOC to TN, AVS, CrRS and TS.
Chapter 6 – Selenium geochemistry
6.3.4
136
Forms and binding phases of selenium in Red Beach sediments
Two sequential extraction (fractionation) procedures were performed to measure the forms
and binding phases of the selenium in Red Beach sediment cores. Sequential extraction
procedure SEP 1 was carried out on Cores B1-B6 (2004 sampling, < 63 µm sediment) and
the procedure SEP 2 was carried out on Cores C1-C4 (2005 sampling, whole sediment).
Only sequential extracts of C cores were analysed for co-extracted metal concentrations and
all the SEP results are presented below.
6.3.4.1
SEP 1 fractionation
The SEP 1 procedure measured selenium associated with: soluble and adsorbed; carbonate;
iron-manganese oxyhydroxide; and organic matter and sulfide fractions. The selenium
remaining in the sediment was defined as residual fraction. Depth profiles of selenium
concentrations (µg/g, d.w.) in the SEP 1 fractions of Cores B1-6 (mean ± SE) are shown in
Figure 6.9, with the selenium SEP 1 fractionation pattern shown in Figure 6.10. Details of
SEP 1 selenium data and individual Cores B1-B6 profiles are given in Appendix B (Table
B.9, Figure B.5, and Figure B.6).
The soluble and adsorbed, carbonate, and iron-manganese oxyhydroxide fractions
comprised negligible selenium fractions in the Red Beach sediment B cores. The average
selenium concentrations in the three labile fractions were below 3 µg/g (d.w.), accounting
for 1.0-3.3, 1.0-2.2 and 0.6-1.3% of the total solid-phase selenium (in <63 µm sediment),
respectively. The depth profiles of the three selenium fractions, in general, reflected the
total solid-phase selenium profile. In the oxic surface layer, the selenium was associated
least with the iron-manganese oxyhydroxide fraction < the soluble and adsorbed fraction <
the carbonate fraction. Interestingly, the soluble and adsorbed selenium fraction increased
in the deeper anoxic sediment and became greater than the carbonate and the ironmanganese oxyhydroxide fractions at depths below 10 cm (to 18 cm). The soluble and
adsorbed selenium profiles seemed to correspond to the porewater selenium profiles (peak
at deeper sediment) discussed in Section 6.3.2.2. This might reflect the equilibrium
between the dissolved phase selenium and the labile ‘soluble and adsorbed’ phase.
137
1.0
3.0
0
4.0
0
0
2
2
4
4
6
6
Depth (cm)
Depth (cm)
0.0
Se (µg/g)
2.0
8
10
50
100
Se (µg/g)
150
200
250
300
8
10
12
12
14
14
16
Soluble & adsorbed
16
Organic matter and sulfides
18
Carbonates
Fe-Mn oxyhydroxides
18
Residual
20
Figure 6.9
Total selenium
20
Selenium concentrations (µg/g d.w.) in different sediment fractions of Cores B1-6 (mean ± SE, n = 6, below 14 cm n=2).
Chapter 6 – Selenium geochemistry
0%
20%
138
40%
60%
80%
100%
1
Depth (cm)
3
5
7
9
11
13
15
17
19
Figure 6.10 Selenium fractionation patterns (SEP 1) in Red Beach sediment
cores (Cores B1-6), as percentages of the total selenium
extracted from sediments.
Organic and sulfide and residual fractions comprised the majority of the selenium in the
Red Beach cores. The average selenium concentrations in the organic matter and sulfide
fraction ranged from 26.1 to 88.9 µg/g (d.w.) (data for individual cores are given in
Appendix B (Table B.9)), and the remaining selenium concentrations as a residual fraction
ranged from 32.8 to 54.0 µg/g (d.w.). As for F1-F3 above, the depth profile of the organic
and sulfides fraction also reflected the total selenium profile.
The SEP 1 selenium fractionation pattern of Cores B1-6, shown in Figure 6.10, indicated
that the solid-phase (<63 µm) selenium was bound predominantly with the organic and
sulfide fraction, accounting for an average of 43-62 %, depending on depth. Large
proportions of selenium (33-54 %) remained in the residual fraction. These results were in
agreement with the sequential extraction studies of Lake Macquarie sediments by Nobbs
and co-workers (1997), who employed the Tessier et al. (1979) and the European
Community Bureau of Reference (BCR) sequential extraction procedures (see Table 6.1).
Nobbs et al. (1997) found that selenium was predominantly bound to organic matter (70100% and 65-100% in the top 8 cm sediment by Tessier’s and BCR methods, respectively).
Chapter 6 – Selenium geochemistry
139
However, the SEP 1 results in this study found much lower percentages of adsorbed (1.03.3%) and carbonate selenium (1.0-2.2%), compared to 0-28% in the adsorbed and 0-18 %
in the carbonate fractions by Tessier’s method, and the combined 26.1% found associated
with the adsorbed and carbonate fractions by the BCR method (Nobbs et al., 1997). The
smallest proportion of selenium (0.6-1.3%) in Red Beach cores was found associated with
the iron-manganese oxyhydroxide fraction. In comparison, no selenium was recovered from
iron-manganese oxyhydroxide fraction using the Tessier’s procedure, while 8.8% selenium
was extracted in iron-manganese oxyhydroxide fraction by the BCR method in Lake
Macquarie sediment cores (Nobbs et al., 1997). The differences in the results might be due
to the differences in the sediment characteristics in Lake Macquarie and Port Kembla
Harbour. The differences in the sequential extraction procedures used (see Table 6.1) might
also affect the results. In the BCR method, a prior hydrogen peroxide oxidation might add
more selenium to the subsequent reducible oxyhydroxide fraction. Selenium extracted by
the Tessier’s iron-manganese oxyhydroxide reagent might be readsorbed into the sediment
particles due to a low pH of the acetic acid matrix (25%). Subsequent phosphate extraction
was performed to correct for any potential re-adsorption in this study as recommended by
Lipton (1991).
The SEP 1 study was carried out during the early fractionation work in 2004. The SEP 1
method was close to the traditional Tessier’s procedure (1979) used to characterize
common cationic trace metals in soils/sediment samples. The procedure was initially
chosen at the time for the selenium fractionation with the aim of simultaneously studying
co-extracted (cationic) trace metals in the SEP 1 fractions. However, the procedure found
large proportions of the selenium to be associated with organic and sulfide and residual
fractions. Selenium is not commonly incorporated into silicate mineral lattices as it has a
large atomic radius. Therefore, other possible selenium forms in the sediment might be
elemental and pyritic selenium. SEP 1 provided less useful information on those selenium
forms as both elemental and pyritic selenium could be included in both organic and sulfide
and residual fractions. The two major selenium fractions required further differentiation and
characterization by an alternate procedure (e.g., SEP 2 below) so no further work was
carried out for the SEP 1 study.
Chapter 6 – Selenium geochemistry
6.3.4.2
140
SEP 2 fractionation
The SEP 2 procedure measured selenium associated with: soluble and adsorbed fraction;
labile organic matter (humic substances); elemental selenium; refractory organic matter and
sulfide fraction; and residual fraction (hot aqua regia digestion). Depth profiles of selenium
concentrations (µg/g, d.w.) in the SEP 2 fractions of individual C cores are shown in
Figure 6.11. Detailed SEP 2 selenium data are given in Appendix B (Table B.10). The
SEP 2 fractionation patterns of selenium in Red Beach sediment cores (mean of Cores C1C4), in comparison to the patterns of other co-extracted elements are shown in Figure 6.12.
SEP 2 selenium fractionation patterns for individual C cores are given in Appendix B
(Figure B.7). The sequential extraction data for co-extracted elements can also be found in
Appendix B (Table B.11 to B.17).
Firstly, selenium in the soluble and adsorbed fraction comprised a minor selenium fraction
in Cores C1-C4 (whole sediment), accounting for averages of 2.4 to 9.1% of the total solidphase selenium (individual samples from 1.5 % in the 6-8 cm section of Core C3 to a
maximum of 17 % in the 10-12 cm section of Core C1). The percentages of the soluble and
adsorbed selenium appeared to increase with depth (see selenium pattern in Figure 6.12),
similar to the dissolved porewater selenium profiles, as observed in the SEP 1 results
above. Chromium and nickel were the two elements co-extracted in large quantities in the
soluble and adsorbed fraction (phosphate pH 8). The correlation analysis (Appendix B,
Table B.21) found the soluble and adsorbed selenium to correlate significantly with coextracted Fe, Pb and Zn, as observed in Cores C2-C4 (Seads-Fe and Seads-Zn in Core C2,
Seads-Zn in Core C3, and Seads-Fe, Seads-Pb and Seads-Zn in Core C4). The soluble and
adsorbed selenium in Core C4 also correlated with TOC, indicating some selenium
desorption from the sediment organic matter and possible solubilisation of some organic
matter in the phosphate buffer pH 8 solution. Note that the unusually high 17 % of soluble
and adsorbed selenium in the 10-12 cm depth of Core C1 may result from mobilization of
organic matter, which also peaked in this section (see the TOC profiles).
Secondly, the selenium associated with labile organic matter (humic substances), extracted
in the sodium hydroxide solutions comprised significantly large percentages of the total
141
Se (µg/g)
Se (µg/g)
0
10
20
30
40
50
0
60
0
0
2
2
10
15
8
10
12
25
30
6
8
10
14
12
16
18
14
C1
C2
Se (µg/g)
0
10
20
30
Se (µg/g)
40
50
60
70
0
0
0
2
3
4
6
Depth (cm)
Depth (cm)
20
4
6
Depth (cm)
Depth (cm)
4
5
6
8
10
80
100
120
140
15
14
21
Figure 6.11
60
12
18
C3
40
9
12
16
20
24
C4
Selenium concentrations (µg/g) in different sequential extracts (SEP 2) of four Red Beach cores (whole sediments):
(
)
142
20%
40%
60%
80%
100%
0%
1
1
3
3
5
5
7
7
Depth (cm)
Depth (cm)
0%
9
11
13
15
20%
100%
80%
100%
13
15
17
19
21
21
23
23
Chromium
S elenium
20%
40%
60%
80%
0%
100%
1
1
3
3
5
5
7
7
Depth (cm)
Depth (cm)
80%
9
19
9
11
13
15
40%
60%
11
13
15
17
19
19
21
21
23
23
Copper
20%
9
17
Figure 6.12
60%
11
17
0%
40%
Iron
Fractionation patterns (SEP 2) of selenium and co-extracted trace elements in Red Beach cores (mean of C1-C4) (cont.).
143
0%
20%
40%
60%
80%
0%
100%
40%
3
5
5
Depth (cm)
3
7
9
11
13
15
80%
100%
60%
80%
100%
7
9
11
13
15
17
17
19
19
21
21
23
23
Nickel
Manganese
0%
20%
40%
60%
80%
100%
0%
1
1
3
3
5
5
7
7
Depth (cm)
Depth (cm)
60%
1
1
Depth (cm)
20%
9
11
13
15
11
13
15
17
19
19
21
21
23
23
Figure 6.12
40%
9
17
Lead
20%
Zinc
Fractionation patterns (SEP 2) of selenium and co-extracted trace elements in Red Beach cores (mean of C1-C4).
Chapter 6 – Selenium geochemistry
144
selenium in the solid phase. The average percentages of organically bound selenium in
Cores C1-C4 ranged from 14 to 28 % and appeared to decrease with depth (see also
selenium pattern in Figure 6.12). The results correspond with the macrocomponent results,
which showed that organic matter was mineralized at deeper depth as the sediment become
anoxic. This may also mean that during the organic matter degradation the organically
bound selenium can be removed from this fraction, possibly to other sediment phases. It
should be noted that the selenium species in sodium hydroxide extracts were identified in
Chapter 4 to be present mainly as selenite in both oxic and anoxic sediments, indicating
strong binding between organic matter and the selenium oxyanion. The major co-extracted
elements in the sodium hydroxide solutions were zinc and lead, with nickel and copper
being present at lesser quantities (yellow bands Figure 6.12). The correlations of the
organically bound selenium with other elements co-extracted in the sodium hydroxide
solution (Appendix B, Table B.22) found significant relationships between Seorg-TC and
Seorg-Ni in Core C1, Seorg-TOC, Seorg-Fe and Seorg-Zn in Core C3, and Seorg-Fe, Seorg-Ni
and Seorg-Pb in Core C4.
Thirdly, the elemental selenium comprised the largest selenium fraction in the solid
sediment phase, accounting for averages of 25 to 53 % in Cores C1-C4 (whole sediment).
The elemental selenium contents increased with depth from the surface to approximately 816 cm and then started to decrease. Interestingly, a slight decrease of the organically bound
selenium and this elemental selenium fraction corresponded well with an increase in the
residual selenium fraction (see selenium pattern Figure 6.12). This might indicate the
selenium solid-phase transformation from the organically bound fraction and the elemental
selenium fraction to the residual fraction upon ageing. In the elemental selenium fraction
(blue-green strips), copper was the only element co-extracted in significant quantities in the
sodium sulfite solution, indicating association between copper and the elemental selenium.
Chromium was also co-extracted in this fraction but at minor quantities. The correlation
analysis of the elemental selenium concentrations with the sediment parameters and coextracted elements (Appendix B, Table B.23), found significant relationships between Se0CrRS and Se0-TS in Core C2, Se0-AVS, Se0-CrRS and Se0-TS in Core C3, and Se0-AVS,
Se0-TC and Se0-TOC in Core C4. A significant correlation between Se0-Cu was observed
in Core C3. These correlation results clearly indicate the association of the reduced
Chapter 6 – Selenium geochemistry
145
elemental selenium species with the reduced sulfide and pyrite sulfur species. The
correlation with TOC again indicated possible coupled reduction of selenium/sulfur upon
the organic matter oxidative degradation.
Fourthly, the selenium associated with the refractory organic matter and sulfides in the
sediment comprised on average 10-19 % of the total solid-phase selenium in Cores C1-C4.
The selenium patterns in this fraction were fluctuated. The upper 16 cm region of this
selenium fraction in Figure 6.12 showed a vase shape with a neck region at 4-6 cm. With
the selenium behaviour knowledge up to this stage, this pattern potentially indicated high
percentages of selenium associated with organic matter above 4 cm region and a gradual
increase in selenium association with sulfide minerals below 6 cm. This pattern was also
observed in individual core results of C2-C4 (not shown). Minimal quantities of other
elements were co-extracted in the refractory organic matter and sulfide fraction. The
elements found present included copper, manganese, lead and zinc, and generally appeared
at lower core region. A correlation analysis (Appendix B, Table B.24) found significant
relationships between Seo&s-Zn in Core C1, Seo&s-AVS and Seo&s-TS in Core C3, and
Seo&s-AVS, Seo&s-TC and Seo&s-TOC (r = 0.940) in Core C4. A strong correlation between
Cu-Zn was also observed in the refractory organic matter and sulfide fractions of Cores C1,
C3 and C4. The selenium association with reduced sulfur species and TOC might be due to
similar processes as for the elemental selenium results. The relationships with zinc and
copper possibly indicate the presence of metal sulfide/selenide species.
Finally, the residual selenium fraction comprised on average 11-37 % in Cores C1-C4. The
pattern of the residual selenium fluctuated and increased with depth clearly at the expense
of the organically bound and elemental selenium fractions. Iron and manganese, and also
chromium and nickel, remained largely in the residual fraction. Lesser quantities of copper,
lead and zinc remained in the residual fraction, interestingly with a decreasing amount in
the deeper core region (below 12 cm). A correlation analysis (Appendix B, Table B.25),
found significant relationships between Seres-Cu in Core C1, Seres-Cu, Seres-Pb and Seres-Zn
in Core C2, and Seres-Ni in Core C3. A strong relationship between Pb-Zn was also
observed in the residual fractions of Cores C2-C4.
Chapter 6 – Selenium geochemistry
6.3.5
146
Selenium geochemical behaviour in Red Beach sediments
Redox potential and pH are known to be important factors in controlling depositional and
diagenetic behaviour of selenium (Masscheleyn et al., 1991; Peters et al., 1997). In this
study, the Red Beach sediment cores can be divided into oxic and anoxic regions based on
the redox boundary at 2 cm depth. The top 2 cm sediment is characterized by the low solidphase selenium concentrations due to minimal selenium input from the source
(decommissioning of the copper smelter) and dilution of the selenium by recent
sedimentation and the overlying water, and the low porewater selenium concentrations.
Selenium present in this surface region may arrive from other harbour sources and from any
remobilization of selenium from deeper sediments. The distinctly different sediment
parameters between the oxic and the deeper anoxic layers were pH, redox potential and C
to S ratios (Figure 6.8). Large variations were observed in results for the surface sediment
layer, such as TOC and AVS results. An anomalous data point in the excess
210
Pb activity
was also found for the top 2 cm layer of Core C4 (see Figure 5.9), indicating dynamic
activity in this sediment layer. In addition to being the redox boundary, which is important
to the selenium behaviour, the surface layer is also important for biological activities.
The anoxic core region between 2-10 cm is characterized by peak solid-phase selenium, but
with low porewater selenium concentrations. This region was enriched with the organic
matter, AVS, organically bound selenium, and elemental selenium species, indicating this
region as being important for organic matter decay processes and hence the reduction of
electron receptors species available in the sediment (evident with the reduced selenium and
sulfur species). The sediment below 10 cm is characterized by moderate solid-phase
selenium, peak porewater selenium concentrations and high soluble and adsorbed selenium
concentrations. The solid-phase selenium was found to become associated with the residual
fraction at the expense of the organically bound and the elemental selenium (formerly
enriched in the upper anoxic 2-10 cm region). The major sulfur species found in this region
was the stable pyritic forms. Also in this core region, lower proportions of copper, lead and
zinc were found to associate with the residual fraction but significant amounts released in
the sodium hydroxide extracts. In particular, copper was (the only major element) released
in great quantities in the sodium sulfite extracts (elemental selenium fraction).
Chapter 6 – Selenium geochemistry
147
The large proportion of reduced elemental selenium form found in the core sediment was in
agreement with the redox and pH conditions of the Red Beach sediment cores (Figure
6.13), which were largely neutral and anoxic. The selenium would most likely be present in
reduced forms in the sediment. Free selenite and selenate anions have not been found when
redox potential is under –200 mV (Peters et al., 1999a). This might be an explanation for
low amounts of selenium found in the soluble and adsorbed and iron-manganese
oxyhydroxide fractions. The sediment characteristics favoured a fast rate of organic matter
degradation, and therefore a fast rate of sulfur and selenium reduction. The trace element
contamination of Red Beach sediments may provide reducing and complexing agents that
facilitate selenium immobilization in the sediment, e.g., iron (II) hydroxide was reported to
have ability to reduce selenite and selenate (abiotically) into elemental selenium but not to
reduce sulfate (Zingaro et al., 1997).
Selenium undergoes reduction to elemental selenium easier than sulfur, and elemental
selenium is stable within larger redox-pH region, compared to elemental sulfur (Figure
6.13). Both selenium and sulfur are predicted to be in reduced hydride forms under the Red
Beach sediment Eh-pH conditions (Brookins, 1988; McNeal and Balistrieri, 1989). In this
study, while the majority of the solid-phase sulfur was found as pyrite (FeS2), a large
percentage of the solid-phase selenium was present in the elemental state. This indicated
the high stability of the elemental selenium toward further reduction to selenides by
bacteria, relative to the less stable elemental sulfur. Selenium hydrides are less stable than
sulfur hydrides (Greenwood and Earnshaw, 1984) so any further reduction of selenium to
the Se–II state would result in the formation of metal selenides. The presence of sulfide
species in Red Beach sediments suggests that metal selenide species could have been
formed. Selenium is not commonly incorporated into silicate minerals due to its large
atomic radius. Some selenium found in the residual fraction may largely be in a form of
pyritic selenium. Pyritic selenium has not commonly been found in marine sediment
systems due to an abundance of sulfur species for pyrite formation (Velinsky and Cutter,
1991). This might be the case in some pristine marine systems where the iron and other
trace elements required for the pyrite formation may be limited. However, transition
elements are not limited in the contaminated Red Beach sediment, Port Kembla Harbour.
Formation of pyritic selenium and metal selenides is possible.
148
Please see print copy for Figure 6.13
Figure 6.13
Comparison of selenium (McNeal and Balistrieri, 1989) and sulfur (Brookins, 1988) Eh-pH diagrams. The
stability region of water is between the two solid lines. The ovals represent the Eh-pH conditions of the Red
Beach cores.
149
In the residual fraction, selenium was correlated significantly with Cu, Pb and Zn, which
may indicate their possible association in forms such as CuSe, PbSe and ZnSe. Copper,
lead and zinc (also Cd, Hg and Mn) are known to form sulfides more rapidly than iron and
are not incorporated into pyrite (Morse and Luther III, 1999). The sulfides CdS, CuS, PbS
and ZnS usually form independent minerals (Morse and Luther III, 1999). Selenium has
similar properties to sulfur so the formation of CuSe, PbSe and ZnSe is possible. The molar
ratios of Cu/Se, Pb/Se and Zn/Se in the residual fraction were 320-698, 41-75 and 200-366,
respectively, indicating a large excess of the transition metals. However, the formation of
selenide minerals is reported to be nonstoichiometric due to variety in valency and minimal
differences between the electronegativity of selenium and transition metals (Greenwood
and Earnshaw, 1984). Further work on characterization of pyritic selenium and metal
selenides is required to verify this hypothesis. The concentrations of metal selenides or
pyritic selenium were not measured in this work due to a special analytical method required
for metal selenide determination. There has been one method of pyritic selenium analysis
reported in the literature (Velinsky and Cutter, 1990) and it requires assessment and
optimisation.
6.3.6
Implications for potential remobilization and bioavailability
The study in Chapter 5 and in this chapter found the very high concentrations of total
selenium in Red Beach sediment cores in the upper core region (6-10 cm peak), which
could potentially remobilize to the sediment surface and into the overlying water or become
assimilated by organisms. However, the analysis of porewater selenium found low
selenium concentrations in the dissolved phase in this upper core region, relative to the high
concentrations of porewater selenium in a deeper core region (below 12 cm). Low
porewater selenium in the upper core region may possibly be because the selenium was
more strongly bound to organic materials as evident in the SEP 2 fractionation study.
The porewater selenium concentrations found in the deeper core region (up to 99.5 µg/L)
were significantly high compared to previous studies in Australian sediments (Peters et al.,
1999b; Doyle et al., 2003). Porewater selenium at such high concentrations has been
reported to cause adverse biological effects, such as disruptive fertilization and larval
Chapter 6 – Selenium geochemistry
150
development, in the sea urchin Heliocidaris tuberculata, if it was present as organic
selenoamino acid species (Doyle et al., 2003). In the Red Beach sediment, the chance of
benthic organism exposure to the porewater selenium may be low due to the presence of
high selenium concentrations deeper in the cores (below 12 cm depth), which is probably
too deep for organisms to dwell. From the four sessions of sediment core sampling for this
research (i.e., for the A, B, C and D cores), many small worms were found in the surface
layer (1-3 cm, abundant at 1 cm layer) of Core D1, collected in January 2006 for the
sediment 210Pb dating. No worms or other organisms were found in the sediment cores from
the other sampling trips.
The solid-phase speciation of selenium in the Red Beach cores revealed the bulk of the
sedimentary selenium was present as elemental selenium, bound to organic matter or as
residual selenium. The elemental selenium form is relatively immobile and therefore less
available to organisms. The organically bound selenium may become remobilized if the
organic matter is oxidised or mineralized. However, the likely fate of post organic-bound
selenium would be in reduced elemental selenium or residual metal selenide forms. It can
be said from this study that the solid-phase selenium in the sediment was mostly
immobilized as reduced forms in the sediment, especially from below 6 cm (where pyrite
was observed to become stabilized in Section 6.3.3). Selenium is relatively resistant to
oxidation (compared to sulfur) (Greenwood and Earnshaw, 1984), providing that the
selenium is protected in the sediment cores with no major redox change; any changes in
natural environment parameters such as salinity or water ionic composition would not
remobilize the reduced selenium upward into the overlying water.
The research in Chapter 5 found that the sedimentary selenium tends to accumulate in the
fine (< 63 µm) sediments. Although the fractionation studies found that the potential
selenium mobilization from the sediment particles was low, it was evident that the <63 µm
particles, as the selenium carrier, were more mobile than larger particles, and thus could be
a pathway for selenium distribution/transport. In addition, the selenium may become
potentially bioavailable to the organisms if the entire sediment particulates are ingested by
organisms (Luoma et al., 1992; Schlekat et al., 2000).
Chapter 6 – Selenium geochemistry
6.4
151
Conclusions
The study of selenium geochemical behaviour in Red Beach cores revealed anoxic
conditions of the sediment cores from below 2 cm depth, with high degree of organic
matter mineralisation that had led to a high degree of sulfur and selenium reduction in the
sediment. The solid-phase selenium in the Red Beach sediment cores was present mainly as
elemental selenium. The second largest quantities of the selenium were bound to the
organic matter in the upper 10 cm region but associated with the residual fraction below 10
cm. Much lower proportions of the solid-phase selenium were in soluble and adsorbed
fractions, with peak concentrations in the lower core region (below 10 cm). Negligible
amounts of selenium were found to associate with iron-manganese oxyhydroxides and
carbonate minerals in the sediment.
The solid-phase selenium correlated with the solid-phase sulfur in Red Beach sediments
and the strong relationship existed through the association of their reduced forms: elemental
selenium, pyrite and possibly as pyritic selenium. The reduced selenium forms (elemental
and residual) correlated significantly with Cu, Pb and Zn, suggesting possible formation of
independent CuSe, PbSe and ZnSe minerals in the sediment. It is concluded that redox
potential, sedimentation rate, organic matter components, sulfur and transition elements are
the important factors affecting the selenium geochemical behaviour in Red Beach cores.
The low redox potentials and the high sedimentation rate provided the anoxic conditions
required for degradation of organic matter and the selenium reduction. The diagenetic
processes and the immobilization of the selenium were further facilitated by the enrichment
of iron, sulfur, organic matter and other transition elements in the contaminated sediment.
The geochemistry of selenium and sulfur is likely to be influenced by microbial processes
and that aspect requires further investigation.
Chapter 7
Conclusions and recommendations
7.1
Introduction
This chapter provides a summary of the research findings and some recommendations for
further work.
7.2
Conclusions
This thesis investigated the spatial distribution, speciation, binding phases and the
geochemical behaviour of selenium in the contaminated marine sediments, Red Beach, Port
Kembla Harbour. It also reports the initial assessment of appropriate sample preparation
and analytical methods for the determination of selenium and its species in marine
sediments.
The evaluation and optimisation of a method for total selenium determination in sediment
samples using a microwave assisted digestion and hydride generation-atomic absorption
spectrometry (Chapter 3) found the ability of hot aqua regia (3HCl: 1HNO3) to effectively
extract total selenium from sediment samples. An aqua regia digestion is very beneficial for
a subsequent selenium analysis by HG-AAS technique as no extra selenate reduction step
was required. An aqua regia matrix contained sufficient nitric acid for oxidative digestion
of any organic materials present in soils/sediments but low enough nitric acid to avoid
nitrogen oxide interferences with the HG-AAS analysis. Nitrogen oxide interferences were
encountered and were overcome by addition of urea. Real samples with high organic
content were found to cause foaming during the HG-AAS analysis as the high content of
organic matter might not be completely ashed by the hot aqua regia (in comparison to
perchloric or nitric acids), but the foaming problem was eliminated by adding an antifoam
solution.
Chapter 7 - Conclusions
153
The study of sediment extraction procedures and selenium speciation methods for
determination of organic and inorganic selenium compounds in sediments based on HPLC
separation and HG-AAS detection (Chapter 4) found an alkaline sodium hydroxide to be
the most effective reagent in extracting labile selenium species from sediments, in
comparison to water, salt, and acid solutions. Four (organic and inorganic) selenium
compounds (selenite, selenate, selenomethionine and selenocystine) in sodium hydroxide
solution at 0.1-mol/L concentrations were separated successfully within 12 min using an
anion exchange column with gradient ammonium phosphate elution. The combination of
the HPLC separation and HG-AAS detection has drawbacks of a lengthy post-column
sample digestion and a high detection limit but was able to measure some selenium
compounds (selenite and selenate) in contaminated sediments. The application of the HPLC
and HG-AAS speciation method to the analysis of selenium species in real sediments from
Port Kembla Harbour and reference materials found selenite and selenate to be present in
the NaOH extracts of both oxic and anoxic sediments, with selenite being the major
species. The organic matter interferences in the NaOH extract matrix prevented the use of
this method to accurately quantify the concentrations of individual selenium compounds in
the real sediments.
The investigation of the spatial distribution of selenium in surface sediments from Port
Kembla Harbour and in cores from the Red Beach area (Chapter 5) found selenium
concentrations in surface sediments from most harbour sites to be low (below 3 µg/g)
except those in sediments from the Red Beach area (up to 9.38 µg/g), which is in close
proximity to a copper refinery. Selenium concentrations in Red Beach sediment cores
ranged from 6 to 1735 µg/g, depending on depth and grain size, with peak selenium
concentrations observed at 6-10 cm and at 14-16 cm depths. The sedimentary selenium was
concentrated in fine (<63 µm) grains that are easily mobile, horizontally in surface
sediments and vertically upward in sediment cores. Selenium was correlated mainly with
Pb, Cu and Zn in the > 250 µm fraction of the surface sediments and in the < 63 µm
fraction of the sediment cores, indicating association from both original ore sources and
through post-depositional transformation.
Chapter 7 - Conclusions
154
The determination of sedimentation rate in Red Beach sediment cores using
210
Pb
radiodating technique (Chapter 5) found that the deeper sediments were not disturbed. The
sedimentation rate estimated from the Constant Initial Concentration (CIC) model was 0.55
± 0.03 cm/year. The sediment
210
Pb dating provided an indication of historical selenium
inputs, potentially from the copper smelter.
The investigation of selenium geochemical forms (Chapter 6) found the solid-phase
selenium in the Red Beach sediment cores to be present mainly as elemental selenium.
Large proportions (slightly lower than that of the elemental selenium) of the solid-phase
selenium were found bound to the organic matter in the upper 10 cm region and associated
with the residual fraction below 10 cm depth. Small proportions of the solid-phase
selenium were in soluble and adsorbed fractions, with peak concentrations in the lower core
region (below 10 cm). Negligible amounts of selenium were found to associate with ironmanganese oxyhydroxides and carbonate minerals in the sediment.
The study of the selenium geochemical behaviour in Red Beach sediment cores, in relation
to the sediment pH, redox potential, pore water anion composition, sediment
macrocomponents, and common trace elements (Chapter 6) found the sediment cores to be
oxic in the top 2 cm and anoxic below 2 cm depths. The oxic (top 2 cm) layer contained
distinctly different pH, redox potential and C to S ratios to the deep anoxic sediment and
contained low solid-phase selenium concentrations and low porewater selenium
concentrations due to minimal recent selenium input from the major source (copper
smelter). The oxic sediment layer showed a high degree of sediment mixing that also
possibly involved biological activity.
The anoxic region between 2-10 cm depth contained peak solid-phase selenium
concentrations, but low porewater selenium concentrations. This region was enriched with
organic matter, AVS, organically bound selenium, and elemental selenium species,
indicating the importance of this core region for organic matter decay processes and hence
the reduction of electron receptors species including sulfur and selenium. The anoxic
sediment below 10 cm contained moderate solid-phase selenium, peak porewater selenium
and high soluble and adsorbed selenium concentrations. Some of the formerly organically
Chapter 7 - Conclusions
155
bound and the elemental selenium in the 2-10 cm anoxic region were found to become
associated more with the residual fraction. The major sulfur species found in this region
were the stable pyritic forms. This below 10 cm region contained lower proportions of
copper, lead and zinc in the residual fraction but significant amounts in the organically
bound fractions (sodium hydroxide extracts). Copper was the only major element released
in great quantities in the sodium sulfite extracts (elemental selenium fraction).
The solid-phase selenium correlated with the solid-phase sulfur in Red Beach sediments
and a strong relationship existed through the association of their reduced forms: elemental
selenium, pyrite and possibly as pyritic selenium. The reduced selenium forms (elemental
and residual) correlated significantly with Cu, Pb and Zn, suggesting possible formation of
independent CuSe, PbSe and ZnSe minerals in the sediment. It is concluded that redox
potential, sedimentation rate, organic matter components, sulfur and transition elements are
the important factors affecting the selenium geochemical behaviour in Red Beach cores.
The low redox potentials and the high sedimentation rate provided the anoxic conditions
required for degradation of organic matter and selenium reduction. The diagenetic
processes and the immobilization of the selenium were further facilitated by the enrichment
of iron, sulfur, organic matter and other transition elements in the contaminated sediment
The evidence of the selenium geochemical study suggested that the solid-phase selenium in
Red Beach sediment cores was mostly immobilized as reduced forms and potentially with
the reduced sulfur species acting as sacrificial reducing agent (i.e., giving electrons to other
redox species before selenium) that protect the selenium from become oxidised. So long as
the conditions of the sediment cores remain anoxic, the reduced selenium is unlikely to
remobilize upward into the overlying water (not taking into account any volatilization
processes). This thesis, to best of my knowledge, is first to report the interesting
geochemical association between selenium and copper, lead, and zinc in real marine
sediments. The outcomes of this study are hoped to be beneficial to the management of
selenium-contaminated sediments in Port Kembla Harbour, as well as to contribute to
selenium analytical and biogeochemical knowledge and to provide supporting information
for regulatory needs. One management option is to leave the contaminated sediments
undisturbed as the selenium is in relatively immobile forms at present conditions.
Chapter 7 - Conclusions
7.3
156
Recommendations for future research
There are several aspects of this selenium research that have not been resolved within the
timeframe of the study. Some issues that warrant further studies include:
•
Interferences of organic matter in the selenium speciation analysis in the sodium
hydroxide extracts, as encountered in Chapter 4. Solving this issue, possibly with a use
of XAD resin to remove the dissolved organic matter, might provide a higher method
confidence level for accurate quantification of individual selenium compounds. The
high detection limit of the HPLC and HG-AAS speciation technique may be improved
by direct coupling of the HPLC with an ICP-MS.
•
The field sample study in Chapter 5, found large grain (> 250 µm) sediments (e.g.,
from sites 18, 19 and 20) to contain significantly high selenium concentrations. Specific
selenium behaviour in the large grain fraction was not closely investigated in this study
due to high variation in analysis results that made data interpretation difficult. A closer
examination of selenium behaviour in this fraction may be interesting, providing that
the issue of sample homogeneity is resolved.
•
Large proportions of the solid-phase selenium were bound to the sediment organic
matter in the upper core region, which corresponded to the disappearance of the
porewater selenium. More studies of selenium behaviour in relation to the organic
matter (both dissolved and solid-phase organic matter) could be beneficial to verify the
relationship between organic matter binding and the porewater selenium in the upper
core region.
•
There was some evidence of pyritic selenium present in the residual selenium fraction.
Measurement of pyritic selenium or metal selenide concentrations would be interesting
for the Red Beach cores. This work would also involve an assessment of an effective
method for pyritic selenium analysis.
•
The solid-phase selenium was found mostly immobilized in the sediment. Further
studies into selenium volatilization processes might be useful to ensure that no or
minimal selenium is remobilized to the sediment surface via microbial volatilization or
other mechanisms. The transfer of selenium from sediments into the harbour foodchain
is also an issue that requires investigation.
References
Abdel-Moati, M. A. R. (1998). Speciation of selenium in a Nile Delta lagoon and SE
Mediterranean Sea mixing zone. Estuarine, Coastal and Shelf Science 46: 621-628.
Abdulah, R., Miyazaki, K., Nakazawa, M. and Koyama, H. (2005). Chemical forms of
selenium for cancer prevention. Journal of Trace Elements in Medicine and Biology
19: 141-150.
Agazzi, A. and Pirola, C. (2000). Fundamentals, methods and future trends of
environmental microwave sample preparation. Microchemical Journal 67: 337-341.
ANZECC/ARMCANZ (2000). Australian and New Zealand Guidelines for Fresh and
Marine Water Quality. Canberra, Australian and New Zealand Environment and
Conservation Council and Agriculture and Resource Management Council of
Australia and New Zealand.
ANZFA (1992). Australia New Zealand Food Standards Code. Australia New Zealand
Food Authority. Australian Government Publishing Service, Canberra.
Appleby, P. G. and Oldfield, F. (1992). Application of Lead-210 to sedimentation studies.
In: M. Ivanovich and R.S. Harmon. Uranium-series Disequilibrium: Applications to
Earth, Marine, and Environmental Sciences, Clarendon Press, Oxford: 731-778.
Apte, S. C., Batley, G. E., Szymczak, R., Rendell, P. S., Lee, R. and Waite, T. D. (1998).
Baseline trace metal concentrations in New South Wales coastal waters. Journal of
Marine and Freshwater Research 49: 203-214.
Baldwin, S., Maher, W., Kleber, E. and Krikowa, F. (1996). Selenium in Marine Organisms
of Seagrass Habitats (Posidonia australis) of Jervis Bay, Australia. Marine Pollution
Bulletin 32: 310-316.
Baldwin, S. and Maher, W. (1997). Spatial and temporal variation of selenium
concentration in five species of intertidal molluscs from Jervis Bay, Australia.
Marine Environmental Research 44: 243-262.
Balistrieri, L. S. and Chao, T. T. (1990). Adsorption of selenium by amorphous iron
oxyhydroxide and manganese dioxide. Geochimica et Cosmochimica Acta 54: 739751.
Barwick, M. and Maher, W. (2003). Biotransference and biomagnification of selenium
copper, cadmium, zinc, arsenic and lead in a temperate seagrass ecosystem from
Lake Macquarie Estuary, NSW, Australia. Marine Environmental Research 56:
471-502.
158
Batley, G. E. (1987). Heavy metal speciation in waters, sediments and biota from Lake
Macquarie, NSW. Australian Journal of Marine and Freshwater Research 38: 591606.
Beavington, F., Cawse, P. A. and Wakenshaw, A. (2004). Comparative studies of
atmospheric trace elements: improvements in air quality near a copper smelter.
Science of the Total Environment 332: 39-49.
Bebbington, G. N., Mackay, N. J., Chvojka, R., Williams, R. J., Dunn, A. and Auty, E. H.
(1977). Heavy metals, selenium and arsenic in nine species of Australian
commercial fish. Australian Journal of Marine and Freshwater Research 28: 277286.
Belzile, N., Chen, Y. W. and Xu, R. (2000). Early diagenetic behaviour of selenium in
freshwater sediments. Applied Geochemistry 15: 1439-1454.
Bhattacharya, B., Sarkar, S. K. and Das, R. (2003). Seasonal variations and inherent
variability of selenium in marine biota of a tropical wetland ecosystem: implications
for bioindicator species. Ecological Indicators 2: 367-375.
Blackmore, D. P. T. (2002). Application of Molybdenum Adsorption Phenomena in Natural
and Waste Water Treatment Systems. PhD Thesis, Department of Environmental
Science, University of Wollongong.
Blowes, D. W. and Jambor, J. L. (1990). The pore-water geochemistry and the mineralogy
of the vadose zone of sulfide tailings, Waite Amulet, Quebec, Canada. Applied
Geochemistry 5: 327-346.
Borella, P., Bargellini, A., Caselgrandi, E., Menditto, A., Patriarca, M., Taylor, A. and
Vivoli, G. (1998). Selenium determination in biological matrices. Microchemical
Journal 58: 325-336.
Brenneisen, P., Steinbrenner, H. and Sies, H. (2005). Selenium, oxidative stress, and health
aspects. Molecular Aspects of Medicine 26: 256-267.
Brimmer, S. P., Fawcett, W. R. and Kulhavy, K. A. (1987). Quantitative reduction of
selenate ion to selenite in aqueous samples. Analytical Chemistry 59: 1470-1471.
Brookins, D. G. (1988). Eh-pH Diagrams for Geochemistry, Springer-Verlag, Berlin.
Brown, R. D. J. (2000). Selenium and Tellurium. U.S. Geological Survey Minerals Year
Book -2000.
159
Brugam, R. B. (1978). Pollen indicators of land-use change in Southern Connecticut.
Quaternary Research 9: 349-362.
Brunori, C., de la Calle-Guntinas, M. B. and Morabito, R. (1998). Optimization of the
reduction of Se(VI) to Se(IV) in a microwave oven. Fresenius Journal of Analytical
Chemistry 360: 26-30.
Bufflap, S. E. and Allen, H. E. (1995). Sediment pore water collection methods for trace
metal analysis: A review. Water Research 29: 165-177.
Bujdos, M., Kubova, J. and Stresko, V. (2000). Problems of selenium fractionation in soils
rich in organic matter. Analytica Chimica Acta 408: 103-109.
Bujdos, M., Mulova, A., Kubova, J. and Medved, J. (2005). Selenium fractionation and
speciation in rocks, soils, waters and plants in polluted surface mine environment.
Environmental Geology 47: 353-360.
Canfield, D. E. (1993). Organic matter oxidation in marine sediments. Interactions of C, N,
P and S Biogeochemical Cycles and Global Change. Wollast, R., Mackenzie, F. T.
and Chou, L. Springer-Verlag, Berlin Heidelberg: 333-363.
Canton, S. P. and Vanderveer, W. D. (1997). Selenium toxicity to aquatic life - an argument
for sediment-based water quality criteria. Environmental Toxicology and Chemistry
16: 1255-1259.
Cao, T. H., Cooney, R. A., Woznichak, M. M., May, S. W. and Browner, R. F. (2001).
Speciation and identification of organoselenium metabolites in human urine using
inductively coupled plasma mass spectrometry and tandem mass spectrometry.
Analytical Chemistry 73: 2898-2902.
Capelo, J. L., Fernandez, C., Pedras, B., Santos, P., Gonzalez, P. and Vaz, C. (2006).
Trends in selenium determination/speciation by hyphenated techniques based on
AAS or AFS. Talanta 68: 1442-1447.
Carroll, B. I., Nobbs, D. M. and Smith, G. (1996). Characterization of selenium
bioavailability by two sequential extraction methods in sediments from an estuarine
lake subject to anthropogenic selenium inputs. Sustainable Energy and
Environmental Technology, Proceedings of the Asia-Pacific Conference, Singapore,
June 19-21, 1996: 72-79.
Carroll, B. I. (1999). Microbial and geochemical aspects of selenium cycling in an
estuarine system - Lake Macquarie, NSW. PhD Thesis, Department of Chemical
Engineering, University of Sydney, Australia.
160
Casiot, C., Szpunar, J., Lobinski, R. and Potin-Gautier, M. (1999). Sample preparation and
HPLC separation approaches to speciation analysis of selenium in yeast by ICPMS. Journal of Analytical Atomic Spectrometry 14: 645-650.
Cha, H. J., Lee, C. B., Kim, B. S., Choi, M. S. and Ruttenberg, K. C. (2005). Early
diagenetic redistribution and burial of phosphorus in the sediments of the
southwestern East Sea (Japan Sea). Marine Geology 216: 127-143.
Chatterjee, A., Shibata, Y. and Morita, M. (2001). Determination of selenomethionine by
high performance liquid chromatography-direct hydride generation-atomic
absorption spectrometry. Microchemical Journal 69: 179-187.
Chau, Y. K., Wong, P. T. S., Silverberg, B. A., Luxon, P. L. and Bengert, G. A. (1976).
Methylation of selenium in the aquatic environment. Science (Washington, DC,
United States) 192: 1130-1.
Cleland, K. (1995). Environmental upgrade and expansion of a copper smelter at Port
Kembla : proposed by Southern Copper Limited : report to the Honourable Craig
Knowles, Minister for Urban Affairs and Minister for Housing, Commissioners of
Inquiry for Environment and Planning, Sydney.
Cobo-Fernandez, M. G., Palacios, M. A., Chakraborti, D., Quevauviller, P. and Camara, C.
(1995). On line speciation of Se(VI), Se(IV), and trimethylselenium by HPLCmicrowave oven-hydride generation-atomic absorption spectrometry. Fresenius'
Journal of Analytical Chemistry 351: 438-42.
Conde, J. E. and Alaejos, M. S. (1997). Selenium concentrations in natural and
environmental waters. Chemical Reviews 97: 1979-2003.
Crews, H. M., Clarke, P. A., Lewis, D. J., Owen, L. M., Strutt, P. R. and Izquierdo, A.
(1996). Investigation of selenium speciation in in vitro gastrointestinal extracts of
cooked cod by high-performance liquid chromatography inductively coupled
plasma mass spectrometry and electrospray mass spectrometry. Journal of
Analytical Atomic Spectrometry 11: 1177-1182.
Crusius, J. and Thomson, J. (2003). Mobility of authigenic rhenium, silver, and selenium
during postdepositional oxidation in marine sediments. Geochimica et
Cosmochimica Acta 67: 265-273.
Cutter, G. A. and Bruland, K. W. (1984). The marine biogeochemistry of selenium: A
reevaluation. Limnology and Oceanography 29: 1179-1192.
Cutter, G. A. (1985). Determination of selenium speciation in biogenic particles and
sediments. Analytical Chemistry 57: 2951-2955.
161
Cutter, G. A. and Cutter, L. S. (1995). Behaviour of dissolved antimony, arsenic, and
selenium in the Atlantic Ocean. Marine Chemistry 49: 295-306.
Cutter, G. A. and Cutter, L. S. (2004). Selenium biogeochemistry in the San Francisco Bay
estuary: changes in water column behavior. Estuarine, Coastal and Shelf Science
61: 463-476.
De Gregori, I., Lobos, M. G. and Pinochet, H. (2002). Selenium and its redox speciation in
rainwater from sites of Valparaiso region in Chile, impacted by mining activities of
copper ores. Water Research 36: 115-122.
Delaney, J. (1996). An examination of the Port Kembla spoil ground 1996. Honours Master
of Environmental Science Thesis, Faculty of Science, University of Wollongong.
Dhillon, K. S. and Dhillon, S. K. (1999). Adsorption-desorption reactions of selenium in
some soils of India. Geoderma 93: 19-31.
Doyle, C. J., Pablo, F., Lim, R. P. and Hyne, R. V. (2003). Assessment of metal toxicity in
sediment pore water from Lake Macquarie, Australia. Archives of Environmental
Contamination and Toxicology 44: 343-350.
Duc, M., Lefevre, G. and Fedoroff, M. (2006). Sorption of selenite ions on hematite.
Journal of Colloid and Interface Science 298: 556-563.
Ducros, V., Ruffieux, D., Belin, N. and Favier, A. (1994). Comparison of two digestion
methods for the determination of selenium in biological samples. Analyst 119:
1715-1717.
Dungan, R. S., Stork, A. and Frankenberger, W. T. (2000). A wind tunnel for measuring
selenium volatilization under field-like conditions. Journal of Environmental
Quality 29: 460-466.
Eklund, E. and Murray, M. (2000). Copper community: a history of the Electrolytic
Refining and Smelting Company of Australia Ltd. & Southern Copper Ltd., Port
Kembla. Wollongong, N.S.W, University of Wollongong Press.
Ellwood, M. J. and Maher, W. A. (2003). Measurement of arsenic species in marine
sediments by high-performance liquid chromatography-inductively coupled plasma
mass spectrometry. Analytica Chimica Acta 477: 279-291.
Elrashidi, M. A., Adriano, D. C., Workman, S. M. and Lindsay, W. L. (1987). Chemical
equilibria of selenium in soils: a theoretical development. Soil Science 144: 141-52.
162
Elrick, K. A. and Horowitz, A. J. (1986). Analysis of rocks and sediments for Ar, Sn and Se
by wet digestion and hydride generation atomic absorption. Varian Instruments At
Work: January.
Engberg, R. A., Westcot, D. W., Delamore, M. and Holz, D. D. (1998). Federal and state
perspectives on regulation and remediation of irrigation-induced selenium
problems. Environmental Chemistry of Selenium. Frankenburger, W.T., Jr. and
Engberg, R. A. (Eds). Marcel Dekker, New York: 1-25.
Environmental Protection Authority (1994). Contamination in Fish from Port Kembla
Harbour. Bankstown, New South Wales, Technical Report, Environmental
Protection Authority, Water and Catchments Branch.
Ewan, C. R. (1989). Chapter 7: Animal tissues. Occurence and Distribution of Selenium.
(Ed.) Inhat, M. CRC Press, Boca Raton: 121-169.
Fan, T. W.-M., Teh, S. J., Hinton, D. E. and Higashi, R. M. (2002). Selenium
biotransformations into proteinaceous forms by foodweb organisms of seleniumladen drainage waters in California. Aquatic Toxicology 57: 65-84.
Ferri, T. and Sangiorgio, P. (1999). Voltammetric study of the interaction between Se(IV)
and dissolved organic matter in environmental aqueous matrices. Analytica Chimica
Acta 385: 337-343.
Fio, J. L. and Fujii, R. (1990). Selenium speciation methods and application to soil
saturation extracts from San Joaquin Valley, California. Soil Science Society of
America Journal 54: 363-9.
Fishbein, L. (1983). Environmental selenium and its significance. Fundamental and
Applied Toxicology 3: 411-419.
Flury, M., Frankenberger, W. T., Jr. and Jury, W. A. (1997). Long-term depletion of
selenium from Kesterson dewatered sediments. Science of the Total Environment
198: 259-270.
Fox, P. M., LeDuc, D. L., Hussein, H., Lin, Z. Q. and Terry, N. (2003). Selenium
speciation in soils and plants. Biogeochemistry of Environmentally Important Trace
Elements. 835: 339-354.
Frankenberger, W. T. and Karlson, U. (1994). Microbial volatilization of selenium from
soils and sediments. Selenium in the Environment. Frankenburger, W. T. and
Benson, S. (Eds). Marcel Dekker, New York: 369-387.
Gagnon, C., Mucci, A. and Pelletier, E. (1996). Vertical distribution of dissolved sulphur
species in coastal marine sediments. Marine Chemistry 52: 195-209.
163
Gao, S., Tanji, K. K., Peters, D. W. and Herbel, M. J. (2000). Water selenium speciation
and sediment fractionation in a California flow-through wetland system. Journal of
Environmental Quality 29: 1275-1283.
Garbisu, C., Ishii, T., Leighton, T. and Buchanan, B. B. (1996). Bacterial reduction of
selenite to elemental selenium. Chemical Geology 132: 199-204.
Garcia-Hernandez, J., Glenn, E. P., Artiola, J. and Baumgartner, D. J. (2000).
Bioaccumulation of Selenium (Se) in the Cienega de Santa Clara Wetland, Sonora,
Mexico. Ecotoxicology and Environmental Safety 46: 298-304.
Gerritse, R. G. (1999). Sulphur, organic carbon and iron relationships in estuarine and
freshwater sediments: effects of sedimentation rate. Applied Geochemistry 14: 4152.
Gibbs, P. J. and Miskiewicz, A. G. (1995). Heavy metals in fish near a major primary
treatment sewage plant outfall. Marine Pollution Bulletin 30: 667-674.
Gleyzes, C., Tellier, S. and Astruc, M. (2002). Fractionation studies of trace elements in
contaminated soils and sediments: a review of sequential extraction procedures.
Trends in Analytical Chemistry 21: 451-467.
Goh, K.-H. and Lim, T.-T. (2004). Geochemistry of inorganic arsenic and selenium in a
tropical soil: effect of reaction time, pH, and competitive anions on arsenic and
selenium adsorption. Chemosphere 55: 849-859.
Goldberg, S., Martens, D. A., Forster, H. S. and Herbel, M. J. (2006). Speciation of
Selenium(IV) and Selenium(VI) using Coupled Ion Chromatography-Hydride
Generation Atomic Absorption Spectrometry. Soil Science Society of America
Journal 70: 41-47.
Gomez-Ariza, J. L., Sanchez-Rodas, D., Morales, E., Herrgott, O. and Marr, I. L. (1999).
Inorganic and organic selenium compound speciation with coupled HPLC-MWHG-AFS. Applied Organometallic Chemistry 13: 783-787.
Goodfellow, A. (1996). A geochemical investigation of Port Kembla Outer harbour
sediments. Honours Bachelor of Environmental Science Thesis, Faculty of Science,
University of Wollongong.
Gordon, A. N., Pople, A. R. and Ng, J. (1998). Trace metal concentrations in livers and
kidneys of sea turtles from south-eastern Queensland, Australia. Marine and
Freshwater Research 49: 409-414.
Gray, L. A. (1997). Metal contamination of sediments associated with deepwater ocean
sewage outfalls, Sydney, Australia. Marine Pollution Bulletin 33: 182-189.
164
Green, B. (2003). Summary of Port Kembla Harbour water quality results 2003.
Unpublished data, Blue Scope Steel Ltd., Port Kembla, NSW.
Greenwood, N. N. and Earnshaw, A. (1984). Chapter 16 Selenium, tellurium, and
polonium. Chemistry of the Elements, Pergamon press, Oxford, Great Britain: 882919.
Gruebel, K. A., Davis, J. A. and Leckie, J. O. (1988). The feasibility of using sequential
extraction techniques for arsenic and selenium in soils and sediments. Soil Science
Society of America Journal 52: 390-397.
Guerin, T., Astruc, A. and Astruc, M. (1999). Speciation of arsenic and selenium
compounds by HPLC hyphenated to specific detectors: a review of the main
separation techniques. Talanta 50: 1-24.
Hall, G. E. M., Vaive, J. E., Beer, R. and Hoashi, M. (1996). Selective leaches revisited,
with emphasis on the amorphous Fe oxyhydroxide phase extraction. Journal of
Geochemical Exploration 56: 59-78.
Hall, G. E. M. (1997b). Chapter 5: Determination of trace elements in sediments. Manual of
Physico-chemical Analysis of Aquatic Sediments. Mudroch, A., Azcue, J. M. and
Mudroch, P. (Eds). CRC Lewis, Boca Raton: 85-145.
Hamilton, S. J. and Lemly, A. D. (1999). Water-sediment controversy in setting
environmental standards for selenium. Ecotoxicology and Environmental Safety 44:
227-235.
Hamilton, S. J. (2002). Rationale for a tissue-based selenium criterion for aquatic life.
Aquatic Toxicology 57: 85-100.
Hamilton, S. J. (2003). Review of residue-based selenium toxicity thresholds for freshwater
fish. Ecotoxicology and Environmental Safety 56: 201-210.
Hamilton, S. J. (2004). Review of selenium toxicity in the aquatic food chain. Science of
the Total Environment 326: 1-31.
Hansen, D., Duda, P. J., Zayed, A. and Terry, N. (1998). Selenium removal by constructed
wetlands: Role of biological volatilization. Environmental Science and Technology
32: 591-597.
Hatch Engineering (2003). 26 Atchison Street, Wollongong, NSW Australia 2500.
Telephone +61 2 42247222.
165
Haygarth, P. M., Rowland, A. P., Sturup, S. and Jones, K. C. (1993). Comparison of
instrumental methods for the determinaiton of total selenium in environmental
samples. Analyst 118: 1303-1308.
Haynes, D., Leeder, J. and Rayment, P. (1997). A comparison of the bivalve species Donax
deltoides and Mytilus edulis as monitors of metal exposure from effluent discharges
along the Ninety Mile beach, Victoria, Australia. Marine Pollution Bulletin 34: 326331.
He, Q. and Walling, D. E. (1996). Interpretating particle size effects in the adsorption of
137
Cs and unsupported 210Pb by mineral soils and sediments. Journal of
Environmental Radioactivity 30: 117-137.
He, Z. and Morrison, R. J. (2001). Changes in the marine environment of Port Kembla
Harbour, NSW, Australia, 1975-1995: A Review. Marine Pollution Bulletin 42:
193-201.
Herbel, M. J., Blum, J. S. and Oremland, R. S. (2003). Reduction of elemental selenium to
selenide: experiments with anoxic sediments and bacteria that respire Se-oxyanions.
Geomicrobiology Journal 20: 587-602.
Hoai, D. X. (2001). Investigation of selenium in sediment and water in Port Kembla
Harbour, NSW, Australia. CHEM 911 Report, Faculty of Science, University of
Wollongong.
Hoch, M., Alonso-Azcarate, J. and Lischick, M. (2002). Adsorption behavior of toxic
tributyltin to clay-rich sediments under various environmental conditions.
Environmental Toxicology and Chemistry 21: 1390-1397.
Hoenjet, K. M. J. L. F., Dagnelie, P. C., Delaere, K. P. J., Wijckmans, N. E. G., Zambon, J.
V. and Oosterhof, G. O. N. (2005). Effect of a Nutritional Supplement Containing
Vitamin E, Selenium, Vitamin C and Coenzyme Q10 on Serum PSA in Patients
with Hormonally Untreated Carcinoma of the Prostate: A Randomised PlaceboControlled Study. European Urology 47: 433-440.
Huang, H. Q. and Nanson, G. C. (1997). Vegetation and channel variation; a case study of
four small streams in southeastern Australia. Geomorphology 18: 237-249.
Hyne, R. V., Hogan, A. C., Pablo, F. and Roach, A. C. (2002). Toxicity of
Selenomethionine- and Seleno-Contaminated Sediment to the Amphipod
Corophium sp. Ecotoxicology and Environmental Safety 52: 30-37.
Jackson, B. P. and Miller, W. P. (1998). Arsenic and selenium speciation in coal fly ash
extracts by ion chromatography-inductively coupled plasma mass spectrometry.
Journal of Analytical Atomic Spectrometry 13: 1107-1112.
166
Jackson, B. P. and Miller, W. P. (1999). Soluble Arsenic and Selenium Species in Fly
Ash/Organic Waste-Amended Soils Using Ion Chromatography-Inductively
Coupled Plasma Mass Spectrometry. Environmental Science and Technology 33:
270-275.
Jackson, B. P. and Miller, W. P. (2000). Effectiveness of Phosphate and Hydroxide for
Desorption of Arsenic and Selenium Species from Iron Oxides. Soil Science Society
of America Journal 64: 1616-1622.
Jha, S. K., Krishnamoorthy, T. M., Pandit, G. G. and Nambi, K. S. V. (1999). History of
accumulation of mercury and nickel in Thane Creek, Mumbai, using 210Pb dating
technique. Science of the Total Environment 236: 91-99.
Johansson, L., Gafvelin, G. and Arner, E. S. J. (2005). Selenocysteine in proteins properties and biotechnological use. Biochimica et Biophysica Acta 1726: 1-13.
John, D. A. and Leventhal, J. S. (1995). "Bioavailability of metals" in Preliminary
compilation of descriptive geoenvironmental mineral deposit models. Geological
Survey Open-File Report 95-831. du Bray, E. A., U.S. Department of the Interior:
10-18.
Johnson, D. (1996). Determination of metals in a 3% sodium chloride (NaCl) matrix by
axially-viewed ICP-AES. ICP-AES Instruments At Work, Varian Instruments, Sugar
Land, Texas: No. ICP-19, December.
Jolley, D. F. (1999). The Accumulation and Storage of Se in Anadara Trapezia. PhD.
Thesis, Applied Ecology Research Group, Department of Applied Science,
University of Canberra.
Jolley, D. F., Maher, W. A. and Kyd, J. (2004). Selenium accumulation in the cockle
Anadara trapezia. Environmental Pollution 132: 203-212.
Judson, G. J. and Reuter, D. J. (1999). Selenium. In: Soil Analysis an Interpretation
Manual. Peverill, K. I., Sparrow, L. A. and Reuter, D. J. CSIRO Publishing,
Collingwood. VIC: 325-329.
Jung, R. F. and Batley, G. E. (2004). Freezing of sediments inappropriate for pore water
selenium analysis. Marine Pollution Bulletin 49: 295-298.
Kennish, M. J. (1997). Chapter 6: Heavy metals. Practical Handbook of Estuarine and
Marine Pollution. CRC Press, Boca Raton: 253-327.
Kersten, M. and Forstner, U. (1989). Speciation of trace elements in sediments. Trace
Element Speciation: Analytical Methods and Problems. Batley, G. E. CRC Press,
Boca Raton: 245-317.
167
Kirby, J., Maher, W. and Krikowa, F. (2001a). Selenium, cadmium, copper, and zinc
concentrations in sediments and mullet (Mugil cephalus) from the Southern Basin
of Lake Macquarie, NSW, Australia. Archives of Environmental Contamination and
Toxicology 40: 246-256.
Kirby, J., Maher, W. and Harasti, D. (2001b). Changes in selenium, copper, cadmium, and
zinc concentrations in mullet (Mugil cephalus) from the Southern Basin of Lake
Macquarie, Australia, in response to alteration of coal-fired power station fly ash
handling procedures. Archives of Environmental Contamination and Toxicology 41:
171-181.
Kos, V., Veber, M. and Hudnik, V. (1998). Determination of selenium in soil by hydride
generation AAS. Fresenius' Journal of Analytical Chemistry 360: 225-229.
Kotrebai, M., Birringer, M., Tyson, J. F., Block, E. and Uden, P. C. (1999). Selenium
speciation in enriched and natural samples by HPLC- ICP-MS and HPLC-ESI-MS
with perfluorinated carboxylic acid ion- pairing agents. Analyst 125: 71-78.
Koyama, H., Omura, K., Ejima, A., Kasanuma, Y., Watanabe, C. and Satoh, H. (1999).
Separation of Selenium-Containing Proteins in Human and Mouse Plasma Using
Tandem High-Performance Liquid Chromatography Columns Coupled with
Inductively Coupled Plasma-Mass Spectrometry. Analytical Biochemistry 267: 8491.
Lemly, A. D. and Smith, G. J. (1987). Aquatic cycling of selenium: implications for fish
and wildlife. Fish and Wildlife Leaflet, Fish and Wildlife Service, United States
Department of the Interior, Washington D.C. 12: 1-10.
Lemly, A. D. (1993). Teratogenic effects of selenium in natural populations of freshwater
fish. Ecotoxicology and Environmental Safety 26: 181-204.
Lemly, A. D. (1995). A protocol for aquatic hazard assessment of selenium. Ecotoxicology
and Environmental Safety 32: 280-288.
Lemly, A. D. (1997a). Environmental implications of excessive selenium: a review.
Biomedical and Environmental Sciences: BES 10: 415-435.
Lemly, A. D. (1997b). A teratogenic deformity index for evaluating impacts of selenium on
fish populations. Ecotoxicology and Environmental Safety 37: 259-266.
Lemly, A. D. (1998). A position paper on selenium in ecotoxicology: a procedure for
deriving site-specific water quality criteria. Ecotoxicology and Environmental
Safety 39: 1-9.
168
Lemly, A. D. (1999a). Selenium impacts on fish: An insidious time bomb. Human and
Ecological Risk Assessment 5: 1139-1151.
Lemly, A. D. (1999b). Selenium Transport and Bioaccumulation in Aquatic Ecosystems: A
Proposal for Water Quality Criteria Based on Hydrological Units. Ecotoxicology
and Environmental Safety 42: 150-156.
Lemly, A. D. (2002). Symptoms and implications of selenium toxicity in fish: the Belews
Lake case example. Aquatic Toxicology 57: 39-49.
Lemly, A. D. (2004). Aquatic selenium pollution is a global environmental safety issue.
Ecotoxicology and Environmental Safety 59: 44-56.
Li, F. S., Goessler, W. and Irgolic, K. J. (1998). Optimization of microwave digestion for
determination of selenium in human urine by flow injection-hydride generationatomic absorption spectrometry. Analytical Communications 35: 361-364.
Lim, T.-T. and Goh, K.-H. (2005). Selenium extractability from a contaminated fine soil
fraction: implication on soil cleanup. Chemosphere 58: 91-101.
Linville, R. G., Luoma, S. N., Cutter, L. and Cutter, G. A. (2002). Increased selenium threat
as a result of invasion of the exotic bivalve Potamocorbula amurensis into the San
Francisco Bay-Delta. Aquatic Toxicology 57: 51-64.
Lipton, D. S. (1991). Associations of selenium with inorganic and organic constituents in
soils of a semi-arid region. PhD Desertation, Graduate Division of Soil Science,
University of California Berkeley.
Louderback, T. (1976). Selenium and the environment. Mineral Industries Bulletin.
Colorado School of Mines Research Institute: 1-14.
Lovley, D. R. (1995). Microbial reduction of iron, manganese and other metals. Advances
in Agronomy 54: 175-218.
Low, C. (1998). An investigation into the physical and geochemical nature of the
particulate matter found in the wet sediment and airfall deposits within the Port
Kembla Inner harbour area. Honours Bachelor of Environmental Science Thesis,
Faculty of Science, University of Wollongong.
Luoma, S. N., Johns, C., Fisher, N. S., Steinberg, N. A., Oremland, R. S. and Reinfelder, J.
R. (1992). Determination of Selenium bioavailability to a benthic bivalve from
particulate and solute pathways. Environmental Science and Technology 26: 485.
169
Lussier, C., Veiga, V. and Baldwin, S. (2003). The geochemistry of selenium associated
with coal waste in the Elk River Valley, Canada. Environmental Geology 44: 905913.
Lyons, G. H., Judson, G. J., Ortiz-Monasterio, I., Genc, Y., Stangoulis, J. C. R. and
Graham, R. D. (2005). Selenium in Australia: Selenium status and biofortification
of wheat for better health. Journal of Trace Elements in Medicine and Biology 19:
75-82.
MacGregor, R. A. (1997). The Geochemistry of Selenium in Sedimentary Environments:
Examples from the UK and Jordan, PhD Thesis. Postgraduate Research Institute for
Sedimentology. University of Reading.
Magee, R. J. (1996). Selenium: the element with two faces. Main Group Elements and
Their Compounds. Das, K. V. G. Narosa Publishing, New Delhi, India: 387-397.
Maher, W., Baldwin, S., Deaker, M. and Irving, M. (1992). Characteristics of selenium in
Australian marine biota. Applied Organometallic Chemistry 6: 103-112.
Maher, W., Deaker, M., Jolley, D., Krikowa, F. and Roberts, B. (1997). Selenium
occurrence, distribution and speciation in the cockle Anadara trapezia and the
mullet Mugil cephalus. Applied Organometallic Chemistry 11: 313-326.
Maher, W. A. and Batley, G. E. (1990). Organometallics in the nearshore marine
environment of Australia. Applied Organometallic Chemistry 4: 419-37.
Marine Science & Ecology and Coastal Environmental Consultants Pty. Ltd. (1992). Port
Kembla Harbour Study 1991. Prepared for BHP Steel, Slab and Plate Products
Division, Marine Science & Ecology and Coastal Environmental Consultants Pty.
Ltd., Essendon, Victoria.
Marine Science & Ecology (1996). Allans Creek Study 1995. Prepared for BHP Steel SPPD
Environment Department, Marine Science & Ecology, Essendon, Victoria.
Marques, A. N. J., Monna, F., Silva-Filho, E. V., Fernex, F. E. and Simoes-Filho, F. F. L.
(2006). Apparent discrepancy in contamination history of a sub-tropical estuary
evaluated through 210Pb profile and chronostratigraphical markers. Marine Pollution
Bulletin 52: 532-539.
Martens, D. A. and Suarez, D. L. (1997). Selenium speciation of soil/sediment determined
with sequential extractions and hydride-generation atomic absorption
spectrophotometry. Environmental Science and Technology 31: 133-139.
170
Martinez, M., Gimenez, J., de Pablo, J., Rovira, M. and Duro, L. (2006). Sorption of
selenium(IV) and selenium(VI) onto magnetite. Applied Surface Science 252: 37673773.
Martley, E. and Gulson, B. L. (2003). Heavy metal partitioning in soil profiles in the
vicinity of an industrial complex and potential health implications. Journal De
Physique Iv 107: 831-834.
Martley, E., Gulson, B. L. and Pfeifer, H.-R. (2004). Metal concentrations in soils around
the copper smelter and surrounding industrial complex of Port Kembla, NSW,
Australia. Science of the Total Environment 325: 113-127.
Masscheleyn, P. H., Delaune, R. D. and Patrick, W. H., Jr. (1990). Transformations of
selenium as affected by sediment oxidation-reduction potential and pH.
Environmental Science and Technology 24: 91-6.
Masscheleyn, P. H., Delaune, R. D. and Patrick, W. H., Jr. (1991). Biogeochemistry
behaviour of selenium in anoxic soils and sediments: an equilibrium
thermodynamics approach. Journal of Environmental Science and Health 26: 555573.
Masscheleyn, P. H. and Patrick, W. H., Jr. (1993). Biogeochemical processes affecting
selenium cycling in wetlands. Environmental Toxicology and Chemistry 12: 22352243.
Masset, S., Monteil-Rivera, F., Dupont, L., Dumonceau, J. and Aplincourt, M. (2000).
Influence of humic acid on sorption of Co(II), Sr(II), and Se(IV) on goethite.
Agronomie 20: 525-535.
Matthai, C., Birch, G. F., Jenkinson, A. and Heijnis, H. (2001). Physical resuspension and
vertical mixing of sediments on a high energy continental margin (Sydney,
Australia). Journal of Environmental Radioactivity 52: 67-89.
McNaughton, S. A. and Marks, G. C. (2002). Selenium content of Australian foods: A
review of literature values. Journal of Food Composition and Analysis 15: 169-182.
McNeal, J. M. and Balistrieri, L. S. (1989). Geochemistry and Occurrence of Selenium: An
Overview. Selenium in Agriculture and the Environment. Jacobs, L. W. (Ed)
American Society of Agronomy and Soil Science Society of America, Madison,
Wisconsin: 1-13.
Mercone, D., Thomson, J., Croudace, I. W. and Troelstra, S. R. (1999). A coupled natural
immobilisation mechanism for mercury and selenium in deep-sea sediments.
Geochimica et Cosmochimica Acta 63: 1481-1488.
171
Miessler, G. L. (1999). Inorganic chemistry. Upper Saddle River, N.J., Prentice Hall.
Moran, P. J. (1984). Water quality control and its effect on the concentration of heavy
metals in Port Kembla Harbour, N.S.W. Marine Pollution Bulletin 15: 294-297.
Moreno, P., Quijano, M. A., Gutierrez, A. M., Perez-Conde, M. C. and Camara, C. (2004).
Study of selenium species distribution in biological tissues by size exclusion and
ion exchange chromatagraphy inductively coupled plasma-mass spectrometry.
Analytica Chimica Acta 524: 315-327.
Morse, J. W. and Luther III, G. W. (1999). Chemical influences on trace metal-sulfide
interactions in anoxic sediments. Geochimica et Cosmochimica Acta 63: 3373-3378.
Mortimer, M. R. (2000). Pesticide and trace metal concentrations in Queensland estuarine
crabs. Marine Pollution Bulletin 41: 359-366.
Muhammad, D. (2003). Fractionation of arsenic in Port Kembla Harbour sediments. PhD
Thesis, Department of Chemistry, University of Wollongong.
Muller, A. (2002). Organic carbon burial rates, and carbon and sulfur relationships in
coastal sediments of the southern Baltic Sea. Applied Geochemistry 17: 337-352.
Nagpal, N. K. and Howell, K. (2001). Water quality guidelines for selenium: Technical
appendix. Water Protection Branch, Water, Lands and Air Protection, Victoria,
British Columbia.
Nazarenko, I. I. (1972). Analytical Chemistry of Selenium and Tellurium. Jerusalem, Israel
Program for Scientific Translations.
Niedzielski, P. (2005). The new concept of hyphenated analytical system: Simultaneous
determination of inorganic arsenic(III), arsenic(V), selenium(IV) and selenium(VI)
by high performance liquid chromatography-hydride generation-(fast sequential)
atomic absorption spectrometry during single analysis. Analytica Chimica Acta 551:
199-206.
NIWQP (1998). Guidelines for Interpretation of the Biological Effects of Selected
Constituents in Biota, Water, and Sediment: Selenium. Information Report No. 3,
National Irrigation Water Quality Program. United Stated Department of the
Interior, Washington DC.: 139-184.
Nobbs, D. M., Barford, J. P., Carroll, B. I. and Peters, G. M. (1997). An integrated
investigation of anthropogenic selenium contamination in Lake Macquarie, NSW.
Pure and Applied Chemistry 69: 2387-2401.
172
Ochsenkuhn-Petropoulou, M., Michalke, B., Kavouras, D. and Schramel, P. (2003).
Selenium speciation analysis in a sediment using strong anion exchange and
reversed phase chromatography coupled with inductively coupled plasma-mass
spectrometry. Analytica Chimica Acta 478: 219-227.
Ohlendorf, H. M. (2002). The birds of Kesterson Reservoir: a historical perspective.
Aquatic Toxicology 57: 1-10.
Oldfield, F. and Appleby, P. G. (1984). Empirical testing of 210Pb-dating models for lake
sediments. Lake Sediments and Environmental History. Hayworth, E. Y. and Lund,
J. W. G. (Eds). Leicester University Press: 93-124.
Olivas, R. M., Donard, O. F. X., Camara, C. and Quevauviller, P. (1994). Analytical
techniques applied to the speciation of selenium in environmental matrices.
Analytica Chimica Acta 286: 357-370.
Olivas, R. M. and Donard, O. F. X. (1998). Microwave assisted reduction of Se-VI to SeIV and determination by HG/FI-ICP/MS for inorganic selenium speciation. Talanta
45: 1023-1029.
Opresko, D. M. (1993). Toxicity summary for selenium and selenium compounds, Chemical
hazard evaluation group, Biomedical environmental information analysis section,
Health and Safety Research Division, Oak Ridge National Laboratory*, Tennessee,
prepared for Oak Ridge Reservation Environmental Restoration Program.*Managed
by Martin Marietta Energy Systems, Inc., for the U.S. Department of Energy under
Contract No. DE-AC05-84OR21400.
Oremland, R. S. (1994). Biogeochemical transformations of selenium in anoxic
environments. Selenium in the Environment. Frankenburger, W. T. and Benson, S.
(Eds). Marcel Dekker, New York: 389-419.
Oremland, R. S. and Stolz, J. F. (2000). Dissimilatory reduction of selenate and arsenate in
nature. Environmental Microbe-Metal Interactions. Lovley, D. R. (Ed). ASM Press,
Washington D.C.: 199-224.
Orero Iserte, L., Roig-Navarro, A. F. and Hernandez, F. (2004). Simultaneous
determination of arsenic and selenium species in phosphoric acid extracts of
sediment samples by HPLC-ICP-MS. Analytica Chimica Acta 527: 97-104.
Ornemark, U. and Olin, A. (1994). Determination of dissolved selenium (VI) in freshwater.
Talanta 41: 1675-1681.
Orr, P. L., Guiguer, K. R. and Russel, C. K. (2006). Food chain transfer of selenium in
lentic and lotic habitats of a western Canadian watershed. Ecotoxicology and
Environmental Safety 63: 175-188.
173
Peters, G. M., Maher, W. A., Barford, J. P. and Gomes, V. G. (1997). Selenium
associations in estuarine sediments: Redox effects. Water, Air and Soil Pollution 99:
275-282.
Peters, G. M., Maher, W. A., Jolley, D., Carroll, B. I., Gomes, V. G., Jenkinson, A. V. and
McOrist, G. D. (1999a). Selenium contamination, redistribution and remobilisation
in sediments of Lake Macquarie, NSW. Organic Geochemistry 30: 1287-1300.
Peters, G. M., Maher, W. A., Krikowa, F., Roach, A. C., Jeswani, H. K., Barford, J. P.,
Gomes, V. G. and Reible, D. D. (1999b). Selenium in sediments, pore waters and
benthic infauna of Lake Macquarie, New South Wales, Australia. Marine
Environmental Research 47: 491-508.
Pezzarossa, B. and Petruzzelli, G. (2001). Chapter 9 Selenium contamination in soil:
Sorption and desorption processes. Heavy Metals Release in Soils. Magdi Selim, H.
and Sparks, D. L. (Eds). CRC Press, Boca Raton: 191-206.
Phillips, M. (2002). Monitoring and assessing the water and sediment quality of Port
Kembla Harbour according to the ANZECC and ARMCANZ (2000) guidelines.
Honours Bachelor of Environmental Science Thesis, Faculty of Science, University
of Wollongong.
Pilon-Smits, E. A. H., de Souza, M. P., Hong, G., Amini, A., Bravo, R. C., Payabyab, S. T.
and Terry, N. (1999). Selenium volatilization and accumulation by twenty aquatic
plant species. Journal of Environmental Quality 28: 1011-1018.
Potin-Gautier, M., Gilon, N., Astruc, M., De Gregori, I. and Pinochet, H. (1997).
Comparison of selenium extraction procedures for its speciation in biological
materials. International Journal of Environmental Analytical Chemistry 67: 15-25.
Pyrzynska, K. (1995). Separation of Inorganic Selenium Species on Anion-Exchange
Resins. Analyst 120: 1933-1936.
Pyrzynska, K. (1998). Speciation of selenium compounds. Analytical Sciences 14: 479-483.
Radojevic, M. and Bashkin, V. (1999). Practical Environmental Analysis. Cambridge,
Royal Society of Chemistry.
Raessler, M., Michalke, B., Schulte-Hostede, S. and Kettrup, A. (2000). Long-term
monitoring of arsenic and selenium species in contaminated groundwaters by HPLC
and HG-AAS. Science of the Total Environment 258: 171-181.
Rayman, M. (2002). Selenium brought to earth. Chemistry in Britain October: 28-31.
174
Rayman, M. P. (2000). The importance of selenium to human health. Lancet 356: 233-241.
Reinfelds, I. and Nanson, G. C. (2004). Aspects of the hydrogeomorphology of Illawarra
streams: implications for planning and design of urbanising landscapes. Wetlands
(Australia) 21: 220-236.
Rue, E. L., Smith, G. J., Cutter, G. A. and Bruland, K. W. (1997). The response of trace
element redox couples to suboxic conditions in the water column. Deep Sea
Research Part I: Oceanographic Research Papers 44: 113-134.
Saiki, M. K. and Ogle, R. S. (1995). Evidence of impaired reproduction by western
mosquitofish inhabiting seleniferous agricultural drainwater. Transactions of the
American Fisheries Society 124: 578-587.
Sappington, K. G. (2002). Development of aquatic life criteria for selenium: a regulatory
perspective on critical issues and research needs. Aquatic Toxicology 57: 101-113.
Scanes, P. R. and Roach, A. C. (1999). Determining natural "background" concentrations
of trace metals in oysters from New South Wales, Australia. Environmental
Pollution 105: 437-446.
Schlekat, C. E., Dowdle, P. R., Lee, B.-G., Luoma, S. N. and Oremland, R. S. (2000).
Bioavailability of particle-associated Se to the bivalve Potamocorbula amurensis.
Environmental Science and Technology 34: 4504.
Schloske, L., Waldner, H. and Marx, F. (2002). Optimisation of sample pre-treatment in the
HG-AAS selenium analysis. Analytical and Bioanalytical Chemistry 372: 700-704.
Schneider, P. M. and Davey, S. B. (1995). Sediment contaminants off the coast of Sydney,
Australia: a model for their distribution. Marine Pollution Bulletin 31: 262-272.
Schulthess, C. P. and Hu, Z. (2001). Impact of chloride anions on proton and selenium
adsorption by an aluminum oxide. Soil Science Society of America Journal 65: 710718.
Schwarz, K. and Faltz, C. M. (1957). Selenium as an integral part of factor 3 against dietary
necrotic liver degeneration. Journal of the American Chemistry Society 79: 32923293.
Schweizer, U., Braeuer, A. U., Koehrle, J., Nitsch, R. and Savaskan, N. E. (2004).
Selenium and brain function: a poorly recognized liaison. Brain Research Reviews
45: 164-178.
175
Seby, F., Potin Gautier, M., Lespes, G. and Astruc, M. (1997). Selenium speciation in soils
after alkaline extraction. Science of the Total Environment 207: 81-90.
Seby, F., Potin-Gautier, M., Giffaut, E., Borge, G. and Donard, O. F. X. (2001). A critical
review of thermodynamic data for selenium species at 25 C. Chemical Geology 171:
173-194.
Shamberger, R. J. (1983). Biochemistry of Selenium. New York, Plenum Press.
Sharmasarkar, S. and Vance, G. F. (1995). Fractional partitioning for assessing solid-phase
speciation and geochemical transformations of soil selenium. Soil Science 160: 4355.
Sharmasarkar, S. and Vance, G. F. (1997). Extraction and distribution of soil organic and
inorganic selenium in coal mine environments of Wyoming, USA. Environmental
Geology (Berlin) 29: 17-22.
Sheppard, M. I. and Stephenson, M. (1995). Critical evaluation of selective extraction
methods for soil and sediments. Contaminated soils. Paris (France), May 15-19, Ed.
INRA, Paris 1997 (Les Colloques, n 85).
Shrift, A. (1964). A selenium cycle in nature? Nature 201: 1304-1305.
Siddique, T., Okeke, B. C., Zhang, Y., Arshad, M., Han, S. K. and Frankenberger, W. T.,
Jr. (2005). Bacterial Diversity in Selenium Reduction of Agricultural Drainage
Water Amended with Rice Straw. Journal of Environmental Quality 34: 217-226.
Sloss, C. R., Murray-Wallace, C. V., Jones, B. G. and Wallin, T. (2004). Aspartic acid
racemisation dating of mid-Holocene to recent estuarine sedimentation in New
South Wales, Australia: a pilot study. Marine Geology 212: 45-59.
Spallholz, J. E. and Hoffman, D. J. (2002). Selenium toxicity: cause and effects in aquatic
birds. Aquatic Toxicology 57: 27-37.
Spencer, K. L., Cundy, A. B. and Croudace, I. W. (2003). Heavy metal distribution and
early-diagenesis in salt marsh sediments from the Medway Estuary, Kent, UK.
Estuarine Coastal and Shelf Science 57: 43-54.
Stewart, A. R., Luoma, S. N., Schlekat, C. E., Doblin, M. A. and Hieb, K. A. (2004). Food
web pathway determines how selenium affects aquatic ecosystems: a San Francisco
Bay case study. Environmental Science and Technology 38: 4519-4526.
Stolz, J. F. and Oremland, R. S. (1999). Bacterial respiration of arsenic and selenium.
FEMS Microbiology Reviews 23: 615-627.
176
Stone, Y., Ahern, C. R. and Blunden, B. (1998). Acid Sulphate Soil Manual 1998.
ASSMAC, Wollongbar, NSW.
Stumm, W. (1992). Chapter 2 The coordination chemistry of the hydrous oxide-water
interface. Chemistry of the Solid-Water Interface, John Wiley & Sons, New York:
13-41.
Summers, R. N. and Pech, J. D. (1997). Nutrient and metal content of water, sediment and
soils amended with bauxite residue in the catchment of the Peel Inlet and Harvey
Estuary, Western Australia. Agriculture, Ecosystems and Environment 64: 219-232.
Swaine, D. J. (1990). Trace Elements in Coal. Butterworth, London: 148-153.
Szpunar, J. and Lobinski, R. (1999). Speciation in the environmental field - trends in
Analytical Chemistry. Fresenius Journal of Analytical Chemistry 363: 550-557.
Takayanagi, K. and Belzile, N. (1988). Profiles of dissolved and acid-leachable selenium in
a sediment core from the lower St. Lawrence Estuary. Marine Chemistry 24: 30714.
Tam, S.-C., Chow, A. and Hadley, D. (1995). Effects of organic component on the
immobilization of selenium on iron oxyhydroxide. Science of the Total Environment
164: 1-7.
Tessier, A., Campbell, P. G. C. and Bisson, M. (1979). Sequential extraction procedure for
the speciation of particulate trace metals. Analytical Chemistry 51: 844-850.
Tinggi, U. (2003). Essentiality and toxicity of selenium and its status in Australia: a review.
Toxicology Letters 137: 103-110.
Tokunaga, T. K., Lipton, D. S., Benson, S. M., Yee, A. W., Oldfather, J. M., Duckart, E.
C., Johannis, P. W. and Halvorsen, K. E. (1991). Soil selenium fractionation, depth
profiles and time trends in a vegetated site at Kesterson Reservoir. Water, Air, and
Soil Pollution 57-58: 31-41.
Tokunaga, T. K., Brown, G. E. J., Pickering, I. J., Sutton, S. R. and Bajt, S. (1997).
Selenium redox reactions and transport between ponded waters and sediments.
Environmental Science and Technology 31: 1419 -1425.
Trumble, J. T., Kund, G. S. and White, K. K. (1998). Influence of form and quantity of
selenium on the development and survival of an insect herbivore. Environmental
Pollution 101: 175-182.
177
U.S. Environmental Protection Agency (1994). Method 3051: Microwave assisted acid
digestion of sediments, sludges, soils, and oils, U.S. Environmental Protection
Agency, Washington DC.
Uden, P. C. (2002). Modern trends in the speciation of selenium by hyphenated techniques.
Analytical and Bioanalytical Chemistry 373: 422-431.
Van Derveer, W. D. and Canton, S. P. (1997). Selenium sediment toxicity thresholds and
derivation of water quality criteria for freshwater biota of western streams.
Environmental Toxicology and Chemistry 16: 1260-1268.
van Staden, J. F., van der Merwe, L., Kempster, P. L. and van Vliet, H. R. (2000). Rapid
sample preparation using closed-vessel microwave digestion for determining trace
metals in fish tissue and sediment. South African Journal of Chemistry 53: 23-27.
Vanclay, E. (2003). Email communication. AA Product Marketing Manager, Varian
Australia Pty Ltd.
Vassileva, E., Becker, A. and Broekaert, J. A. C. (2001). Determination of arsenic and
selenium species in groundwater and soil extracts by ion chromatography coupled
to inductively coupled plasma mass spectrometry. Analytica Chimica Acta 441:
135-146.
Velinsky, D. J. and Cutter, G. A. (1990). Determination of elemental selenium and pyriteselenium in sediments. Analytica Chimica Acta 235: 419-425.
Velinsky, D. J. and Cutter, G. A. (1991). Geochemistry of selenium in a coastal salt marsh.
Geochimica et Cosmochimica Acta 55: 179-191.
Voth-Beach, L. M. and Shrader, D. E. (1985). Reduction of interferences in the
determination of arsenic and selenium by hydride generation. Spectroscopy Nov:
60.
Wake, B. D., Butler, E. C. V., Featherstone, A. M., Virtue, P., Averty, B. and Michel, P.
(2004). Determination of inorganic selenium species in marine waters by Hydride
Generation-AFS. Australian Journal of Chemistry 57: 937-943.
Wakeham, S. (2002). Chapter 6. Diagenesis of Organic Matter at the Water-Sediment
Interface. Chemistry of Marine Water and Sediments. Gianguzza, A., Pelizzetti, E.
and Sammartano, S. (Eds). Springer, Berlin: 147-164.
Wang, M. C. and Chen, H. M. (2003). Forms and distribution of selenium at different
depths and among particle size fractions of three Taiwan soils. Chemosphere 52:
585-593.
178
Wang, Z., Gao, Y. X. and Belzile, N. (2001). Microwave digestion of environmental and
natural waters for selenium speciation. Analytical Chemistry 73: 4711-4716.
Welz, B. and Schubert-Jacobs, M. (1991). Evaluation of a flow injection system and
optimization of parameters for hydride generation atomic absorption spectrometry.
Atomic Spectroscopy 12: 91-104.
Weres, O., Jaouni, A.-R. and Tsao, L. (1989). The distribution, speciation and geochemical
cycling of selenium in a sedimentary environment, Kesterson Reservoir, California,
U.S.A. Applied Geochemistry 4: 543-563.
Whant, L. M. (2003). Heavy metal contaminants in surface sediments from Port Kembla
Harbour, NSW Australia. CHEM915 Report. Master of Environmental Science
Program, Faculty of Science, University of Wollongong.
White, K. (2001). An assessment of water and sediment quality in creeks of the Southern
Wollongong City Council area, NSW. Honours Bachelor of Environmental Science
Thesis, Faculty of Science, University of Wollongong.
Wlodarczkyk, J. and Beath, K. (1997). Heavy metals in seafood in Lake Macquarie: a
cross-sectional survey. Report to the Hunter Public Health Unit. University of
Newcastle Research Association, Callaghan NSW and John Wlodarczkyk
Consulting Services, New Lambton, NSW.
Wright, M. T., Parker, D. R. and Amrhein, C. (2003). Critical evaluation of the ability of
sequential extraction procedures to quantify discrete forms of selenium in sediments
and soils. Environmental Science and Technology 37: 4709.
Wright, W. G. (1999). Oxidation and mobilization of selenium by nitrate in irrigation
drainage. Journal of Environmental Quality 28: 1182-1187.
Zawislanski, P. T., Benson, S. M., Terberg, R. and Borglin, S. E. (2003). Selenium
speciation, solubility, and mobility in land-disposed dredged sediments.
Environmental Science and Technology 37: 2415-2420.
Zhang, Y. and Moore, J. N. (1996). Selenium Fractionation and Speciation in a Wetland
System. Environmental Science and Technology 30: 2613-2619.
Zhang, Y. and Moore, J. N. (1997). Environmental conditions controlling selenium
volatilization from a wetland system. Environmental Science and Technology 31:
511.
Zhang, Y., Frankenberger Jr., W. T. and Moore, J. N. (1999a). Measurement of selenite in
sediment extracts by using hydride generation atomic absorption spectrometry.
Science of the Total Environment 229: 183-193.
179
Zhang, Y. Q., Moore, J. N. and Frankenberger, W. T. (1999b). Speciation of soluble
selenium in agricultural drainage waters and aqueous soil-sediment extracts using
hydride generation atomic absorption spectrometry. Environmental Science and
Technology 33: 1652-1656.
Zhang, Y. and FrankenbergerJr., W. T. (2000). Effect of moisture and casein on
demethylation of trimethylselenonium in soil. Science of the Total Environment
257: 111-119.
Zhang, Y., Amrhein, C. and Frankenberger, J., William T. (2005). Effect of arsenate and
molybdate on removal of selenate from an aqueous solution by zero-valent iron.
Science of the Total Environment 350: 1-11.
Zhang, Y. Q. and Frankenberger, W. T. (2003). Determination of selenium fractionation
and speciation in wetland sediments by parallel extraction. International Journal of
Environmental Analytical Chemistry 83: 315-326.
Zhang, Y. Q., Siddique, T., Wang, J. and Frankenberger, W. T. (2004a). Selenate reduction
in river water by Citerobacter freundii isolated from a selenium-contaminated
sediment. Journal of Agricultural and Food Chemistry 52: 1594-1600.
Zhang, Y. Q., Zahir, Z. A. and Frankenberger, W. T. (2004b). Fate of colloidal-particulate
elemental selenium in aquatic systems. Journal of Environmental Quality 33: 559564.
Zheng, B., Ding, Z., Huang, R., Zhu, J., Yu, X., Wang, A., Zhou, D., Mao, D. and Su, H.
(1999). Issues of health and disease relating to coal use in southwestern China.
International Journal of Coal Geology 40: 119-132.
Zhou, C. Y., Wong, M. K., Koh, L. L. and Wee, Y. C. (1997). Evaluation of high-pressure
microwave digestion methods for hydride generation atomic absorption
spectrometric determination of total selenium and arsenic in sediment.
Mikrochimica Acta 127: 77-83.
Zhu, J., Zuo, W., Liang, X., Li, S. and Zheng, B. (2004). Occurrence of native selenium in
Yutangba and its environmental implications. Applied Geochemistry 19: 461-467.
Zingaro, R. A., Dufner, D. C., Murphy, A. P. and Moody, C. D. (1997). Reduction of
oxoselenium anions by iron (II) hydroxide. Environment International 23: 299-304.
APPENDIX A
SURFACE SAMPLE DATA
Table A.1
Site
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19a
19b
20
21
22
23
GPS, pH and grain size data for surface sediment samples collected on 7th April 2003.
GPS coordinates
Oxic/
pH
Mean pH
East
North
Anoxic
Oxic
Anoxic
0306766
0306721
0306718
0306407
0306221
0305840
0305872
0306612
0306757
0306835
0306920
0307320
0307628
0307176
0307255
0307424
0307314
0307688
0307839
0307839
0307920
0308290
0308519
0308098
6184424
6184493
6184493
6184717
6184860
6185152
6185498
6185678
6185531
6185058
6184654
6184453
6184445
6184197
6183815
6183750
6183622
6183440
6183194
6183194
6183200
6183801
6184277
6184447
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
-
7.63
7.43
7.37
7.59
7.38
7.38
7.19
7.25
7.31
7.78
7.33
7.42
7.34
7.57
7.49
7.53
7.41
7.55
7.35
7.45
7.44
7.48
7.52
7.48
7.37
7.52
7.67
7.82
7.7
7.42
7.7
7.45
7.67
7.52
7.57
7.41
7.65
7.69
7.41
7.58
7.43
7.61
7.41
7.38
7.19
7.28
7.78
7.34
7.44
7.39
7.53
7.51
7.51
7.39
7.54
7.67
7.82
7.70
7.44
7.67
7.52
7.63
7.41
7.62
7.43
% Grain size (µm)
< 63
63-250
>250
46
83
85
83
80
44
78
66
83
70
83
30
48
21
8
4
32
58
1
17
55
40
1
8
23
13
11
10
17
30
19
20
4
12
16
48
17
46
31
3
4
7
3
26
3
14
14
18
1
22
35
33
92
81
28
8
95
55
8
20
57
25
15
40
34
4
28
37
40
42
66
II
Table A.2
Trace element concentrations (µg/g) in different grain size fractions (µm) of surface sediment samples.
Site
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19a
19b
20
21
22
23
Se
Cr
Cu
Fe
Total
< 63
63-250
>250
Total
<63
63-250
>250
Total
<63
63-250
>250
Total
<63
63-250
>250
1.25
1.29
1.63
1.69
0.90
1.03
1.05
0.79
0.65
0.98
1.07
0.57
0.86
0.61
0.10
0.23
1.13
3.03
1.16
6.45
1.58
0.84
0.07
0.14
1.42
1.36
2.15
1.54
1.23
1.17
1.10
1.07
0.96
1.05
1.05
1.29
1.74
1.66
0.26
2.82
2.03
4.09
1.81
3.71
0.64
1.74
NA
0.84
0.55
1.04
0.58
1.22
0.51
0.50
0.64
0.36
0.40
0.50
0.63
0.40
0.48
0.48
NS
0.18
0.86
1.76
6.93
4.58
4.59
0.54
0.24
NA
1.04
0.67
0.47
1.69
0.48
0.74
0.55
0.33
0.49
0.49
0.82
0.43
0.38
0.25
0.08
0.19
0.38
1.04
0.77
2.43
5.74
0.59
0.37
0.28
100
113
127
123
123
120
159
87
95
95
107
66
54
45
13
11
55
114
18
50
122
52
10
16
141
136
165
124
150
162
166
125
132
135
117
138
127
144
52
122
133
205
147
167
154
99
NA
69
73
70
77
55
75
80
133
44
46
56
79
50
40
38
NS
13
52
93
44
41
95
32
11
NA
73
48
42
64
45
70
96
24
18
24
44
18
5
10
7
8
18
30
19
23
61
24
8
9
270
224
181
125
206
165
194
151
80
202
226
219
175
317
67
58
771
513
300
1398
7123
197
22
30
263
269
3194
102
255
236
204
215
102
239
252
456
369
1038
151
670
1403
663
5814
3955
7996
319
NA
181
113
206
111
221
128
78
139
79
41
116
194
167
150
256
NS
93
930
481
929
1514
5509
117
13
NA
301
125
60
595
78
94
72
51
14
164
145
75
41
46
71
19
132
140
130
489
4836
152
17
19
52282
43019
44210
41950
61345
39002
43304
33147
33440
33898
37447
40890
25707
28468
11912
8907
29714
40297
11007
22001
47870
25658
7389
29013
39344
28268
34989
38150
34315
27402
29984
34893
26530
28784
33142
39005
38271
43450
27746
37160
33158
31348
42899
38766
34854
28053
NA
48453
73468
63557
63451
41757
73649
32389
35756
24531
24395
30532
40891
34652
27777
27638
NS
9061
28714
39887
21917
25546
37231
17401
8801
NA
44737
47699
35992
52663
64376
20950
26326
10562
12016
9669
20713
10008
3584
8649
5706
5071
11204
14531
10731
12786
21862
9955
4977
8957
NS = No sample, NA = Not analysed due to insufficient sample
III
Table A.2
(continued).
Site
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19a
19b
20
21
22
23
Mn
Ni
Pb
Zn
Total
<63
63-250
>250
Total
<63
63-250
>250
Total
<63
63-250
>250
Total
<63
63-250
>250
768
517
478
569
582
523
582
385
243
445
495
406
247
250
144
61
291
350
146
181
370
254
233
403
573
449
452
581
614
538
462
440
262
521
565
564
486
480
382
355
382
397
480
365
398
383
NA
1049
897
753
680
391
552
443
593
308
227
426
544
402
275
272
NS
79
355
309
182
201
323
186
173
NA
727
682
471
483
430
397
320
126
98
248
330
195
40
109
66
36
134
134
65
76
126
146
266
234
23
28
27
24
31
37
35
22
25
24
26
16
12
10
4
3
12
26
11
29
314
12
2
5
29
31
102
24
31
37
31
28
31
29
28
28
25
27
8
25
26
37
88
90
407
22
NA
17
20
25
26
17
24
34
38
16
17
21
26
14
10
10
NS
3
13
28
27
28
221
8
3
NA
17
19
15
29
69
28
32
8
10
21
18
7
3
4
3
2
5
10
10
18
504
5
2
3
205
191
182
218
179
165
192
150
94
187
179
170
155
147
55
27
187
308
145
344
2003
123
18
46
259
246
1232
197
253
272
230
223
135
235
197
257
328
373
90
278
364
492
913
1059
2885
203
NA
93
115
156
123
100
124
80
137
89
59
120
178
180
146
144
NS
30
202
308
254
339
1500
89
17
NA
266
140
70
381
79
95
89
53
21
231
127
85
23
47
46
11
84
107
144
217
1779
95
17
27
1219
732
842
1525
640
745
781
607
585
779
613
618
475
492
250
68
546
905
425
687
2374
375
56
219
1180
837
1827
1402
827
1017
875
843
811
775
674
922
1032
1244
402
732
1004
1336
1657
1526
2931
610
NA
595
632
614
638
509
482
473
720
409
296
461
555
647
456
513
NS
69
640
848
705
724
1884
288
57
NA
2193
646
430
2152
386
557
552
282
125
1516
411
238
72
141
205
28
240
260
631
480
1522
206
45
95
NS = No sample, NA = Not analysed due to insufficient sample.
IV
Table A.3
Correlations (r)* between selenium and common trace metals in different grain size fractions of surface sediments from
Port Kembla Harbour sites, excluding Red Beach area (Sites 18, 19 and 20).
> 250 µm (n = 20)
Whole surface sediment (n = 20)
Se
Cr
0.814
Cr
Cu
0.402
Fe
0.708
Mn
0.682
Ni
0.722
Pb
0.889
Zn
0.858
0.142
0.835
0.787
0.969
0.810
0.771
0.249
0.129
0.105
0.529
0.205
0.877
0.815
0.809
0.741
0.758
0.744
0.789
0.744
0.683
Cu
Fe
Mn
Ni
Pb
Table A.3
Cr
0.661
Cu
0.909
Fe
0.638
Mn
0.651
Ni
0.392
Pb
0.874
Zn
0.802
0.483
0.686
0.730
0.635
0.565
0.592
0.545
0.504
0.276
0.942
0.859
0.835
0.799
0.579
0.570
0.525
0.624
0.663
0.358
0.348
0.837
0.955
(Continued).
63-250 µm (n = 18)†
Se
Cr
Cu
Fe
Mn
Ni
Cr
0.398
< 63 µm (n = 18)†
Cu
0.464
Fe
0.448
Mn
0.451
Ni
0.358
Pb
0.486
Zn
0.569
0.032
0.597
0.742
0.947
0.457
0.771
0.006
0.015
-0.081
0.617
0.389
0.904
0.571
0.424
0.661
0.707
0.464
0.755
0.333
0.670
Pb
0.840
Cr
0.104
Cu
0.567
Fe
0.118
Mn
-0.429
Ni
0.280
Pb
0.484
Zn
0.353
0.351
-0.399
-0.458
0.532
0.475
0.554
0.110
-0.198
0.851
0.952
0.705
0.649
-0.136
0.028
0.181
-0.185
-0.197
-0.134
0.948
0.725
0.782
* Values highlighted in bold are significant at P < 0.0001 level.
†
No samples/data for Sites 15 and 22.
V
1
4
2
3
10
11
5
6
7
18
19b
8
12
9
13
21
14
19a
17
15
16
22
23
Figure A.1
Se
Cu
Pb
Zn
Cr
Fe
Mn
Ni
Se
Cu
Pb
Cr
Ni
Fe
Mn
Zn
1
4
19b
2
3
11
6
7
10
5
8
12
9
17
21
18
19a
13
16
14
15
22
23
Dendrograms showing correlation patterns of selenium and common trace metals in whole (left) and > 250 µm (right)
fractions of surface sediments from Port Kembla Harbour, excluding Site 20.
VI
1
2
3
5
4
10
12
17
6
11
7
18
8
9
13
14
21
16
22
19a
19b
20
Figure A.2
Se
Cr
Cu
Ni
Pb
Zn
Fe
Mn
Se
Cu
Pb
Ni
Zn
Cr
Fe
Mn
1
12
13
4
14
2
10
6
7
5
8
11
9
21
16
17
18
23
3
19b
19a
Dendrograms (hierarchical clustering analysis) of selenium and common trace metals in 63-250 (left) and <63
µm (right) fractions of Port Kembla Harbour surface sediments.
VII
APPENDIX B
CORE SAMPLE DATA
Table B.1
Sampling
Date
Summary of all core samples.
Core
25/2/03
A1
25/6/03
A2
A3
16/4/04
B1
B2
B3
B4
B5
B6
29/7/05
C1
C2
C3
C4
16/1/06
D1
GPS
UTM
(E/N)
Length
(cm)
pH
18
~ Distance
from the
Darcy Rd
drain (m)
80
030 7889
618 3192
030 7853
618 3222
030 7950
618 3214
030 7890
618 3200
030 7875
618 3183
030 7812
618 3212
030 7891
618 3220
030 7904
618 3193
030 7832
618 3214
030 7791
618 3252
030 7784
618 3251
030 7971
618 3192
030 7925
618 3186
030 7924
618 3208
Sample treatment and analysis
20
Sieved
Porewater
Total Se
SEP 1
SEP 2
x
Redox
potential
-
Macrocomponents
x
-
x
-
-
90 West
x
-
x
-
x
-
-
22
140 East
x
-
x
-
x
-
-
20
90
x
x
x
x
x
x
-
-
14
65
x
x
x
x
x
x
-
-
14
80 West
x
x
x
x
x
x
-
-
16
110
x
x
x
x
x
x
-
-
14
100 East
x
x
x
x
x
x
-
-
20
70
x
x
x
x
x
x
-
-
18
120 West
x
x
-
x
x
-
x
x
14
120 West
x
x
-
x
x
-
x
x
16
150 East
x
x
-
x
x
-
x
x
24
90 East
x
x
-
x
x
-
x
x
36
120 East
Pb-210 dating
IX
Table B.2
pH values of core samples.
Depth (cm)
A1
7.60
7.71
7.74
7.66
7.78
7.99
8.02
7.92
7.86
1
3
5
7
9
11
13
15
17
19
21
23
Table B.3
2003 cores
A2
7.69
7.74
8.06
8.11
8.16
8.22
8.19
8.03
8.02
8.12
A3
7.34
7.52
7.41
7.52
7.67
7.63
7.92
7.99
8.09
7.83
8.07
B1
7.47
7.55
7.63
7.65
7.69
7.84
7.83
7.79
7.71
7.64
B2
7.65
7.85
7.86
7.86
7.82
7.85
7.97
2004 cores
B3
B4
7.53
7.74
7.87
7.79
7.85
7.77
7.75
7.79
7.77
7.87
7.75
7.8
7.75
7.85
7.85
B5
7.81
7.94
7.88
7.9
7.88
8.15
8.13
B6
7.49
7.67
7.65
7.79
7.79
7.72
7.81
7.67
7.8
7.89
C1
7.65
7.75
7.76
7.77
7.82
7.79
7.77
7.85
7.82
2004 cores
B3
B4
-330
-182
-350
-394
-341
-394
-369
-384
-363
-401
-365
-387
-357
-406
-377
B5
-285
-359
-396
-362
-382
-387
-394
B6
-235
-382
-409
-412
-373
-387
-392
-358
-374
-346
C1
+303
-228
-290
-287
-340
-326
-310
-337
-358
2005 cores
C2
C3
7.54
7.7
7.61
7.72
7.76
7.72
7.73
7.85
7.67
7.9
7.69
7.89
7.67
7.9
7.88
C4
7.61
7.64
7.71
7.75
7.82
7.82
7.8
7.76
7.75
7.78
7.8
7.75
Redox potentials (mV) of core samples*.
Depth (cm)
1
3
5
7
9
11
13
15
17
19
21
23
* - : not analysed.
A1
-
2003 cores
A2
A3
-
-
B1
-305
-400
-384
-382
-395
-410
-407
-393
-409
-371
B2
-278
-370
-412
-406
-422
-416
-416
2005 cores
C2
C3
+197
+224
-318
-362
-352
-374
-342
-382
-379
-421
-342
-397
-359
-404
-375
C4
+201
-337
-387
-392
-412
-428
-402
-376
-362
-305
-333
-313
X
Table B.4
Percentage grain size of < 63 µm, 63-250 µm and > 250 µm fractions in sediment A, B and C cores.
2003 cores
Depth
A1
A2
A3
(cm) <63µm 63-250µm >250µm <63µm 63-250µm >250µm <63µm 63-250µm >250µm
1
24
20
56
63
24
13
54
36
10
3
37
13
50
54
33
14
64
28
8
5
11
9
81
64
26
10
71
23
6
7
4
9
87
50
33
17
66
23
11
9
6
13
80
62
24
14
58
22
20
11
48
24
28
39
18
44
37
34
28
13
50
21
29
34
19
47
48
25
27
15
20
27
54
35
31
35
45
30
25
17
23
26
51
41
30
29
51
21
27
19
67
25
9
62
21
17
21
42
22
73
Table B.4
B1
45
41
49
54
55
59
53
55
59
38
2004 cores (all < 63 µm)
B2
B3
B4
B5
42
45
39
50
58
48
45
73
76
78
78
75
66
58
64
76
76
78
45
63
81
93
68
70
61
71
51
56
51
B6
63
59
45
37
31
47
43
40
68
81
(continued).
2005 cores
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
<63µm
36
46
48
66
65
71
55
48
72
C1
63-250µm
30
27
27
24
19
16
30
36
23
>250µm
35
27
25
11
16
14
15
16
5
<63µm
28
33
41
44
36
36
24
C2
63-250µm
36
34
30
30
37
28
41
>250µm
36
33
28
26
27
36
35
<63µm
21
24
23
29
29
29
36
18
C3
63-250µm
44
43
38
41
45
40
39
48
>250µm
35
34
39
29
25
31
25
35
<63µm
43
30
33
42
44
48
40
19
8
12
21
20
C4
63-250µm
32
36
36
24
25
21
22
27
23
21
30
26
>250µm
25
34
31
34
31
31
38
54
69
67
49
54
XI
Table B.5
Depth
Porewater sulfate and phosphate concentrations (mg/L) in Cores C1-C4*. Total porewater volume extracted (mL) vs the
sample solid wt. before the porewater extraction (g) are included for information.
Core C1
SO4 PO4 Vol. Solid wt
2851 ND 15.0
70.3
2767 ND 13.5
48.9
2585 43 15.0
62.4
2437 48 13.0
59.9
2341 47 11.0
69.3
2036 47 6.5
57.6
1469 53 9.0
68.5
905
19 12.0
78.2
451
20 13.0
69.0
1
3
5
7
9
11
13
15
17
19
21
23
*NS = No Sample, ND = Not Detected.
Table B.6
Core C2
PO4 Vol. Solid wt
ND 14.0
78.7
ND 12.5
81.9
ND 11.0
62.9
ND 10.0
67.7
37 9.5
89.4
35 4.5
61.8
41 5.0
93.9
Core C3
SO4 PO4 Vol. Solid wt
2876 ND 11.0
68.5
2833 ND 6.5
63.9
2861 ND 7.5
63.8
2657 ND 6.8
64.7
2287 47 7.5
76.3
2028 45 6.5
57.7
1589 52 8.5
65.4
NS
NS 1.5
40.6
Core C4
SO4 PO4 Vol. Solid wt
2671 ND 13.0
68.0
2741 ND 8.5
64.3
2487 47 13.0
71.7
1864 55 11.0
57.6
1120 56 8.5
57.4
586 59 11.0
70.4
406 59 8.0
68.3
411 47 3.5
70.9
NS NS NS
68.9
NS NS 2.0
76.2
842 ND 3.0
82.0
1019 45 4.0
97.9
Porewater selenium concentrations (µg/L) in Cores B1-B6 and C1-C4.
Depth (cm)
1
3
5
7
9
11
13
15
17
19
21
23
SO4
2876
2869
2940
2833
2788
2879
2671
B1
22.0
24.5
10.0
13.5
17.0
20.5
19.5
33.0
77.5
28.0
B2
11.0
14.0
15.5
10.5
10.5
10.0
10.0
2004 Cores
B3
B4
11.5
6.0
6.0
51.5
7.5
99.5
10.5
39.5
31.0
37.0
13.0
28.5
15.0
20.5
13.0
B5
4.0
6.0
10.5
9.5
9.5
13.0
9.5
B6
25.5
23.5
26.0
82.5
59.5
31.0
14.5
16.5
20.0
19.0
C1
1.00
1.50
1.75
1.50
1.75
4.75
4.25
3.25
3.00
2005 Cores
C2
C3
1.25
1.50
0.75
1.25
0.75
0.50
0.50
1.25
0.75
2.00
5.25
7.50
2.25
3.00
2.25
C4
0.75
2.25
2.50
5.50
7.00
3.25
2.00
24.25
33.80
43.25
26.50
7.00
XII
Table B.7
Depth
(cm)
1
3
5
7
9
11
13
15
17
Table B.7
Sediment macrocomponent concentrations (% dry wt, whole sediment) in Cores C1-C4*.
AVS
0.122
0.228
0.223
0.307
0.282
0.158
0.102
0.348
0.670
CrRS
0.984
1.369
1.435
1.626
1.753
0.691
1.556
1.684
2.107
Core C1
TC
TOC
6.05
2.73
8.78
2.32
11.60
2.85
6.40
2.06
5.29
1.56
2.89
2.32
3.59
2.64
3.66
2.99
4.60
3.44
TS
1.09
2.63
1.84
2.87
2.13
0.86
2.18
2.57
3.32
TN
0.24
0.39
0.25
0.27
0.22
0.13
0.18
0.18
0.25
AVS
0.044
0.057
0.160
0.168
0.113
0.152
0.120
CrRS
0.480
0.645
0.932
0.752
0.513
0.532
0.435
Core C2
TC
TOC
13.70
2.43
13.49
2.61
11.69
2.80
6.54
2.26
3.33
1.56
6.56
1.50
3.69
1.59
TS
0.60
1.05
1.50
1.13
0.68
0.90
0.72
TN
0.32
0.29
0.36
0.23
0.14
0.16
0.23
(continued).
Depth
Core C3
Core C4
(cm)
TC
TOC
TS
TN
TC
TOC
TS
TN
AVS
CrRS
AVS
CrRS
0.022
0.387
0.061
0.378
1
13.18
2.25
0.68
0.41
6.01
2.10
0.60
0.28
0.024
0.511
0.098
0.871
3
12.97
2.09
0.77
0.42
9.01
1.60
1.02
0.24
0.064
0.709
0.064
0.749
5
21.09
2.50
1.04
0.49
13.86
2.85
1.17
0.41
0.074
1.005
0.182
1.363
7
18.22
3.50
1.42
0.52
21.57
4.65
2.58
0.60
0.101
1.648
0.458
1.948
9
25.06
3.03
2.43
0.52
37.65
6.40
3.08
0.69
0.134
1.358
0.328
2.673
11
28.27
4.48
2.24
0.68
19.21
5.45
4.70
0.40
0.166
1.956
0.133
2.427
13
14.24
3.84
3.18
0.42
24.60
3.44
4.32
0.44
0.078
2.142
0.024
1.897
15
18.50
2.32
2.33
0.33
19.57
4.30
2.98
0.26
0.020
0.954
17
1.28
0.58
1.07
0.09
0.016
0.672
19
8.95
0.77
1.19
0.13
0.016
0.893
21
2.35
1.28
1.53
0.10
0.030
1.693
23
13.81
1.46
2.60
0.31
*AVS = Acid Volatile Sulfides; CrRS = Chromium Reducible Sulfides (pyrites); TC = Total Carbon; TOC = Total Organic Carbon; TS = Total
Sulfur; and TN = Total Nitrogen.
XIII
Table B.8
Total selenium concentrations (µg/g, d.w.) in different grain sizes of the sediment A, B and C cores.
Depth (cm)
1
3
5
7
9
11
13
15
17
19
21
Table B.8
<63
44
127
428
1735
1620
230
87
49
71
A1
63-250
11
66
61
85
101
67
36
17
16
Whole
6
16
14
170
19
10
14
40
17
12
<63
187
177
149
212
81
36
95
152
89
134
A3
63-250
80
73
84
126
88
8
7
8
11
15
>250
-
Whole
21
19
20
18
11
18
15
58
19
15
10
<63
54
54
57
54
35
95
107
310
65
23
43
63-250
10
13
16
19
7
13
7
12
16
8
4
>250
-
(continued).
Depth (cm)
1
3
5
7
9
11
13
15
17
19
21
23
A2
>250
-
B1
70
171
218
263
207
141
142
66
29
54
B2
37
107
128
169
211
113
44
Core Bs (all in < 63 µm sediment)
B3
B4
B5
128
71
124
136
188
139
121
490
182
72
73
100
53
33
62
39
56
32
31
36
31
17
B6
331
140
94
70
80
74
53
70
98
86
C1
20.7
38.5
56.8
59.1
28.0
19.1
31.5
20.8
36.5
Core Cs (all in whole sediment)
C2
C3
C4
13.8
13.5
18.3
23.6
19.7
31.3
27.7
16.9
57.3
23.1
40.0
131.4
19.1
64.2
88.3
14.7
47.7
83.2
13.7
47.3
41.3
46.4
65.9
14.1
13.2
19.2
20.4
XIV
Table B.9
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
Table B.9
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
Concentrations of Selenium (µg/g, d.w.) in sequential fractions (SEP 1) of Cores B1-B6 samples (< 63 µm sediment).
Soluble &
adsorbed
2.00
2.81
2.28
0.53
2.87
0.93
0.51
1.55
0.59
0.95
Core B1
Carbonate Fe-Mn Org &
oxides sulfides
2.98
0.92
39.1
6.31
1.11
95.7
3.65
1.08
95.2
1.61
0.62
56.1
4.46
2.52
140.5
1.19
0.77
50.4
0.65
0.59
35.7
1.00
0.95
51.9
0.28
0.30
19.4
0.48
0.71
33.5
Soluble &
adsorbed
0.23
1.13
0.45
0.80
0.90
0.84
0.84
Core B2
Carbonate Fe-Mn Org &
oxides sulfides
0.31
0.07
14.7
1.20
0.91
52.0
1.10
1.11
54.1
1.43
0.91
55.6
1.24
1.46
76.1
0.77
0.89
47.7
0.32
0.36
29.7
Core B3
Soluble & Carbonate Fe-Mn Org &
adsorbed
oxides sulfides
1.34
2.49
0.93
73.8
0.88
1.43
1.02
81.0
0.99
0.76
0.79
56.7
0.71
0.83
0.57
42.5
0.34
0.42
0.37
31.0
0.53
0.46
0.50
23.3
0.70
0.40
0.47
21.0
Soluble &
adsorbed
1.33
2.22
1.36
2.02
1.34
1.93
0.64
Core B5
Carbonate Fe-Mn Org &
oxides sulfides
3.09
1.19
89.1
2.06
1.29
83.3
1.72
1.47
73.0
1.54
1.11
53.7
0.70
0.47
31.3
0.34
0.33
21.3
0.24
0.18
21.0
Core B6
Soluble & Carbonate Fe-Mn Org &
oxides sulfides
adsorbed
6.29
3.22
2.75
150.2
1.44
1.21
1.02
47.2
0.85
0.86
0.46
40.6
1.63
0.93
0.50
45.3
2.18
1.43
0.72
67.7
1.53
0.80
0.58
53.0
0.99
0.60
0.47
28.7
0.66
0.80
0.61
28.8
1.12
1.16
1.51
48.8
0.96
1.06
1.05
42.0
(continued).
Soluble &
adsorbed
1.59
2.60
9.08
0.71
0.84
1.14
1.25
1.14
Core B4
Carbonate Fe-Mn Org &
oxides sulfides
1.80
0.92
56.6
1.93
1.10
89.2
5.78
3.72
213.6
0.50
0.27
32.9
0.48
0.24
24.4
1.06
0.58
27.4
1.01
0.61
20.3
0.32
0.14
11.3
XV
Table B.10
Depth
(cm)
1
3
5
7
9
11
13
15
17
Table B.10
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
Concentrations of Selenium (µg/g, d.w.) in sequential fractions (SEP 2) of Cores C1-C4 samples (whole sediment).
Core C1
Soluble & Organically Elemental Org & Residual Total Se
selenium sulfides
adsorbed
bound
0.5
6.4
9.2
0.8
5.1
20.7
1.5
11.1
22.2
1.6
7.9
38.5
1.3
9.9
28.9
8.5
4.5
56.8
1.9
9.3
34.2
12.4
5.8
59.1
0.8
5.0
17.7
4.9
3.6
28.0
3.4
4.5
8.5
0.9
2.5
19.1
2.8
6.9
21.7
6.0
3.0
31.5
1.8
5.3
14.3
1.8
2.8
20.8
2.9
8.4
20.3
10.9
2.2
36.5
Core C2
Soluble & Organically Elemental Org & Residual Total Se
selenium sulfides
adsorbed
bound
0.3
4.6
6.1
2.4
2.7
13.8
0.6
6.1
10.9
7.6
2.6
23.6
0.7
6.0
17.3
2.6
5.6
27.7
0.5
2.2
10.7
3.9
3.3
23.1
0.5
3.6
10.5
4.0
1.3
19.1
0.4
2.7
5.9
3.4
1.6
14.7
0.5
2.7
6.7
1.9
2.2
13.7
(continued).
Core C3
Soluble & Organically Elemental Org & Residual Total Se
selenium sulfides
adsorbed
bound
0.4
2.9
7.2
1.0
2.7
13.5
0.5
3.5
8.3
2.0
3.2
19.7
0.7
4.8
8.0
1.0
4.6
16.9
0.5
9.1
10.4
1.8
13.5
40.0
1.0
8.5
23.4
10.5
12.2
64.2
1.6
10.5
23.4
10.5
6.1
47.7
3.2
8.6
28.9
11.8
5.1
47.3
2.4
5.8
28.1
4.9
5.5
46.4
Core C4
Soluble & Organically Elemental Org & Residual Total Se
selenium sulfides
adsorbed
bound
0.7
7.5
6.9
3.3
3.1
18.3
0.7
6.8
12.4
4.6
4.5
31.3
2.3
22.5
13.2
4.4
7.2
57.3
3.4
29.6
61.0
12.8
13.0
131.4
2.8
9.2
45.3
11.7
9.6
88.3
4.0
15.4
35.7
10.1
6.1
83.2
1.7
7.2
22.4
6.1
6.6
41.3
2.5
7.7
24.0
9.0
11.6
65.9
1.2
2.2
4.9
1.9
4.0
14.1
1.1
1.9
3.1
1.9
4.1
13.2
1.2
2.2
4.1
2.5
5.8
19.2
1.2
2.9
5.3
2.8
6.7
20.4
XVI
Table B.11
Depth
(cm)
1
3
5
7
9
11
13
15
17
Table B.11
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
Concentrations of Chromium (µg/g, d.w.) co-extracted in the sequential extracts (SEP 2) of Cores C1-C4 samples.
Core C1
Soluble & Organically Elemental Org & Residual Total Cr
selenium sulfides
adsorbed
bound
1.93
0.47
ND
ND
29.1
28.8
1.95
0.77
ND
ND
32.3
31.4
1.86
1.68
ND
ND
42.0
45.3
1.92
2.16
ND
ND
64.1
75.2
1.91
3.12
ND
ND
38.9
39.9
1.13
0.07
ND
ND
25.3
23.5
2.56
2.87
ND
ND
43.0
45.3
1.41
0.28
ND
ND
53.1
59.0
3.20
0.05
ND
ND
67.3
59.5
Core C2
Soluble & Organically Elemental Org & Residual Total Cr
selenium sulfides
adsorbed
bound
2.17
0.15
ND
ND
35.2
35.4
2.17
0.72
ND
ND
35.5
33.4
2.77
1.80
ND
ND
48.4
48.5
1.18
0.58
ND
ND
53.2
45.5
2.16
0.52
ND
ND
52.5
43.2
1.99
0.63
ND
ND
24.3
24.1
1.90
0.02
ND
ND
46.3
39.0
(continued).
Core C3
Soluble & Organically Elemental Org & Residual Total Cr
selenium sulfides
adsorbed
bound
5.05
0.55
ND
ND
46.7
53.7
5.12
0.69
ND
ND
45.2
65.2
4.70
0.56
ND
ND
50.4
66.3
5.93
5.22
ND
ND
53.0
67.4
4.64
0.72
ND
ND
73.4
103.7
5.92
2.70
ND
ND
80.2
81.7
5.02
2.19
ND
ND
92.8
98.9
5.15
1.83
ND
ND
47.7
67.2
Core C4
Soluble & Organically Elemental Org & Residual Total Cr
selenium sulfides
adsorbed
bound
4.76
0.70
4.02
ND
54.8
63.3
3.77
0.26
3.50
ND
43.7
53.2
4.78
1.37
3.87
ND
48.7
61.8
5.49
1.36
4.80
ND
74.3
82.1
3.24
2.08
4.05
ND
81.9
83.9
3.69
3.22
4.59
ND
89.1
94.7
2.78
2.25
3.96
ND
70.1
82.7
3.12
1.43
3.78
ND
36.8
61.9
2.67
0.41
3.29
ND
20.3
31.6
2.57
0.18
3.41
ND
27.7
33.3
2.91
0.01
3.85
ND
23.8
42.8
2.42
0.06
3.11
ND
36.1
40.1
XVII
Table B.12
Depth
(cm)
1
3
5
7
9
11
13
15
17
Table B.12
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
Concentrations of Copper (µg/g, d.w.) co-extracted in the sequential extracts (SEP 2) of Cores C1-C4 samples.
Core C1
Soluble & Organically Elemental Org & Residual Total Cu
selenium sulfides
adsorbed
bound
10.81
155.8
185.7
44.0
2131
2206
10.08
126.0
218.2
30.7
2616
2944
9.07
94.7
191.8
176.1
2408
3316
10.33
63.3
287.3
375.5
3007
3578
5.28
82.4
109.2
107.7
1828
1842
3.30
30.0
30.4
12.0
893
891
5.10
55.4
89.6
22.0
1106
1079
4.09
18.0
94.4
24.8
1342
1286
5.23
6.2
362.6
38.5
1966
2003
Core C2
Soluble & Organically Elemental Org & Residual Total Cu
selenium sulfides
adsorbed
bound
5.87
79.7
109.0
34.5
1105
1169
7.31
60.0
101.2
46.6
1415
1701
8.58
61.8
282.5
22.3
1798
1701
5.18
33.5
74.0
33.5
1308
1245
3.53
24.8
87.9
4.8
725
747
3.23
25.5
68.0
6.0
782
770
3.97
22.5
143.3
2.8
560
689
(continued).
Core C3
Soluble & Organically Elemental Org & Residual Total Cu
selenium sulfides
adsorbed
bound
9.54
153.2
72.8
12.7
1155
1251
8.62
121.9
84.8
15.5
1528
1607
7.30
105.7
81.3
28.0
1564
1695
9.86
114.0
438.5
25.0
2378
2468
6.55
40.1
293.5
9.1
2331
3480
7.60
70.9
405.2
53.5
2389
3367
7.55
38.5
429.2
166.7
2161
3359
7.87
77.2
538.7
293.8
1388
2035
Core C4
Soluble & Organically Elemental Org & Residual Total Cu
selenium sulfides
adsorbed
bound
9.04
150.8
105.8
57.6
1982
1876
7.53
86.8
123.9
61.1
1769
1885
11.67
96.8
578.8
153.9
2468
4009
9.06
86.7
359.2
190.1
3671
5450
4.67
35.5
270.5
75.9
2549
3630
5.94
39.2
359.1
118.7
2825
4003
4.37
46.6
291.5
571.4
2050
2925
5.33
84.5
292.3
594.6
1607
2278
4.46
134.2
139.2
136.6
584
867
3.85
66.7
132.1
21.1
411
525
3.95
45.7
111.5
9.4
638
733
3.74
41.9
151.6
63.1
998
1192
XVIII
Table B.13
Depth
(cm)
1
3
5
7
9
11
13
15
17
Table B.13
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
Concentrations of Iron (µg/g, d.w.) co-extracted in the sequential extracts (SEP 2) of Cores C1-C4 samples.
Core C1
Soluble & Organically Elemental Org & Residual Total Fe
selenium sulfides
adsorbed
bound
7.1
18.9
1.28
ND
19595
19607
4.2
21.0
0.03
ND
19331
20269
13.9
21.8
1.19
ND
17903
20696
61.3
22.2
0.93
ND
22204
24741
15.0
18.3
0.41
ND
20913
22136
1.6
14.2
0.63
ND
20048
25402
1.8
24.6
0.63
ND
21127
24178
62.8
16.2
0.58
ND
20310
24886
186.0
25.8
1.05
ND
28597
33077
Core C2
Soluble & Organically Elemental Org & Residual Total Fe
selenium sulfides
adsorbed
bound
2.3
14.3
0.48
ND
17391
19219
11.2
14.1
0.43
ND
17060
19504
16.1
18.6
0.14
ND
23240
22445
9.9
14.1
0.11
ND
16335
21357
5.8
13.4
0.1
ND
18766
19714
6.6
13.9
0.12
ND
18195
17942
4.4
14.9
0.02
ND
23117
26280
(continued).
Core C3
Soluble & Organically Elemental Org & Residual Total Fe
selenium sulfides
adsorbed
bound
0.6
18.1
1.97
ND
17787
17557
0.4
17.1
0.21
ND
17952
21779
0.3
19.2
0.08
ND
17826
21910
2.2
23.0
0.39
ND
21734
21847
11.8
21.7
0.55
ND
20961
24740
3.4
26.6
0.14
ND
21922
21694
33.2
19.5
0.02
ND
19789
26024
3.9
21.8
0.02
ND
18508
19656
Core C4
Soluble & Organically Elemental Org & Residual Total Fe
selenium sulfides
adsorbed
bound
1.1
16.7
8.51
0.05
18078
18463
0.6
12.8
5.68
0.40
16168
16603
7.2
22.0
7.47
0.64
18234
18259
8.1
27.1
9.27
0.37
21808
20035
5.8
19.1
7.22
0.06
18893
17887
47.1
19.9
11.13
3.51
20907
21504
2.0
17.8
6.87
3.51
18768
19347
0.2
25.8
6.05
96.9
17672
14989
0.9
12.4
6.02
0.22
8363
8484
1.1
14.6
4.76
1.63
8776
9334
0.5
11.9
5.28
1.83
12088
13489
0.1
15.4
4.25
1.55
11877
14955
XIX
Table B.14
Depth
(cm)
1
3
5
7
9
11
13
15
17
Table B.14
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
Concentrations of Manganese (µg/g, d.w.) co-extracted in the sequential extracts (SEP 2) of Cores C1-C4 samples.
Core C1
Soluble & Organically Elemental Org & Residual Total Mn
selenium sulfides
adsorbed
bound
ND
ND
ND
ND
64
62
ND
ND
ND
ND
64
78
ND
ND
ND
ND
72
86
ND
ND
ND
ND
100
103
ND
ND
ND
ND
32
30
ND
ND
ND
ND
30
26
ND
ND
ND
ND
91
98
ND
ND
ND
ND
114
143
ND
ND
ND
5.50
149
166
Core C2
Soluble & Organically Elemental Org & Residual Total Mn
selenium sulfides
adsorbed
bound
ND
ND
ND
ND
91
101
ND
ND
ND
ND
86
81
ND
ND
ND
ND
98
85
ND
ND
ND
ND
111
125
ND
ND
ND
ND
116
168
ND
ND
ND
ND
84
94
ND
ND
ND
ND
111
146
(continued).
Core C3
Soluble & Organically Elemental Org & Residual Total Mn
selenium sulfides
adsorbed
bound
ND
ND
ND
ND
86
130
ND
ND
ND
ND
93
153
ND
ND
ND
ND
84
88
ND
ND
ND
ND
81
111
ND
ND
ND
ND
113
122
ND
ND
ND
ND
90
87
ND
ND
ND
ND
94
115
ND
ND
ND
3.82
49
49
Core C4
Soluble & Organically Elemental Org & Residual Total Mn
selenium sulfides
adsorbed
bound
ND
ND
0.93
ND
78
111
ND
ND
0.74
ND
67
66
ND
ND
0.87
ND
71
61
ND
ND
1.17
ND
90
82
ND
ND
1.02
ND
32
30
ND
ND
1.14
2.51
76
75
ND
ND
0.92
3.49
63
72
ND
ND
1.42
4.86
31
29
ND
ND
1.29
1.17
14
14
ND
ND
0.93
0.94
10
9
ND
ND
0.83
0.86
12
14
ND
ND
0.83
4.33
89
104
XX
Table B.15
Depth
(cm)
1
3
5
7
9
11
13
15
17
Table B.15
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
Concentrations of Nickel (µg/g, d.w.) co-extracted in the sequential extracts (SEP 2) of Cores C1-C4 samples.
Core C1
Soluble & Organically Elemental Org & Residual Total Ni
selenium sulfides
adsorbed
bound
2.50
2.05
ND
ND
26
26
1.20
3.64
ND
ND
52
53
1.65
3.10
0.09
ND
55
52
2.64
3.08
ND
ND
31
34
1.89
2.69
ND
ND
14
15
1.75
1.39
ND
ND
10
11
1.62
2.74
ND
ND
31
32
1.91
2.01
ND
ND
22
23
3.76
2.86
ND
ND
33
33
Core C2
Soluble & Organically Elemental Org & Residual Total Ni
selenium sulfides
adsorbed
bound
1.58
1.79
ND
ND
31
36
1.84
2.01
ND
ND
45
43
1.80
2.55
ND
ND
33
32
1.52
1.54
ND
ND
17
17
1.51
1.63
ND
ND
7
10
2.02
1.59
ND
ND
6
8
1.36
2.01
ND
ND
31
44
(continued).
Core C3
Soluble & Organically Elemental Org & Residual Total Ni
selenium sulfides
adsorbed
bound
4.50
3.49
ND
ND
57
67
3.22
4.47
ND
ND
61
90
3.98
3.27
ND
ND
59
96
4.68
6.02
ND
ND
109
116
3.96
3.67
ND
ND
108
112
4.63
3.75
ND
ND
90
99
4.74
5.04
ND
ND
73
99
3.29
3.79
ND
ND
54
87
Core C4
Soluble & Organically Elemental Org & Residual Total Ni
selenium sulfides
adsorbed
bound
4.20
3.58
1.51
ND
76
93
3.37
5.88
1.67
ND
120
130
6.81
6.74
3.31
ND
466
438
6.34
9.72
2.82
ND
254
268
3.82
7.29
1.93
ND
109
129
3.03
4.47
1.65
ND
91
114
3.24
4.01
1.15
ND
68
98
2.82
2.73
1.21
ND
62
97
2.22
3.01
1.37
ND
32
54
2.02
3.62
0.73
ND
27
56
3.21
1.11
1.07
ND
44
77
1.66
2.47
0.79
ND
49
59
XXI
Table B.16
Depth
(cm)
1
3
5
7
9
11
13
15
17
Table B.16
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
Concentrations of Lead (µg/g, d.w.) co-extracted in the sequential extracts (SEP 2) of Cores C1-C4 samples.
Core C1
Soluble & Organically Elemental Org & Residual Total Pb
selenium sulfides
adsorbed
bound
6.5
65.2
ND
ND
425
454
6.4
108.6
ND
ND
715
743
11.6
100.2
ND
ND
955
941
40.1
32.2
ND
1.6
1193
1171
16.7
77.2
ND
ND
582
589
4.3
59.4
ND
ND
597
673
8.1
170.5
ND
ND
819
871
47.8
10.5
ND
ND
1526
1504
458.1
4.1
ND
21.7
3358
3622
Core C2
Soluble & Organically Elemental Org & Residual Total Pb
selenium sulfides
adsorbed
bound
5.6
52.8
ND
ND
326
354
10.3
66.9
ND
ND
522
513
13.8
70.6
ND
ND
607
586
2.8
27.4
ND
ND
432
453
6.6
31.8
1.7
ND
251
244
10.4
50.0
ND
ND
292
304
6.4
69.6
0.12
ND
334
448
(continued).
Core C3
Soluble & Organically Elemental Org & Residual Total Pb
selenium sulfides
adsorbed
bound
2.9
49.8
2.80
ND
305
387
2.9
54.7
ND
ND
374
466
4.4
49.6
ND
ND
424
509
9.6
134.0
ND
ND
557
714
12.9
81.3
ND
ND
870
943
3.3
137.2
ND
ND
1048
1072
22.4
46.8
ND
ND
991
1089
4.8
105.9
ND
230
283
568
Core C4
Soluble & Organically Elemental Org & Residual Total Pb
selenium sulfides
adsorbed
bound
3.7
71.9
8.89
ND
364
455
4.5
71.6
6.24
ND
441
495
10.5
166.3
7.20
ND
753
881
13.9
367.7
8.96
ND
2007
2038
8.1
143.6
8.66
ND
1696
1513
37.8
167.0
8.89
0.2
1650
1740
5.8
192.8
7.60
342
690
1003
4.8
218.8
8.97
281
271
653
4.9
88.4
6.73
34.7
137
231
2.1
71.9
6.26
2.4
182
217
4.4
54.1
7.79
0.9
174
255
2.6
99.9
5.39
18.0
326
347
XXII
Table B.17
Depth
(cm)
1
3
5
7
9
11
13
15
17
Table B.17
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
Concentrations of Zinc (µg/g, d.w.) co-extracted in the sequential extracts (SEP 2) of Cores C1-C4 samples.
Core C1
Soluble & Organically Elemental Org & Residual Total Zn
selenium sulfides
adsorbed
bound
1.47
74
0.68
4.5
541
635
1.56
213
0.48
4.6
770
945
2.83
301
1.01
4.0
999
1213
10.08
590
0.88
88.2
1267
1786
2.95
238
0.64
5.3
789
903
1.13
81
0.67
4.8
509
660
2.40
397
0.53
6.3
669
894
10.31
516
1.05
3.3
860
1173
46.06
389
1.05
49.1
1310
1684
Core C2
Soluble & Organically Elemental Org & Residual Total Zn
selenium sulfides
adsorbed
bound
1.29
89
0.67
3.6
552
623
2.74
206
0.43
3.7
788
908
4.10
334
0.61
5.2
962
1113
2.25
168
0.48
3.9
680
815
1.95
137
0.3
11.8
391
503
1.74
139
0.54
4.6
392
503
1.67
157
0.11
4.6
360
581
(continued).
Core C3
Soluble & Organically Elemental Org & Residual Total Zn
selenium sulfides
adsorbed
bound
1.65
45
0.39
5.9
568
707
1.98
45
0.39
5.5
731
774
1.73
63
0.40
4.2
687
815
2.60
252
0.51
4.9
831
1155
3.96
446
0.77
16.5
1156
1710
3.64
733
0.48
19.8
1384
1806
6.94
495
0.31
22.6
1258
1746
2.97
356
0.75
168.0
284
905
Core C4
Soluble & Organically Elemental Org & Residual Total Zn
selenium sulfides
adsorbed
bound
1.84
106
3.60
4.7
567
722
1.78
101
3.41
5.4
614
734
2.47
190
2.20
2.7
873
1104
4.03
785
4.52
5.9
1550
2122
3.01
675
3.52
3.0
1382
1695
8.44
693
3.21
125.3
1218
2085
2.91
558
3.56
275.9
513
1474
3.42
495
8.53
190.5
349
1084
2.48
173
8.84
43.6
506
617
1.79
165
3.47
22.4
591
631
2.25
113
2.16
15.5
340
474
2.99
234
1.46
59.8
489
762
XXIII
Table B.18
Core A1
1
3
5
7
9
11
13
15
17
Core A2
1
3
5
7
9
11
13
15
17
19
Core A3
1
3
5
7
9
11
13
15
17
19
21
Concentrations of total trace metals (µg/g, d.w.) in < 63 µm fractions of Cores A1-A3 samples.
As
74
312
422
496
657
234
101
126
212
As
457
222
134
125
37
15
103
165
411
860
As
55
50
64
61
100
175
380
250
57
46
199
Cd
4.1
14.8
33.0
86.0
58.9
34.3
36.9
25.7
35.5
Cd
30.9
27.9
25.4
29.3
29.3
10.7
7.8
11.0
34.9
54.5
Cd
5.0
13.1
21.2
21.1
10.3
13.5
11.8
8.8
11.2
2.5
4.7
Cr
145
145
171
173
185
186
184
96
89
Cr
183
176
204
200
193
93
121
219
227
149
Cr
174
207
211
171
127
249
210
222
167
112
109
Cu
4009
7841
14694
24421
29082
10310
6271
3535
7489
Cu
16616
11696
8820
8477
5740
2218
2932
4325
4584
5432
Cu
4006
4039
4331
2897
2072
5672
4694
4946
2934
1323
2493
Fe
58468
56395
51758
57063
60436
55308
53109
62678
63697
Fe
56183
56108
58437
56781
53935
48520
52708
57852
75878
74769
Fe
57171
62716
61126
60849
56848
73241
74957
69547
58773
51182
64589
Mn
444
406
345
321
380
356
338
572
647
Mn
378
392
372
347
327
189
264
531
558
518
Mn
456
472
538
545
499
695
746
630
450
417
645
Ni
84
321
428
315
435
143
56
49
63
Ni
492
232
91
73
51
29
85
133
77
72
Ni
51
49
50
38
34
56
64
77
35
29
40
Pb
1343
2315
5734
10989
10252
3429
2285
1714
2865
Pb
5049
3705
3271
3362
2264
985
2165
3916
5686
9530
Pb
1389
1574
1869
1360
1376
2449
4805
3776
1059
794
2715
Sb
14
23
53
135
109
57
37
37
63
Sb
62
56
44
46
34
14
29
46
94
180
Sb
15
19
22
17
15
34
54
40
17
10
22
Se
44
127
428
1735
1620
230
87
49
71
Se
187
177
149
212
81
36
95
152
89
134
Se
54
54
57
54
35
95
107
310
65
23
43
Zn
1976
2786
4634
7905
7502
4317
3047
2047
2830
Zn
5251
4734
4107
4655
3292
1600
2525
3921
4221
4521
Zn
2142
2602
2926
2165
1820
3749
3793
4132
1888
1384
2061
XXIV
Table B.19
Depth
(cm)
1
3
5
7
9
11
13
15
17
Table B.19
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
Concentrations of trace metals (µg/g, d.w.) co-extracted in the reactive iron fraction of Cores C1-C4 samples.
Cr
29.2
26.1
28.8
41.9
29.1
21.5
21.5
23.8
24.2
Cu
1456
1374
1212
1316
982
673
561
793
898
Fe
17887
17183
17419
21325
16114
14105
10110
14758
21630
Core C1
Mn
0.69
0.71
7.71
4.24
0.65
0.61
6.26
16.14
36.48
Ni
16.3
25.8
23.5
23.1
11.8
11.1
15.5
15.8
17.3
Fe
13878
17854
18924
18979
20213
20697
20959
10423
Core C3
Mn
9.99
0.64
3.90
0.66
0.67
0.73
12.45
0.17
Ni
52.2
72.8
68.0
100.4
98.3
75.6
60.4
56.2
Pb
371
505
638
725
437
470
530
983
2221
Zn
508
706
904
1256
738
491
599
880
1172
Cr
21.8
23.1
33.4
36.7
23.9
26.5
20.8
Pb
244
307
342
487
691
765
764
356
Zn
515
588
639
935
1569
1611
1408
701
Cr
56.3
46.7
43.9
62.0
85.4
71.3
54.5
34.9
22.7
23.7
23.4
26.7
Cu
837
936
1112
954
552
547
529
Fe
14022
14089
18536
17248
12404
12446
11991
Core C2
Mn
7.21
16.05
8.78
1.07
31.25
0.72
26.48
Ni
26.2
25.6
27.4
17.0
9.6
8.5
46.8
Pb
285
345
423
337
209
278
295
Zn
520
721
920
666
407
395
378
(continued).
Cr
33.3
42.9
43.7
52.4
67.9
61.7
66.1
43.6
Cu
808
1004
1089
1368
1761
1695
1536
1127
Cu
1211
1178
1728
2016
2291
1585
1265
906
445
344
389
501
Fe
17127
15140
14847
20186
20325
18067
11767
6713
4188
5508
5521
6717
Core C4
Mn
0.20
0.06
0.50
11.82
0.44
0.59
0.27
ND
ND
0.30
0.20
7.77
Ni
94.5
125.5
292.8
207.7
130.0
78.6
60.3
49.4
36.2
42.1
39.5
43.3
Pb
356
414
630
1853
1645
1433
782
514
194
191
244
339
Zn
589
613
935
2188
1850
2084
1320
913
479
492
449
740
XXV
Table B.20
Depth
(cm)
1
3
5
7
9
11
13
15
17
Sediment macrocomponent ratios (weight ratio, unless indicated) of Red Beach cores (C1-C4).
Core C1
TS/TSe
TS/TSe TC/TOC TC/TN TC/TS TC/AVS TC/CrRS TOC/TN TOC/TS TOC/AVS TOC/CrRS TS/AVS TS/CrRS AVS/CrRS
(molar ratio)
1296
526
1678
681
798
324
1195
485
1876
762
1103
448
1700
690
3040
1235
2244
911
Table B.20
2
4
4
3
3
1
1
1
1
25
22
46
24
24
23
20
20
19
6
3
6
2
2
3
2
1
1
49
39
52
21
19
18
35
11
7
6
6
8
4
3
4
2
2
2
11
6
11
8
7
18
15
16
14
2.5
0.9
1.5
0.7
0.7
2.7
1.2
1.2
1.0
22
10
13
7
6
15
26
9
5
3
2
2
1
1
3
2
2
2
9
12
8
9
8
5
21
7
5
1
2
1
2
1
1
1
2
2
0.12
0.17
0.16
0.19
0.16
0.23
0.07
0.21
0.32
(continued).
Depth
(cm)
TS/TSe
1
3
5
7
9
11
13
(molar ratio)
1069
1092
1334
1199
869
1507
1299
Core C2
TS/TSe TC/TOC TC/TN TC/TS TC/AVS TC/CrRS TOC/TN TOC/TS TOC/AVS TOC/CrRS TS/AVS TS/CrRS AVS/CrRS
434
443
542
487
353
612
527
6
5
4
3
2
4
2
43
47
33
28
24
42
16
23
13
8
6
5
7
5
313
239
73
39
30
43
31
29
21
13
9
6
12
8
8
9
8
10
11
10
7
4.1
2.5
1.9
2.0
2.3
1.7
2.2
56
46
17
13
14
10
13
5
4
3
3
3
3
4
14
19
9
7
6
6
6
1
2
2
1
1
2
2
0.09
0.09
0.17
0.22
0.22
0.29
0.28
XXVI
Table B.20
(continued).
Depth
(cm)
TS/TSe
1
3
5
7
9
11
13
15
(molar ratio)
1232
962
1511
872
932
1153
1657
1234
Core C3
Table B.20
Depth
(cm)
1
3
5
7
9
11
13
15
17
19
21
23
TS/TSe TC/TOC TC/TN TC/TS TC/AVS TC/CrRS TOC/TN TOC/TS TOC/AVS TOC/CrRS TS/AVS TS/CrRS AVS/CrRS
500
391
613
354
379
468
673
501
6
6
8
5
8
6
4
8
32
31
43
35
48
41
34
56
20
17
20
13
10
13
4
8
595
531
329
245
249
211
86
238
34
25
30
18
15
21
7
9
6
5
5
7
6
7
9
7
3.3
2.7
2.4
2.5
1.2
2.0
1.2
1.0
101
86
39
47
30
33
23
30
6
4
4
3
2
3
2
1
30
31
16
19
24
17
19
30
2
2
1
1
1
2
2
1
0.06
0.05
0.09
0.07
0.06
0.10
0.08
0.04
(continued).
Core C4
TS/TSe
TS/TSe TC/TOC TC/TN TC/TS TC/AVS TC/CrRS TOC/TN TOC/TS TOC/AVS TOC/CrRS TS/AVS TS/CrRS AVS/CrRS
(molar ratio)
809
328
803
326
502
204
483
196
858
349
1389
564
2573
1045
1114
452
1864
757
2216
900
1964
798
3126
1269
3
6
5
5
6
4
7
5
2
12
2
9
21
38
33
36
55
48
56
76
14
67
24
44
10
9
12
8
12
4
6
7
1
8
2
5
98
92
215
118
82
59
185
821
65
553
145
464
16
10
19
16
19
7
10
10
1
13
3
8
7
7
7
8
9
14
8
17
7
6
13
5
3.5
1.6
2.4
1.8
2.1
1.2
0.8
1.4
0.5
0.6
0.8
0.6
34
16
44
26
14
17
26
180
30
48
79
49
6
2
4
3
3
2
1
2
1
1
1
1
10
10
18
14
7
14
33
125
54
73
94
87
2
1
2
2
2
2
2
2
1
2
2
2
0.16
0.11
0.09
0.13
0.24
0.12
0.05
0.01
0.02
0.02
0.02
0.02
XXVII
Table B.21
TOC
Correlations (r)* between sediment parameters and co-extracted elements in the soluble and adsorbed fraction of Cores
C1-C4.
Core C1
Core C2
Ads Se Ads Cr Ads Cu Ads Fe Ads Ni Ads Pb Ads Zn
Ads Se Ads Cr Ads Cu Ads Fe Ads Ni Ads Pb Ads Zn
0.306
Ads Se
0.430
-0.116
0.580
0.461
0.610
0.606
0.157
-0.626
0.328
0.179
0.381
0.377
0.070
0.608
0.559
0.720
0.687
-0.204
-0.030
-0.239
-0.230
0.868
0.953
0.986
0.847
0.863
Ads Cr
Ads Cu
Ads Fe
Ads Ni
Ads Pb
Ads Se
Ads Cr
0.343
0.964
0.623
0.197
0.308
0.652
0.378
0.606
0.887
0.120
0.569
0.911
0.511
0.324
0.407
0.816
0.527
0.722
0.244
0.499
0.788
0.403
0.612
0.961
0.748
0.330
0.691
0.989
Table B.21
TOC
0.437
(continued).
Core C3
Core C4
Ads Se Ads Cr Ads Cu Ads Fe Ads Ni Ads Pb Ads Zn
Ads Se Ads Cr Ads Cu Ads Fe Ads Ni Ads Pb Ads Zn
0.414
0.570
-0.195
0.450
0.719
0.444
0.624
0.002
-0.439
0.773
0.087
0.614
0.833
0.522
-0.226
0.372
-0.199
-0.033
-0.403
0.199
-0.314
-0.448
0.357
0.943
0.961
0.418
0.377
Ads Cu
Ads Fe
Ads Ni
Ads Pb
0.893
0.858
0.319
0.163
0.529
0.365
0.587
0.617
0.340
0.190
0.728
0.383
0.798
0.819
0.919
0.200
0.888
0.301
0.117
0.107
0.885
0.191
-0.017
0.077
0.986
0.944
0.197
0.003
0.951
*Values in bold are significant at P < 0.01 level.
XXVIII
Table B.22
Correlations (r)* between sediment parameters and co-extracted elements in the organically bound selenium fraction
of Cores C1-C4.
Core C1
TOC
TC
-0.018
TOC
Core C2
Org Se Org Cr Org Cu Org Fe Org Ni Org Pb Org Zn
0.766
0.141
0.590
0.270
0.668
0.237
-0.064
0.173
-0.612
-0.321
0.354
-0.082
-0.271
0.167
0.032
0.327
0.647
0.872
0.176
0.246
0.175
0.294
0.400
0.569
0.266
-0.026
0.299
0.435
-0.510
0.713
0.229
0.422
0.296
0.346
Org Se
Org Cr
Org Cu
Org Fe
Org Ni
Table B.22
TOC
Org Se
Org Cr
0.859
0.794
0.342
0.950
0.343
0.444
0.442
0.278
0.767
0.565
0.843
0.570
0.597
0.344
0.580
0.566
0.766
0.524
0.731
0.597
0.572
0.264
0.785
0.656
0.251
0.907
0.365
0.447
0.371
0.197
0.906
0.593
0.871
0.812
0.832
-0.270
Org Pb
TC
TOC Org Se Org Cr Org Cu Org Fe Org Ni Org Pb Org Zn
0.484
(continued).
Core C3
Core C4
TOC Org Se Org Cr Org Cu Org Fe Org Ni Org Pb Org Zn
TOC Org Se Org Cr Org Cu Org Fe Org Ni Org Pb Org Zn
0.548
0.883
0.659
0.136
-0.486
0.815
0.921
0.603
-0.556
0.760
0.654
-0.705
-0.078
Org Cu
Org Fe
Org Ni
Org Pb
-0.328
0.599
0.662
0.347
0.544
0.850
0.835
0.387
0.674
0.880
0.598
0.802
0.750
0.369
-0.420
-0.038
-0.153
-0.781
0.086
0.900
0.813
0.303
0.070
0.572
0.387
0.564
0.716
-0.537
0.592
0.552
0.568
0.857
-0.389
0.717
0.556
0.644
0.891
0.487
0.067
0.782
0.820
0.808
0.565
-0.335
0.565
0.379
0.530
0.806
-0.019
0.029
-0.088
-0.437
0.573
0.895
0.714
0.657
0.541
0.828
0.805
*Values in bold are significant at P < 0.01 level.
XXIX
Table B.23
Correlations (r)* between sediment parameters and co-extracted elements in the elemental Se fraction of Cores C1-C4.
Core C1
AVS CrRS
<63
TC
TOC
TS
0.489 0.277 -0.435 -0.184 0.267
0.748 -0.096
AVS
0.019
CrRS
TC
Core C2
Ele Se Ele Cu Ele Fe Ele Pb Ele Zn
AVS CrRS
TC
0.141
0.123
-0.091
0.00
0.161
0.651 0.784
0.004
0.194
0.682
0.135
0.00
0.658
TOC
TS
0.111
-0.239
0.044
0.424
0.489 -0.532 -0.227 0.546 0.353
0.226
-0.837
-0.027
-0.065
0.449
0.743
0.224
0.896
0.502
0.552
-0.023
0.00
0.451
0.708 0.963 0.898
0.628
-0.069
-0.282
0.458
-0.018 0.066
0.565
0.376
0.160
0.00
0.103
0.859 0.317 0.254
0.308
0.862
-0.542
0.725
0.225 -0.106
0.352
0.479
0.00
0.576
0.622 0.627
0.531
0.591
-0.475
0.602
0.602
0.672
-0.184
0.00
0.367
0.842
0.635
-0.153
-0.389
0.354
0.546
0.080
0.00
0.235
0.729
-0.117
0.062
0.214
0.360
0.00
0.403
-0.123
-0.206
0.173
0.00
0.566
-0.281
0.552
TOC
TS
Ele Se
Ele Cu
Ele Fe
0.368
0.000
Ele Pb
Table B.23
-0.400
(continued).
Core C3
<63
AVS
CrRS
TC
TOC
Ele Se Ele Cu Ele Fe Ele Pb Ele Zn
0.297 0.710 0.637
TC
TOC
TS
Core C4
AVS
CrRS
Ele Se
Ele Cu
Ele Fe
Ele Pb
Ele Zn
AVS CrRS
0.765
0.306
0.156 0.777 0.593
0.369
0.308
-0.293
-0.362
-0.298
0.728 0.398 0.597 0.737 0.470
0.657
0.436
0.800
0.623
-0.386
0.754
0.441 0.837 0.919
0.809
0.670
-0.520
-0.492
0.030
0.583 0.794 0.845 0.584
0.745
0.325
0.594
0.491
-0.169
0.311 0.403 0.936
0.964
0.854
-0.529
-0.503
0.528
0.699 0.691 0.983
0.533
0.328
0.347
0.305
0.044
0.548 0.349
0.350
0.304
-0.318
-0.415
0.492
0.883 0.704
0.770
0.486
0.339
0.428
-0.083
0.623
0.491
0.565
-0.326
-0.351
-0.109
0.713
0.871
0.577
0.665
0.736
0.021
0.958
0.777
-0.494
-0.480
0.311
0.578
0.371
0.421
0.371
-0.033
0.789
-0.462
-0.422
0.435
0.504
0.650
0.651
0.081
-0.481
-0.471
0.455
0.518
0.335
-0.080
0.959
-0.124
0.740
-0.009
TS
Ele Se
Ele Cu
Ele Fe
Ele Pb
-0.259
TC
TOC
TS
Ele Se Ele Cu Ele Fe Ele Pb Ele Zn
0.252
*Values in bold are significant at P < 0.02 level.
XXX
Table B.24
Correlations (r)* between sediment parameters and co-extracted elements in the organic matter and sulfide fraction of
Cores C1-C4.
Core C1
<63
AVS
CrRS
0.489
0.277
0.748
AVS
CrRS
TC
O&S Se
O&S Cu
O&S Zn
AVS CrRS
TC
-0.435 -0.184 0.267
0.487
0.202
0.504
0.651 0.784
0.004
-0.096
0.449
0.743
0.537
0.065
0.499
0.019
0.224
0.896
0.648
0.208
0.415
-0.018 0.066
0.230
0.396
0.225
0.124
0.624
TC
TOC
TOC
TS
Core C2
TS
O&S Se
O&S Se
O&S Cu
O&S Zn
0.297 0.710
0.196
0.200
0.128
0.489 -0.532 -0.227 0.546
-0.346
-0.455
0.107
0.708 0.963
0.139
0.393
-0.169
-0.027
0.859 0.317
0.324
0.813
-0.556
-0.360
-0.016
0.622
0.279
0.831
-0.462
0.274
0.571
0.172
0.309
-0.260
0.704
0.798
0.582
-0.028
O&S Cu
0.368
TOC
-0.541
0.748
Table B.24
(continued).
Core C3
<63
AVS
CrRS
TC
TOC
TS
AVS
CrRS
0.765
0.306
0.754
TC
TOC
TS
Core C4
O&S Se
O&S Cu
O&S Zn
0.156 0.777 0.593
0.685
-0.165
-0.482
0.441 0.837 0.919
0.889
0.360
0.311 0.403 0.936
0.738
0.548 0.349
0.623
TS
O&S Se
O&S Cu
O&S Se
O&S Cu
O&S Zn
0.728 0.398 0.597 0.737 0.470
0.654
0.059
0.075
0.080
0.583 0.794 0.845 0.584
0.761
-0.070
-0.006
0.770
0.656
0.699 0.691 0.983
0.616
0.530
0.719
0.471
-0.120
0.020
0.883 0.704
0.814
0.384
0.315
0.704
-0.010
-0.221
0.713
0.940
0.337
0.254
0.902
0.579
0.370
0.643
0.523
0.716
0.266
0.085
0.331
0.180
0.901
AVS
CrRS
TC
TOC
TS
0.867
*Values in bold are significant at P < 0.02 level.
XXXI
Table B.25
Correlations (r)* between sediment parameters and co-extracted elements in the residual fraction of Cores C1-C4.
Core C1
Core C2
AVS CrRS Res Se Res Cr Res Cu Res Fe Res Mn Res Ni Res Pb Res Zn
AVS CrRS Res Se Res Cr Res Cu Res Fe Res Mn Res Ni Res Pb Res Zn
<63 µm 0.489 0.277 -0.485
0.402
-0.199
0.622
0.098
-0.430
0.438
0.402
0.651 0.784
0.439
0.324
0.578
-0.245
0.101
-0.345
0.416
0.523
0.748 -0.301
0.774
0.187
0.845
0.714
0.047
0.940
0.787
0.489
0.301
0.340
0.077
0.315
0.355
-0.527
0.136
0.068
AVS
-0.175
CrRS
Res Se
0.828
0.247
0.638
0.698
0.202
0.704
0.770
-0.219
0.772
-0.419
-0.225
0.593
-0.411
0.009
0.307
0.714
0.859
0.124
0.797
0.915
-0.002
0.155
0.623
0.053
0.567
0.694
-0.158
0.883
0.625
0.267
0.856
0.718
0.109
0.357
Res Cr
Res Cu
Res Fe
Res Mn
Res Ni
0.875
0.342
0.899
0.162
-0.045
0.229
0.864
0.896
0.309
0.867
0.409
-0.071
0.479
0.869
0.875
0.144
0.271
0.898
-0.037
0.173
0.187
-0.012
-0.300
0.513
0.914
0.989
0.261
0.207
0.236
0.069
-0.311
-0.243
-0.261
0.672
0.591
0.748
Res Pb
Table B.25
0.959
(continued).
Core C3
Core C4
AVS CrRS Res Se Res Cr Res Cu Res Fe Res Mn Res Ni Res Pb Res Zn
AVS CrRS Res Se Res Cr Res Cu Res Fe Res Mn Res Ni Res Pb Res Zn
<63 µm 0.765 0.306
0.385
0.861
0.814
0.643
0.620
0.609
0.864
0.879
0.728 0.398
0.257
0.931
0.866
0.899
0.633
0.345
0.766
0.670
0.754
0.263
0.936
0.717
0.621
0.193
0.402
0.877
0.739
0.583
0.344
0.860
0.654
0.606
0.185
0.100
0.844
0.801
0.288
0.580
0.394
0.381
-0.227
0.199
0.461
0.223
0.438
0.621
0.380
0.459
0.203
-0.182
0.497
0.299
0.198
0.738
0.764
0.244
0.910
0.340
0.298
0.392
0.588
0.556
0.184
0.336
0.593
0.508
0.716
0.596
0.482
0.439
0.955
0.881
0.873
0.862
0.545
0.263
0.885
0.787
0.480
0.896
0.848
0.826
0.310
0.909
0.752
0.709
0.512
0.578
0.725
0.617
0.627
AVS
CrRS
Res Se
Res Cr
Res Cu
Res Fe
Res Mn
Res Ni
Res Pb
0.940
0.942
0.665
0.587
0.882
0.819
0.630
0.472
0.754
0.630
0.432
0.440
0.381
0.388
0.455
0.965
0.963
*Values in bold are significant at P < 0.02 level.
XXXII
50%
75%
100%
0%
1
Depth (cm)
Depth (cm)
5
9
13
17
25%
25%
50%
0%
100%
1
3
3
5
7
75%
11
11
13
13
0%
25%
50%
75%
100%
0%
Depth (cm)
15
13
C ore B 4
Figure B.1
75%
100%
5
7
11
50%
3
5
13
25%
1
9
11
100%
C ore B 3
3
9
75%
7
9
100%
3
7
50%
5
9
1
5
25%
C ore B 2
1
Depth (cm)
75%
1
C ore B 1
0%
50%
Depth (cm)
25%
Depth (cm)
0%
7
9
11
13
15
17
19
C ore B 5
Grain size distribution in Cores B1-B6 collected in April 2005 (
C ore B 6
< 63 µm and
> 63 µm)
XXVI
0%
20%
40%
60%
80%
0%
100%
1
40%
80%
100%
80%
100%
3
Depth (cm)
5
7
9
11
5
7
9
13
11
15
13
17
Core C 2
Core C 1
0%
20%
40%
60%
80%
0%
100%
20%
40%
60%
1
1
3
3
5
Depth (cm)
Depth (cm)
60%
1
3
Depth (cm)
20%
5
7
9
7
9
11
13
15
11
17
19
13
21
15
23
Core C 3
Figure B.2
Grain size distribution in Cores C1-C4, collected in July 2005 (
Core C 4
< 63µm,
63-250 µm,
> 250 µm).
XXVII
40
60
80
0
20
0
60
0
3
3
3
6
6
6
9
12
Depth (cm)
0
9
12
15
18
18
18
21
Se conc
30
60
90
0
0
3
Depth (cm)
6
9
12
15
18
21
Figure B.3
B4
10
20
40
Se conc
30
0
40
0
0
3
3
6
6
9
12
18
18
Porewater selenium (
, µg/L) and total selenium (
Red Beach cores: B1-B6, collected in April 2004.
21
, µg/g,
40
60
80
100
12
15
B5
20
9
15
21
30
B3
Se conc
120
Depth (cm)
0
21
B2
20
12
15
B1
10
9
15
21
Depth (cm)
40
0
Depth (cm)
Depth (cm)
20
Se conc
Se conc
Se conc
0
B6
<63 µm, five times scale reduction) in individual
XXVIII
T otal Se ( µg/g)
20
40
60
0
80
10
30
40
0
T otal Se ( µg/g)
20
40
60
T otal Se ( µg/g)
80
0
0
3
3
3
3
6
6
6
6
9
12
9
12
Depth (cm)
0
Depth (cm)
0
9
12
15
18
18
18
18
21
21
21
21
24
C1
P orewater Se, µg/L
0
2
24
C2
4
0
6
2
4
6
24
C3
Porewater Se, µg/L
0
P orewater Se, µg/L
2
4
6
0
8
3
3
3
3
6
6
6
6
12
12
Depth (cm)
0
Depth (cm)
0
Depth (cm)
0
9
9
12
15
18
18
18
18
21
21
21
21
Figure B.4
C1
24
Porewater selenium (
collected in July 2005.
C2
, µg/L)
and total selenium (
24
, µg/g,
C3
40
60
12
15
24
20
9
15
15
C4
Porewater Se, µg/L
0
9
100 150 200 250
12
15
24
50
9
15
15
Depth (cm)
20
0
Depth (cm)
Depth (cm)
0
T otal Se ( µg/g)
24
C4
whole sediment) in individual Red Beach cores: C1-C4,
XXIX
Se (µg/g)
4.0
2.0
Se (µ g/g)
6.0
8.0
0.0
0.5
1.5
Se (µg/g)
2.0
0.0
0
3
3
3
6
6
6
9
12
Depth (cm)
0
9
12
15
18
18
18
21
0
2
4
6
8
21
B2
Se (µg/g)
0.0
1.0
2.0
0
4.0
3
3
6
6
6
Depth (cm)
3
Depth (cm)
0
0
12
12
15
18
18
18
21
Figure B.5
B5
3.0
2
4
6
8
12
15
B4
2.5
9
15
21
2.0
Se (µg/g)
3.0
0
9
1.5
B3
Se (µ g/g)
10
9
1.0
12
15
B1
0.5
9
15
21
Depth (cm)
1.0
0
Depth (cm)
Depth (cm)
0.0
21
B6
Selenium concentrations (µg/g d.w.) in labile fractions of Cores B1-6.
XXX
25%
50%
75%
100%
0%
1
3
Depth (cm)
7
9
11
17
19
25%
50%
0%
1
3
3
5
7
11
13
13
B2
75%
100%
0%
25%
50%
3
Depth (cm)
9
75%
100%
0%
25%
50%
75%
100%
3
5
5
7
7
9
11
9
13
13
11
15
15
13
Figure B.6
100%
1
11
B4
75%
B3
3
7
50%
7
11
1
5
25%
5
9
1
Depth (cm)
100%
1
B1
0%
75%
9
13
15
50%
Depth (cm)
Depth (cm)
5
25%
Depth (cm)
0%
17
19
B5
B6
Selenium fractionation patterns (SEP 1) in individual Red Beach cores (B1-B6), Port Kembla Harbour.
XXXI
0%
20%
40%
60%
80%
100%
0%
1
40%
60%
80%
100%
60%
80%
100%
1
3
3
5
Depth (cm)
Depth (cm)
20%
7
9
11
5
7
9
13
11
15
13
17
C1
20%
40%
60%
80%
0%
100%
1
1
3
5
5
7
9
40%
7
9
11
13
15
11
17
13
19
15
23
21
C3
Figure B.7
20%
3
Depth (cm)
Depth (cm)
0%
C2
C4
Selenium fractionation patterns (SEP 2) in individual Red Beach cores (C1-C4), Port Kembla Harbour.
XXXII