sources, mobility and bioaccessibility of potentially...

Post on 30-Mar-2020

0 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

TRANSCRIPT

Sources, Mobility and Bioaccessibility of Potentially Harmful Elements in UK Soils

© NERC All rights reserved

Mark CaveBritish Geological Survey

Sources of Potentially Harmful Elements in soils

• Natural geogenic sources• Anthropogenic pollution

• Point source (single identifiable source)

© NERC All rights reserved

identifiable source)• Diffuse pollution

(dispersed over a wide area)

Census Area Statistical Wards 2003

© NERC All rights reserved

Domain Area

(km2)

Area

(%)

NBC

(mg/kg)

n

Ironstone 1,300 1 220 437

Mineralisation 2,300 2 290 187

Principal 129,30

0

97 32 41,50

9

© NERC All rights reserved

Establishing a Normal Background

Concentration (NBC)

© NERC All rights reserved

Arsenic in the Ironstone domain

Domain Area

(km2)

Area

(%)

NBC

(mg/kg)

n

Urban 5,400 4 820 7,529

Mineralisation 2,900 2 2,400 347

Principal 124,60

0

94 180 34,25

7

© NERC All rights reserved

Domain Area

(km2)

Area

(%)

NBC

(mg/kg)

n

Urban 5,200 4 2.1 9,308

Min. Group 2 500 <1 17 224

Min. Group 1 1,600 1 2.9 95

Chalk South 6,900 5 2.5 265

Principal 118,70

0

89 1.0 4,418

© NERC All rights reserved

How do we measure PHE mobility?

• Geo availability – PHE fractionation, mineralogy, sequential extraction, SEM, XAFS,XANES

CarbonatesMn oxides

© NERC All rights reserved

• Bioaccessibility/bioavailability – measure in-vivo or mimic inhalation, ingestion, dermal contact.

Iron Oxides

Alumino Silicates

Organic

Ingestion

What are we trying to achieve?

© NERC All rights reserved

Absorption

Exposure biomarkersBiological markers (biomarkers) can be utilised to estimate

levels of exposure to harmful substances.

Following exposure, soluble arsenic is adsorbed from the gastro-intestinal tract and distributed to all bodily systems in

the blood, accumulating in many body parts.

© NERC All rights reserved

Toenails Fingernails Hair

Long-term (past exposure)

Blood Urine

Short-term (recent exposure)

Bioaccessibility : Unified BARGE Method (UBM)

Saliva

pH = 6.5 ± 0.5

Stomach

extractantpH = 0.9/1.0

pH =1.2 +/-.05

Centrifuge (3000 g,

5 min). Analysis by

© NERC All rights reserved

0.6 g soil

0.6 g soil

End-over-end

1 hour

Intestinal

extractant

End-over-end

4 hours

pH =6.3 ±

0.5pH = 6.3 ± 0.5

5 min). Analysis by

ICP-OESGastric sample

Gastro-Intestinal sample

The PBET method

Stomach and Intestine reagents are Soil samples are weighed into Soils are extracted with gastric

© NERC All rights reserved

Stomach and Intestine reagents are prepared according to the protocol

Soil samples are weighed into centrifuge tubes

Soils are extracted with gastric and intestine solutions in a water

bath at 370 C

Samples are CentrifugedDecanted samples are diluted and preserved in 0.1 M HNO3

Samples are analysed by ICP-

AES

Comparison of in vivo and in vitro data for NIST 2710 for the UBM inter-laboratory trial (2006/2007)

© NERC All rights reserved

2012,46, pages 6252-6260

© NERC All rights reserved

bone

urin

e

| |

| |

| |

| |

bone

urin

e

| |

| |

| |

| |

bone

urin

e

| |

| |

| |

| |

Summary of the RBA vs RBAc regression statistics for the four end points for As. Black squares show data for the ‘stomach’ phase and white triangles for the ‘stomach & intestine’

phase. Error bars represent 95% confidence limits dotted lines show benchmark values.

© NERC All rights reserved

0 2 4 6 8Intercept

kidn

eyli

ver

| |

| |

| |

| |

0.2 0.6 1.0 1.4Slope

kidn

eyli

ver

| |

| |

| |

| |

0.6 0.8 1.0r squared

kidn

eyli

ver

| |

| |

| |

| |

bone

urin

e

| |

| |

| |

| |

bone

urin

e

| |

| |

| |

| |

bone

urin

e

| |

| |

| |

| |

Summary of the RBA vs RBAc regression statistics for the four end points for Cd. Black squares show data for the ‘stomach’ phase and white triangles for the ‘stomach & intestine’

phase. Error bars represent 95% confidence limits, dotted lines show benchmark values

© NERC All rights reserved

-10 10 30Intercept

kidn

eyli

ver

bone

| |

| |

| |

| |

| |

0.4 0.8 1.2Slope

kidn

eyli

ver

bone

| |

| |

| |

| |

| |

0.6 0.8 1.0r squared

kidn

eyli

ver

bone

| |

| |

| |

| |

| |

bone

urin

e

| |

| |

| |

| |

bone

urin

e

| |

| |

| |

| |

bone

urin

e

| |

| |

| |

| |

Summary of the RBA vs RBAc regression statistics for the four end points for Pb. Black squares show data for the ‘stomach’ phase and white triangles for the ‘stomach & intestine’

phase. Error bars represent 95% confidence limits, dotted lines show benchmark values.

© NERC All rights reserved

-5 5 15Intercept

kidn

eyli

ver

| |

| |

| |

| |

0.2 0.6 1.0 1.4Slope

kidn

eyli

ver

| |

| |

| |

| |

0.6 0.8 1.0r squared

kidn

eyli

ver

| |

| |

| |

| |

Rel

ativ

e bi

oacc

essi

bilit

y % 20

4060

8010

0

(a)

Stomach

2040

6080

100

(a)

Stomach & Intestine

RBAc against RBA for (a) Pb and (b) Cd for the ‘stomach’ and ‘stomach & intestine’ phases for the kidney endpoint. Bold dashed dotted line is the line of equivalence, dashed lines are

the 95% confidence intervals and the solid lines is the best line of fit

Pb

© NERC All rights reservedRelative bioavailability %

Rel

ativ

e bi

oacc

essi

bilit

y %

0 20 40 60 80 100

0

0 20 40 60 80 100

0

0 20 40 60 80 100

020

4060

8010

0

(b)

0 20 40 60 80 100

020

4060

8010

0

(b)

Cd

Rel

ativ

e bi

oacc

essi

bilit

y % 20

4060

8010

0

(c)

Stomach

2040

6080

100

(c)

Stomach & Intestine

RBAc against RBA for (c) As and (d) Sb for the ‘stomach’ and ‘stomach & intestine’ phases for the urine end point. Bold dashed dotted line is the line of equivalence, dashed

lines are the 95% confidence intervals and the solid line is the best line of fit.

As

© NERC All rights reservedRelative bioavailability %

Rel

ativ

e bi

oacc

essi

bilit

y %

0 20 40 60 80 100

0

0 20 40 60 80 100

0

0 5 10 15 20

05

1015

20

(d)

0 5 10 15 20

05

1015

20

(d)

Sb

Comparison of the Relative Bioaccessibility of As in the UK

6

5

4

s

GLAS

HUMB

LOND

NORT

SWAN

AREA

Scatterplot of LnBSAs vs LnAs: UBM analysis

© NERC All rights reserved

8765432

3

2

1

0

LnAs

LnBSAs

LondonSwansea

Northampton

Glasgow

Hull

Don’t just rely on the bioaccessibility test

• Always use geochemical tests to back up your bioaccessibility results.

• Helps the risk assessor put the bioaccessibility value in context.

Centrifugation

© NERC All rights reserved

bioaccessibility value in context.• Bioaccessibility is no longer just a ‘black

box’ or a black art in the eyes of the regulators and policy makers

• Allows decisions to be made regarding current and future land use.

• Gives the regulator added confidence in the risk assessment.

0.45 µµµµfilter

membrane leachate

How are PHE distributed in the soil components?

CISED TestChemometric Identification of Substrates and Element

Distributions

•Separate aliquots of aqua regia of increasing concentration. Centrifugation

© NERC All rights reserved

increasing concentration.

•Passed through the sample under centrifugal force.

•Determination by ICP-AES.

•Chemometric data processing .

•Identification of physico-chemical hosts and the metal distributions within

the sample under test.

Centrifugation

0.45 µµµµfilter

membrane leachate

Example of CISED

Fe

Fe-Al

Pb

Fe

Fe-Al

As

Fe oxide

Fe oxide

© NERC All rights reserved

0 20 40 60 80

Ca

K -Ca

Mn-Al

Al-Fe

Ca-Al

Fe-Na-Si-P

mg kg-10 5 10 15 20 25

Ca

K -Ca

Mn-Al

Al-Fe

Ca-Al

Fe-Na-Si-P

mg kg-1

Mn oxide

Carbonate

Al oxide

Carbonate2

Residual solutes

Organics

Northampton• Large Market town in central England

• Population of c. 200,000• Busy Road and Rail links

• Primary industrial activities were shoe making and other leather industries• Now a hub for finance and distribution

© NERC All rights reserved

• Now a hub for finance and distribution industries

• BGS surveyed the area as part of the G-BASE programme

• Ironstone soils, naturally elevated in arsenic• 45% of the soils have As concentrations above

the residential SGV of 32 mg mg-1

© NERC All rights reserved

© NERC All rights reserved

Measurement of the diffuse reflectance spectra of soils using a mug light.

© NERC All rights reserved

2.5

3

3.5

4A

bsor

banc

e

281 spectra Northampton urban soils

© NERC All rights reserved

500 1000 1500 2000 25000.5

1

1.5

2

Wavelength, nm

Abs

orba

nce

Coefficient Value Standard Error P value % Variance

explained

Intercept -0.56 1.22 <0.64 -

pH 0.41 0.12 <0.05 22.6

MLR model of bioaccessible As using NIR spectral components and Major element compositions

© NERC All rights reserved

As 0.05 0.01 <0.05 37.6

Mg 0.00 0.00 <0.05 4.9

SC1 -21.1 3.6 <0.05 2.5

SC2 16.2 2.5 <0.05 16.5

R square = 0.84

© NERC All rights reserved

0.2

0.3

0.4

Den

sity

Bio

acce

ssib

le A

s

0.01

00.

020

0.03

0

Den

sity

Tot

al A

s

Bioaccessible AsTotal As

© NERC All rights reserved

0 20 40 60 80 100

0.0

0.1

As mg kg−1

Den

sity

Bio

acce

ssib

le A

s

0.00

00.

010

Den

sity

Tot

al A

s

© NERC All rights reserved

A Structured Approach

© NERC All rights reserved

top related