method development studies for hexavalent chromium … · method development studies for hexavalent...

Method Development Studies for Method Development Studies for Hexavalent Chromium in Ambient Hexavalent Chromium in Ambient

Air SamplesAir Samples

Julie L. Swift

Eastern Research Group, Inc.

[email protected]

1

Introduction

� History of EPA Hexavalent Chromium (Cr(VI) Method)

http://www.epa.gov/ttnamti1/files/ambient/airtox/hexchro

msop.pdf

� Analytical preparation method change

� Improvements in precision after method change

� Evaluation of EPA Method vs. other analytical techniques (IC/UV vs. IC/ICPMS)

� Evaluation of EPA/NATTS sampler vs. sampler developed by NYS and Clarkson University

2

2003 - ERG started working on method

2004 - EPA contracted ERG to study CARB 039

method

- ERG authored Method Development paper

2006 - ERG authored an SOP

(http://www.epa.gov/ttnamti1/airtox.html)

2008 - ERG modified NATTS sampler to add chiller

2009 - ERG modified filter preparation technique before

sampling – stable <15°C (60°F) for 3 days

2011 - ERG modified sample preparation technique –

sonication vs. shaking

2012 - New study with NJ DEP CTI Grant3

Timeline of EPA Method Variations

� Incorporated analytical procedure to obtain lowest Method Detection Limits (MDL)

� Investigated filter media

– Cellulose, Binderless Quartz, PVC, Teflon

– Cellulose showed best retention but had high

background – had to acid wash in order to obtain low

MDL (current MDL = 0.0034 ng/mL)

� Investigated interferences

– No interference of Cr (III), Fe, Mg

4

ERG Initial Study (2003-2004)

� Added chiller to keep samples frozen while sampling and up to 3 days after sampling

– Works in laboratory conditions, but collects water at sites with high humidity/high temperature

� ERG modified filter preparation technique before sampling – stable for up to 3 days at <15°C (60°F)

5

ERG Sampler Study: Chiller (2008-2009)

T o

A t m o s p h e r e

F l o w

C o n t r o l

V a l v e

S a m p le

F l o w

R o t a m e t e r

S a m p le

P u m p

O U T S I D E

I N S I D E

F i l t e rH o l d e r

F u n n e lA s s e m b l y

F i l t e r

F i l t e rH o l d e r

F u n n e lA s s e m b l y

F i l t e r

S a m p l e A i r i n F r o m A t m o s p h e r e

C h i l l e r


6

Sample Air in From Atmosphere

OutsideInside

ChillerSample Pump

Sample

Flow

Rotameter

To Atmosphere

Flow Control

Valve

0

10

20

30

40

50

60

70

80

90

100

Day 0 Day 1 Day 3 Day 5

NATTS Chiller, 15°C Chiller,18°C Chiller, 21°C7


% R

ecovery

� Community Toxics Initiative Grant

– NJ DEP CTI grant (funds from US EPA)• Dr. Linda Bonanno – Principal Investigator at NJDEP

“Evaluation of Two Analytical Methods and Sampling

Trains for the Measurement of Hexavalent Chromium in

Ambient Air”

– In conjunction with University of Medicine and Dentistry of NJ (EOHSI), Clarkson University and ERG

• Compare Analytical Instrumentation

– IC/UV

– IC/ICPMS

• Compare Sampling Systems (added ERG Prototype)

– EPA/NATTS Sampler

– Clarkson/State of NY

– ERG Prototype8

NJ DEP CTI: Newest Study (2009 - present)

Preliminary/Draft

� New Jersey Department of Environmental Protection:

– Dr. Linda Bonanno

� Clarkson University:

– Dr. Philip Hopke, Dr. Lin Lin, Mehdi Amouei

Torkmahalleh

� EOHSI:

– Dr. Tina Fan, Chang Ho Yu

� ERG:

– Julie Swift, Victoria Genther, Dr. Laura Krnavek, Randy

Mercurio, Ariel Atkinson, Donna Tedder9

NJ DEP CTI:Research Team

Preliminary/Draft

� Evaluate the 2 analytical methods

– All cellulose filters are prepared at ERG

– Spiked filters were prepared by EOHSI and sent to

Clarkson, ERG and kept in-house

– Presented at 2011 National Air Toxics Monitoring &

Data Analysis Workshop

http://www.epa.gov/ttn/amtic/airtox-daw-2011.html

10

NJ DEP CTI: New Study Analytical Module Objective

Preliminary/Draft

NJ DEP CTI: Sample Preparation Changes

� Previously prepared samples for analysis by sonicating filters in sodium bicarbonate solution

– Studies with NJDEP/EOHSI/Clarkson have detected

Cr(VI) when Cr(III) was spiked on filters

– In previous ERG studies, we did not see this problem

– However, spiking concentration in new study is higher

• Recoveries showed need to reevaluate the preparation procedure

• Dr. Phil Hopke suggested that the presence of hydroxyl ions may cause conversion of Cr(III) to Cr(VI) during sonication

11 Preliminary/Draft

� If sonication is causing a problem, how should samples be prepared?

� Study sample extraction via sonication versus wrist-action shaker

� Data collected comparing sonication to shaking with liquid-spiked filters

– Cr(III) only

– Cr(VI) only

– Cr(VI) and Cr(III)

12

ERG Study:Sonicator vs. Shaker

0%

1%

2%

3%

4%

5%

15 30 45 60 75

Pe

rce

nt

rec

ove

ry,

Cr6

+

Time / minutes

Sonic Cr(III) / 1.10 ng/mL Shake Cr(III) / 1.10 ng/mL

ERG Study Sonicator vs. Shaker: Cr(III) only

13 Note: All recoveries for the shaker were zero.

80%

85%

90%

95%

100%

105%

110%

115%

120%

125%

130%

135%

140%

145%

150%

15 30 45 60 75

% R

eco

very

Time / minutes

Sonic Cr(VI) / 0.05 Sonic Cr(VI) / 0.1 Shake Cr(VI) / 0.05 Shake Cr(VI) / 0.1

ERG Study Sonicator vs. Shaker: Cr(VI) only

(ng/mL Cr6+)

14

ERG Study Sonicator vs. Shaker: Cr(III) and Cr(VI)

80%

90%

100%

110%

120%

130%

140%

150%

160%

170%

180%

190%

200%

15 30 45 60 75

% R

eco

very

Time (minutes)

Sonic O / 0.05 Sonic O / 0.1 Shake O / 0.05 Shake O / 0.1

(ng/mL Cr6+)

15

ERG Study Sonicator vs. Shaker: Increased Precision

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2002 2003 2005 2006 2007 2008 2009 2010 2011 son 2011 shake

143 182 140 115 118 113

% R

PD

66

45

Maximum values for %RPD are in blue16

� Detected Cr(VI) on filters spiked with high levels of Cr(III) extracted via sonication

� Found that filters spiked with both Cr(III) and Cr(VI) extracted via shaking show acceptable recoveries and variability

� Duplicate filters extracted via shaker show good %RPD

� Changed extraction method to shaking for 45 minutes instead of sonicating for one hour

ERG Study Sonicator vs. Shaker: Conclusions

17

� Compared the IC/UV to IC/ICP-MS

– Standard NATTS method uses IC/UV

– New method uses IC/ICP-MS

� IC/ICP-MS method uses same filter as IC/UV (Sodium bicarbonate coated cellulose filters)

– Coated after acid washed, clean enough for low

concentration ambient samples

– Causes background on IC/ICP-MS for Cr(VI)

– Sodium Bicarbonate causes Cr(III) to precipitate (only

able to detect Cr(VI), not Cr(III) and Cr(VI) on

IC/ICPMS)

18

NJ DEP CTI: Analytical Instrumentation

Preliminary/Draft

NJ DEP CTI: Initial Laboratory Comparisons

� Audits were put in place in order to confidently evaluate the different analytical techniques

– External audits obtained by Wibby Environmental

– Internal audits prepared by ERG

� Method Detection Limits were preformed on each analytical system.

– IC/UV lower than IC/ICPMS (background interference

on IC/ICPMS does not allow lower MDL)

– ERG IC/UV results 3-4 times lower

� Laboratory techniques improved over time


0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0

EOSHI IC/ICPMS Clarkson IC/ICPMS EOSHI IC Clarkson IC ERG IC

low high low w/Cr3+ high w/Cr3+

NJ DEP CTI:Wibby Audit (August 2011)

Perc

ent

Recovery


0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0


low high low w/Cr3+ high w/Cr3

NJ DEP CTI:ERG Audit (September 2011)

Perc

ent

Recovery


0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0


low high low w/Cr3+ high w/Cr3

NJ DEP CTI:ERG Audit (October 2011)

Perc

ent

Recovery


0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0


low high

NJ DEP CTI:ERG Audit (January 2012)

Perc

ent

Recovery


� Sampler set up– Sampler installation – 6 samplers (8 ports)

• Plus ERG Prototype sampler

– Instrument design challenges

• Clarkson sampler also had water on filters but was a sampler error and was easily corrected

• ERG prototype collected water and was removed from study

� Not enough valid samples but did get some information– Use only Teflon screens in samplers (Clarkson)

– Tighten all filter holders (NATTS)

24

NJ DEP CTI:Summer 2011 Field Results

Preliminary/Draft

25

NJ DEP CTI: NYS-Clarkson Sampler

Preliminary/Draft

� All samples taken were sent to each laboratory– Each set included 8 samples

• 2 NATTS samplers – each sampler able to sample primary and collocated

• 4 Clarkson samplers – for study, 2 considered primary, 2 considered collocated

– Two whole sets sent to each laboratory

• EOHSI and Clarkson analyzed by IC/UV and IC/ICPMS

• ERG analyzed by IC/UV only

– Two sets sent to all 3 laboratories

• EOHSI analyzed by IC/UV and IC/ICPMS

• Clarkson and ERG analyzed by IC/UV only

26

NJ DEP CTI:Winter 2012 Field Results

Preliminary/Draft

� Only the IC/ICPMS & IC/UV final Cr(VI) results can be compared

� At this point, have not been able to compare the interconversion of Cr(VI)/Cr(III) by IC/ICPMS and IC/UV– Different sample preparation techniques used

• IC/ICPMS – acidified Nitric Acid extraction before analysis

• IC/UV – basic Sodium Bicarbonate extraction before analysis

� Blanks – IC/ICPMS appear to have interferences at low levels with the sodium bicarbonate filters – Could this be a problem for IC/ICPMS for low concentration

samples?27

NJ DEP CTI: Field Results - Discussion

Preliminary/Draft

� EPA Approved Cr(VI) Method developed by ERG was approved as an ASTM International Standard– Began process in 2008

– Modified draft in 2011 to incorporate new preparation technique (shaker instead of sonicator)

– Received approval on May 7, 2012!

Standard Test Method for

Determination of Total Suspended Particulate (TSP) Hexavalent Chromium in Ambient Air Analyzed By Ion Chromatography (IC) and SpectrophotometricMeasurements

28

ERG Recent Development (2012)

Conclusions

� ERG Prototype sampler collects water in humid/rainy conditions. It does give good recoveries for samples left out for multiple days

� Shaker converts less Cr(III) to Cr(VI) for spiked filters that contain Cr(III) over time

� More studies are needed to compare Clarkson and NATTS sampler Cr(VI) recoveries– Summer study not conclusive – too many variables

but did learn from it

– Winter study had low concentrations. Need more samples to obtain any definitive conclusions

� Now have a ASTM Cr(VI) method29 Preliminary/Draft

Acknowledgements

� The Community Toxics Initiative Grant was funded in part by the US Environmental Protection Agency through a grant to the New Jersey Department of Environmental Protection

� We would like to thank the following people at U.S. EPA:

– David Shelow, Mike Jones, Joann Rice, Dennis Mikel

30

Questions?

31

method development studies for hexavalent chromium … · method development studies for hexavalent...

Documents