close

Вход

Забыли?

вход по аккаунту

?

Identification of standard method biases in chromium(VI) species analysis and microwave improvements in extraction efficiency

код для вставкиСкачать
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
film s the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type o f computer printer.
T h e quality o f this reproduction is dependent upon th e quality o f the
cop y subm itted.
Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted.
Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
O versize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back o f the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy.
Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI
A Bell & Howell Information Company
300 North Zeeb Road, Ann Arbor MI 48106-1346 USA
313/761-4700 800/521-0600
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IDENTIFICATION OF STANDARD METHOD BIASES
IN CHROMIUM(VI) SPECIES ANALYSIS
AND
MICROWAVE IMPROVEMENTS IN EXTRACTION
EFFICIENCY
A Thesis Presented to the
Bayer School o f Natural and Environmental Science
o f Duquesne University
As Partial Fulfillment o f the Requirements
for the degree o f Master o f Science
by
Yusheng Lu
July, 1998
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
UMI Number: 1391227
UMI Microform 1391227
Copyright 1998, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MI 48103
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
Name:
Thesis Title:
Yusheng LU
identification o f Standard Method Biases In Chromium(Vi) Species Analysis and
Microwave Improvements In Extraction Efficiency
Degree:
Master o f Science
Date:
August 6, 1998
Approved^.
£
y Dr. H.M. (Skip) Kingston, Advisor
Department of Chemistry and Biochemistry
Approved:
Dr. Omar Steward. Committee Member
Department o f Chemistry and Biochemistry
Approved^,
Q . C .
Dr. Thomas Isenhour, Chairman
Department o f Chemistry and Biochemistry
Approved^.
[y.QoJyiC{
Dr. Heinz Machatzke, Dean
Bayer School o f Natural and Environmental Sciences
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
ACKNOWLEDGMENTS
In the completion o f my thesis project, there are many people to
whom I would like to express my thanks.
First. I would like to thank my advisor professor H. M. (Skip)
Kingston with the deepest gratitude for his exceptional guidance. I would
like to thank Dr. Omar Steward, my second reader, for his help and his
valuable time.
I would like to express my gratitude to Dengwei. my husband for his
helpful discussions and the SIDMS data he provided.
Moreover, I would like to thank all the members o f the Kingston
research group, especially Dengwei, Stuart. Elke, Dirk, and Helen for their
kindness and help.
And I would like to thank the Department o f Chemistry and
Biochemistry for their support.
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
ABSTRACT
This thesis consists of two parts: the identification o f standard method
biases in chromium six species analysis, and the method improvements in
efficiency. The paired methods in EPA RCRA SW-846 Update III for
environmental measurement of Cr(VI). 3060A (alkaline extraction) and
7196A (UV-Vis detection) may cause some analytical biases o f Cr(VI) by
different mechanisms during alkaline extraction, neutralization, storage, and
detection. The newly developed Speciated Isotope Dilution Mass
Spectrometry (SIDMS) method, which accurately measured the
concentrations o f Cr(VI) and Cr(III) in sample extracts, was applied in these
experiments to evaluate these EPA methods. SIDMS provided the means to
trap errors, evaluate matrix component and isolate method biases. For the
method improvement, a microwave enhanced closed vessel extraction
procedure was developed to replace the conventional hot plate process (EPA
3060A). This extraction method greatly shortened the extraction time,
reduced the oxidation o f Cr(III) to Cr(VI), and enhanced the precision of
analysis results.
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS---------------------------------------------------------------- ii
ABSTRACT---------------------------------------------------------------------------------iii
TABLE OF CONTENTS-----------------------------------------------------------------iv
LIST OF TABLES (PART I ) ---------------------------------------------------------- vii
LIST OF TABLES (PART I I)--------------------------------------------------------- vii
LIST OF FIGURES (PART I ) --------------------------------------------------------- vii
LIST OF FIGURES (PART II)------------------------------------------------------- viii
iv
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
PARTI
IDENTIFICATION OF STANDARD METHOD BIASES
IN CHROMIUM(VI) SPECIES ANALYSIS
1. INTRODUCTION................................................................................................. 2
1.1 Chromium and Chromium Speciation..........................................................2
1.1.1 Chromium and Its Chemistry..................................................................2
1.1.2 Toxicity o f Chromium and Its Presence................................................3
1.2 Applied M ethods.............................................................................................5
1.2.1 EPA Method 3C60A.................................................................................5
1.2.2 EPA Method 7 196A.................................................................................7
1.2.3 Method Speciated Isotope Dilution Mass Spectroscopy (SIDMS) - 8
1.3 Research G oals...............................................................................................10
2. EXPERIMENTAL SECTION...........................................................................12
2.1 Reagents.......................................................................................................... 12
2.2 Sample Preparation........................................................................................14
2.3 Equipment....................................................................................................... 15
3. RESULTS AND DISCUSSION........................................................................17
3.1 Biases in Method 3060A .............................................................................. 17
3.1.1 Oxidation o f Cr(III) During Extraction............................................... 17
3.1.2 Loss o f Insoluble Cr(VI) During Neutralization............................... 21
3.1.3 Reduction o f Cr(VI) During Neutralization.......................................30
3.1.4 Suggestion and M odification............................................................... 31
3.2 Biases in EPA Method 7196A ..................................................................... 32
3.2.1 Detection o f Cr(VI) With Method 7196A .......................................... 33
3.2.2 Comparable Results for COPR Samples.............................................36
3.2.3 Lower Recoveries of Cr(VI) for Some Sand and Soil...................... 38
3.2.4 Bias Sources in Method 7196A............................................................41
3.2.5 Method of Standard Addition................................................................54
3.2.6 Cr(VI) Wastes and Ascorbic Acid........................................................55
3.2.7 Suggestion............................................................................................... 55
4. SUM M ARY......................................................................................................... 57
5. REFEREN CES....................................................................................................58
v
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
PART n
MICROWAVE IMPROVEMENT IN EXTRACTION
EFFICIENCY
1. INTRODUCTION.................................................................................................63
1.1 Applications o f Microwave Assisted Sample Preparation......................63
1.2 Microwave Assisted Closed Vessel System .............................................. 64
1.3 Microwave Heating M echanism .................................................................. 65
1.4 Microwave Instrumentation.......................................................................... 72
1.5 Research G oal..................................................................................................74
2. EXPERIMENTAL SECTION............................................................................78
2.1 Reagents............................................................................................................78
2.2 Equipment.........................................................................................................79
3. RESULTS AND DISCUSSION........................................................................ 83
3.1 Extraction o f Cr(VI) S o lid s.......................................................................... 83
3.2 Oxidation o f Cr(III) During Microwave Extraction................................. 85
3.2.1 Oxidation o f Cr(III) at 90-95°C............................................................. 85
3.2.2 Oxidation o f Cr(III) at 120°C and 150°C...........................................88
3.2.3 Quantities o f Cr(VI) Oxidized from Soluble C r(III)......................... 90
3.3 Extraction o f PbCrtTt at 150°C.....................................................................92
3.4 Bias from Neutralization P rocess
...............................................•••••94
3.5 Future Study and Suggestions.......................................................................97
4. SUMMARY...........................................................................................................99
5. REFERENCES................................................................................................... 100
vi
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
LIST OF TABLES (P iM tT ^
Table 1 Oxidation o f Cr(III) during extraction with hot plate system ............20
Table 2 Cr(VI) recoveries o f COPR samples using 7196A and SID M S.......37
Table 3 Recovery o f spiked Cr(VI) in sand and soil extracts.........................
LIST OF FIGURES (PART I)
Figure I Distributions of chromium species upon pH and potential at 25°C..4
Figure 2 Flow chart of the whole procedure in paired EPA Methods
3060A/7196A and the possible analytical bias sources.....................11
Figure 3 Relationship between the quantity o f lead chromate extracted and
the pH at which precipitate started to occur during neutralization.......... 24
Figure 4 Distributions of the three species o f CO: in aqueous solution
in pH range 2.0-12.2................................................................................25
Figure 5 Calibration curve o f Cr(VT)....................................................................35
Figure 6 Effect of some soil matrix components on Cr(VI) recoveries......... 43
Figure 7 Spectra o f Fe(II)-phen in Fe(III), phen. and DPC mixture................. 51
Figure 8 Spectrum o f DPC oxidized by FeJ~ and
sub-boiling distilled nitric a c id ..............................................................53
LIST OF TABLES (PART II)
Table 1 Dissipation factors o f different materials.............................................68
Table 2 Microwave extraction o f Cr(VT) Standards..........................................84
Table 3 Microwave extraction o f P b C r0 4 at 150°C...........................................87
Table 4 Detection o f neutralized P b C r0 4 by UV-Vis and SIDMS................. 87
vii
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
LIST OF FIGURES (PART II)
Figure 1 Electromagnetic spectrum .....................................................................67
Figure 2 One of the microwave heating mechanisms: dipole rotation
o f water m olecules.................................................................................. 70
Figure 3 Components of microwave unit............................................................ 73
Figure 4 Milestone Unit: (I) exhaust module EM-45;
(II) M LS-1200 MEGA; (III) control terminal 240.............................80
Figure 5 Oxidation o f Cr(III) under different conditions................................. 91
Figure 6 Amount o f Cr(VI) oxidized from soluble Cr(III) and
the amount of soluble Cr(III).................................................................93
viii
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN CH RO M IU M (V I) SPECIES ANALYSIS
PARTI
IDENTIFICATION OF STANDARD METHOD BIASES
IN CHROMIUM(VI) SPECIES ANALYSIS
1. INTRODUCTION................................................................................................. 2
1.1 Chromium and Chromium Speciation.......................................................... 2
1.1.1 Chromium and Its Chem istry.................................................................. 2
1.1.2 Toxicity o f Chromium and Its Presence................................................ 3
1.2 Applied M ethods.............................................................................................. 5
1.2.1 EPA Method 3060A .................................................................................. 5
1.2.2 EPA Method 7196A.................................................................................. 7
1.2.3 Method Speciated Isotope Dilution Mass Spectroscopy (SIDM S).. 8
1.3 Research G oals................................................................................................10
2. EXPERIMENTAL SECTION............................................................................12
2.1 Reagents............................................................................................................12
2.2 Sample Preparation......................................................................................... 14
2.3 Equipment........................................................................................................ 15
3. RESULTS AND DISCUSSION........................................................................ 17
3.1 Biases in Method 3060A ............................................................................... 17
3.1.1 Oxidation o f Cr(III) During Extraction................................................17
3.1.2 Loss o f Insoluble Cr(VI) During N eutralization................................21
3.1.3 Reduction o f Cr(VI) During N eutralization........................................30
3.1.4 Suggestion and M odification................................................................ 31
3.2 Biases in EPA Method 7196A ......................................................................33
3.2.1 Detection o f Cr(VI) With Method 7196A ........................................... 33
3.2.2 Comparable Results for COPR Sam ples............................................. 36
3.2.3 Lower Recoveries o f Cr(VI) for Some Sand and Soil.......................38
3.2.4 Bias Sources in Method 7196A.............................................................40
3.2.5 Method o f Standard Addition................................................................ 54
3.2.6 Cr(VI) Wastes and Ascorbic Acid.........................................................55
3.2.7 Suggestion................................................................................................ 55
4. SUMMARY.......................................................................................................... 57
5. REFERENCES.................................................................................................... 58
Yusheng Lu 1:14 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
1
PA R T I IDENTIFICATION O F STAN DARD METHOD BIASES IN CHRO M IUM (VI) SPECIES AN ALYSIS
L
INTRODUCTION
1.1 Chromium and Chromium Speciation
1.1.1 Chromium and Its Chemistry
Chromium with an atomic number o f 24 belongs to the first series o f
the transition elements. It can occur in every one o f the oxidation states from
-2 to +6. Among these oxidation states. Cr(III) and Cr(VI) are the two
dominant oxidation states in natural environment and the m ost important
oxidation states in solutions.
Cr(VI) can exists as chromate (C r04:'). dichromate (C r20 7: ), and
hydrogen chromate (HCrOf) in solutions. Above pH 8, C r 0 42' is the major
species. When the pH decreases into range 2-6, the equilibrium shifts to
dichromate (1). Cr(VI) is a strong oxidant, and the redox potentials o f
Cr(VI)/Cr(III) highly depend on pH (Figure I). In acidic solutions, the
following reaction may be involved:
Cr20 72- + 14H~ + 6e' -► 2C r^ + 7H20
(E,° =1.33 eV),
The corresponding N em st equation is
In basic solutions, reaction
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
2
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
Cr(OH)3 + 50H- - 3e* -> C r0 42' + 4H20
( E / = -0.13 eV)
may take place, and the Nemst equation is
E
2
=E
„
RT
\OH~V
I s 7 ----------- *-=■
2 nF*[crOi-}
To simplify the calculation, we use concentrations instead of activities
in the above equations.
Based on these two Nem st equations, both redox potentials E, and E,
are related to pH and temperature. Besides these two factors, E, and E2also
depend on concentrations o f Cr20 72' and Cr3*, and concentration of C r0 42'
respectively. All these factors will be discussed and presented later in the
first part of the thesis. For brief introduction, the distributions of chromium
species ( including Cr(III), Cr(VI), and other chromium species) upon pH and
potential at 25°C are shown in Figure 1(1).
1.1.2 Toxicity o f Chromium and Its Presence
Chromium is ubiquitous in the universe: its compounds are frequently
encountered in minerals and in geochemical deposits; it ranks as the sixth
metal in the earth’s crust, ahead of nickel, copper, lead, and zinc in
abundance, fifteenth in sea water, and fifteenth in human bodies (2).
Chromium and its compounds have been widely applied in industry as
pigments, dyes, steels, refractory material, and in leather tanning,
electroplating processes, as well as in chemical reactions as catalysts.
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
PA RT I IDENTIFICATION OF STANDARD M ETHOD BIASES IN CHROM IUM {VI) SPECIES ANALYSIS
E (eV)
HCrOj
CrfOFX
0
2
4
6
8
10
12
14
pH
Figure 1 Distributions o f chromium species upon pH and potential at 25°C
(1)
Yusheng Lu 1:14 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
4
PART I IDENTIFICATION OF STAN DARD METHOD BIASES IN CHRO M IUM (VI) SPECIES ANALYSIS
Because o f the extensive applications of chromium and its compounds, large
quantities o f this element are being discharged into the environment.
However. Cr(VI) is a human carcinogen via inhalation, while trace
Cr(III) is an essential nutrient for humans and other mammals. Therefore,
there is a growing concern about the fate and effects o f chromium in the
environment. For analytical chemists, it is very important to quantify these
two species o f chromium especially the toxic Cr(VI) in water (3.4), soil
samples (5-7). food (8). biological tissues (9. 10). and human blood and urine
(11). Such quantification is also helpful for the environmental protection and
the remediation o f the chromium-contaminated environment (12).
1.2
Applied Methods
The determination o f Cr(VI) in solid samples requires complete
extraction and accurate detection. EPA method pairs 3060A and 7196A w-ere
considered as such methods. However, our studies in these two methods
indicated that these conventional EPA methods had analytical biases when
they were applied to some environmental samples. As a diagnostic tool.
SIDMS, a newly developed method for accurately detecting Cr( VI), was also
applied in our research to test and study some o f these biases. All the details
about these three methods are described in the following sections.
1.2.1 EPA Method 3060A
Yusheng Lu 1:14 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
PA RT I IDENTIFICATION OF STANDARD METHOD BIASES IN CH RO M IU M (V I) SPECIES ANALYSIS
Cr(VI) in solid samples is extracted prior to measurements. Both
water-soluble and water-insoluble Cr(VI) compounds may exist in such
environmental samples as sands, soils, and sediments. Barium chromate and
lead chromate are among the most water-insoluble Cr(VI) salts. Therefore,
an extraction procedure should be dynamic enough to completely extract as
many o f the Cr(VI) compounds as possible .
RCRA SW-846 EPA Method 3060A is the currently accepted
extraction method o f Cr(Vl). In Method 3060A (13). the alkaline extraction
solution (0.28 M Na:CO3/0.5 M NaOH) is used to extract both water-soluble
and water-insoluble Cr(VI) compounds by heating to 90-95°C to accelerate
the dissolution. This temperature should be maintained for one hour (14).
A fter the extracted solution is cooled down to room temperature, it is filtered
with a 0.45pm membrane filter. Then the filtrate is neutralized to pH 7.5 ±
0.5 with concentrated nitric acid. The neutralized extract can be stored in a
bottle at 4°C for further analysis. In our experiments, all the detection
procedures were performed right after the extraction.
Method 3060A is a currently accepted protocol (15, 16); however,
Cr(III) may be partially oxidized during extraction resulting in positive
results. The oxidation of soluble Cr(III) in a soil sample has been suggested
as a possible Method 3060A deficiency (17, 18). Additionally, the
neutralization process could cause negative results o f Cr(Vl).
Y usheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN CH RO M IU M (V l) SPECIES ANALYSIS
1.2.2 EPA Method 7 196A
The complete extraction and the accurate detection o f Cr(VI) are two
important requirements for Cr(VI) determination in real environmental
samples. In conjunction with the extractionjirocess, EPA Method 7196A is
generally suggested as an appropriate detection method for Cr(VI).
Current methods for detecting Cr(VI) include UV-Vis detection (1921), chromatography (22-24), and electrochemistry (7. 21). Among these
methods. Method 7196A, an inexpensive UV-Vis detection method of Cr(VI)
with reasonable sensitivity, has been most extensively applied in the
quantification o f Cr( VI) or total chromium in water samples (4), biological
tissues (10). and soil matrices (6) by detecting the absorbance at 542 nm in a
pH 2 solution adjusted with 10% sulfuric acid. The reactions in this method
can also be used to determine Fe (25) and ascorbic acid in pharmaceuticals
and fruits (26).
However, several problems have been demonstrated when this method
is applied to real samples with complex matrices (20). For example, large
quantities o f reducing agents may coexist with Cr(VI) in some samples and
may result in low recoveries of Cr(VI). Our experiments indicate that the
analytical biases in Method 7196A may be caused by different mechanisms.
Several common soil matrix components were evaluated for their effect on
the biases o f Method 7196 A.
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
PART I IDENTIFICATION OF STANDARD M ETHOD BIASES IN CHROM IUM fVI) SPECIES ANALYSIS
1.2.3
Method Speciated Isotope Dilution Mass Spectroscopy (SIDMS)
As described previously, Method pairs 3060A/7196A includes the
alkaline extraction o f CrfVI) from solid samples, the neutralization of
extracts, and the UV-detection in acidic solutions. During the procedures, the
pH decreases from 12 (extraction) to 7.5 (neutralization), then to 2 (UV-Vis
detection). With the decrease in pH. the redox potentials of C r0 1: 'Cr(0H)3
and Cr;0 7:7Cr* may change dramatically (see Section 1.1.1), and these
changes in these two potentials may cause the interconversion between
Cr(III) and Cr(VI) resulting in method biases. For example, the oxidation of
Cr(III) may take place during the extraction, and the reduction o f Cr( VI) may
occur during the detection in Method 7196A.
Besides these two potential analytical biases, the EPA method pairs
may cause some other biases of Cr(VI) which can not be identified by these
EPA methods themselves. The new method Speciated Isotope Dilution Mass
Spectroscopy (SIDMS) (5. 27. 28) was applied in our experiments as a
diagnostic tool to identify the method biases in Methods 3060A and 7196A.
SIDMS corrects the interconversion between such species as Cr(III)
and Cr(VI) during manipulation and/or measurement procedures. It can
exactly measure the concentration o f the two species at the point o f spiking
with one or more separated stable isotopes that behave chemically like the
natural-occurring species. Thus it is also a diagnostic tool for identifying the
most error-prone steps in the speciated measurement, storage, and sample
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
8
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN CHRO M IUM (VI) SPECIES ANALYSIS
preparation. This new procedure has become an update(IV) EPA method in
the RCRA program in 1998 (29).
Chromium has four naturally occurring isotopes: 50Cr. ^Cr, n Cr, and
<JCr. ' :Cr has the highest isotopic abundance in nature, while ^C r has the
lowest (Table). In our experiments, a ^Cr enriched spike for Cr(III) and a
'3Cr enriched spike for Cr(VI) were utilized. The concentrations o f these two
spikes were calibrated with inverse isotope dilution approach. The following
table shows the isotope abundances o f chromium for natural and isotopicallyenriched materials.
Natural Isotopic
Abundance
'Cr(III) and n“C r ( \rl)
Enriched Isotope :0Cr*
Enriched Isotope :jCrb
;0Cr( III) spike
53Cr(VI) spike
4.35%
50
93.1%
50
0.03%
83.79%
52
6.8%
52
2.19%
53
9.50%
53
0.1%
53
97.7%
54
2.36%
54
0%
54
0.08%
50
Isotec Inc. Lot #2 6 9 1 . b Isotec Inc. Lot #2692
The aqueous sample was collected, followed by double-spiking with
both 50Cr(III) and 5JCr(VI) isotopic spike. The natural chromium in the
aqueous sample was equilibrated with the isotopic spikes. A n anionexchange chromatography connected an ICP-MS physically was used to
separate and measure Cr(III) and Cr(VI) in the sample.
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
PART I IDENTIFICATION O F STANDARD M ETHOD BIASES IN CHROMIUM(VI) SPECIES ANALYSIS
1.3 Research Goals
Based on the two EPA method pairs and the chemical properties o f
chromium, potential biases may occur in both methods as described above.
My research goals for the first part o f my thesis are to study the possible
biases in the two methods: (1) the oxidation o f Cr(III) during extraction: (2)
the biases during the neutralization: and (3) the reduction of Cr(VI') during
the UV-Vis detection. SIDMS will be applied to identify the potential errors
in each step. Evaluation o f both method conditions and soil matrix
components that may cause these biases is the overall objective in this
research.
The flow chart and some o f the potential biases o f these paired methods
is shown in Figure 2.
Yusheng Lu 1:14 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
10
PART I IDENTIFICATION OF STANDARD M ETHOD BIASES IN CHROMIUM(VI) SPECIES ANALYSIS
M ethod 3060
M ethod 719G&
(alkaline Extraction)
(UV-Vis detectioq)
Extract Cr(VI) with 0.28M
Add extract and DPC
Na:C 0 3 /0.5 M NaOH at
reagent in a volumetric
90-95°C for 60 min.
flask. Adjust the solutioa
to pH 2 with 10% H;S 0 4
Potential errors
Cr(III)
Cr(VI)
Filter the extract
UV-Vis detection at
after it is cool to
542nm
room temperature
P oten tial errors
Cr(VI)
Neutralize the filtrate with
nitric acid to pH 7.5± 0.5 &
store it for determination
P oten tial errors
1. Loss o f Cr(VI)
2. Cr(VI) — ^Cr(III)
F igu re 2. Flow Chart of EPA Methods 3060A/7196A and Some Possible
Analytical Bias Sources
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
11
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHROMIUM (VI) SPECIES ANALYSIS
2.
EXPERIM ENTAL SECTION
2.1 Reagents
Deionized (DI) water (18MQ) prepared by a NANOpure Ultrapure
Water System (Bamstead) (Dubuque. Iowa) was used in the preparation of all
solutions. Sub-boiled nitric acid was prepared from quartz stills (Milestone,
Sorisole (BG). Italy).
•
The extraction solution (0.5 M NaOH/0.28M N a,C 0 3) in Method 3060A
was made by dissolving 20 g of NaOH (98%. Fisher. Fair Lawn. NJ) and
30 g anhydrous Na:C 0 3 (99.6%. Fisher, Fair Lawn, NJ) in 500 mL of
deionized water and then diluting to I L.
•
Lead chromate solid (Special for micro analysis. Fisher. Fair Lawn. NJ)
and barium chromate solid (98+%. Aldrich. Milwaukee. WI) were
extracted by Method 3060A.
•
Natural Cr(VI) standard solution was prepared by dissolving dried
analytical reagent grade potassium dichromate (99.4%. J.T. Baker Inc.,
Phillipsburgh, NJ) in deionized water.
•
Natural Cr(III) standard solution was prepared by dissolving chromium
metal (99.995%, Aldrich Chemical Co., Milwaukee, WI) in a minimum
amount o f 6M HC1 prepared from sub-boiled HC1 and the solution was
diluted with 1% H N 0 3.
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
12
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHRO M HJM (VI) SPECIES ANALYSIS
•
Isotope 50Cr(III) standard solution (5) was made by dissolving chromium
metal enriched in ^C r (Lot#269l, Isotec Inc.. M iamisburgh. OH) in 6 M
HC1 in Teflon vessel, and diluting the solution with 1% H N 0 3.
•
Isotope 53Cr(VI) standard solution (5) was prepared using chromium
oxide enriched in 5jCr (Lot#269l. Isotec Inc.. Miamisburgh. OH) as the
source material. 5,Cr enriched oxide and concentrated HC104 (67-70%,
Optima) were added into a Teflon vessel and slowly heated until all solid
was dissolved. After the solution was cooled. H:0 : and NH4OH were
added, and the vessel was heated slowly again to oxidize all Cr to Cr( VI).
The solution was then boiled for 20 minutes to remove the excess H:0 :.
•
Reagent diphenvlcarbazide (DPC) was made by dissolving 0.5g DPC
(Fisher. Fair Lawn. NJ) in 100 ml acetone and stored in a brown bottle.
•
Concentrated nitric acid (Fisher. Pittsburgh. PA) was used for
neutralizing the extracts.
•
10% (v/v) sulfuric acid was made by slowly adding 10 ml concentrated
sulfuric acid (Fisher, Pittsburgh. PA) to 90 ml DDI water with continuous
stir.
•
K M n04 solution containing approximate 10 pg/'g Mn was prepared bydissolving KJVlnOj (J. T. Baker Inc., Philipsburgh. NJ) in deionized water.
•
Fe3* solutions (10 ppm and 1000 ppm) were made by dissolving ferric
nitrate (100.4%, Fisher, Fair Lawn, NJ) in deionized water, and the pH o f
the solution was adjusted to 3.4 with diluted sulfuric acid.
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
13
PART I IDENTIFICATION O F STANDARD M ETHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
•
Fe:t solution (1 .8x 10* M) was prepared by dissolving F e S 0 4 (Fisher, Fair
Lawn, NJ) in deionized water. The pH o f the solution was adjusted to 3.4
with diluted sulfuric acid (10%).
•
1.10-phenanthroline solution (1.0 x 10'5 M) was made by dissolving 1.10phenanthroline (Aldrich. 98%) in deionized water.
•
Phenol solution (5.0 x 10‘: M) was made by dissolving phenol
(Mallinckrodt. Crystal) in deionized water.
•
Na.S solution (0.01 M) was made by dissolving Sodium Sulfide (Fisher.
Certified. 98%) in deionized water.
•
Phthalic acid solution (0.05 M) was prepared dissolving phthalic acid
(Aldrich. 98%) in deionized water.
•
Tartaric acid solution (0.05 M) prepared dissolving tartaric acid (Fisher,
Certified ACS) in deionized water.
All the other chemicals were analytical or ACS reagents.
2.2 Sample Preparation
•
The synthesized sample utilized in Section 3.2.4 of this part was made by
simply adding natural Cr(VI) solution and a individual soil matrix
component into the deionized water. This sample was then detected using
Method 7196 A.
•
Several COPR (Chromite Ore Processing Residue) samples were
determined in Section 3.2.2. According to the EPA M ethod 3060A, 2.5 g
Yusheng Lu 1:14 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
14
PART I IDENTIFICATION O F STANDARD M ETHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
COPR sample was weighted into a beaker, 50 ml extraction solution was
then added. After extraction, filtration, and neutralization, the COPR
extract was detected using both Method 7196A and SIDMS.
•
The same procedure was performed to sand samples as COPR samples.
•
The same extraction and filtration were processed on the soil samples
analyzed in Section 3.2.3. After filtration, some natural Cr(VI) was added
into the soil extracts before neutralization, and a portion of the solution
was taken out for SIDMS analysis. The rest o f the solution was
neutralized with concentrated nitric acid according to Method 3060A, and
detected by Method 7196A.
2.3 Equipment
•
A hot plate with stirring capability was used in Method 3060A for the
extraction of Cr(Vl).
•
A Cary UV-Vis Spectrometer (IE , Varian) was used to detect Cr(Vl) and
study some mechanisms.
•
For SIDMS measurement o f Cr(VI), a VG PlasmaQuad system (Fisons,
Winford, UK) (equipped with a water-cooled spray chamber and v-groove
nebulizer) and a continuous dynode multiplier as the detector were used
to detect Cr(VI) and C r(III)); a CostaMetric 4100Bio/MS pump (Thermo
Separation Products, Riviera Beach, FL) and a Cetac ANX 4605 Cr
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
15
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CH RO M JU M (V n SPECIES ANALYSIS
anion-exchange column (CETAC Corporation, Omaha, NE) were used to
separate Cr(VI) and Cr(IIT).
•
A Perkin Elmer 1100 Atomic Absorption Spectrophotometer was used to
detect iron concentrations in filtered soil extract and COPR 1 extract.
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
16
PART f IDENTIFICATION O F STANDARD M ETHOD BIASES IN CHRO M IUM (VI) SPECIES ANALYSIS
3.
RESULTS AND DISCUSSION
As stated previously, during the procedure o f the EPA method paire,
the pH o f the analyzed solution changes from pH 12 (alkaline extraction) to
7.5 (neutralization) to pH 2 (UV-Vis detection). The redox potentials o f
CnO-^/Cr3* and C rO ^C rfO H ^ change rapidly with the decrease in pH,
which may result in analytical biases o f Cr( VI) in the above steps. The
sources o f the potential biases were evaluated and analyzed step by step in
our experiments.
3.1 Biases in Method 3060A
A good extraction method o f Cr(VI) not only requires complete
extraction of Cr(VI) from environmental samples, but also limits analytical
biases during the whole procedure. However. Method 3060A may cause
biases in the following major steps: alkaline extraction, neutralization, and
storage. All these steps will be evaluated and discussed in the following
studies.
3.1.1 Oxidation o f Cr(III) During Extraction
Method 3060A effectively extracts Cr(VI) in the strong alkaline
solution at the specified temperature 90-95°C. Strong basic solution prevents
the reduction o f Cr(VI) to Cr(III); high temperature enhances the extraction
Yusheng Lu 1:14 PM 07/29/98
o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
17
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES AN ALYSIS
efficiency (15). Different from Cr(VI). Cr(III) is unstable in alkaline
solutions: and the oxidation o f Cr(III) may take place during the extraction o f
Cr(VI). The oxidation o f Cr(III). however, highly depends on the forms or
compounds o f Cr(III) (17). For example, aged Cr(OH)3 precipitate decreased
the tendency ofCr(OFT)3 to oxidize: water-insoluble Cr:0 3 was inert to
oxidation: freshly precipitated Cr(OHT, was partially oxidized to Cr(VI). In
this alkaline extraction solution, only a small fraction of Cr(III) exists as
Cr(OHV in equilibrium with insoluble Cr(OH)-,. When water-soluble Cr(IID
standard solution (such as C r(N 03)3 solution) is added to this extraction
solution, fresh Cr(OH)- precipitate would be formed. Although this freshly
precipitated Cr(OH)3 is not the representative o f most soil chromium(III) in
the field, the possibility of the oxidation o f fresh Cr( OH)3 precipitate or free
Cr3' cannot be excluded. Free Cr3' ion and freshly precipitated Cr(OH)3 may
be the two most easily oxidizable Cr(III) species, thus study o f the conversion
o f soluble Cr(III) in extraction solution would help us to estimate the possible
method biases during extraction.
Cr(III) standard solution was used as a sample to study the oxidation o f
the freshly precipitated Cr(OH)3 and/or free Cr3' during extraction. Because
soluble Cr(ni) does not interfere with the detection o f Cr(VI) in Method
7196A, this detection should be accurate. The detected quantities o f Cr(VI)
by Method 7196A demonstrated that soluble Cr(III) was oxidized to Cr(VI)
during the extraction procedure (Table 1).
Yusheng Lu 1:14 PM 07/29/98
o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
18
PART I IDENTIFICATION O F STANDARD M ETHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
When 102 p.g o f soluble Cr(III) was extracted, 11.9% o f oxidation was
obtained; while 5.6% o f Cr(IID was oxidized to Cr(VI) when 200 pg o f
soluble Cr(III) was extracted. Although there was a large difference between
these two percentages, the overlapped total quantities o f Cr(VI) oxidized
from Cr(III) were at the same level. The average quantities o f Cr( VI) in these
two cases were 12.2 pg and 1 l.l pg, respectively (Table 1). Further
experiments indicated that increasing the soluble Cr(III) slightly increased the
quantities of Cr(VI) oxidized from Cr(III) during extraction.
As described in Section 1.1.1, the redox potential o f C r0 4:7C r(0H )5
would be slightly changed from 25°C to 95°C. Besides temperature, the
redox potential also depends on the concentration of C r0 4:‘ and pH value. It
increases with the increase in the concentration o f C r0 4: . For example,
assuming the pH is 12 and the concentrations o f C r0 4: are 10‘5 M
(corresponding to 0.52 ppm Cr(VI) which is in the detection range in Method
7 196A) and 10'3 M (corresponding to 52 ppm Cr(VI) which may be the
concentration of Cr(VI) in extraction solution), the formal redox potentials at
95°C are calculated as 0.22 eV and 0.36 eV, respectively. The standard redox
potential o f 0 2/OH' in alkaline solutions is 0.401 eV, which means oxygen is
capable o f oxidizing Cr(III) to Cr(VI) in this extraction solution. Since there
was no matrix in these synthesized samples, oxygen was a potential oxidant.
The results in Table 1 suggested that the quantity o f the Cr(VI)
converted from Cr(III) was limited, no matter how much water-soluble
Yusheng Lu 1:14 PM 07/29/98
o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
19
PART I IDENTIFICATION OF STAN DARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
e
0
co
>»
•4—*
CO
0
.2 3
C3 C/5
2 +i
x
—
C5
m
vO
o
+1
VO
CN
rz»
+1
Ov
iy~»
tri
CN
**
js
CL
O
X
-C
c
o
o
0
X
1JJ
05
o ^
<u ^
CD c
cC c
C
m
VO
vo vn vr>
o C3
o *o
<5 x
ft. O
CV O
cn r -
Cv
r^
tri i n
cn
c
3
Q
•u.
c r cjd
> 3
O
c
o
0
■g
x
O
0
.Q
u
1
Cv
m ov Cv (N
io (N in
cv r —• —*
CN —' — O >
o "53
^ Q
*o
<u
"O
T3
< /co
0
CN (N
O
O
(N
O
OS
OO
O
Cv GV O
— — CN
u
Yusheng Lu 1:14 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
20
PART I IDENTIFICATION O F STANDARD M ETHOD BIASES IN C H RO M IU M fV I) SPEC IES ANALYSIS
Cr(III) was added in the 50 ml o f extraction solution. Therefore, the amount
o f Cr(VI) oxidized from Cr(III) during the extraction could be estimated in
these synthesized samples. However, the oxidation o f Cr(III) may vary with
the sample matrix. For example, some matrices may cause more oxidation of
Cr(IID during the extraction if some oxidants are present in samples.
3.1.2 Loss of Insoluble Cr(VI) During Neutralization
3.1.2.1
Observations
The quantification of toxic Cr(VI) in a sample requires the complete
extraction of Cr(VI) from water-soluble forms, as well as from waterinsoluble forms. Such solids as potassium dichiomate. barium chromate. and
lead chromate standards were studied to evaluate the effectiveness o f this
alkaline extraction method. Both methods 7196A and SIDMS were applied
to detect Cr(VI) in the extracts of these three solids. Because no matrix was
present in the detection solution, reduction o f Cr(VI) would not take place
during UV-Vis detection. The SIDMS results in pH 12 extracts indicated that
Method 3060A could effectively extract both water-soluble and waterinsoluble Cr(VI) compounds.
K2C r0 7 is water-soluble. Its solid could be easily dissolved in the
extraction solution, and 100% recoveries o f Cr(VI) were obtained by both
detection methods as expected.
Yusheng Lu 1:14 PM 07/29/98
perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
21
PART I IDENTIFICATION OF STAN DARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
B aC r04 and PbC r04 are among the most water-insoluble Cr(VI)
compounds. Based on our experiments, at least 0.0244 g o f BaCrO,
(corresponding to 5010 pg o f Cr(VI)) could be completely extracted.
Approximate 100% recoveries o f Cr(VI) were obtained by both detection
methods, although some white precipitate was observed on the filter paper
after the filtration. The 100% recoveries indicated that the white precipitate
was not a chromium compound.
In contrast to this result, when P bC r04 solid with a larger quantity than
0.017 g (corresponding to 2790 pg o f Cr(VI)) was extracted, white and
yellow precipitates were observed during neutralization. The pH value at the
point that the precipitation started to occur depended on the quantity of
PbCr04 solid extracted. More PbCr04 solid corresponded to higher pH
(Figure 3). During the detection by Method 7 196 A. if another filtration was
not performed after neutralization, poor precision was caused; if the filtration
was done to get rid o f the precipitate, low recoveries o f Cr(VI) were obtained.
However, when the filtrate o f PbCr04 (pH~12) before neutralization was
analyzed by SIDMS, 100% recoveries o f Cr(VI) were achieved, which
indicated the complete extraction of Cr(VI) from P bC r04 solid. Low
recoveries o f Cr(VI) by UV-Vis detection could be caused in the
neutralization step due to the reprecipitation.
Yusheng Lu 1:14 PM 07/29/98
o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
22
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
3.1.2.2 Mechanism on the Precipitation
We proposed a hypothesis to explain the above phenom ena we
observed. Before the discussion of the hypothesis, we need to describe the
influence o f pH on the distribution of inorganic carbonate species in the
extraction solution.
Figure 4 shows the distributions of three inorganic carbonate species
(C 0 32\ H CO ,\ and H2C 0 3) when the activity coefficients o f ions in the
aqueous solution are not taken into account. The pKa, and pKa: o f the
dibasic acid H2C 0 3are 6.35 and 10.33, respectively. When pH is higher than
10.33. the main specie in the solution is C 0 3:' : H C 0 3' is the dominant specie
in the pH range 6.35-10.33: and H2C 0 3 has the largest distribution when pH
is less than 6.33.
The extraction solution contains 0.28 M Na2C 0 3 and 0.5 M NaOH. In
this case, the solubility of C 0 2 in the extraction solution can be ignored.
Anion C 0 32' dissociated from Na2C 0 3 can shift to two other forms H C 0 3' and
H2C 0 3 in aqueous solutions. In the pH 12 extraction solution, there are 0.27
M (98% of the total concentration of these three species) o f C 0 32' and 0.01 M
o f OH'. During neutralization, the distribution o f C 0 32' dramatically
decreases to 50% at pH 10.3 and less than 1% at pH 8; while the
concentration o f OH* decreased to 10-6 M at pH 8. These calculations are
based on the following equations:
Yusheng Lu 1:14 PM 07/29/98
o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
23
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHRO M IUM (VI) SPECIES ANALYSIS
12.5 1
12
-
O 10.5<D
Q-
-i n -
o. 9 . 5 “
98.5-
0
7
6
5
2
4
3
Mass of Lead Chromate Extracted (x 0.01 g)
1
F igure 3 . R elationship betw een th e quantity o f lead chrom ate
extracted and th e pH at which precipitate started to occu r during the
neutralization
Yusheng Lu 2:36 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
24
8
PA RT I IDENTIFICATION OF STANDARD METHOD BIASES IN CHRO M IUM (VI) SPECIES ANALYSIS
0 .9
0.8
0 .7
H CO.
0.6
£
0 .5
0 .4
CO.
0 .3
0.2
0.1
2
4
6
8
10
12
pH
F ig u r e 4 . Distributions o f the three s p e c ie s o f c o 2 in a q u eo u s solution
in pH range 2 .0 -1 2 .2
Yusheng Lu 2:36 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
25
PART I IDENTIFICATION OF STANDARD M ETHOD BIASES IN CHROM IUM (Vl) SPECIES ANALYSIS
m'";
[ H- ] 2 + K ' \ H ‘] + K m K , 2 ’
[n-\~ + K „ \H '\+
The water-insoluble BaC r04 (solubility product K,p = 1.2 x 10'10 at
25°C) could be dissolved in this extraction
rBaCr04(s)cs> Ba2* +- C r0 42'
solution. The dissolution o f B aC r04 in
I
H C 03*ce> CO,2 -r H*
!
I
aqueous solutions could be expressed as:
Z
B aC 03(s)
B aC r04 <=> Ba2* +- C r0 42*. This equilibrium
can be shifted to the right by decreasing either the concentration o f Ba2* or
the concentration of C r0 42'. In this extraction procedure of Cr(VI), only the
concentration o f Ba2* can be decreased to enhance the solubility o f B aC r04.
Based on the previous discussion. C 0 32' is the dominate carbonate species at
pH 12, and its concentration is approximately 0.27 M in the extraction
solution. This highly concentrated C 0 32' causes the BaCO- precipitate
although the
of B aC 03 (5.1 x 10'9 at 25°C) was higher than that of
BaC r04 (1.2 x 10'10 at 25°C). The precipitation o f B aC 0 3 greatly decreases
the concentration of free Ba2* in solutions, driving the dissolution o f BaC r04.
Therefore in our experiments, almost all the barium could be precipitated as
BaCOj after extraction, releasing C r0 42' as a free anion in solutions. This
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
26
PART I IDENTIFICATION OF STAN DARD M ETHOD BIASES IN CHROMIUM(VI) SPECIES ANALYSIS
B aC 0 5 precipitate was removed during the filtration, and only a very small
quantity of barium was left in the filtered solution. When neutralizing the
solution from 12 to 7.5. the distribution o f inorganic carbonate changed
dramatically. A rough calculation indicated that the concentration o f CO;.:'
could be lower than 0.003 M. Despite such a low concentration o f CO ,:\
C r0 4:' could not reprecipitate because barium had been removed during the
filtration. Therefore, the alkaline extraction took advantage o f the formation
o f BaCO-, to release C r04:' from B aC r04 and the subsequent filtration to
remove barium to prevent the reprecipitation o f BaCr04.
PbCr04 (K,p = 2.8 x I O'13 at 25°C) is more water-insoluble than
B aC r04. However. PbC r04 can be dissolved in strong basic solutions. For
example, up to 12 mg of PbCrO, solid (corresponding to 1970 pg o f Cr(VI))
could be dissolved in a 50 ml o f pH 12 NaOH (0.01 M) solution using the
same procedures of Method 3060A. The formation of Pb(OH): (K,p = 1.2 x
10'15 at 25°C) or/and Pb(OH)4:‘ drove the dissolution of PbC r04 in this pH 12
NaOH solution. Different from PbC r04, B aC r04 could only be slightly
r
P bC 03 (s)
Z
HCOj- <=> C 0 3:- + H*
dissolved in this solution due to
the water-soluble Ba(OH)2.
- i.
PbC r04(s) <=> Pb:~+ C r0 42'
+
20H '
z
2 0H ' + Pb(OH)2(s) c=> Pb(OH)42'
Therefore, besides the
similar reactions involved in
BaCr04 extraction, two other
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
27
PART I IDENTIFICATION OF STANDARD M ETHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
main reactions also took place in the whole procedure o f P bC r04 extraction:
Pb2* ^ 20H- <=> Pb(OH)2 and Pb(OH)2 ^ 2 0 H ' e=> Pb(OH)42'. During
extraction, highly concentrated C 0 32' and OH' made the formation o f PbCO,
(K ^ = 7.4 x 10'l4at 25°C). Pb(OH);. and Pb(OH)4:'. These produced
chemicals significantly reduced the concentration of Pb2' in the solution and
drove the dissolution of P bC r04.
Although the precipitates o f PbCO-. and Pb(OH), could be removed by
the filtration, lead could still be present in the solution as Pb(OH)42'. During
the neutralization o f PbCrO, extract, equilibrium 1 shifted to the left as pH
decreased, and Pb(OH); precipitate was produced. Continuously decreased
pH. Pb2' was released from Pb(OH)2: then P b C r0 4 (s) and PbCO:. (s) could be
precipitated due to the released Pb‘‘. However, not much P bC 03 would exist
in the system at pH 7.5 because of the very low concentration of C 0 32' as
described previously. Moreover, the K,p o f PbCrO, was lower than that of
P b C 0 3. Therefore, some o f the dissolved P bC r04 compound was forced to
reprecipitate. That was the reason for the loss o f Cr(VI) as PbCr04 during the
neutralization. The more PbCrU4 solid was extracted, the more Pb(OH)42' and
C r0 42' could be present in the filtrate, and the sooner the precipitation o f
Pb(OH)2, then P b C 0 3 and P bC r04 occurred during the neutralization (Figure
3).
This hypothesis explained the complete extraction o f B aC r04 and
P b C r0 4 in our experiments and the reasons for the loss o f P bC r04 during the
Yusheng Lu 1:14 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
28
PART 1 IDENTIFICATION O F STANDARD M ETHOD BIASES IN CHROMIUM (VI) SPEC IES ANALYSIS
neutralization. If the concentrations of Crt VI) in real samples are low
enough, the loss of Cr(VI) with Method 3060A may not take place.
However, in soil samples, the total chromium content usually ranges from
100 to 300 pg/g. but may vary from trace to 4000 pg/g; and Cr(VI)
distribution varies from soil to soil (30). According to Method 3060A. 2.5g
solid sample should be weighed for each extraction. If Cr(VI) as P bC r04
compound in a soil sample is 1120 pg/g, the Cr(VI) in 2.5g sample should be
2800 pg. In this case, the reprecipitation of PbCrO4 may occur during the
neutralization. Increasing the extraction solution would be a possible proper
method to avoid or decrease the loss o f water-insoluble Cr(VI) compounds
during the neutralization. Decreasing the sample size, however, may cause
poor precision due to the heterogeneous nature of most environmental
samples.
Additionally, besides the reprecipitation of PbC r04, some other
compounds may also precipitate during the neutralization. For example,
some yellow or brown precipitates were observed when a soil extract which
contained a low concentration o f Cr(VI) was neutralized. The large quantities
o f humic acids and fiilvic acids in soil samples may be extracted by the
alkaline extraction solution. During the neutralization o f the extract,
however, some of these water-insoluble acids, especially the humic acids,
may precipitate as the pH decreases to 7.5. The precipitated organic acids
may result in the heterogeneity o f the extracts and influence the UV-Vis
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
29
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHRO M IUM (VI) SPECIES ANALYSIS
detection. A second filtration may help to improve the precision of the
detection o f Cr(VI) if the precipitates in the neutralized extracts are organic
acids. If the precipitates contain some water-insoluble Cr(VI) compounds,
however, the second filtration would cause the loss o f Cr(VI) resulting in
negative errors.
3.1.3
Reduction o f Cr(VI) During Neutralization
Besides the loss of some insoluble Cr(VI) compounds especially
PbC r04 solid, the reduction o f Cr(VI) could also be caused in the
neutralization step (Table 3 in Section 3.2.3). A soil sample was
expeiimented to test this reduction of Cr(VI) during neutralization.
The soil extract was made by using Method 3060A, and the filtered
extract was spiked with Cr(VI) standard. A portion o f the pH 12 filtrate and a
portion o f neutralized extract (pH 7.5) were spiked with 5jCr(VI) and 50Cr(III)
isotopic standards for SIDMS detection. As the data in T able 3 indicated, the
recoveries o f Cr(VI) in the pH 7.5 extracts were about 2-4% lower than those
in the pH 12 extracts, implying the reduction o f Cr(VI) to Cr(III) during the
neutralization o f the soil extract.
In acidic solutions, Cr(VI) is a common oxidant (Section 1.1.1). Its
oxidizing capability increases with the decrease in pH value. In the pH 7.5 ±
0.5 extract, both Cr(VI) and Cr(III) were believed stable under this condition.
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
30
PART I IDENTIFICATION O F STANDARD M ETHOD BIASES IN C H RO M IU M (V I) SPECIES ANALYSIS
The experiments, however, indicated that the reduction o f Cr(VI) may occur
although it may be kinetically slow. The detail experiments are described in
Section 3.2.3. Additionally, according to Method 3060A, the extract o f a
sample can be stored after neutralization for future analysis. The extended
storage time may cause more significant loss o f Cr(VI). For this reason,
immediate detection after neutralization is suggested.
3.1.4
Suggestion and Modification
Because sample preparation is a very important step in sample analysis,
and it is gaining more attention as a part o f the entire sample analysis.
Method 3060A. as the extraction method o f Cr(VI) from solid samples, may
cause analytical biases o f Cr(VI) during the whole procedure. For instances,
Method 3060A may cause the oxidation o f Cr(VI) during the extraction, the
loss o f some water-insoluble Cr(VI) compounds, and the reduction of Cr(VI)
during the neutralization. Besides these shortcomings. Method 3060A is also
time-consuming. This method takes 60 minutes for each extraction and
nearly 30 minutes to heat the extraction solution to 90-95 °C. Thus about 90
minutes is needed for each sample preparation using Method 3060A.
Statistically, at least three parallel analysis for each sample should be taken.
For this reason, at least 4.5 hours should be spent only on the extraction for
each sample. In addition, there is a doubt whether all the Cr(VI) forms can be
completely extracted by this method or not. Although the most water-
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
31
PART I IDENTIFICATION O F STANDARD M ETHOD BIASES IN CHROMIUM(VT) SPECIES AN ALYSIS
insoluble PbCrO, and BaC r04solids were completely extracted with this
method, some other forms o f Cr(VI) which possible exist in samples may not
be extracted at 90-95 °C with a hot plate. For example, if some Cr(VI) are
contained in crystals or coated by some extremely insoluble oxides, it will be
very hard to extract these Cr(VI) by Method 3060A.
Although we can increase the volume o f the extraction solution to
avoid or decrease the loss o f some water-insoluble Cr(VI) compounds during
the neutralization, it may increase the oxidation of Cr(III) during the
extraction. To overcome the shortcomings o f Method 3060A. we developed
a microwave-assisted closed vessel extraction method, a more robust
extraction methods of Cr(VI) from environmental samples.
In this newly developed method, the Milestone M LS-1200 microwave
unit was used to extract Cr(VI) from samples. This microwave-assisted
closed vessel extraction system can extract ten samples simultaneously in ten
minutes. Compared with the conventional Method 3060A. this method not
only greatly reduces the extraction time, but also decreases errors;
additionally, those non-extractable Cr(VI) forms in Method 3060A m ay be
extracted by this system. The microwave-assisted closed vessel extraction
method will be discussed in the second half o f the thesis as a specific project
to develop a more efficient Cr(VI) extraction.
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
32
PART I IDENTIFICATION OF STANDARD M ETHOD BIASES IN CHROMIUM(VI) SPECIES ANALYSIS
3.2 Biases in EPA Method 7196A
Method 7 196A, the detection method in this EPA method pairs, may
also cause analytical biases o f Cr(VI). Compared with SIDMS. Method
7 196A obtained lower concentrations or lower recoveries of Cr(VI) for some
sands and soil extracts. Comparable results were obtained for Chromite Ore
Processing Residue (COPR) samples. Sources o f the biases in Method
7196A were identified by studying the effects o f some soil matrix
components on Cr(VI) recoveries.
3.2.1 Detection of Cr(VI) W ith Method 7 196A
3.2.1.1 Mechanism of M ethod 7196A
The mechanism of Method 7196 A can be expressed as (31):
cnvn
This mechanism, however, has not been verified because the complex CrDPCO had never been isolated (31).
Yusheng Lu
1 :14
PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
33
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CH RO M IU M fV I) SPECIES ANALYSIS
The complex is produced by adjusting the solution containing
diphenylcarbazide (DPC) and Cr(VI) to pH 2 with 10% H2S 0 4 and detected
at 542 nm, the maximum absorption wavelength. According to this
mechanism, the strong oxidant Cr(VI) oxidizes DPC, then the produced
Cr(III) combines with the oxidized DPC, diphenylcarbazone (DPCO), to form
the red-violet complex Cr(IID-DPCO. These redox and complexation
reactions take place simultaneously (31). Our experiments confirmed that the
molecule o f complex Cr(III)-DPCO contained two DPC and one Cr (31).
Different from many other color-development reactions, a redox reaction as
well as a complexation reaction is involved in the mechanism. Since there is
a redox reaction involved, both reducing and oxidizing agents may cause
interferences: reducing agents may compete with DPC to reduce Cr(VI);
oxidizing agents may compete with Cr(VI) to oxidize DPC.
3.2.1.1
Calibration Curve
When 0.5 ml DPC solution was added into a 25 ml volumetric flask, a
calibration curve with a correlation coefficient between 0.999-1.000 was
obtained when the concentration of Cr(VI) was between 0.1 and 1.0 |J.g/g. A
typical calibration curve in the experiment is shown in F igure 5. According
to Beer’s Law :
A = e c d, where A is the absorbance, s is the molar
extinction coefficient, c is the molar concentration, d is the light path. Under
our detection condition, e and d are constants; and A is proportional to c.
Yusheng Lu 1:14 PM 07/29/98
o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
34
PART I IDENTIFICATION OF STAN DARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
Q8
8
c
Q6
(0
nL_
0
V)
2< 0.4
Q2
0
0.4
0.2
Q8
Q6
Concentration o f Cr(VI) (jig/g)
F igu re 5.
1
Calibration curve of Cr(VI)
Yusheng Lu 2:38 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
35
PART I IDENTIFICATION OF STANDARD METHOD BIA SES IN CHRO M tUM (V I) SPECIES ANALYSTS
According to the calibration curve in Figure 5, s was calculated as 4.1 x 104
Lcm 'V ofV
3.2.1.2 Limit o f Detection
After the absorbance o f ten blank solutions was measured at 542 nm,
the standard deviation (5) o f the measurements were calculated. For a
reliable detection. D. L. = ^ (where S is the sensitivity). The detection limit
for the solutions without matrix was then calculated as 0 .0 1 p.g/g.
3.2.2 Comparable Results for COPR Samples
Three different COPR samples were extracted by using Method 3060A.
Each filtered and neutralized extract (pH~7.5) was prepared for both 7196A
and SIDMS detection. A portion o f the pH 7.5 extract was spiked with
5jCr(VI) and "°Cr(III) isotopic standards for SIDMS detection: another
portion o f the neutralized extract was measured directly by Method 7196 A.
The concentrations o f Cr(VI) for the three CO PR samples detected by both
methods were shown in Table 2.
As the data in T able 2 suggested, the Cr(VI) concentrations in these
three COPR samples varied significantly. CO PR 1 contained the highest
36
with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
- 3 "S'
00 CO
CO
o
r-
-H
^—-V r-~
Q
CO
CT c5
u
-a ^ <
o v r ?=
O
n
vo
-H
-H
-H
o
std: standard deviation
PART I IDENTIFICATION O F STANDARD M ETHOD BIASES IN CHRO M IUM (VI) SPECIES ANALYSIS
oo
oc
oo
oo
cj ^
-a
c
03
<
CO
<
/)
(33
JD
CL
£
c5 „—.
cn CD o o
c"~
— 'S
i)
hO
)
C
cn
o cn — ^ r O NC o c
—
c
n r-i o
o
un cn
’ ( /)
cn
oc
*
3
-
cn
n-
Z)
CO
_0
u
f-
a.
E
!~
03
CO
.co - a
m
00 "on oo
3
-h
-H
O
'
CL
O
O
co
03
‘l.
0
~o
o
>
cr
>
o
CM
_0
oc
o
cn
£§
oc
ao U ^
cn
0
on
-H
on
of—
>
o
v C o
—• ca
O
']
I
>
3
a.
£
C/3 Ofi o o o
— o
c 'Hi) m
cn Tt- on
<N• wn• — •
— cn
cn
oo o c
CN
O
0 —
0
Tf "3-
>
O
J
--
.a
(0
*
G
.
B
ca
cn
a,
O
U
cn
ai,
cl
O
CJ
,'5t*
CL
o
CJ
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
37
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN C H RO M lU M (V t) SPECIES ANALYSIS
concentration o f Cr(VI) (larger than 1300 ug/g); COPR 4 contained about
400 fJ.g/g Cr(VI); and COPR 3 had the lowest concentrations o f Cr(VI) (less
than 100 pg/g). Although the heterogeneity o f these samples probably
resulted in the decreased precision, these two detection results were
comparable for all the three COPR extracts (pH—7.5). implying that Method
7196A did produce a suitable detection method o f Cr(VI) for such samples as
these COPRs.
3.2.3 Lower Recoveries of Cr(VT) for Some Sand and Soil
As the results for the above three COPR samples demonstrated, the
inexpensive Method 7196A was a good detection method for some samples.
However, it caused low recoveries o f Cr(VI) for some sand and soil samples
compared with SIDMS (Table 3).
All the sand and soil samples were extracted using Method 3060A.
After filtration, the extracts (pH~12) were spiked with different quantities of
Cr(VI) standard solution. These spiked solutions were used as sample
solutions for further analysis with both Methods 7196A and SIDMS. A
portion o f the spiked extract (p H -12) before neutralization was sampled for
SIDMS analysis, and the rest of the solution was neutralized to pH 7.5 with
concentrated nitric acid. The neutralized extracts were then analyzed with
both Methods. All the measurements were performed immediately after the
neutralization.
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
38
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
For Method 7196A. it achieved approximately 100% recoveries o f
Cr( VI) for the blank solution, but obtained low recoveries o f Cr( VI) for the
sand and soil extracts at pH 7.5. For SIDMS detection, it achieved
approximately 100% recoveries o f Cr(VI) for all the blanks with different pH
and the sand and soil extracts at pH 12; however, the recoveries o f Cr(VI) for
the sand and soil extracts at pH 7.5 were approximately 2-4 % lower than
those for the extracts at pH 12.
As the results showed. lower recoveries o f Cr(VI) could be caused by
using Method 7196A if the sample matrix was present. In addition, the
reduction of Cr(VI) could take place during the neutralization.
In our experiments, the ratios o f soil matrix (g) to Cr(VI) added (ug/'g)
were approximately 2. I. 0.5. and 0 for soil samples I. 2. 3. and blank,
respectively; and the recoveries o f Cr(VI) by Method 7196A were 71.6%.
81.9%. 91.8%. and 101%. respectively. As the disparity o f the recovery
indicated, increasing the ratio of soil matrix to Cr(VI) resulted in lower
recoveries of Cr(VI) in Method 7196A. In other words, more concentrated
matrix could cause lower recoveries o f Cr(VI) in Method 7196A. However,
the concentration and the complexity of the soil matrix did not affect the
accuracy of Method SIDMS for pH 12 extracts. In addition, the ratio of
matrix (g) to Cr(VI) (pg/g) for the sand sample was approximately 2.5, which
was larger than 2.0, the ratio for soil sample 1; if the matrix effects for these
two samples were the same, lower recoveries o f Cr(VI) for the sand sample
Yusheng Lu 1:14 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
39
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN C H RO M IU M (V I) SPECIES ANALYSIS
3
t/J
C/3
o
in
C/3
■q
CD
L—
-H
C
ed
<D
P
m
O
o '
<
+1
+1 +1 +1
2
-C
/-S
in
in
_
p
in
On
—;
O
o
«0 o
vo c—
C n On ’ 1
X
X
> 3
LU
c
CD
T3
£Z
CD
CO
*5
GO
C/3
m
—*
+1
-H
5 =
Q (D
00 sS
n
nT
o
CN
ON
m
o'
+1
m
CN
CN
m
o'
+1
m
CN
+1
VO
i- '
—
—
+1
CN
,__i
OC
CN
in
CN
CN
_
o
"■“
m
o ’
+1
o
o
cn
o ‘
+1
—
+1
oc
,_'
CN
o
+1
CN
O
cn
p
cn
o
O
cC
o
>
o
T3
CD
‘CL
CO
n
tin
>
-H
1
>
3
co
C
ed
<D
£
N°
>—1
in i
!+ i
,GN
is d
IOO
0
‘l .
0
>
O
o
0
ct
,--- V “O
<D
u.
u
OP 1CN
"co IO
CL
ZL
cn
p
cn
CO w
CO
_©
n
CD
I-
_
co
O s—'
l—
cn
cn
ed
<D
CL
£
cd
C/3
—
CL
£cd
CO
IO
Im
j CN
!* c
• c
1 ed
1m
CN
m”
‘o
V3
VO
O
en
m
in
CN
—
‘o
c/ j
m
—
’o
Vi
O
M
C
cd
X)
c standard deviation; d not analyzed.
T3
a SIDMS results of pH 12 filtrates; b SIDMS results of pi 17.5 neutralized extract.
o
CD
Yusheng Lu 2:44 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
40
PA RT I IDENTIFICATION O F STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES AN ALYSIS
was expected. In contrast to this expectation, 86.9% recoveries of Cr(VI)
were obtained for the sand samples, which was higher than that for soil
sample I (71.6%). This result implied that the sand matrix had less effect on
Cr(VI) recoveries than this soil matrix.
The detection limit o f Method 7196A for solutions without matrix was
0 .0 1 ug/g. Based on the above discussion, limited Cr(VI) can be completely
reduced by sample matrix during the measurement in some cases, and the
detection limit for real soil samples would be larger than 0.01 ug/g due to the
negative effect o f matrix on Cr( VI) under the detection condition. For
example. 0.03 p-g/g was the detection limit o f Cr(VI) in soil extracts (20). In
these cases, Method 7196A will not be an accurate method.
The complicated soil matrix contains large quantities o f organic and
inorganic compounds. Some o f the individual soil matrix components were
evaluated in the following sections to study the bias sources in Method
7196 A.
3.2.4 Bias Sources in Method 7196A
In our following experiments, individual soil matrix component were
tested to evaluate their effects on Cr(VI) detection. Adding an individual soil
matrix component and the standard Cr(VI) solution to a volumetric flask to
make a synthesized sample, and the Cr(VI) in this synthesized sample was
detected with only Method 7196A. As the experiments indicated, some soil
Yusheng Lu 2:44 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
41
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHROM IUM (VU SPECIES ANALYSIS
organic compounds, such reducing agents as Fe:" and some sulfides, and the
oxidant Fe3* shown in F igure 6 could cause negative effect on Cr(VI).
3.2.4.1 Effects o f Organic Cheraicafe
The extraction solution in Method 3060A contains a large quantity of
NaOH which can extract such organic acids as fulvic acids and humic acids
from soil samples. In some cases, such fulvic acids in soil samples as acetic
and citric acids could reduce Cr(VI) to Cr(III) (30). However, when the
concentration o f Cr(VI) in each synthesized sample was 0.50 ug/g, 8 x 10° M
o f acetic, 1.0
x
10'3 M o f oxalic. 1.0 x 10" M of citric. 2.0
x
10'’ M of
salicylic, 2.0x10'3 M o f maleic and l.OxlO"1 M of benzoic acids were not
observed to interfere with the detection o f Cr(VI). In the experiments, only
phthalic acid (2.0x10'3 M), tartaric acid (2.0x10'3 M). and phenol (4.0x10"
M) slightly decreased the recoveries o f Cr(VI) by 3.5%. 3.4%. and 2.6%.
respectively.
The number and the concentrations o f organic chemicals in soils vary
with different soil types. The concentrations o f most organic compounds
used in the experiments were in the typical ranges o f corresponding
components in soil solution, which is defined as the aqueous liquid phase at
field moisture contents (32). Among the tested organic chemicals at the
specified concentrations, we only found three organic chemicals, phthalic
acid, tartaric acid, and phenol (33) could decrease the recoveries o f Cr(VI).
Yusheng Lu 2:44 PM 07/29/98
perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
42
PART 1 IDENTIFICATION OF STANDARD METHOD BIASES IN CHROMIUM (VI) SPECIES ANALYSIS
5237
m
<P
s
70
Soil Matrix C om pon en t
F ig u re 6. Effect of s o m e soil matrix co m p o n en ts on Cr(VI) reco v eries
The error bars show the standard deviation of Cr(VI) recoveries for
several different synthesized solutions. The concentration of Cr(VI) was 0.5
pg/g. The concentrations o f these soil components were: (1) 2.0 x 10'J M
phthalic acid; (2) 2.0 x 10'3 M tartaric acid; (3) 4.0 x 10"* M phenol;
(4) 67.2 pg/g S2*; (5) 131.2 pg/g S2'; (6) 0.4 pg/g Fe2*;
(7) 1.6 pg/g Fe2*; (8) 2.0 pg/g Fe3*; (9) 8.0 ug/g Fe3*.
Yusheng Lu 2:44 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
4j
PA RT I IDENTIFICATION OF STANDARD METHOD BIASES IN CHRO M IUM (VI) SPECIES ANALYSIS
In the alkaline soil extracts, however, the concentrations o f these organic
compounds, especially water-insoluble humic acids may be much higher than
those in the soil solution, so those chemicals which did not show effect on
Cr(VI) at relatively low concentrations used in our experiments may cause
biases o f Cr(VI) when their concentrations are higher. In addition, many
other untested organic chemicals that might be present in soils may also cause
analytical biases in Method 7196A.
3.2.4.2 Effects o f Fe2'a n d S2'
As the experiments on such reducing agents as Fe2' and S2' indicated,
these agents decreased the recoveries o f Cr( VI') significantly (Figure 6).
The quantitative redox reaction between Fe2' and Cr(VI) has been used
to detect Fe and Cr(VI) (25. 34). In the following experiments, we will study
and discuss the influence of the reductant Fe2' on the quantification o f Cr(VI)
in Method 7196A.
Fe2' could reduce Cr(VI). The standard redox potential o f Fe'VFe2' in
acidic solution is 0.77 eV; the potential o f Fe37F e2' decreases with the
increase in Fe2' and the decrease in Fe3*. W hen the concentrations o f Fe2' and
Fe3' are the same, the redox potential is equal to the standard potential. 0.77
eV. On the other hand, the redox potential o f Cr20 - 2'/Cr3' can be calculated
as 1.05 eV at pH 2 solutions if other conditions are the same. It decreases
with the increase in Cr3' and the decrease in Cr20 72'. In the very beginning o f
Yusheng Lu 2:44 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
44
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN CHRO M IUM fVI) SPECIES ANALYSIS
the experiment, when the standard Cr(VI) and Fe2" solution were mixed, the
concentrations o f Fe3" and Cr3' were zero, the potential o f FeJ7Fe2"was very
low. while potential o f Cr:0 - 2/Cr3" was very high. Therefore, the reduction
o f Cr(VI) by Fe2" must be taken place at that time.
Another inorganic reducing agent. S2' may also reduce Cr(VI) during
either the neutralization or the detection by Method 7196A. Assuming the
pH of the neutralized extract is 7. the redox potential o f Cr.Oy VCr3" in the
solution will be 0.37 eV if other conditions are unchanged: while the standard
redox potential o f S/S2' in neutral solutions is -0.508 eV. As these data
indicate, Cr(VI) may be reduced by S2' during the neutralization and storage.
In acid solutions, the redox potential of S/S2' is 0.141 eV. which indicates that
S2' is still a very good reductant. The redox potential o f Cr20 7'VCr" is 1.05
eV at pH 2 solutions if other conditions are maintained at the same. Thus, the
strong oxidant. Cr(VI) in the pH 2 detection solution may be easily reduced
by S2'.
In addition, doubling the concentration o f S2' to 131.2 pg/g caused the
decrease in the recoveries o f Cr(VI): from 93.2 to 82.1%: while doubling Fe'*
to 1.6 pg/g decreased the recoveries from 95.6 to 84.1%. Increasing the ratios
of these ions to Cr(VI) decreased the recoveries o f Cr(VI), which matched the
results discussed in the last section: more matrix caused lower recovery of
Cr(VT).
Yusheng Lu 2:44 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
45
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
The concentrations of free Fe:* and Fe3*" ions are very low in the
alkaline extraction solution because of the precipitation o f Fe(OH), (K^=l .64
x 10‘l4at 18°C) and Fe(OH)3 (Ksp= l . 1 x 10'36at 18°C). However, this does
not mean that the total concentration o f iron is low since Fe m ay exist as
complexes that are soluble in the alkaline extraction solution and remain in
the extracts after filtration. For example. AA results show that 2.7 ug/g o f
total iron is present in the filtered soil extract (in Section 3.2.3), which was
much higher than 9.2 x IO41 pg/g calculated using the K_p o f Fe(OH):. Soil
samples contain numerous types o f organic acids that can combine with
Fe(II) or Fe(III). As the pH decreases. H* may associate w ith some of the
anions o f organic acids in the extract replacing Fe(II) or Fe(III) in their
complexes and releasing Fe2* or Fe'-* to the solution. These free Fe2* and Fe'*
in solutions may interfere with Cr(VI) detection as we have seen above and
will see in the next section.
3.2.4.3 Effect o f Fe3*
Interestingly, we observed that FeJ* also significantly reduced the
recoveries of Cr(VI) (Figure 6). The concentration o f Cr(VI) in each sample
was 0.50 pg/g. Increasing the concentration o f FeJ~decreases the recovery o f
Cr(VI). For example, recoveries o f 93.1% and 66.9% were obtained when
the concentration o f Fe3* was 2.0 pg/g and 8.0 pg/g, respectively.
Yusheng Lu 2:44 PM 07/29/98
perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
46
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN CH RO M IU M (V I) SPECIES ANALYSIS
The interference o f FeJ* with the measurement o f Cr(VI) in Method
7196A was thought to be resulted from the spectral overlap: Fe3'- combined
with DPC to form a colored complex (10). Such spectral overlap should
cause positive errors rather than negative errors, which we observed in our
experiment. To understand the effect of Fe3\ we need to study the
mechanism o f Method 7196A described in Section 3.2.1. As we have seen,
the color development reaction includes two steps: Cr(VI) oxidizes DPC to
DPCO. then the produced Cr(III) combines with DPCO to form the
detectable red-violet complex Cr(III)-DPCO at pH 2. These redox and
complexation reactions take place in-situ. Reagents that can interfere with
either step may cause interferences. For the first step, either oxidizing agents
or reducing agents may cause negative errors of Cr(VI).
In other words,
oxidizing agents may compete with Cr(VI) to react with DPC: while reducing
agents may compete with DPC to react with Cr(Vl). Both agents may cause
low recoveries o f Cr(VI). As an oxidant. FeJ~ may oxidize DPC resulting in
negative errors. To verify this hypothesis, we designed the following
experiments.
Reagent phenanthroline (phen) was introduced to our experiments. The
complexation o f Fe2‘ and phen develops an orange colored complex
Fe(phen)32‘ with the maximum absorption at 510 nm (25) in the pH 2
detection condition o f Cr(VI).
Yusheng Lu 2:44 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
47
PA RT I IDENTIFICATION OF STANDARD METHOD BIASES IN CHRO M IUM (VI) SPECIES AN ALYSIS
Fe2* + 3 phen —> Fe(Phen)32*
At pH 2. no color changes was observed when Fe2* was mixed with
DPC, or Fe3* was mixed with phen (the complex Fe(phen)33* may be formed
in the solution). However, when FeJ\ phen, and DPC were mixed together
under the same condition, an orange chemical which had the specific Fe(II)phen absorption spectrum was quickly produced. As these results indicated, a
redox reaction took place in this mixture between Fe(III) and DPC. Fe(III)
oxidized some DPC. then the formed Fe(II) combined with phen to produce
the orange Fe(II)-phen complex. After 15 minutes, the absorbance o f Fe(II)phen at 510 nm was increased significantly, implying more Fe(II) produced
from Fe(III) (F igure 7). In this mixture solution. Fe(phen);,‘* may be
produced by two paths:
Fe3* - r DPC -► Fe2*
DPCO
3 phen
1
Fe(phen)32'
or
Fefphen)/* + DPC —> Fe(phen)32* + DPCO
Fe(III) may be more easily reduced to Fe(II) in the presence o f phen
because the redox potential o f Fe(phen)337Fe(phen)32* in 1M sulfuric acid.
1.056 eV is higher than 0.77 eV, the standard reduction potential o f FeJ7 F e_*.
Therefore, there are two possibilities for the reduction o f Fe(III) in the
mixture solution o f Fe3*, DPC, and phen. First, only the second path is
involved in the solution: the complex o f Fe(phen)33* may be quickly formed
YushengLu 2:44 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
48
Q16
QOB
QOB
004
Q04
380
430
480
530
500
v«elerTjh(m)
630
680
730
38D
Q16
Q04
430
48D
530
5BD63068D7D
W ute$h(m )
380430480530980630680730
V\Men^h(m)
Figure 7. (a) S p ectru m o f s p e c ific F e(ll)-phen s y s te m a t pH2 (A.m= 508nm ). T his s y s te m co n ta in ed 2 .(^ g /g Fe(ll);
(b) Sp ectru m o f Fe(lll), DPC, and p h en s y s te m a t pH2 (Xm= 508nm ). In th e m ixture s o lu tio n 4 .0 |ig /g Fe(lll) w a s
ad d ed and 0.84|ag/g Fe(ll) w a s p ro d u ced by th e c a lcu la tio n b a s e d o n calibration c u rv e o f Fe(l!)-phen reaction ;
an d
(c) S p ectru m o f (b) in 15 m in u te s later, w h e n 1 .3 0 |ig /g Fe(ll) had b e e n p ro d u ced .
The s h a p e s o f t h e s e three sp e c tr a w ere e x a c tly th e s a m e . T he r ea g e n t DPC its e lf had a very sm a ll a b s o r b a n c e
w h ich c o u ld b e ign ored in th e Fe(ll) d eterm in ation .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Q16
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
in the mixture, and this complex may react with DPC to form the orange
complex Fe(phen)32*. Second, both paths are involved: the complex
Fe(phen)33* and free Fe3‘ may oxidize DPC simultaneously to form the final
product, complex Fe(phen)32\ As the negative effect o f Fe3" on Cr(VI)
recoveries in Figure 6 (those solutions did not contain phen) implied, free
Fe3* could oxidize DPC in pH 2 solutions resulting in low recoveries o f
Cr(VI).
Another experiment supplied further evidence to prove the oxidation o f
DPC by Fe3*.
A sub-boiling distilled nitric acid with a brown color instead of the
regular concentrated nitric acid was used for neutralizing a blank extraction
solution to pH 7.5. After DPC was added in the neutralized solution, the
solution was adjusted to pH 2 by 10% sulfuric acid. A yellow chemical was
produced in the solution immediately. Then the standard Cr(VI) solution was
added to this yellow solution, but no expected Cr(III)-DPCO complex was
observed. After several more portions o f DPC were added, this complex
started to be produced. The yellow chemical was the destroyed DPC because
they did not react with Cr(VI) to form Cr(III)-DPCO complex.
However, when the regular concentrated nitric acid and sulfuric acid
were used for neutralizing the blank extraction solution, none o f them caused
the DPC solution yellow; the addition o f standard Cr(VI) to this DPC
solution produced Cr(III)-DPCO complex immediately as expected. The
Yusheng Lu 2:44 PM 07/29/98
o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
50
PART I IDENTIFICATION OF STAN DARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
above phenomena indicated that the sub-distilled nitric acid could oxidize
DPC in pH 2 solutions.
Fe3" could oxidize DPC as the sub-distilled nitric acid did. The
following procedure about Fe3" was similar to the above process.
A blank extraction solution was neutralized to pH 7.5 with the regular
concentrated nitric acid because it did not affect the detection o f Cr(VI).
Some concentrated Fe3" solution was added to the extract as the only matrix
(the concentration o f Fe3" was about 40 ppm). A portion o f DPC was added
to the neutralized extract, the solution was then acidified to pH 2 with 10%
sulfuric acid. During this pH adjustment, the colorless solution turned to
yellow. After that, a standard Cr(VI) solution was added to this yellow
solution; but no expected red-violet Cr(III)-DPCO complex was produced,
and no color change was observed. Adding two more portions of DPC, the
solution turned more yellow; when another two portions o f DPC were added,
the Cr(III)-DPCO complex started to occur. As the experiment indicated, the
first four portions o f DPC were completely destroyed.
The destroyed yellow DPC should be the oxidized DPC by Fe3' in this
case. Further experiments demonstrated that K M n04 could oxidize DPC as
well as Fe3'and the sub-boiling distilled nitric acid.
Spectral data provided further evidence on the oxidation o f DPC. We
prepared several solutions in these experiments, each o f them was adjusted to
pH 2 with 10% sulfuric acid before the UV-Vis spectral scan. The first
Yusheng Lu 2:44 PM 07/29/98
perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
51
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CH RO M IU M fV I) SPECIES ANALYSIS
solution was made by neutralizing a blank extraction solution with sub­
distilled nitric acid followed by adding DPC. The second solution which
contained 10 ppm Fe3* was prepared by neutralizing a blank extraction
solution with the regular concentrated nitric acid followed by adding Fe3* and
DPC. From 310 nm to 700 nm, these two solutions had alm ost the same UVVis spectrum which was the curve (a) in F ig u re 8. The third solution was
made by neutralizing a blank extraction solution with the regular
concentrated nitric acid followed by adding only DPC. This solution had
lower absorbance in this range (curve (b) in Figure 8). As the spectra
indicated, this quantity of Fe3' completely oxidized the DPC. as did the sub­
boiling distilled nitric acid. Since 10 ppm Fe3' had limited absorption in this
range which could be ignored, the spectrum o f the oxidized DPC could be
obtained by deducting curve (b) from curve (a). This spectrum with the
maximum absorbance at 340 nm did not interfere much with the detection of
Cr(VI) due to the large distance between 340 nm and 542 nm (the maximum
absorption wavelength of Cr(III)-DPCO). Therefore, the effect of Fe3' on
Cr(VI) detection was dominantly contributed by the oxidized DPC, not by the
spectral overlap. The chemicals tested in our experiments are only a part of
the components existing in environmental samples. Som e other organic
compounds, reducing agents, and oxidizing agents might influence on Cr(VI)
detection in Method 7196 A. In addition, higher concentrations of the
chemicals which effect on Cr(VI) was not observed in our experiments at
Y ushengL u 2:44 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
52
I3-
<D
a
<Q
r~
c
NJ
•fe.
■fe
a>
o
c
(0
ni_
o
U)
n
<
OF STANDARD
D
§
O
2 .5 -
PART I IDENTIFICATION
£
00
0.5-
METHOD
310 330 350 370 390 410 430 450 470 490 510
wavelength(nm)
F igure 8. (a) S p ectru m o f y e llo w s o lu tio n co n ta in in g DPC, c o n c en tr a ted Fe(lll), and blank ex tra ct neutralized w ith
regu lar c o n c en tr a ted nitric acid , o r th e sp e c tr u m o f s o lu tio n c o n ta in in g DPC and blank ex tra ct n eutralized with
su b -b o ilin g d istilled nitric acid; (b) S p ectru m o f th e c lea r s o lu tio n co n ta in in g DPC and blank e x tr a ct neutralized
w ith th e regular c o n c en tr a ted nitric a cid a n d , very sim ilar to th e sp ectru m o f so lu tio n c o n ta in in g on ly th e blank
ex tra ct neutralized w ith th e regular nitric a cid or su b -b o ilin g d istilled nitric acid; and (c) S p ectru m o f th e o x id iz e d
DPC with th e m axim um a b so rp tio n at a b o u t 340 nm . S p ectru m (c) w a s o b ta in ed by th e d ed u ctio n (b) (torn (a),
b e c a u s e both ty p e s o f nitric a c id s had a lm o s t th e s a m e sp e ctru m in th is w a v e len g th range, and th e a b so rp tio n if
DPC w a s very little w h ich c o u ld b e ig n o red in th is w a v e le n g th range.
lOppm Fe(III) solution was slightly yellow colored with limited absorption. Therefore, Fe(Ill) and sub-boiling distilled nitric
acid had the very sim ilar effect on DPC.
BIASES IN CHROMIUM(VI) SPECIES ANALYSIS
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 . 5-1
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN CHRO M IUM (VI) SPECIES ANALYSIS
those specific concentrations might also cause interference with Cr(VI). This
trend demonstrates several plausible general mechanisms o f biases in Method
7196 A.
3.2.5
Method o f Standard Addition
When the inexpensive UV-Vis detection method has to be applied to
determine Cr(VI) in real samples, standard addition method o f 7196A is
probably suitable to increase the accuracy.
Additional study on Method 7196A showed that the absorbance o f
complex Cr(III)-DPCO slightly varied with pH value in the modified specific
range (pH 1.6-2.2). which could also cause analytical biases o f Cr(VI).
Standard addition was applied in our experiments trying to get more
accurate results. The same soil extracts used in Section 3.2.3 were detected
by standard addition with UV-Vis measurement. Data in the table 4 showed
that better recoveries o f Cr(VI) were obtained compared with the direct
Method 7196A. However, different from the SIDMS method, this “standard
addition” could only partially correct the biases.
Based on the relationship between the interference and the ratio o f
matrix to Cr(VI) discussed in Section 3.2.3. when the standard addition
method was performed to those naturally spiked Cr(VI) soil samples, extra
standard Cr(VI) solution was added to the soil matrix for the UV-Vis
detection o f the naturally spiked Cr(VI). Less ratio o f matrix to Cr(VI) was
Yusheng Lu 2:44 PM 07/29/98
perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
54
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
then obtained in the detection solution, and higher recovery of Cr(VI) was
expected for each measurement in Method 7196A. The more Cr(VI) was
added during the "standard addition", the higher recovery of Cr(VI) could be
measured; and the obtained working curve should be non-linear. However, a
line with a higher slope was plotted for the calculation, which caused lower
than 100% recoveries o f the spiked Cr(VI) in the soil sample.
3.2.6 Cr(VI) Wastes and Ascorbic Acid
Because o f the toxicity o f Cr(VI). Cr(VI) wastes should be properly dealt
with after experiments. Our experiments indicated that ascorbic acid
(Vitamin C) can completely reduce the toxic Cr(VI) to much less toxic Cr(III)
in acidic solutions. The same procedure as that in Section 3.2.4 was
performed to ascorbic acid. When the concentration o f Cr(VI) was 0.5 p.g/g.
and the concentrations o f ascorbic acid were 4.0
x
10"° M. 1.0 x 10° M. and
5.0 x 10'5 M. respectively, the corresponding recoveries o f Cr(VI) in these
cases were (93.6 ± 3.5)%. (84.7 ± 2.2)%, and (53.0 ± 0.9)%. As the data
indicated, ascorbic acid could greatly reduce Cr(VI) in pH 2 solutions. Large
quantity o f ascorbic acid in low pH solutions could completely reduce Cr(VI)
in its wastes. Therefore, ascorbic acid was used in our lab to reduce Cr(VI) in
Cr(VI) wastes and decrease the toxicity o f the wastes.
Yusheng Lu 2:44 PM 07/29/98
perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
55
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
3.2.7
Suggestion
SIDMS is a method which can be used to improve the accuracy in the
detection o f Cr(VI). However, when Method 7196A has to be applied in the
detection o f Cr(VI) in a matrix with much oxidizing agents ( for example,
Fe3"). a larger size o f reagent DPC is suggested. Due to the special
mechanism on the color-development of complex Cr(III)-DPCO, some
oxidizing agents may consume DPC and result in low recoveries o f Cr(VI).
More DPC can decrease such interference with Cr(VI). Additionally,
standard addition may be helpful for increasing the accuracy. In our
experiments, small quantities of extra Cr(VI) were added to detect the
naturally spiked Cr( VI) soil sample, and higher recoveries o f Cr( VI) were
achieved with the standard addition method.
Yusheng Lu 2:44 PM 07/29/98
perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
56
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CH RO M IU M (V I) SPECIES ANALYSIS
4.
SUM MARY
Method biases exist in the extraction, neutralization and measurement
steps in the analysis of Cr(VI) in environmental soil samples using the EPA
method pairs 3060A/7196A. The oxidation o f Cr(III) to Cr(VI) may occns
during the extraction resulting in positive errors of Cr(VI), due to the
oxidizing agents in the samples or the oxygen dissolved in the extraction
solution. The loss o f water-insoluble P bC r04 may occur during
neutralization resulting in negative errors due to the reprecipitation o f
PbC r04. The reduction o f Cr( VI) to Cr(VI) may take place during the
neutralization and the UV-Vis measurement resulting in negative errors.
With the decrease in pH, the formal redox potential o f Cr( VI)/Cr(III)
increases, and Cr(VI) could be reduced to Cr(III) by reducing agents during
the neutralization and detection. Besides such reducing agents as some
organic compounds, inorganic reducing agents Fe2' and S2\ such oxidizing
agent as Fe3' could also cause negative errors during detection due to a
different mechanism. F e" could compete with Cr(VI) to oxidize the reagent
DPC, and the decrease in DPC caused low recoveries o f Cr(VI).
To identify some o f these errors, Speciated Isotope Dilution Mass
Spectrometry (SIDMS), a diagnostic tool, was applied in such real samples as
sand, soil, and COPR samples.
Yusheng Lu 2:44 PM 07/29/98
perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
57
PART I IDENTIFICATION O F STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
5.
(1)
REFERENCES
Weckhuysen. B. M ; Wachs. I. E.; Schoonheydt, R. A. Chem. Rev.
1996, 3327-334*.
(2)
Burrows. D. Chromium: Metabolism and Toxicity:; CRC Press, Inc.:
Boca Raton, 1983.
(3)
Kong. Q. A.: Wu. Q. F.; Guo. X. H. Fenxi Huaxue 1996, 24. 1-5.
(4)
Zhang, H. S.: Yang, X. C.; Wu. L. P. Fenxi Huaxue 1995, 23, 11481150.
(5)
Kingston. H. M ; Huo. D.: Lu. Y. Spectrochimica Acta Part B 1998, 53.
299.
(6)
Grate. J. W.: Taylor. R. H. Field Anal Chem Technol 1996, /. 39-48.
(7)
Flores Velez. L. M.; Gutierrez Ruiz. M. E.; Reyes Salas. O.; Cram
Hevdrich. S.: Baeza Reyes. A. Int J Environ Anal Chem 1996, 61. 177187.
(8)
Harrington, C. F.; Fairman. B. E.; Catterick, T. Spectrosc Eur. Jan Feb
1997; 9(1): 10 1997, 14. 16.
(9)
Lam, J. W. H.; McLaren, J. W.; Methven, B. A. J. J Anal At Spectrom.
Aug 1995, 10. 551-554.
Yusheng Lu 2:44 PM 07/29/98
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
58
PART I IDENTIFICATION O F STANDARD M ETHOD BIASES IN C H R O M IU M (V I) SPEC IES ANALYSIS
(10) Brvson. W. G.; Goodall. C. M. Analytical Chimica Acta 1981, 124,
391-401.
(11) Granadillo, V. A.; Parra de Machado, L.; Romero, R. A. Anal Chem
1994, 66, 3624-3631.
(12) James, B. R. Envir. Sci. & Tech. 1996, 30, 248.
(13) EPA SW-846 Method 3060: Alkaline Digestion o f Hexavalent
Chromium, 3rd ed.; U.S. Environmental Protection Agency:
Washington, DC. 1995.
(14) James, B. R.: Petura, J. C.; Vitale. R. J.; Mussoline, G. R.
Environmental Science & Technology 1995, 29, 2377-2381.
(15) Vitale, R. J. Am. Enivron. Lab. 1995, 7, 1.
(16) Broekaert. J. A. C. 1998 Winter Conference on Plasm Spectrochemistry
1998.
(17) Vitale, R. J. M., J. C. Peura, B. R. James Journal o f Environmental
quality 1994,23, 1249-1256.
(18) Vitale, R. J.; Mussoline, G. R.; Rinehimer, K. A.; Petura. J. C.; James,
B. R. Environ Sci Technol. Feb 1997, 31, 390-394.
(19) Eckert, J. M.; Jud, R. J.; Lay, P. A.; Symons, A. D. Anal. Chim. Acta.
1991, 255(7), 31-33.
(20) Milacic, R.; Stupar, J.; Kozuh, N.; Korosin, J. Analyst (London) 1992,
117, 125-130.
Yusheng Lu 2:44 PM 07/29/98
perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
59
PART [ IDENTIFICATION O F STANDARD M ETHOD BI.ASES IN CHROM IUM (VI) SPECIES .ANALYSIS
(21) Paniagua. A. R.; Vazquez. M. D.: Tascon. M. L.: Sanchez-Batanero, P.
Electroanalysis (N. Y) 1993, 5(2), 155-163.
(22) Tian, S.; Schwedt, G. Fresenius' J Anal Chem. Feb 1996, 354, 447-450.
(23) Inoue, Y.; Sakai. T.: Kumagai, H. J. Chromatogr.. A 1995, ~06, 127136.
(24) Byrdy, F. A.; Olson. L. K.; Vela. N. P.: Caruso. J. A. J. Chromatogr.
1995, 7/2. 311-320.
(25) Teshima. N\: Ayukawa. K.: Kawashima. T. Talanta 1996, 43. 17551760.
(26) Arya. S. P.; Mahajan, M. Chem. Pharm. Bull. 1996, 44. 1561-1563.
(27) Kingston. H. M. Method ofSpeciated Isotope Dilution Mass
Spectrometry. US Patent Number: 5.414.259. 1995
(28) Kingston. H. M.; Huo. D.; Lu. Y. Applied Spectroscopy 1998 (in
review).
(29) EPA SW-846 Method 6800: Elemental and Speciated Isotope Dilution
Mass Spectrometry, Test Method fo r Evaluating Solid Waste, 4: U.S.
Environmental Protection Agency: W ashington D.C.. 1998.
(30) Nriagu. J. O.; Nieboer, E. Chromium in the Natural and Human
Environments', John Wiley & Sons, 1988.
(31) Dionex In Dionex Ion Chromatography Recipe Book', Dionex
Corporation: Sunnyvale, CA, 1990; Vol. Technical Note 26, pp 7.
Yusheng Lu 2:44 PM 07/29/98
perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
60
PART I IDENTIFICATION OF STANDARD METHOD BIASES IN CHROM IUM (VI) SPECIES ANALYSIS
(32) Wolt, J. D. Soil Solution Chemistry>—Application to Environmental
Science and Agricultures John Wiley & Sons, 1994.
(33) Elovitz, M. S.; Fish, W. Environ. Sci. Technol. 1994, 28, 2161-2169.
(34) Itabashi. H.: Teshima, N.; Kawashima, T. Anal. Sci. 1995, II, 693-694.
Yusheng Lu 2:44 PM 07/29/98
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
61
Part*
PART II
MICROWAVE IMPROVEMENT IN EXTRACTION
EFFICIENCY
1. INTRODUCTION............................................................................................... 63
1.1 Applications o f Microwave Assisted Sample Preparation..................... 63
1.2 Microwave Assisted Closed Vessel System.............................................. 64
1.3 Microwave Heating Mechanism.................................................................. 65
1.4 Microwave Instrumentation.......................................................................... 72
1.5 Research Goal................................................................................................. 74
2. EXPERIMENTAL SECTIO N ........................................................................... 78
2.1 Reagents........................................................................................................... 78
2.2 Equipment........................................................................................................ 79
3. RESULTS AND DISCUSSION........................................................................83
3.1 Extraction o f Cr(VF) S olids.......................................................................... 83
3.2 Oxidation o f Crflll) During Microwave Extraction.................................85
3.2.1 Oxidation o f Cr(III) at 90-95°C ............................................................85
3.2.2 Oxidation o f Cr(III) at 120°C and 150°C.......................................... 88
3.2.3 Quantities o f Cr(VI) Oxidized from Soluble C r(III).........................90
3.3 Extraction o f P bC r04 at 150°C.................................................................... 92
3.4 Bias from Neutralization Process................................................................ 94
3.5 Future Study and Suggestions......................................................................97
4. SUM MARY.......................................................................................................... 99
5. REFERENCES...................................................................................................100
62
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
1.
IN TR O D U C TfO H
I. I Applications o f M icrowave Assisted Sample Preparation
Most measurement methods are performed to analytes o f interest i s
solutions, thus sample preparation is very important for sample analysis.
Inefficient sample preparation methods may increase errors and costs, and
slow down the whole sample analysis. Conventional sample preparation
usually requires hours or days to prepare samples for analysis, which
becomes the most time-consuming step in the entire instrumental analysis (1)
because o f the development o f computers and instruments: while the
microwave assisted digestion generally needs 10 minutes. This method is not
only very rapid, safe, and convenient, but also precise due to the decrease in
the loss o f the analyzed elements and in contamination (2).
The earliest acid microwave digestion occurred in I970's using a
domestic microwave oven to heat acids and samples in open beakers (3).
Since then, microwave sample preparation has been applied for over two
decades. The numerous applications o f microwave in sample preparation
include ashing (3), drying (4), leaching (5, 6). digestion (7-10), and extraction
(11-13). The new application o f microwave in chemistry is on the
microwave-assisted reactions including inorganic reactions (14. 15) and
organic reactions (16-19). Among all these applications in microwave
sample preparation, microwave assisted digestion and extraction are the two
main applications currently. Microwave assisted digestion has been
63
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
extensively used for the analysis of elements in such samples as geological
samples (20). sediments (7, 8). soils (21-23). mud (24). biological samples
including animal tissues (25-28) and botanical samples (29-31). and food (10,
32). Besides the applications for measuring elements in samples, microwave
assisted extraction has also been applied for the analysis o f organic
compounds or pesticides from soils and sediments (33-35).
As the development o f microwave assisted sample preparation, the
number of such regulations as RCRA ( the Resource Conservation and
Recovery Act) and CERCLA (the Comprehensive Environmental Response
Compensation and Liability Act) has significantly increased during the past
25 years due to the rapidly growing environmental pollution. Because o f the
advantages o f microwave sample preparation, many methods have been
developed for microwave assisted environmental sample preparation. Such
EPA (Environmental Protection Agency) sample preparation methods as
3015. 3051. and 3052 are microwave assisted acid digestion methods o f some
metal elements in environmental samples.
1.2 Microwave Assisted Closed Vessel System
Microwave heating briefly consists o f two systems: open vessel and
closed vessel systems. The sample in an open vessel is not isolated from the
around environment, while the sample in a closed vessel can be prepared in
an isolated system with a different pressure and temperature. A microwave
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
Part [I
heating method is usually designed according to the properties and
concentration o f the analyte o f interest, the sample matrix, and the detection
method utilized for the analyte. Compared with conventional methods, both
open vessel and closed vessel microwave system can reduce analytical errors
and sample preparation time. However, closed vessel systems, in general,
have a number o f advantages over open vessel systems.
•
Closed vessel sample preparation can achieve higher temperature. A high
pressure in a closed vessel can be produced, and the boiling point o f the
liquid in the vessel is raised as the pressure. Because o f this, the time
required for the sample preparation can be greatly reduced.
•
The loss o f volatile elements during the sample preparation is virtually
eliminated in a closed vessel because there is little vapor loss.
•
The solution required for the sample preparation can be less due to the
reduced vapor loss in a closed vessel.
•
The influence from the environment during the sample preparation can be
substantially reduced.
1.3 Microwave Heating Mechanism
Microwave heating involves direct absorption o f microwave energy
by the substance to be heated. In the electromagnetic spectrum, the
microwave region lies between infra-red (IR) radiation and radio frequencies
with corresponding wavelengths from several centimeter to one meter. And
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Pan II
the microwave energy has a frequency range from 300 to 300,000 MHz
(Figure 1). By convention and treaty, four frequencies are used in scientific
and industrial microwave applications: 915±25, 2450±13, 5800±75, and
22,125±125 MHz. Among these frequencies, 2450 M Hz is most commonly
used for the industrial, medical, and scientific use (15. 16).
The energy o f microwave radiation at 2,450 M Hz is much lower than
that o f the UV and Visible radiation, and it is less than the chemical bond
energy. However, microwave radiation causes heating of solutions through
two mechanisms: molecular motion by migration o f ions and rotation o f
dipoles (15), but it does not directly cause changes in molecular structure.
The microwave heating efficiency of a sample partially depends on
the dissipation factor o f the sample. The dissipation factor (tan<5) is defined
as the ratio o f the sample's dielectric loss or "loss" factor (e” ) to its dielectric
constant (s ’). The dielectric constant measures a sam ple's ability to obstruct
the microwave energy as it passes through, and the loss factor describes the
sample's ability to dissipate that energy. Large dissipation factor o f a
material indicates better microwave absorption of the material. The
dissipation factors o f several typical laboratory materials is shown T able 1.
Water and sodium chloride solution have very large dissipation
factors, while the dissipation factors o f other materials in the table are very
small. As the data indicate, water and ionic solutions have much better
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
X-Rays
Microwaves I'd' rmSombvm
U ltrav io let
L a s e r R a d ia tio n
_ J _ -----------L _ _
io ° °
to -9
L
3x1Q1*
107®
to7'
3x1010
io 7 3 107 5 i o 7* t o 73
W a v e L e n g th (m e te rs )
3x10®
io *
10-’
_L
3x10®
3X104
3x10*
F r e q u e n c y (M Hz)
tA A A l
M o le c u la r
v ib ra tio n s
Inner-shell
electrons
(v a le n c e )
e le c tr o n s
M o le c u la r
r o ta tio n s
F ig u re 1 Electromagnetic Spectrum (15)
Part II
On
Part II
microwave absorption than other materials. Generally, there are three classes
microwave interactions: reflective, transparent, and absorptive. Metals reflect
Table 1 Dissipation factors o f some materials (3000 MHz at 25°C) (15, 16)
Material
water
Aqueous Sodium Chloride (0.1M)
Borosilicate Glass
Quartz Fused
Teflon™ PFA
Methanol
Ethanol
Ethylene glycol
tan5 (x 104)
1570
2400
12-75
0.6
1.5
6400
2500
10.000
microwave energy, which are not heated by microwave. Many materials such
as Borosilicate glass and Teflon™ PFA with small dissipation factors are
transparent to microwave energy, which can not be heated by microwaves
either. Some other materials between these two. such as water and sodium
chloride solution, absorb much microwave energy and are heated by
microwave. Similar to water and sodium chloride solution, some inorganic
acids and bases can be heated by migration o f ions or rotation o f dipoles or
both o f the mechanisms. The transparent materials to microwave are usually
selected as the vessel materials for quick heating and energy save, and the
absorptive material can be heated with the sample during microwave sample
preparation.
The typical time required to complete a wet digestion by conductive
heating is from 1 to 8 hours, or longer. However, an open vessel dissolution
68
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
by microwave heating can be completed in 5-15 minutes. The difference is
due to the different heating mechanisms. It takes time to heat the vessel and
transfer the heat to the solution by conductivity o f the vessel in conductive
heating, and a thermal gradient is established by convection currents ; on the
other hand, microwave heats the sample liquid of a typical analytical sample
size (10-50 ml) without heating the vessel directly.
Generally, microwave energy is lost to the sample by being dissipated
as heat by two mechanisms: ionic conduction and dipole rotation. Ionic
conduction is the conductive of dissolved ions in the applied electromagnetic
field. The flow of current caused by the ionic migration in the
electromagnetic field results in heat production due to the resistance to ion
flow. The microwave energy losses due to this ionic migration depend on the
conductivity including the concentration, mobility (size and charge) of the
dissolved ions in the medium and temperature. Dipole rotation refers to the
alignment o f molecules in the sample that have permanent or induced dipole
moments. As the microwave field increases, it aligns the polarized
molecules; as the field decreases, the reduced disorder is restored; as the
field is removed, thermal agitation returns the molecules to disorder, and
thermal energy is released during the relaxation time (Figure 2). At 2450
MHz, the alignment o f the molecules occurs 4.9x109 times per second, and
results in very rapid heating. The efficiency o f heating by dipole rotation
depends on the sample’s characteristic dielectric relaxation time. The
69
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
( 1)
H*
-
o-
/* *
/ H»
V
\
^
V/
■C
~
^
.
1
"x A f
*
~
1I
✓
-
F ig ure 2 One o f the microwave heating mechanisms: dipole rotation o f water molecuhw.
(1) polarized molecules aligned with the poles o f the electromagnetic field; (2) thermally
induced disorder as electromagnetic field is removed (15).
Part II
o
A
^
<
Part II
dielectric relaxation time is defined as the time that the molecules take to
achieve 63% o f their return to disorder (15). A material with less relaxation
time has a larger dissipation factor, and short heating time is needed. The
relaxation tim e depends on temperature and the viscosity o f the sample. At a
higher temperature, material molecules have longer relaxation tim e because
they are harder to be aligned and return to disorder due to the extra energy of
each molecule. The viscosity o f a sample affects its ability to absorb
microwave energy (dissipation factor) because it affects molecular rotation.
For example, increasing the temperature of water from 0 to 27 °C increases
the dissipation factor from 2.7x10"* to 12.2 at 2450 MHz (15). However, as
the water became more fluid, the dielectric relaxation time has a great impact
on the dissipation factor o f the water.
In general, temperature determines the relative contributions o f these
two energy conversion mechanisms. When an ionic sample is heated by
microwave, the dielectric loss to the sample is initially dominated by the
contribution o f dipole rotation; the dielectric loss is dominated by ionic
conduction as the temperature increases. The percent contributions o f these
two mechanisms depend on the mobility and concentration o f each ion, and
the relaxation time o f the sample. If the ion mobility and concentration o f the
sample ions are low, the sample heating is entirely dominated by dipole
rotation. On the other hand, as the mobility and concentration o f the sample
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
Part II
ions increase, it will be dominated by ionic conduction and the heating time
will be independent o f the relaxation time o f the solution.
Besides the conductivity o f the solution and the dielectric relaxation
time of the sample material, heating time also depends on the microwave
system designed and the sample size. The last two factors can be modified to
optimize the microwave sample preparation technique. Small sample size
causes a considerable amount o f energy to be unabsorbed. A relatively larger
sample size (50-200 ml) with a larger dissipation factor can be applied for
sample preparation with microwave procedure.
1.4 Microwave Instrumentation
Single-mode and multi-mode cavity type microwave instruments are
applied in microwave assisted sample preparation. A typical multi-mode
cavity type microwave instrument for analytical sample preparation consists
of six major components: magnetron (microwave generator), wave guide,
microwave cavity, mode stirrer, a circulator and a turntable (Figure 3 ). The
wave guide transports the microwave produced by the magnetron to the
cavity, and the mode stirrer distributes the incoming microwave in the cavity.
Vessels are placed on a turntable, which alternately rotates back and forth
180° to 360° in the cavity to expose the vessels to a uniform heating energy
field. And the circulator reflects the microwave to a dummy load to prevent
the magnetron from damage. The power output of the magnetron is either
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission
Figure 3 Components of microwave unit
Part II
73
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
Part II
directly modulated or more generally controlled by cycling the magnetron at
full power in a time chopped mode to achieve an average power level. The
duty cycle o f a magnetron is the time the magnetron is on divided by the time
base. For example, a time on o f 5 s with a time base o f 10 s is a duty cycle o f
0.5. Similarly, a time on o f 0.5 s with a time base o f I s is also a duty cycle
o f 0.5. Usually, the time base for the magnetron used for sample preparation
is 1 s. The typical time base for appliance grade microwave ovens is 10 s.
This relatively long time base is not desired for analytical analysis. Most
magnetrons for analytical use operate at a time base of I second; for example,
if the magnetron is on and off for 0.5s, 50% o f output power is obtained.
1.5 Research Goal
The research goal in this part o f my thesis is to evaluate and develop a
microwave assisted extraction method for Cr(VI) in environmental samples.
The extraction methods of Cr(VI) which have been applied in samples
include solvent extractions (41-44), acid extractions (45, 46), and alkaline
extractions (46-49). The current accepted extraction method of Cr(VI) from
environmental samples is EPA Method 3060A (alkaline extraction method),
which is very efficient compared with some other methods (47, 48). This
method may completely extract both water-soluble and water-insoluble
Cr(VI) compounds by heating the extraction solution to 90-95°C with a hot
plate and maintaining the temperature for 60 minutes (50).
74
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
Part II
Compared with several other methods, Method 3060A is currently regarded
as the most proper method for extracting Cr(VT) from environmental sotr3
samples (see ref)- However, Method 3060A has many shortcomings. To
measure the Cr(VI) in a sample, at least three replicates should be done for
the statistical analysis. In other words, at least 4 and half hours is needed for
one sample preparation. This time-consuming step significantly decreases
the entire analysis speed. Moreover, the heating temperature o f this method
is usually restricted in 90-95 °C. Because of the limited boiling point of the
sample solution (water solution in this case) under the atmospheric pressure,
the extraction temperature using Method 3060A can not be reached higher
than 100 °C. Besides the most water insoluble lead chromate and barium
chromate, the complex environmental matrices may contain other forms of
Cr(VI), which m ight exist in extremely insoluble crystals or might be coated
with some insoluble oxides. The extraction o f these forms o f Cr(VI) may
need higher extraction temperature. However. M ethod 3060 may not meet
the requirement, and there is a doubt that not all the Cr(VI) forms are
completely extracted at 90-95°C with Method 3060A. Additionally, Method
3 060A may cause errors during extraction due to the open vessel system as
described in part I.
Since M ethod 3060A has so many shortcomings, developing new
methods or improving the existing EPA method using the same extraction
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
Part II
solution is always o f great interest to enhance the accuracy and/or to save the
labor and time.
Although few microwave assisted alkaline digestion or extraction
have been reported, the alkaline extraction (0.28 M Na;C 0 3/ 0.5 M NaOH)
solution should have a very large dissipation factor due to the large
concentration o f free ions and large amount o f water based on the microwave
heating mechanism. This extraction solution can be heated rapidly by both of
the two heating mechanisms: dipole rotation (water) and ionic conduction
(ions). The dissipation factors o f water and 0. IM sodium chloride in table 1
are 1570 and 2400 at 25°C respectively. The alkaline extraction solution
contains approximate 1.06 M o f N a \ 0.27 M o f C 0 3:\ and 0.01 M of OH';
therefore, the ionic conductivity o f the extraction solution should be much
better than that of 0 .1M sodium chloride solution. The large resistance
produced by the flow o f current makes the temperature o f the solution rise
rapidly. Therefore, the alkaline extraction procedure is very suitable to be
performed by using the microwave heating system, and a microwave assisted
extraction method o f Cr(VI) can be developed based on the hot plate
procedure.
Compared with the open vessel system, microwave assisted closed
vessel sample preparation has some advantages. It can achieve higher
temperature because the boiling po«nt o f the extraction can be raised by the
pressure produced in the vessel. Because o f this, the tim e required for the
76
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
extraction o f Cr(VI) from environmental samples can be greatly reduced, and
the other Cr(VI) forms in environmental samples which are very difficult to
be extracted may be completely extracted. In addition, the contamination
from the environment during the extraction may be substantially reduced in
closed vessels.
Based on these considerations, microwave enhanced closed vessel
system was selected to replace the hot plate system trying to increase the
extraction efficiency, shorten the extraction time, and reduce analytical errors.
Similar procedure to Method 3060A was performed to the three
standard water-soluble and water-insoluble Cr(VI) solids as used in Part I to
test the feasibility of the microwave extraction o f Cr(VI), and complete
extraction o f these three Cr(VI) solids was obtained. The oxidation o f Cr(III)
using the same closed vessel procedure was then studied. The less oxidation
o f Cr(III) with closed vessel than that with hot plate system indicated that the
closed vessel could be better than the hot plate. Further experiments on
oxidation o f Cr(III) during extraction were then done with closed vessel
system by increasing temperature and shortening extraction time to select an
appropriate working condition. Under this condition, the most waterinsoluble lead chromate was completely extracted.
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
Part II
2.
EXPERIMENTAL SECTION
2 .1 Reqgenls
Deionized (DI) water (18MQ) prepared by a N A NOpure U1trap use
Water System (Bamstead) (Apple Scientific Inc., Chesterland, OH) was used
in the preparation o f all solutions.
•
The extraction solution (0.5 M NaOH/0.28M Na^CO^) in Method 3060A
was made by dissolving 20 g o f NaOH (98%, Fisher, Fair Lawn, NJ) and
30 g anhydrous N a,C 03 (99.6%, Fisher, Fair Lawn, NJ) in 500 mL o f
deionized water and then diluting to 1 L.
•
Lead chromate solid (Special for micro analysis, Fisher, Fair Lawn. NJ)
and barium chromate solid (98+%, Aldrich, M ilwaukee, WI) were
extracted by Method 3060A. Natural Cr(VI) standard solution was
prepared by dissolving dried analytical reagent grade potassium
dichromate (99.4%, J.T. Baker Inc., Phillipsburgh. NJ) in deionized
water.
•
Natural Cr(III) standard solution was prepared by dissolving chromium
metal (99.995%, Aldrich Chemical Co., M ilwaukee, WI) in a minimum
amount o f 6M HC1 prepared from sub-boiled HC1 and the solution was
diluted with 1% H N 03.
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
•
Reagent diphenylcarbazide (DPC) was made by dissolving 0.5g DPC
(Residue after ignition 0.01%, Fisher, Fair Lawn, NJ) in 100 ml acetone
and stored in a brown bottle.
•
Concentrated nitric acid (Fisher, Pittsburgh, PA) was used for
neutralizing the extract.
•
10% (v/v) sulfuric acid was made by slowly adding 10 ml concentrated
sulfuric acid (Fisher, Pittsburgh, PA) to 90 ml DDI water with continuous
stir.
2.2 Equipment
A modified Milestone MLS-1200 MEGA (microwave digestion
system with microwave digestion rotor) was equipped with temperature
feedback and a stirring capability. This unit, equipped with EM-45 exhaust
module and a Model 240 control terminal was used in our experiments
(F igure 4).
The new microwave digestion rotor (MDR) technology has brought
microwave digestion technology from the pioneer phase to the world o f
routine analytical procedures. It guarantees fast, simple, reliable, and very
safe digestion of samples, as well as extraction and dissolution.
Besides the above configurations, another modification o f this
Milestone unit is the use o f Weflon. Weflon, a new PTFE-C compound
produced by Milestone, a teflon impregnated with carbon to produce partial
79
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
• A
rn
1 o
■'f
CN
03 3
r -■
J J
r—1
r—
I—<
5J
—
*
C
i.
n
rt
c
X
03
*o
•_
c
Q
u
03
_w
•
"E
ZD <
<u
c 03
o
a
Crt
O
O
o
Tf C/3
«
U
5d'—
•-
SO
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
Part II
microwave absorption. The advantages o f Weflon liners included:
homogenous temperature distribution on the vessel surface; higher
temperature o f solutions at reduced microwave power and therefore increased
Magnetron life; the possibility of heating other microwave transparent
materials. In laboratory practice, condensation o f vapors on the cooler vessel
walls makes that high pressure and high temperatures are difficult to attain;
therefore, heating a small sample by microwave in a closed vessel is difficult.
This condensation also makes evaporation o f solution in an open vessel
extremely difficult due to the reflux effect. The use of material Weflon
solved these problems and eliminated the negative effects o f decreased
absorption at higher temperature. When heating a solution in a microwave
field, the temperatures that can be reached are a function o f the dielectric
factor of the reagent and the volume o f the reagent. The value of the
dielectric factor is a function o f the temperature o f the material. The
absorption of the microwaves decreases with the temperature, which means
that a large amount o f microwave power is required for a minimal
temperature increase at higher temperatures. The higher the volume o f the
reagent the larger the microwave absorption will be, and therefore the higher
the temperature. In our experiments, 50 ml o f extraction solution was used in
each 100 ml o f closed vessel. The volume o f the solution was large enough
to quickly heat to 150°C.
81
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
MDR-600/10 was used in the microwave enhanced alkaline extraction
o f Cr(VI). Up to 10 samples can be extracted simultaneously with this
digestion equipment. The volume o f the vessels is 100 ml with the maximum
pressure 30 bar.
For M LS-1200 MEGA unit, the power o f microwave generating
magnetron is 1200 W and 1000 W delivered inside the working chamber.
Power emission is microprocessor controlled from 10 to 1000 W, in 10 W
increments. Usually, pulsed microwave emission is used for adjustable
fractions o f a cycle ranging from 1/10 to 50/10.
In the experiments, a solid sample was weighed into a vessel, then a
stirring bar and 50 mL o f the alkaline extraction solution were added in this
vessel. The samples were inserted in the digestion rotor which was placed in
the microwave oven. The stirring mode was on full. Two heating steps are
involved in the extraction: heating the sample and the extraction solution
from room temperature to a specific extraction temperature; keeping the
extraction temperature for a specific time. The power o f microwave
instrument in each heating step should be selected.
perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
3.
RESULTS AND DISCUSSiOH
Compared with the open vessel system, the closed vessel microwave
sample preparation offers many advantages. Samples in closed vessels can
be isolated from the around environment reducing sample loss and
contamination. Because the most used solvent is water, the highest
temperature which can be obtained in conventional open vessel is
approximate 100°C. However, in general, much higher temperature can be
achieved in closed vessel because the boiling points o f solutions increase
with the increase in pressure produced inside the vessels. Generally, much
shorter time is required for the closed vessel sample preparation due to the
higher temperature and pressure supplied.
3.1 Extraction of Cr(VI) Solids
Three standard Cr(VI) solids: potassium dichromate, barium
chromate, and lead chromate, were extracted with closed vessel microwave
procedure which is very sim ilar to EPA Method 3060A.
Among the three Cr(VI) compounds, the first one (K:Cr20 7) is watersoluble, the other two are insoluble in water. As the data in Table 2
indicated, the microwave enhanced extraction achieved approximately 100%
recoveries o f Cr(VI) for all these three compounds. Besides the deferent
heating mechanisms, another difference between the hot plate and the
83
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Tabic 2 MW Extraction of Cr(VI) Standards
Cr(VI)
Compound
K2Cr20 7
BaCr04
PbCr04
Cr(Vl)
(mg)
(Eg/g)
Recovery
Cr(VI)
Recovery (%) (average ± std)
10.7
75.6
103
11.1
78.5
104
11.9
84.1
101
14.4
59.1
98.3
24.3
99.8
97.3
24.4
100
100
8.2
26.4
99.9
13.8
44.4
102
16.0
51.5
96.1
(%)
Recovery
(average ± std) (%)
with Hot plate (50)
103 ± 1
103 t 6
(K2Cr04 solution)
99 ± 2
93 ± 9
99 ± 3
102 ± 2
Part II
00
Mass of Solid
Part II
microwave assisted extraction process is the heating time. It took only 8
minutes to heat the sample to 92°C using this microwave extraction
procedure; while, it took about 30 minutes on a hot plate. The hot plate
extraction had been performed to K2C r0 4 solution, B aC r04 and P bC r04
solids, and approximately 100% recoveries were obtained for all these three
Cr(VI) compounds. Compared with the hot plate results, better precision and
accuracy in the table indicated that the microwave extraction method may
replace the conventional hot plate procedure.
3.2 Oxidation o f Cr(lII) During Microwave Extraction
3.2.1 Oxidation o f Cr(III) at 90-95°C
Oxidation o f Cr(III) to Cr(VI) may take place in high pH solutions,
and Method 3060A requires the pH of the extraction solution larger than
11.5 for efficient extraction o f Cr(VI). In such high pH solutions. Cr(III) may
be oxidized to Cr(VI) if some oxidants are present. The oxidation of Cr(IIl)
during the hot plate extraction was detected in Part I. Similar procedure was
performed by using microwave assisted closed vessel extraction to test the
oxidation o f Cr(III) during the microwave extraction.
In section 3.1.1 o f Part I, standard Cr(III) solution was extracted with
hot piate and the oxidation o f Cr(III) was detected. The potential oxidant, in
this case, was the oxygen dissolved in the extraction solution and/or some
oxygen from the air during the extraction. Because o f this, argon gas was
85
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
Part II
then filled in some o f the vessels trying to remove the air and reduce the
oxidation. For the comparison on the oxidation o f Cr(III), standard Cr(III)
solution was extracted by hot plate, closed vessel system, and closed vessel
system filled with argon gas with the same procedures by keeping the
temperature at 92°C for an hour.
The extracts o f these samples were detected only by Method 7196A.
Free Cr3* tends to form Cr(OH)3 precipitate in this alkaline solution. Very
limited Cr3* can exist in solution ( ) . which does not interfere with the UVVis detection o f Cr(VI), therefore, Method 7196A should be an accurate
measurement in these cases.
Approximately 200 pg Cr(III) was added to a vessel or a beaker, and
almost the same extraction procedure (heating to 92°C for 60 minutes) was
performed to the synthesized samples with these three different extraction
systems. Oxidation o f Cr(III) was observed for all these three systems (see
F ig u re 5). Hot plate procedure achieved 5.6% oxidation o f Cr(III); however,
both closed vessel procedures obtained 1.7% oxidation o f Cr(III). The same
percentages o f Cr(III) oxidation with these two closed vessel microwave
extraction systems suggested that the dissolved oxygen in extraction solution
could be the only oxidant in closed vessels.
The oxidation o f Cr(III) during extraction affected the accuracy o f the
Cr(VI) determination in solid environmental samples. To avoid or reduce
such oxidation of Cr(III), Mg2* and phosphate buffer have been used in the
86
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
T ab ic 3 Extraction o f P b C r0 4 (150°C/10 min.)
Mass of
Solid (mg)
36.7
46.1
97.3
119
Cr(VI)
Element (mg)
5.90
7.42
15.7
19.1
UV-Vis Recovery o f
Cr(VI)
Detected (mg]1
pH 12 extract (%)
5.87
7.44
15.5
19.0
99.5
100
99.2
99.5
Recovery
(mean ± std)
(%)
99.6 ±0.5
T ab le 4 Detection o f Neutralized P b C r0 4 by UV-Vis and SIDMS
UV-Vis Recovery
(%)
SIDMS Recovery
of pH 12 (%)
99.9
101
98.0
96.1
77.8
78.9
101
101
101
103
98.8
81.5
Part [I
Mass o f PbC r04 Cr(VI) Element (pg/g)
(mg)
26.4
8.2
44.4
14.3
16.0
49.7
104
32.2
188
58.5
236
73.2
Part II
alkaline extraction procedure trying to precipitate soluble Cr(III) as phosphate
from solutions since soluble Cr(III) and freshly precipitated Cr(OH)3 are
easily oxidized, while insoluble Cr(III) and Cr(III) complexes are difficult ( ).
However, Mg'* and phosphate buffer did make limited improvement to
reduce the oxidation o f Cr(III). In comparison with the application o f Mg2* in
Method 3060A. closed vessel heating system has been proved to be a better
method with much less Cr(III) oxidation and great precision and accuracy.
3.2.2 Oxidation o f Cr(III) at 120°C and 150°C
Speeding up the extraction is one o f the advantages o f microwave
assisted digestion procedure. Theoretically, increasing the temperature of the
extraction solution to over 100 °C in closed vessel system, which is
impossible in the hot plate procedure, can accelerate the extraction as well as
the Cr(III) oxidation reactions. Generally, the reaction rate increases by 2-3
times with the 10°C increase in temperature. If the extraction temperature
increases from 90°C (the hot plate extraction temperature) to 120°C, the
extraction rate will increase by at least 8 times. If the temperature goes up
from 90°C to 150°C, the extraction rate will increase by at least 64 times.
Increasing the extraction temperature, the extraction rate o f Cr(VI) as
well as the oxidation rate o f Cr(III) will increase. Therefore, the oxidation o f
Cr(III) should be studied at higher temperatures. When the microwave
extraction was performed to the Cr(III) standard solution at 120°C for 60
88
R eproduced w ith perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
Part II
minutes, the oxidation of Cr(III) was increased as expected; but it was still
less than that in hot plate procedure. The extraction time can be shortened
due to the increase in extraction temperature. Thus, another procedure was
performed to Cr(III) standard at 150°C for 10 minutes, and almost the same
percentages o f oxidation o f Cr(III) as the hot plate system were obtained.
For extracting the same quantities o f standard Cr(III), heating the
extraction solution to 120°C and maintaining it for 60 minutes significantly
increased the oxidation percentages of Cr(III) from 1.7% to 4.3% (F igure 5).
However, 4.3% is less than 5.6%, the result o f the hot plate procedure in EPA
M ethod 3060A. The increase in temperature by 30°C will increase the rate of
the oxidation o f Cr(III) as well as the rate o f the extraction o f Cr(VI) by at
least 8 times. However, the total quantity o f Cr(VI) detected (4.3% ) in
120°C extracts was not as 8 times as that (1.7%) in 90°C extracts.
Continuously increasing the temperature to 150°C, but reducing the
extraction time from 60 minutes to 10 minutes. A slightly higher oxidation
percentage o f Cr(III) was obtained (5.1%) compared with that in the
procedure at I20°C for 60 minutes (4.3%), but it was still less than the hot
plate results (5.6%) (see F igure 5). The data implied that maintaining the
temperature around 150°C for about 10 minutes could be a suitable procedure
for extracting Cr(VI) from samples.
Theoretically, this procedure should be better than Method 3060A in
efficiency. The extraction rate increases by 64-729 times as the temperature
89
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
P an II
is raised from 90 to 150°C. Even if the extraction time is cut from 60
minutes to 10 minutes (1/6 of the original time), the extraction efficiency o f
Cr(VI) at should still be much better than that of EPA Method 3060A.
3.2.3 Quantities o f Cr(VI) Oxidized from Soluble Cr(III)
After the extraction o f approximate 50 or 200 pg o f standard Cr(III) at
120°C for 60 minutes and at 150°C for 10 minutes, the Cr(VI) oxidized from
Cr(III) in these synthesized samples was detected by Method 7196A. As the
experiments in Part I o f the thesis demonstrated, the total detected Cr(VI)
with Method 3060A was not much related to the quantity o f Cr(III) in the
extraction solution if the concentration o f standard Cr(ili) added in the
extraction solution was high enough. The same conclusion could be drawn
for these two microwave assisted closed vessel extraction o f Cr(VI).
For the first microwave extraction procedure at 120°C for 60 minutes,
we hypothesized that the detected Cr(VI) oxidized from 200 pg o f Cr(III)
should be more than that from 50 pg o f Cr(III). However, the total quantities
o f Cr(VI) detected (average ± standard deviation) were 8.63 ± 0.50 pg and
8.68 ± 0.54 pg, respectively (n=3). The overlapped data did not show the
statistic difference. Similarly, 50 pg and 200 pg Cr(III) were extracted at
150°C forlO minutes, 9.60 ± 0.83 pg andl0.3 ± 0.39 pg o f Cr(VI) were
obtained respectively. These data were also overlapped. No significant
difference on the total quantities o f Cr(VI) oxidized from Cr(III) during the
90
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
F i g u r e 5 Oxidation ot'Cr(III) under Different Conditions
Part II
1. Hot plate 92°C for 60 min.; 2. MW filled with argon 92°C for 60 min.;
3. MW 92°C for 60 min.; 4. MW 120°C for 60 min.; 5. MW 150°C for 10 min.,
The soluble Cr(IIl) added in all these cases was 200 pg. (n=3 for each condition)
Part II
extraction in these two cases, although there was a big difference (four times)
between the amounts o f the soluble Cr(III) added (F igure 7). Therefore, it
could be concluded that the amount o f Cr(VI) oxidized from soluble Cr(III)
or freshly precipitated Cr(OH)3 was limited in the synthesized samples for a
certain extraction procedure, regardless the amount o f Cr(III) added. This
conclusion may help us estimate the oxidation o f Cr(III) during sample
extraction.
However, the oxidation o f Cr(III) during extraction could vary with
the sample matrix. Many chemicals in the sample matrix may oxidize Cr(III)
in the alkaline extraction solution resulting in more oxidation of Cr(III). And
different matrix may affect the extraction with different ways.
3.3 Extraction of PbC r04 at 150°C
We already know that the neutralization process in Method 3060A
could cause the loss o f Cr(VI) when large quantity o f PbC r04 was extracted.
A better extraction method o f Cr(VI) was developed with a microwave
assisted closed vessel system by increasing the temperature to 150°C and
decreasing the extraction time to 10 minutes. This method could achieve
complete extraction o f Cr(VI) from three standard Cr(VI) solids and greatly
shorten the extraction time. However, the neutralization process could also
cause the loss o f PbC r04just like it took place in Method 3060A: after
extraction and filtration, precipitation occurred during the neutralization, and
92
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
(n =3 for each case)
Ia: 50 f.ig Cr(III) 120°C / 60 min.;
Ila: 50 ng Cr(III) 150°C / 10 min.;
Ib: 200 ^tg Cr(III) 120°C / 60 min.;
lib: 200 >ig Cr(Ul) 150°C /10 min..
Part II
Part II
low recoveries o f Cr(VI) were caused by Method 7196A. The conclusion is:
the neutralization step in either the hot plate or the microwave extraction
procedure could cause the loss of Cr(VI) if the concentration of P b C r0 4 was
higher enough in the extract
The extract o f PbCrO ,at pH~-I2 was used as a sample and detected
with Method 7196A to avoid the loss o f Cr(VI) during neutralization and test
the extraction efficiency. The complete recoveries o f these samples indicated
that the extraction procedure was proper for PbC r04; and the extracts at
pH~12 instead o f the pH~7.5extracts could be detected to avoid the loss of
Cr(VI) during the neutralization. Because the concentrations of Cr(VI) could
be very high in some samples, detecting the pH 12 extracts could be a better
way for obtaining a greater accuracy. For example, the highest concentration
o f PbC r04 in the extract in our experiments was about 382 ug'g, 99.5%
recovery o f Cr(VI) was achieved in this case(Table 3).
3.4 Bias from Neutralization Process
Although 100% recoveries o f the three standard Cr(VI) compounds in
table 2 and less oxidation o f Cr(III) in F igure 5 using closed vessel system
were obtained, a problem in the neutralization step o f Method 3060A existed
in hot plate procedure as well as in closed vessel process when the quantity of
lead chromate was large, as mentioned previously.
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
Precipitates were formed during the neutralization process in Method
3060A with the increase in the quantity o f lead chromate although all the
solutions were clear before and after filtration. More PbC r04 solid added in
the extraction solution could cause the reprecipitate to occur earlier (higher
pH) during neutralization. The phenomena and the extraction mechanisms
have been presented in Section 3.1.2 o f Part [.
After the microwave extraction at 150°C, the filtered extract (pH~12)
before neutralization was sampled for SIDMS detection (as the diagnostic
tool) to check the neutralization process, and 100% recoveries o f Cr(VI)
were achieved for all these samples. The neutralized extracts (pH~7.5) were
detected with UV-Vis measurement, almost 100% recoveries o f Cr(VI) were
obtained for the first three samples which were clear solutions after
neutralization: however. 77%-82% recoveries o f Cr(VI) for the last three
samples in Table 4 with larger quantities were obtained due to the
precipitation. The precipitated lead chromate could be dissolved during UVVis measurement because the detection solution was acidic and lead
chromate could be dissolved in acidic solution. For the last three samples,
particles of yellow precipitate formed during the neutralization step quickly
precipitated to the bottom of the container resulting in poor accuracy and
precision, although some o f the extract was sampled for UV-Vis detection
right after the extract was swirled. The comparison o f the results o f SIDMS
and UV-Vis in Table 3 demonstrated that this microwave enhanced alkaline
95
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
extraction procedure o f Cr(VI) could be an excellent extraction method, but
the neutralization step could cause a problem in closed vessel extraction as
well as in hot plate procedure.
In addition, similar precipitate (yellow or brown) was observed whea
neutralization process was performed to soil extract. Besides the above
reason, another possible reason was that the solubility of humic acids and
fulvic acids decreased with the decrease in pH. The alkaline extraction
procedure can extract a large amount o f organic acids from soils as well as
some inorganic compounds including Cr(VI). When the extract was
neutralized with concentrated nitric acid, some o f these water insoluble acids
precipitated from the solution at pH 7.5. The particles o f the insoluble acids
may result in the heterogeneity of the extracts and influence the UV-Vis
detection.
Because of the precipitate o f organic acids in neutralized extracts, the
second filtration may be helpful to improve the precision of analysis.
However, if the precipitate contains some water-insoluble Cr(VI) compounds,
the second filtration would cause the loss o f Cr(VI). Those organic acids like
other chemicals would not affect the SIDMS measurement; therefore, the
modification of the whole procedure o f Cr(VI) analysis in environmental
samples including extraction and detection is suggested as: using microwave
enhanced closed vessel extraction procedure at 150°C to extract Cr(VI) from
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
samples, then using SIDMS detection method to measure Cr(VI) in the pH 12
filtrates.
SIDMS method does not need the neutralization process. The extract
(pH~12) right after extraction and filtration can be spiked with isotope
species o f chromium immediately to detect the concentration o f Cr(VI) at this
point, even though there is an interconversion between Cr(III) and Cr(VT)
during the operations.
3.5 Future Study and Suggestions
Some of the research work on microwave assisted closed vessel
extraction o f Cr(VI) has been done in this part o f the thesis. The microwave
extraction procedure at 150°C for 10 minutes not only completely extract
Cr(VI) from three standard Cr(VI) solids, greatly reduced the extraction time,
but also limited the oxidation o f soluble Cr(III). Up to 10 samples can be
extracted simultaneously with this Milestone unit, and the total extraction
time was less than 30 minutes (including the time to heat the sample to
150CC, extraction time, and venting time); while these 10 samples need 900
minutes (15 hours) to be extracted with a hot plate.
This new extraction method o f Cr(VI) has many advantages; however,
oxidation o f Cr(IH) did take place during extraction. The conventional
methods, such as chromatography, electrochemistry, and UV-Vis detection
can not correct for such interconversion between the chromium species. For
97
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
example, as the detection method o f Cr(VI) in EPA method pairs
3060A/7196A, Method 7196A could not correct for the oxidation of Cr(III)
during the extraction; additionally, this method could cause analytical biases
of Cr(VI) during the detection. Currently, only the new isotopic method
based on Method SIDMS applied in this thesis can accurately determine the
concentration of Cr( VI) in real samples. Therefore, Method SIDMS will be
applied in our further research to detect Cr(VI) instead o f Method 7196A.
Real solid samples need to be experimented to test the efficiency o f
this microwave assisted closed vessel extraction of Cr(VI). and the
concentrations of Cr(VI) should be detected by Method SIDMS. This
indicates that much research work has to be done in the future to verify or
further improve this microwave extraction method, which could not be
finished in this thesis.
98
R eproduced w ith perm ission o f the copyright owner. F urth er reproduction prohibited w itho ut perm ission.
Part II
4.
SUMMARY
The good extraction method should not only obtain complete extraction,
but also decrease the errors during the extraction procedure. Compared with
EPA Method 3060A, a conventional extraction method o f Cr(VI) with hot
plate heating system, the microwave assisted closed vessel alkaline extraction
method o f Cr(VI) has many advantages. This method can reduce the
oxidation o f Cr(III) during extraction, greatly shorten the extraction time
from 60 minutes to 10 minutes by increasing the extraction temperature to
150°C, and increase the precision and accuracy o f the results.
Because this microwave extraction method o f Cr(VI) can significantly
enhance the extraction efficiency, it is believed that it will play a very
important role in the further for the analysis o f Cr(VI) in environmental
samples. Further studies and experiments will be based on this extraction
method and SIDMS detection to determine the concentration o f Cr(VI) in real
solid samples quickly and accurately.
R eproduced with perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
5.
REFERENCES
(1)
Majors, R. E. LC-GC 1991, 9, 16-28.
(2)
Matusiewicz. H.; Sturgeon. R. E. Prog. Anal. Spectrosc. 1989, 12, 2139.
(3)
Abu Samra, A.; Morris, J. S.; Koirtyohann, S. R. Anal. Chem. 1975,
47, 1475-1477.
(4)
Barbano, D. M.; Della Valle. M. E. J. FoodProt. 1984, 47. 272-278.
(5)
Donard. O. F. X.; Lalere. B.: Martin. F.; Lobinski. R. Anal. Chem.
1995, 67, 4250-4254.
(6)
Jassie, L. B.; Hasty, E.; Revesz, R.; Kingston, H. M ., Arlington, VA
1992; US-EPA; 500-508.
(7)
Scifres, J.; Cheema, V.; Wasko. M.; McDaniel, W. American
Environmental Laboratory 1995, 7, 1-8.
(8)
Scifres, J.; Cheema, V.; Wasko, M.; McDaniel, W. American
Environmental Laboratory 1995, 7, 34-38.
(9)
Benson, R. L.; McKelvie, I. D.; Hart, B. T. Anal. Chim. Acta 1994,
291, 233-242.
(10)
Fang, Y. PittCon poster 1995.
(11)
Baneijee, S. K.; Makdisi, W. F.; Weston, A. P.; Mitchell, S. M.;
Campbell, D. R. BioTechniques 1995, 18, 768-774.
100
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part [I
(12)
Hasty, E.; Revesz, R. American Laboratory 1995, 66-74.
(13)
Renoe, B. W. American Laboratory 1994, 26, 34-40.
(14)
Baghurst, D. R.; Mingos, D. M. P. J. Chem. Soc. Chem. Commun.
1988, 829-830.
(15)
Kingston. H. M.; Jassie. L.B. Introduction to Microwave Sample
Preparation: Theory and Practice; American Chemical Society:
Washington, DC, 1988.
(16)
Kingston. H. M.; Haswell., S.J. Microwave-Enhanced Chemistry:
Fundamentals, Sample Preparation, and Applications'.: American
Chemical Society: Washington. DC, 1997.
(17)
Gedye. R.; Smith, F.; Westaway, K. C . , New Orleans 1988;Paper 58.
(18)
Gedye. R.; Smith, F.; Westaway, K. C. J. Micro. Power 1991, 26, 317.
(19)
Gedye, R. N.; Rank. W.; Westaway, K. C. Can. J. Chem. 1991, 69.
706-711.
(20)
Alexander, W. R.; Shimmield, T. M. J. Radioanal. Mucl. Chem. 1990,
145, 301-310.
(21)
Pagano, S. T.; Smith, B. W.; Winefordner, J. D. Talanta 1994, 41,
2073-2078.
(22)
H ew itt A. D.; Reynolds, C. M. Atomic Spectroscopy 1 9 9 0 , 11, 187192.
101
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Pan II
(23)
Hewitt, A. D .; Reynolds, C. M .; U.S. Army Corps o f Engineers Cold
Regions Research & Engineering Laboratory, 1990.
(24)
Alvarado. J.; Petrola, A. J. Anal. At. Spectrom. 1989, 4, 411-414.
(25)
Aduna de Paz, L.; Alegria. A.; Barbera, R.; Farre, R.; Lagarda, M. J.
Food Chem. Jan Feb 1997, 58, 169-172.
(26)
Sun. D. H.; Waters, J. K.; Mawhinney, T. P. J Anal At Spectrom. Jan
1997, 12, 675-679.
(27)
Lan. W. G.: Wong, M. K.: Sin. Y. M. Talanta 1994, 41, 53-58.
(28)
Lan, W. G.; Wong, M. K.; Sin. Y. M. Talanta 1994, 41, 195-200.
(29)
Chakraborty, R.; Das. A. K.: Cervera. M. L.: De La Guardia. M. J.
Anal. At. Spectrom. 1995, 10, 353-358.
(30)
Allen. L. B.; Siitonen, P. H.; Thompson, H. C.. Jr. J Agric Food
Chem. Jan 1997, 45, 162-165.
(31)
Barnes. K. W. A t Spectrosc. May Jun 1997, 18, 84-101.
(32)
Hocquellet. P.; Candillier, M. P. Analyst 1991, 116. 505-509.
(33)
Lopez Avila, V.; Young, R.; Beckert, W. F. Anal. Chem. 1994, 66,
1097-1106.
(34)
Lopez Avila, V.; Young, R. ; United States Environmental Protection
Agency, 1994.
(35)
Onuska, F. I.; Terry, K. A. Chromatographia 1993, 36, 191-194.
(36)
Llompart, M. P.; Lorenzo, R. A.; Cela, R.; Pare, J. R. J.; Belanger, J.
M. R.; Li, K. J Chromatogr, A 1997, 757, 153-164.
102
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
(37)
Demesmay. C.; Olle, M .; Service Central d'Analyse du Centre
National de Recherche Scientifique, Vemaison.
(38)
Nyomora. A. M. S.; Sah, R. N.; Brovvn, P. H.; Miller, R. O.
Fresenius' J Anal Chem. Apr 1997, 357, 1185-1191.
(39)
Vazquez, M. J.; Carro. A. M.; Lorenzo, R. A.; Cela, R. Anal Chem
1997, 69, 221-225.
(40)
Mahan. K. I.; Foderaro, T. A.; Garza, T. L.; Martinez, R. M.;
Maronev, G. A.; Trivisonno, M. R.; Willging, E. M. Anal. Chem.
1987, 59. 938-945.
(41)
Manzoori, J. L.; Shemirani, F. J. Anal. At. Spectrom. 1995, 10. 881 883.
(42)
Yamada. M.; fCusakabe, S.; Prekopova, J.; Sekine, T. Anal Sci. Jun
1996, 12. 405-410.
(43)
Yamada, M.; Sekine. T. Solvent Extr Res Dev, Jpn 1996. 399-108.
(44)
Tarafder, P. K..; Singh, S. B.; Rathore. D. P. S. Fresenius' J Anal
Chem. Jan 1996, 354, 124-125.
(45)
ICamburova, M.; Aleksandrov, A. ZhAnal Khim. Dec 1995, 50. 12921295.
(46)
Grate, J. W.; Taylor, R. H. Field Anal Chem Technol 1996, I, 39-48.
(47)
James, B. R.; Petura, J. C.; Vitale, R. J.; Mussoline, G. R.
Environmental Science & Technology 1995, 29, 2377-2381.
103
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
Part II
(48)
Vitale, R. J. M., J. C. Peura, B. R. James Journal o f Environmental
quality 1994, 23, 1249-1256.
(49)
Vitale, R. J.; Mussoline, G. R.; Rinehimer, fC. A.; Petura, J. C.; James,
B. R. Environ Sci Technol. Feb 1997, 31, 390-394.
(50)
Vitale, R. J. American Environmental Laboratory 1995, 7, I.
104
R eproduced w ith perm ission o f the copyright owner. F urther reproduction prohibited w itho ut perm ission.
*
IMAGE EVALUATION
TEST TARGET ( Q A - 3 )
✓
.'135-
y '
,b.
b.
5^
A
t
HM
1.0 ^1-i
1« - 2 2
us
■jo
l.l
Ifi
1.8
1.25
1.4
1.6
150mm
A P P L IE D j IM/4GE . Inc
■
1653 E ast Main Street
- = ~- Rochester. NY 14609 USA
-= = = r —
Phone: 716/4820300
- = r- =
Fax: 716/288-5989
0 1993. A p plied Im age. Inc.. All R ights Reserved
/
perm ission o f the copyright owner. Further reproduction prohibited w itho ut perm ission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
3 866 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа