1
PHASE EQUILIBRIUM AND PARTITION COEFFICIENT OF SELECTED
PESTICIDES AND NITROPHENOLS IN DOUBLE AND TRIPLE-PHASE
LIQUID-PHASE MICROEXTRACTION
NURUL AUNI BINTI ZAINAL ABIDIN
UNIVERSITI TEKNOLOGI MALAYSIA
2
PHASE EQUILIBRIUM AND PARTITION COEFFICIENT OF SELECTED
PESTICIDES AND NITROPHENOLS IN DOUBLE AND TRIPLE-PHASE
LIQUID-PHASE MICROEXTRACTION
NURUL AUNI BINTI ZAINAL ABIDIN
A thesis submitted in fulfillment of the
requirements for the award of the degree of
Master of Science (Chemistry)
Faculty of Science
Universiti Teknologi Malaysia
OCTOBER 2009
5
Dedicated to my wonderful mother, my beloved siblings, family, and friends….
6
ACKNOWLEDGEMENTS
First of all, in a humble way I wish to give all the Praise to Allah, the
Almighty God for with His mercy has given me the strength, and time to complete
my research.
I am deeply indebted to my supervisor, Professor Dr. Mohd Marsin Sanagi
for giving me the opportunity to carry out research in the interesting area of liquid
phase microextraction. I thank him for his helpful suggestion, direction, advice and
scientific guidance that enabled me to approach work positively.
I also would like to thank my co-supervisor, Associate Professor Wan Aini
Wan Ibrahim for her helpful comments and advices. Thanks also to the all the staff
from the Department of Chemistry and Ibnu Sina Institute for Fundamental Science
Studies, UTM for their prompt and timely help during my study and research. I also
gratefully acknowledge to the Ministry of Science, Technology and Innovation
(MOSTI) Malaysia for its financial support through National Science Fellowship
(NSF).
I would like to say thanks to all the group members in the Separation
Sciences Research Group (SSRG) for their supports and advices, also to everyone
who has encouraged me and assisted me in completing this research.
To my sweetheart thank you for your support. Last but not least, heartfelt
appreciation is given to my beloved family for their endless patient, encouragement,
support and prayer for me.
7
PREFACE
This thesis is the result of my work carried out in the Department of Chemistry,
Universiti Teknologi Malaysia between December 2006 to December 2008 under
supervision of Prof. Dr. Mohd Marsin Sanagi and Assoc. Prof. Dr. Wan Aini Wan
Ibrahim. Part of my work described in this thesis has been reported in the following
publications or presentations:
1. Nurul Auni Zainal Abidin, M. Marsin Sanagi, and Wan Aini Wan Ibrahim,
“Double-phase Liquid Membrane Extraction for the Analysis of Pesticides”,
poster presentation at 12th Asian Chemical Congress, Putra World Trade
Centre, KL, 23-25th August 2007.
2. Nurul Auni Zainal Abidin, Mohd Marsin Sanagi, and Wan Aini Wan
Ibrahim, “Application of Double-Phase Liquid Phase Microextraction to the
Analysis of some Pesticides in Water Samples and their Partition
Coefficient.”, poster presented at the Regional Annual Fundamental Science
Seminar 2008 (RAFSS 2008), IIS, UTM, JB Johor, 28-20th May 2008.
3. Nurul Auni Zainal Abidin, Mohd Marsin Sanagi, and Wan Aini Wan
Ibrahim, “Determination of Nitrophenols in Water Samples using ThreePhase Liquid Phase Microextraction Coupled with High Performance Liquid
Chromatography”, Paper presented at the 21st Malaysian Symposium on
Analytical Chemistry (SKAM-21), 25-27th November 2008, Universiti
Malaysia Sabah, Kota Kinabalu, Sabah.
8
ABSTRACT
A two-phase and three-phase hollow fiber-protected liquid-phase
microextraction method (HF-LPME) has been developed and used for the
determination of partition coefficient and analysis of selected pesticides and
nitrophenols in water samples. For two-phase HF-LPME, the analysis was
performed by gas chromatography-electron capture detector (GC-ECD). Extraction
conditions of the selected pesticides were optimized. The following conditions have
been selected as the optimum extraction conditions: toluene as organic solvent, 4 mL
water samples, 20 min extraction time, 850 rpm stirring rate, 1.8 cm HF length, 1.5%
(w/v) NaCl content, and pH 3. The analytes were extracted from 4 mL donor phase
through 4 µL of an organic solvent immobilized in the pores of a porous
polypropylene hollow fiber and then into the acceptor phase present inside the
hollow fiber. Correlation coefficients (r2) of 0.996-0.997 and limits of detection
(LOD) of 0.0026-0.0028 µg/L were achieved under the optimized conditions. The
proposed method provided good average enrichment factors of up to 350-fold and
successfully determined the partition coefficient for the selected analytes that were
found to be directly correlated to the enrichment factor. In the three-phase HF-LPME
technique coupled with HPLC, nitrophenols were extracted from 2.5 mL aqueous
solution with the adjustment of pH in the range of 3.0-5.0 (donor solution) into an
organic phase (1-hexanol) immobilized in the pores of the hollow fiber and finally
back-extracted into 5.0 µL of the acceptor microdrop (pH 12.0) located at the end of
the microsyringe needle. After a prescribed back-extraction time, the acceptor
microdrop was withdrawn into the microsyringe and directly injected into the HPLC
system under the optimum conditions (donor solution: 1.0 M H3PO4, pH 3.0-5.0;
organic solvent: 1-hexanol; acceptor solution: 5 µL of 0.1 M NaOH, pH 8.0-12.0;
agitation rate: 1050 rpm; extraction time: 15 min). The calibration curve for these
analytes was linear in the range of 0.6-200 µg/L with r2 > 0.9994 was achieved under
the optimized conditions. The enrichment factors of up to 500-fold were obtained. In
this work, determination of analytes partition coefficients in three-phase HF-LPME
has been successfully achieved. This study found that the partition coefficient (Ka/d)
values were high for 2-nitrophenol and 3-nitrophenol and the individual partition
coefficients (Korg/d and Ka/org) promoted efficient simultaneous extraction from the
donor through the organic phase and further into the acceptor phase. Both methods
were successfully applied for the analysis of water samples.
9
ABSTRAK
Kaedah pengekstrakan mikro fasa cecair menggunakan gentian berongga
(HF-LPME) yang terdiri daripada 2-fasa dan 3–fasa telah dibangunkan dan
digunakan untuk menentukan pemalar partisi bagi pestisid dan nitrofenol di dalam
sampel air. Bagi HF-LPME 2-fasa, analisis dijalankan menggunakan gas
kromatografi-pengesan penangkapan elektron (GC-ECD). Parameter-parameter
penting yang mempengaruhi pengekstrakan pestisid telah dioptimumkan: pelarut
organik (toluena), isipadu sampel air (4 mL), masa pengekstrakan (20 min),
kandungan NaCl (1.5% (w/v), dan pH (3). Analit yang diekstrak daripada 4 mL fasa
penderma akan melalui 4 µL pelarut organik yang terserap pada liang-liang rongga
gentian ke dalam fasa penerima yang terdapat di dalam rongga gentian tersebut.
Nilai pekali kolerasi (r2) dan had pengesanan terendah (LOD) masing-masing ialah
0.996-0.997 dan 0.0026-0.0028 µg/L. Teknik ini menghasilkan faktor pemekatan
yang baik, iaitu melebihi 350-lipat dan dikenalpasti mempunyai hubungan dengan
pemalar partisi bagi pestisid yang dianalisis. Bagi HF-LPME 3-fasa pula, analisis
dijalankan menggunakan kromatografi cecair prestasi tinggi (HPLC). Nitrofenol
diekstrak daripada 2.5 mL fasa penderma dengan mengubah pH di antara julat 3 – 5
melalui fasa organik (1-heksanol) yang terjerap pada liang-liang gentian ke dalam
5.0 µL fasa penerima (pH 12.0) yang terletak pada hujung penyuntik mikro.
Kemudian, titisan mikro tadi disuntik ke dalam HPLC dengan parameter yang telah
dioptimumkan; pelarut sampel (1.0 M H3PO4), pH (3.0-5.0), pelarut organik (1heksanol), 5 µL larutan penerima (0.1 M NaOH, pH 8.0-12.0), kadar pengadukan
(1050 rpm), dan masa pengekstrakan (15 min). Graf kalibrasi adalah linear di dalam
julat 0.6-200 µg/L dengan r2 > 0.9994. Faktor pemekatan melebihi 500-lipat telah
berjaya diperolehi. Penentuan pemalar partisi (Ka/d) bagi nitrofenol telah berjaya
dikaji. Kajian mendapati nilai Korg/d bagi 2-nitrofenol and 3-nitrofenol serta nilai
partisi individu (Korg/d dan Ka/org) yang tinggi telah menggalakkan proses
pengekstrakan analit daripada fasa penderma ke dalam fasa organik dan seterusnya
ke dalam fasa penerima. Kedua-dua teknik ini telah berjaya digunakan untuk
menganalisis sampel air.
10
TABLE OF CONTENT
CHAPTER
TITLE
PAGE
11
1
2
TITLE
i
DECLARATION
ii
DEDICATION
iii
ACKNOWLEDGEMENTS
iv
PREFACE
v
ASTRACT
vi
ABSTRAK
vii
TABLE OF CONTENTS
viii
LIST OF TABLES
xiii
LIST OF FIGURES
xv
LIST OF ABBREVIATIONS
xx
INTRODUCTION
1.1 Research Background
1
1.2 Statement of Problem
4
1.3 Research Objectives
4
1.4 Scope of Study
6
1.5 Outline of the Thesis
6
LITERATURE REVIEW
2.1 Compounds Studied
7
2.1.1
Pesticides
7
2.1.2
Nitrophenols
9
2.2 Development of Liquid-Phase Microextraction
11
2.3 Comparison of Liquid-Phase Microextraction with Solid-
13
Phase Microextraction
2.4 Hollow-Fiber based Liquid-phase Microextraction
2.4.1
Hollow-Fiber Liquid-phase Microextraction
14
14
Sampling Modes
2.4.2
Hollow Fiber Configurations
17
12
2.5 Analysis of Liquid-phase Microextraction Extracts
19
2.5.1
Gas Chromatography
19
2.5.2
High Performance Liquid Chromatography
22
2.6 Application of Two-phase Hollow-Fiber Liquid-Phase
23
Microextraction
2.7 Application of Three-Phase Hollow-Fiber Liquid-Phase
26
Microextraction
3
EXPERIMENTAL
3.1 Materials and Chemicals
29
3.2 Instrumentation
32
3.2.1
Gas Chromatography-Electron Captured
32
Detector
3.2.2
High Performance Liquid Chromatography
32
3.3 Sampling and Pretreatment of Samples
33
3.4 Hollow-Fiber Liquid-phase Microextraction Procedure
34
3.4.1
Extraction of Pesticides based on Two-phase
34
Hollow-Fiber Liquid-phase Microextraction
3.4.2
Extraction of Nitrophenol Compounds based on
35
Three-phase Hollow-Fiber Liquid-phase
Microextraction
3.5 Method Optimization: Parameters affecting Hollow-Fiber
36
Liquid-phase Microextraction
3.5.1 Organic Solvent
36
3.5.2
Agitation Rate
37
3.5.3
Extraction Time
37
3.5.4
Influence of Hollow Fiber Length
37
3.5.5
Volume of Donor and Acceptor Solutions
38
3.5.6
Salt Addition
38
3.5.7
Adjustment of the pH
39
3.6 Fundamental Theory
3.6.1
Determination of Partition Coefficient,
40
40
13
Recovery, and Enrichment Factor in Two-phase
Hollow-Fiber Liquid-phase Microextraction
3.6.2
Determination of Partition Coefficient,
41
Recovery, and Enrichment Factor in Threephase Hollow-Fiber Liquid-phase
Microextraction
4
APPLICATION OF DOUBLE-PHASE LIQUID PHASE
MICROEXTRACTION TO THE ANALYSIS OF
PESTICIDES
4.1 Preliminary Investigation on the Separation of Pesticides
43
Using GC-ECD
4.1.1
Conformation of Individual Peaks
43
4.2 Optimization of Hollow Fiber-Liquid Phase
45
Microextraction in Aqueous Samples
4.2.1
Effect of Organic Solvent on Analytes Behavior
45
in the Two-Phase LPME System
4.2.2
Effect of Agitation Rate on Analytes Behavior
46
in the Two-Phase LPME System
4.2.3
Effect of Extraction Time Profile on Analytes
47
Behavior in the Two-Phase LPME System
.2.4
Effect of Sample Volume on Analytes Behavior
48
in the Two-Phase LPME System
4.2.5
Effect of Length of Hollow Fiber on Analytes
49
Behavior in the Two-Phase LPME System
4.2.6
Effect of Salt Addition on Analytes Behavior in
50
the Two-Phase LPME System
4.2.7
Effect of pH on Analytes Behavior in the Two-
51
Phase LPME System
4.3 Performance of HF-LPME Procedure
53
4.3.1
Precision
53
4.3.2
Linearity
54
14
4.3.3
5
Detection Limits
55
4.4 Partition Coefficient and Enrichment Factor
56
4.5 Application of HF-LPME in Real Water Samples
58
APPLICATION OF THREE-PHASE LIQUID PHASE
MICROEXTRACTION TO THE ANALYSIS OF
NITROPHENOLS
5.1 Preliminary Investigation on the Separation of
63
Nitrophenols using HPLC-UV
5.1.1
Conformation of Individual Peaks
63
5.1.2
Evaluation of Flow Rate on the Separation
64
Efficiency of Nitrophenols
5.1.3
Evaluation of Mobile Phase Compositions on
66
the Separation Efficiency of Nitrophenols
5.1.4
Evaluation of Concentration and pH of
68
Phosphate Buffer on the Separation Efficiency
of Nitrophenols
5.1.5
Evaluation of Wavelength Absorbance on the
72
Separation Efficiency of Nitrophenols
5.2 Basic Principle of Extraction using Three-Phase LPME
73
System
5.3 Optimization of Three-phase LPME in Aqueous Samples
75
75
15
5.3.1
Effect of Organic Solvent on Analytes Behavior
in the Three-Phase LPME System
5.3.2
Effect of Composition of Donor and Acceptor
76
Phases on Analytes Behavior in the Three-Phase
LPME System
5.3.3
Effect of Extraction Time on Analytes Behavior
78
in the Three-Phase LPME System
5.3.4
Effect of Salt Addition on Analytes Behavior in
79
the Three-Phase LPME System
5.3.5
Effect of Agitation Rate on Analytes Behavior
80
16
in the Three-Phase LPME System
5.3.6
Effect of Acceptor Volume on Analytes
82
Behavior in the Three-Phase LPME System
5.4 Analytical Performance of LPME
82
5.5 Determination of Partition Coefficients and Verification
84
of Experimental Results
5.6 Application of Three-Phase LPME in Sea Water Samples
6
84
CONCLUSION AND SUGGESTIONS FOR FURTHER
STUDY
6.1 Conclusion
87
6.2 Suggestions for Further Study
90
92
REFERENCES
LIST OF TABLES
TABLE NO.
2.1
TITLE
PAGE
Types of pesticides and functions (Fest and Schmidt,
8
17
1982)
3.1
Properties of the target analytes
30
3.2
Properties of the nitrophenol compounds.
31
4.1
Retention times of pesticides studied
44
4.2
Summary of the performance of the developed methods
53
4.3
Analytical performance in terms of enrichment factor
57
and partition coefficient
4.4
Application performance of HF-LPME in farm water
62
samples
4.5
Application performance of HF-LPME in drinking
62
water samples
5.1
Effect of different flow rates on the retention time and
65
retention factor of the nitrophenols
5.2
Effect
of
different
mobile
phase
compositions
67
(phosphate buffer:acetonitrile) on the resolutions and
separation factor of the nitrophenols
5.3
Effect of compositions of donor and acceptor phases on
78
the enrichment factor
5.4
Optimization of phase volume on LLLMEa
82
5.5
Performance of LLLMEa
83
5.6
Experimental
partition
coefficients,
theoretical
84
recovery, and experimental recovery of 2-nitrophenol
and 3-nitrophenol for three-phase LPME
5.7
Results of the relative recovery for determining 2nitrophenol and 3-nitrophenol in the concentrations of
spiked sea water sample (1.0, 5.0, and 10.0 µg/L) in
intra-day (n = 3) and inter-day measurements (n = 5)
86
18
LIST OF FIGURES
FIGURE NO.
1.1
TITLE
Flow chart of work involved in this study
PAGE
5
19
2.1
Cross section of the hollow fiber inside the aqueous
15
sample during (i) two-phase and (ii) three-phase LPME
2.2
Schematic
representation
of
the
hollow-fiber
17
configuration when two syringe needles are connected to
each other with a hollow fiber. Adapted from ref.
(Psillakis and Kalogerakis, 2003)
2.3
Schematic representation of configuration where a
18
microsyringe is used for supporting the hollow-fiber,
introducing or collecting the acceptor solution, and
serves as a sample introduction device for subsequent
analysis. (Adapted from ref. (Lee et al., 2001))
2.4
Schematic of an ECD
20
2.5
Schematic illustration of the setup for HF-LPME.
28
Adapted from ref. (Bårdstu et al., 2007)
3.1
Basic setup for HF-LPME
34
4.1
GC chromatogram of pesticides studied from direct
44
injection of 20 ppm solution. HP-5 column, 30 m × 0.32
mm i.d., and 0.25 µm film thickness. Detector: Electron
Capture Detection (ECD). Carrier gas: Helium at flow
rate of 1.0 mL/min. Injector temperature: 300
o
C.
Temperature programming: 200 oC held for 1 min, 6
o
C/min to 225 oC, held for 1 min, 4 oC/min to 240 oC,
held for 1 min. Peaks identification: (1) Solvent, (2)
Vinclozolin (internal standard), (3) Quinalphos, (4)
Methidathion, and (5) Hexaconazole
4.2
Effect of extraction solvent on LPME efficiency
46
4.3
Effect of agitation rate on LPME efficiency
47
4.4
Effect of extraction time on LPME efficiency
48
4.5
Effect of sample volume on LPME efficiency
49
4.6
Effect of hollow fiber length on LPME efficiency
50
4.7
Effect of salt on LPME efficiency
51
4.8
Effect of pH on LPME efficiency
52
4.9
Calibration graph of hexaconazole using HF-LPME
54
20
4.10
Calibration graph of methidathion using HF-LPME
55
4.11
Calibration graph of quinalphos using HF-LPME
55
4.12
Example of GC chromatogram of blank. GC conditions:
59
HP-5 column, 30 m × 0.32 mm i.d., and 0.25 µm film
thickness. Detector: Electron Capture Detection (ECD).
Carrier gas: Helium at flow rate of 1.0 mL/min. Injector
temperature: 300 oC. Temperature programming: 200 oC
held for 1 min, 6 oC/min to 225 oC, held for 1 min, 4
o
4.13
C/min to 240 oC, held for 1 min
GC chromatogram from farm water extracts using HF-
59
LPME. HP-5 column, 30 m × 0.32 mm i.d., and 0.25 µm
film thickness. Detector: Electron Capture Detection
(ECD). Carrier gas: Helium at flow rate of 1.0 mL/min.
Injector temperature: 300 oC. Temperature programming:
200 oC held for 1 min, 6 oC/min to 225 oC, held for 1
min, 4
o
C/min to 240
o
C, held for 1 min. Peaks
identification: (1) Solvent and (2) Vinclozolin (internal
standard)
4.14
GC chromatogram from drinking water extracts using
60
HF-LPME. HP-5 column, 30 m × 0.32 mm i.d., and 0.25
µm film thickness. Detector: Electron Capture Detection
(ECD). Carrier gas: Helium at flow rate of 1.0 mL/min.
Injector temperature: 300 oC. Temperature programming:
200 oC held for 1 min, 6 oC/min to 225 oC, held for 1
min, 4
o
C/min to 240
o
C, held for 1 min. Peaks
identification: (1) Solvent and (2) Vinclozolin (internal
standard)
4.15
GC chromatogram of spiked standard pesticides from
farm water sample concentration level of 20.00 µg/L.
HP-5 column, 30 m × 0.32 mm i.d., and 0.25 µm film
thickness. Detector: Electron Capture Detection (ECD).
Carrier gas: Helium at flow rate of 1.0 mL/min. Injector
temperature: 300 oC. Temperature programming: 200 oC
61
21
held for 1 min, 6 oC/min to 225 oC, held for 1 min, 4
o
C/min to 240 oC, held for 1 min. Peaks identification:
(1) Solvent, (2) Vinclozolin (internal standard), (3)
Quinalphos, (4) Methidathion, and (5) Hexaconazole
4.16
GC chromatogram of spiked standard pesticides from
61
drinking water sample concentration level of 20.00 µg/L.
HP-5 column, 30 m × 0.32 mm i.d., and 0.25 µm film
thickness. Detector: Electron Capture Detection (ECD).
Carrier gas: Helium at flow rate of 1.0 mL/min. Injector
temperature: 300 oC. Temperature programming: 200 oC
held for 1 min, 6 oC/min to 225 oC, held for 1 min, 4
o
C/min to 240 oC, held for 1 min. Peaks identification:
(1) Solvent, (2) Vinclozolin (internal standard), (3)
Quinalphos, (4) Methidathion, and (5) Hexaconazole
5.1
Separations of nitrophenols at 220 nm using phosphate
64
buffer (pH 4, 50 mM)-acetonitrile 60:40 (v/v) as eluent.
Separation conditions: C18 column (100 mm × 4.6 mm I.
D.); flow rate: 1.0 mL/min; peaks identification: (1) 2,4dinitrophenol (IS), (2) 3-nitrophenol, (3) 2-nitrophenol;
injection volume:1 µL
5.2
Separations of nitrophenols at different flow rate (a) 1.5
66
mL/min; (b) 1.0 mL/min; (c) 0.7 mL/min without adding
phosphate buffer. Separation conditions: C18 column
(100 mm × 4.6 mm I. D.); mobile phase: acetonitrilewater 60:40 (v/v); peaks identification: (1) 2,4dinitrophenol (IS), (2) 3-nitrophenol, (3) 2-nitrophenol;
UV absorbance at 220 nm; injection volume:1 µL
5.3
Separations of nitrophenols at different mobile phase
compositions (phosphate buffer:acetonitrile). Separation
conditions: C18 column (100 mm × 4.6 mm I. D.) mobile
phase: phosphate buffer:acetonitrile (a) 70:30; (b) 60:40;
(c) 50:50; (d) 40:60; and (e) 30:70 (v/v); flow rate: 1.0
mL/min; peaks identification: (1) 2,4-dinitrophenol (IS),
68
22
(2) 3-nitrophenol, (3) 2-nitrophenol; detection: UV
absorbance at 220 nm; injection volume:1 µL
5.4
Separations
of
nitrophenols
at
different
buffer
70
concentrations (a) 75 mM; (b) 50 mM; (c) 25 mM; (d) 0
mM using phosphate buffer-acetonitrile 60:40 (v/v) as
eluent. Separation conditions: C18 column (100 mm ×
4.6 mm I. D.); flow rate: 1.0 mL/min; peaks
identification:
(1)
2,4-dinitrophenol
(IS),
(2)
3-
nitrophenol, (3) 2-nitrophenol; detection: UV absorbance
at 220 nm; injection volume:1 µL
5.5
Separations of nitrophenols at different pH of mobile
71
phase (a) pH 5.0; (b) pH 4.5; (c) pH 4.0; (d) pH 3.5; (e)
pH 3.0 using phosphate buffer-acetonitrile 60:40 (v/v) as
eluent. Separation conditions: C18 column (100 mm ×
4.6 mm I. D.); flow rate: 1.0 mL/min; peaks
identification:
(1)
2,4-dinitrophenol
(IS),
(2)
3-
nitrophenol, (3) 2-nitrophenol; detection: UV absorbance
at 220 nm; injection volume: 1 µL
5.6
Separations of nitrophenols at different wavelength
72
absorbance (a) 254 nm; (b) 235 nm; (c) 220 nm using
phosphate buffer (pH 4, 50 mM)-acetonitrile 60:40 (v/v)
as eluent. Separation conditions: C18 column (100 mm ×
4.6
mm
I.D.);
identification:
(1)
flow
rate:
1.0
2,4-dinitrophenol
mL/min;
(IS),
peaks
(2)
3-
nitrophenol, (3) 2-nitrophenol; injection volume:1 µL
5.7
Effect of the organic solvent on the extraction of
76
analytes. Extraction conditions: 50 µg/L of analytes;
donor phase: 2.5 mL of solution containing 0.1 M H3PO4
(pH 3); donor phase temperature: 25 oC; acceptor phase:
5 µL of the aqueous solution containing 0.01 M NaOH
(pH 10); extraction time: 20 min
5.8
Effect of the extraction time on the extraction of
analytes. Extraction conditions: 50 µg/L of analytes;
79
23
donor phase: 2.5 mL of solution containing 1.0 M H3PO4
(pH 3); donor phase temperature: 25 oC; organic phase:
20 µL of 1-hexanol; acceptor phase: 5 µL of the aqueous
solution containing 0.1 M NaOH (pH 12); extraction
time: 20 min
5.9
Effect of the NaCl on the extraction of analytes.
80
Extraction conditions: 50 μg/L of analytes; donor phase:
2.5 mL of solution containing 1.0 M H3PO4 (pH 3);
donor phase temperature: 25 oC; organic phase: 20 μL of
1-hexanol; acceptor phase: 5 μL of the aqueous solution
containing 0.1 M NaOH (pH 12); extraction time: 15 min
5.10
Effect of the stirring rate on the extraction of analytes.
81
Extraction conditions: 50 μg/L of analytes; donor phase:
2.5 mL of solution containing 1.0 M H3PO4 (pH 3);
donor phase temperature: 25 oC; organic phase: 20 μL of
1-hexanol; acceptor phase: 5 μL of the aqueous solution
containing 0.1 M NaOH (pH 12); extraction time: 15 min
5.11
The chromatograms of nitrophenols: (A) 5 μg/L standard
solution; (B) sea water blank sample; (C) 5 μg/L spiked
sea water sample
LIST OF ABBREVIATIONS
EU
European Union
MRLs
Maximum Residue Levels
85
24
GC
Gas Chromatography
GC-ECD
Gas Chromatography-Electron Capture Detection
GC-MS
Gas Cromatography Mass Spectrometry
HPLC
High Performance Liquid Chromatography
CE
Capillary Electrophoresis
LLE
Liquid-liquid Extraction
SPE
Solid-phase Extraction
SPME
Solid-phase Microextraction
LPME
Liquid-phase Microextraction
HF-LPME
Hollow Fiber Liquid-phase Microextraction
EPA
Environmental Protection Agency
MDL
Method Detection Limit
Korg/d
Partition ratio at equilibrium between the organic phase and the
donor solution
Ka/org
Partition ratio at equilibrium between the acceptor solution and the
organic phase
AED
Atomic Emission Detector
FT-IR
Fourier transforms infrared spectrometry
TCD
Thermal Conductivity Detector
FID
Flame Ionization Detector
HAAs
Haloacetic acids
LLLME
Liquid-liquid-liquid microextraction
PAA
Phenylacetic acid
PPA
Phenylpropionic acid
AR
Androgen receptor
TMPAH
Trimethylphenylammonium hydroxide
TMSH
Trimethylsulfonium hydroxide
D
Distribution coefficient
SLM
Supported Liquid Membrane
LOD
Limit of Detection
RSD
Relative Standard Deviation
EF
Enrichment factor
Rs
Resolution
k
Retention factor
25
tR
Retention time
α
Separation factor
CHAPTER 1
INTRODUCTION
26
1.1
Research Background
According to the status list of all active pesticide substances on the European
Union (EU) market more than 1100 pesticides are currently registered (Hercegová et
al., 2007). The pesticides industry in Malaysia is made up of about 140 companies,
both multinational and local companies that are involved in manufacturing,
formulating or trading activities. The majority of pesticides are imported as technical
materials, which are then blended, diluted or formulated. However, in recent years an
increasing variety of pesticides are manufactured in Malaysia. Pesticides are widely
used in Malaysia for agricultural activities due to their relatively low price and high
effective ability to control pests, weeds, and diseases (Xiong and Hu, 2008). The
increasing production of pesticides for agricultural and non-agricultural purposes has
caused the pollution of air, soil, ground, and surface water which involves a serious
risk to the environment and as well as human health due to either direct exposure or
through residues in food and drinking water (Pan and Ho, 2004). The need for
accurate determination of pesticides at the trace levels in the environmental samples
is therefore obvious. With the improvement of self-safeguard consciousness and the
development in analytical instruments, levels of pesticides in vegetables and fruits
are currently regulated by international and national organizations and maximum
residue levels (MRLs) have been established in many countries (Pan et al., 2008).
Phenols and substituted phenols are important pollutants in water because of
their wide use in many industrial processes such as the manufacture of plastics, dyes,
drugs, antioxidants, and pesticides. Nitrophenols are formed photochemically in the
atmosphere from vehicle exhaust. They are very toxic and have a diverse effect on
the taste and odour of drinking water at low concentrations, so they are
environmentally of particular interest and concern. Gas chromatography (GC) and
high-performance liquid chromatography (HPLC) are the most common analytical
techniques used for the determination of phenols. However, in GC, derivatization is
needed to analyze phenols in order to avoid peak tailing.
27
Sample preparation is normally required to isolate and concentrate
compounds of interest from the sample matrix, before analysis (Xu et al., 2007).
Ultimately, the concentration of target compounds is enhanced (enrichment) and the
presence of matrix components is reduced (sample clean up). In order to achieve a
low detection limit, an enrichment step should be conducted prior to analysis (Liu et
al., 2007).
Liquid-liquid extraction (LLE) and solid phase extraction (SPE) are the
classical techniques (Liu et al., 2007) for sample pre-concentration and isolation in
analytical chemistry. However, LLE and SPE are time-consuming, generally labourintensive, and requires use of large amounts of expensive high-purity organic
solvents, which are often hazardous. During the last 10 years, some interest has been
focused on the miniaturizing of analytical LLE. The major idea behind this has been
to facilitate automation, to speed up extractions, and to reduce the consumption of
organic solvents.
An attractive alternative pre-treatment method to the traditional technique is
solid-phase microextraction (SPME). SPME was developed by Pawliszyn and
coworkers (Gallardo et al., 2006; Lee et al., 2007). SPME is a solvent-free extraction
technique that incorporates sample pre-treatment, concentration, and sample
introduction into a single procedure (Lambropoulou and Albanis, 2006; BeceiroGonzalez et al., 2007). But the extraction fiber is expensive, fragile, and has a
limited lifetime. Miniaturized LLE or liquid-phase microextraction (LPME), was
first introduced in 1996 by Jeannot and Cantwell (Xu et al., 2007), and was based on
a droplet of organic solvent hanging at the end of a micro syringe needle. Although
hanging drop LPME is very simple and efficient, and reduces the consumption of
organic solvents per sample to a few µL, it is still used only in limited number of
research laboratories. One reason for this may be the low stability of the hanging
drop, which is easily lost into sample during extraction.
Alternatively, miniaturized LLE may be accomplished by hollow fiber
protected LPME. In these systems, the small volume of extracting liquid is contained
within the lumen of a porous hollow fiber. The major advantage of this is that the
28
extracting liquid is mechanically protected, and it is prevented from leaking into the
sample during extraction. This is especially important since LPME is conducted with
strong agitation of the sample to speed up the extractions. Hollow fiber LPME can
be accomplished both in the two- and three-phase modes.
Compared with LLE and SPE, LPME gives a comparable and satisfactory
sensitivity and much better enrichment of analytes, and the consumption of solvent is
significantly reduced by up to several hundred or several thousand times (Xiao et al.,
2007). LPME is applicable to neutral compounds with the two-phase system, and the
acidic and basic substances utilizing either the two- or three-phase concept.
In the present study, the determination of the partition coefficient of selected
analytes using double-phase and triple-phase LPME in water samples was carried
out. Two-phase LPME was applied combined with gas chromatography-electron
captured detection (GC-ECD) for the extraction and preconcentration of pesticides,
while three-phase LPME was applied combined with HPLC-UV for the extraction
and preconcentration of nitrophenols. Different aspects of the extraction procedure
such as the kinds of organic solvent for the immobilization, compositions of the
acceptor and donor phase, the extraction time, the agitation rate, and the volume of
acceptor phase were investigated. The feasibility of this methodology is also
evaluated by determining the enrichment factor, linearity, detection limit and
recovery.
1.2
Statement of Problem
The present work therefore focuses on the development, validation and
application of a HF-LPME method prior to HPLC or GC for the analysis of
pesticides in water samples. It is expected that, the developed method will eliminate
problems related to carry-over effects because hollow fibers utilized were used only
once for every experiment. This single use adds high demands on the reproducibility
of the manufacturing of the hollow fiber. Conventional extraction methods such as
29
SPE and LLE need more time-consuming operation and using specialized apparatus.
In contrast, HF-LPME is inexpensive and there is considerable freedom in selecting
appropriate solvents for extraction of different analytes. Since very little solvent is
used, there is minimal exposure to toxic organic solvent for the operator. The
optimization of several parameters influencing the efficiency of pesticides extraction
such as extraction solvent, agitation speed, ionic strength of the aqueous sample and
exposure time are also explored.
1.3
Research Objectives
The objectives of this research are:
i.
To study two-phase (liquid-[membrane]-liquid) liquid membrane
extraction mechanism based on phase equilibrium and analyte
partition distribution using non-ionizable compounds.
ii.
To study triple-phase (liquid-liquid [membrane]-liquid) liquid
membrane extraction mechanism based on equilibrium and analyte
partition distribution using ionizable compounds.
iii.
To apply and optimize the critical parameters for efficient extraction
and pre-concentration of pesticides and nitrophenols.
Figure 1.1 shows a flow chart of work involved in this research to achieve the
objectives.
1. Separations of selected test compounds (pesticides and nitrophenols)
by GC, and HPLC. Repeatability, reproducibility, linearity, and limit
of detections on the separations of test compounds are carried out.
2. Optimization: using the selected test compounds, the standard (1050 ug/L) is spiked into double-distilled deionized water. Optimization
parameter: type of membrane, organic solvent as acceptor solution,
agitation of sample, salting out effect, volume of donor and acceptor
phase, adjustment of pH, and extraction time.
3. Elucidation of possible mechanisms in liquid-phase microextraction
(LPME). The extraction condition will be examined according to its
linearity range, limits of detection, enrichment factors results, and
extraction recovery study.
PART I
Preliminary
separations and
elucidation of
possible
mechanisms
30
Figure 1.1: Flow chart of work involved in this study
1.4
Scope of Study
The study involves the use of hollow fiber liquid membrane in order to
protect the extracting solvent, thus permitting extraction only on the surface of the
solvent immobilized in the membrane pores. The research is divided into two main
areas; to develop the HF-LPME using two-phase mode, (liquid-[membrane]-liquid)
and three-phase mode, (liquid-liquid [membrane]-liquid).
31
For every mode of extraction, the condition of extraction will be optimized.
The parameters of optimization consisted of type of membrane, organic solvent as
acceptor solution, agitation of sample, salting out effect, volume of donor and
acceptor phase, adjustment of pH, and extraction time.
1.5
Outline of the Thesis
This thesis consists of six chapters. Chapter 1 presents general introduction,
research background, statement of problems, research objectives, and scope of study.
Chapter 2 compiles the literature review and theoretical background on compounds
studied, liquid-phase microextraction (LPME), analytical instruments for LPME, and
application of LPME including double and triple-phase LPME. The procedures for
LPME and chemicals used in this work are presented in Chapter 3. Chapter 4 and
Chapter 5 report the results and discuss about the application of two-phase and threephase LPME in water samples respectively. The concluding Chapter 6 summarizes
this thesis by presenting the overall conclusions and suggestions for future study.
CHAPTER 2
LITERATURE REVIEW
32
2.1
Compounds Studied
In the presented study, environmental samples of water such as drinking
water, farm water and sea water were investigated for pesticides and nitrophenol
compounds. Pesticides have been under intensive study, since the late 1960s. Phenol
compounds have been widely studied for many years, whereas the intensive study of
nitrophenols has only recently begun. From their various sources these compounds
are transported in air and water throughout the world. They enter the soil in rain
droplets and adsorb to particle containing organic matter. Eventually they end up in
seawater and in sea-bottom sediments. Because of the effective transport, these
compounds are found everywhere in the world.
2.1.1
Pesticides
A pesticide is any substance or mixture of substances intended for
preventing, destroying or repelling pests. The term pesticide applies to insecticides,
herbicides, fungicides, and various other substances used to control pests. A
pesticide is also any substance or mixture of substances intended for use as a plant
regulator, defoliant or desiccant. Pests are living organisms that occur where they are
not wanted or that cause damage to crops, humans or other animals. Examples are
insects, mice, and other animals, weeds, fungi, and microorganisms such as bacteria
and viruses. Table 2.1 shows the types of these pesticides and their functions. The
hydrophilicity or lipophilicity of a pesticide has an impact on how it penetrates cell
membranes and other barriers in the target. If effects in both hydrophilic and
lipophilic areas of the target are required the pesticide can be dissolved into suitable
transporter or solvent.
Table 2.1 Types of pesticides and their functions (Fest and Schmidt, 1982)
Type
of
Functions
Examples
33
pesticide
Insecticides
Chemicals used to control insects. It may kill Diazinon,
insect by skin contact or it may have to be fenthion,
swallowed to be effective. Broad spectrum malathion,
insecticides are wide range killers. Not broad methidathion
spectrum insecticides kill specific insects.
Herbicides
Chemicals used to control unwanted plants and Dicamba,
kill or slow growth of some plants. Selective cycloate,
herbicides kill some plants with little or no butylate
injury to other plants. Non-selective herbicides
are toxic to all plants.
Fungicides
Chemicals used to control fungi, which cause Myclobutanil,
molds, rots, and plant disease. All fungicides tebucanazole
work by coming in contact with the fungi.
Rodenticides
Chemicals used to control rats, mice, bats, and Bromdiolone,
other rodents.
Pesticides
include
many
difethialone
chemically
diverse
groups,
such
as
organophosphorous compounds, chlorinated hydrocarbons, carbamates, pyrethrins
and pyrethrinoids, and phenoxyacids. The physico-chemical properties of pesticides
affect their distribution, transport in air and water, absorbance onto particles or
organisms, and toxicity. Pesticides affect their target by different mechanisms. Some
chlorinated pesticides affect the nervous system and disturb signal transport.
Organophosphorous compounds, on the other hand, disturb the function of
acetylcholine esterase, a transmitter in the nervous system. Carbamates are designed
to cause intoxication, where acetylcholine accumulates in the nervous system
causing the target animal to become overactive. Convulsions ensue, then paralysis
and eventually death. Some systemic fungicides that are transported with water
interfere with the formation of DNA by preventing protein synthesis and in that way
also the normal growth of the target fungi. Phenoxyherbicides affect the formation
of auxine hormone in weeds and stop their growth.
34
For the analysis of pesticides in food and environmental samples, sample
preparation is a critical step in the analytical procedure. Routine methods for the
determination of pesticides residue in environmental and food samples typically
involve several sample preparation such as extraction, clean-up, and concentration
before instrumental analysis (Hu and Xiong, 2008).
In general, the desired properties of the sample preparation method for
pesticide residue analysis are inclusion of as high number of pesticides as possible
single procedure (multidimensional assay), recoveries as close as possible to 100%,
sufficient removal of potential interferents from the sample to increase the selectivity
and avoid undesirable matrix effects, increased concentration of the analytes and
hence the sensitivity of the assay, good precision, low costs, and safe. Most
pesticides are volatile and thermally stable and therefore are amenable to GC with
specific detectors or with mass spectrometry (MS).
2.1.2 Nitrophenols
Nitrophenols are man-made pollutants that mostly originate from wastewater
discharges from the dye, pesticide, plastics, paper, and ammunition industries as well
as from various chemical-manufacturing plants (Zhao and Lee, 2001). They are also
found in diesel exhaust particles. Nitrophenols can be formed in the air as a result of
the breakdown of many other manufactured chemicals. Thousands of tons of these
agents are produced yearly by countries around the world.
Registered as priority pollutants by the Environmental Protection Agency
(EPA), they are toxic to aquatic life. Most goes to water and soil, little goes to the
air. Very little is known about the fate of nitrophenols in air. They readily break
down in surface water. They produce immediate toxic effects to the nervous system,
and some reports have implicated them as possible endocrine disruptors.
Concentrations in the range of 4.6-100 μg/L have been found in rain water and in the
tropospheric atmosphere (Bishop and Mitra, 2007).
35
The importance of nitrophenols in the environment calls for sensitive and
reliable methods for their determination. Several EPA methods have been published
for the detection of nitrophenols, which utilize liquid-liquid extraction (LLE) and
derivatization, followed by GC detection. For example, EPA Method 625 couples a
liquid-liquid extraction with gas chromatography-mass spectrometry (GC-MS) to
increase sensitivity and EPA Method 604 uses methylene chloride extraction, solvent
exchange with 2-propanol as a concentration step, followed by GC with flameionization. EPA Method 604 also utilizes derivatization and electron capture
detection GC for qualitative conformation and increased analytical sensitivity. Both
EPA Methods 604 and 625 are used to measure trace organics from municipal and
industrial wastewaters and these methods are able to determine three nitrophenols:
2,4-dinitrophenol (DNP), 2-nitrophenol (2-NP), and 4-nitrophenol (4-NP) (Mitra and
Bishop, 2007). Low method detection limits (MDL) are obtained at the cost of
lengthy and cumbersome extraction-derivatization procedures typically using
methylene chloride, which is toxic and carcinogenic.
Gas chromatography (GC) and high performance liquid chromatography
(HPLC) are the most common analytical techniques used for the determination of
phenols. In GC, derivatization is needed to analyze phenols in order to avoid peak
tailing. Compared with GC, HPLC is a good alternative technique, in which isocratic
or gradient elution can be used to separate these compounds. A preconcentration step
is always necessary prior to liquid chromatography due to the low concentration of
phenols in water (Lee et al., 2001).
2.2
Development in Liquid-phase Microextraction
In recent years, the development of fast, precise, accurate, and sensitive
methodology has become an important issue. Despite the great technological
advances, most analytical instruments cannot handle sample matrices directly and, as
a result, a sample preparation step is commonly introduced. For organic trace
analysis, the step mainly comprises extractions, which serve to isolate compounds of
interest from a sample matrix. Eventually, the concentration of target compounds is
36
enriched and the presence of matrix components is reduced because of sample clean
up (Psillakis and Kalogerakis, 2003).
Traditional liquid-liquid extraction (LLE) is still among the most popular
procedure in routine sample preparation. LLE is recognized as an attractive method
for screening tests of pesticides not only because of its simplicity, robustness,
minimal operator training, efficiency, and a wealth of available analytical data, but
also because of its wide acceptance in many standard methods (Lambropoulou and
Albanis, 2007).
In the present era of “green chemistry”, sample preparation methods such as
LLE suffer from the disadvantage of being time-consuming, expensive, and
requiring large volumes of toxic organic solvents (Jönsson and Mathiasson, 1999). In
contrast, solid-phase extraction (SPE) techniques typically require reduced amounts
of organic solvents relative to LLE, but SPE can be tedious, time-consuming, and
suffer
analyte
breakthrough
when
large
sample
volumes
are
analyzed
(Lambropoulou and Albanis, 2004).
Solid-phase microextraction (SPME), introduced by Arthur and Pawliszyn,
successfully redressed the limitations inherent in the traditional LLE method. It
rapidly gained high popularity, as it incorporated sampling, extraction, concentration
and sample introduction into a single solvent-free step. With SPME, a small amount
of extractant phase, dispersed on a solid support (fiber), is exposed to the sample.
Target analytes partition between the sample matrix and the extractant phase and,
after a well-defined period of time, the fiber is transferred to a gas chromatograph
(GC) or to the SPME-HPLC interface for analysis. More than a decade after its
introduction, the main problems commonly encountered with SPME include the
limited lifetime of the SPME fibers, their relatively fragile nature and the possibility
of carry-over between analyses (Hu and Xiong, 2008). Nevertheless, SPME accounts
for numerous reports in a wide range of applications, such as environmental, food,
clinical and forensic analysis (Psillakis and Kalogerakis, 2003).
37
For the reduction of solvent usage in sample preparation, a miniaturized
format of LLE called liquid-phase microextraction (LPME) has been introduced
(Yamini et al., 2007). Compared with LLE and SPE, LPME gives a comparable and
satisfactory sensitivity and much better enrichment of analytes, and the consumption
of solvent is significantly reduced by up to several hundred or several thousand
times. LPME has emerged as an attractive alternative for sample preparation. LPME
can be performed by using a single drop of solvent or a small length of porous
hollow fiber-protected solvent. The LPME technique is simple, fast, and inexpensive
(Sanagi et al., 2007). However, LPME based on hanging droplets is not very robust,
and the microdrop suspended on the needle of microsyringe is easily dislodged
during extraction, especially the case when samples are stirred vigorously (Xiao et
al., 2007).
Pedersen-Bjergaard and Rasmussen introduced an alternative concept for
hollow fiber-liquid phase microextraction (HF-LPME) using porous hollow fibers
made of polypropylene. HF–LPME was considered to be more robust, because the
organic solvent is contained within the lumen of a porous hollow fiber, so the
organic solvent is not in direct contact with the aqueous solution (Xiao et al., 2007;
Hu et al., 2008). The advantages of using this microporous membrane include
protection of the acceptor phase as well as efficient sample microfiltration through
the pores of the hollow fiber (Psillakis et al., 2008).
In brief, there are two modes used: two-phase HF-LPME and three-phase
HF-LPME. A piece of porous polypropylene hollow fiber is firstly placed in the
aqueous sample and the analytes are extracted by passive diffusion from the sample
into the hydrophobic organic solvent supported by the fiber (acceptor phase, twophase HF-LPME). Or the analytes are extracted through an organic solvent
immobilized in the pores of fiber and further into a new aqueous phase in the lumen
of fiber (acceptor phase, three-phase HF-LPME). In general, two-phase HF-LPME is
applied for analytes with a high solubility in non-polar organic solvents and threephase HF-LPME is applied for basic or acidic analytes with ionizable functionalities.
Moreover, prior to the extraction, it is necessary to adjust the sample pH for both
two- and three-phase LPME, if the analytes are basic or acidic.
38
2.3
Comparison
of
Liquid-Phase
Microextraction
with
Solid-Phase
Microextraction
SPME is a simple, solventless method that can be automated easily. It has been
successfully applied to the determination of a wide variety of volatile and semivolatile analytes using either immersion or headspace SPME sampling. There are
several types of SPME fibers, and selectivity of the method towards classes of
compounds depends on the polarity and the film thickness of the coating phase.
SPME fibers are rather expensive and have a limited life-time, as the length and the
coating character of each SPME fiber many differ from lot to lot; variations in
analytes enrichment may be observed from fiber to fiber. Before using a fiber for the
first time, a thermal conditioning step is required. Even when this step is carefully
done, partial loss of the coating may occur resulting into extra peaks during the
chromatographic analysis, thus affecting the performance of the method. In addition,
sample carry-over between runs has often been reported with SPME and, unless an
extra-cleaning step is introduced in the sampling protocol, the results are invalid
(Chen et al., 2006). Ugland et al. (2003) also reported that, the time to reach
equilibrium applied in bioanalysis of drugs is quite long using SPME; in addition the
recovery is low.
A major advantage of LPME over SPME is that the range of compounds
amenable to this technique can be extended by simply changing from the two-phase
to the three-phase LPME mode and by adjusting the composition of the different
phases. The price of each extraction unit is low and each extraction device is used
only for a single extraction. Thus, a carry-over effect between extractions is
eliminated (Ugland et al., 2003). However, as the hollow fiber segments are cut
manually, variations in length are very likely. In addition, variations in wall
thickness are possible and such fluctuations may alter analyte enrichment, especially
when two-phase approach is used.
39
Another advantage of LPME over SPME is that the small pore size ensures
microfiltration, thus yielding very clean extracts. On the contrary, when the SPME
approach is used in complex matrices, a sample pre-treatment step (such as filtration)
or a modification of the sampling protocol (such as using membrane-protected
SPME) is often required. Damage to the SPME fiber, leading to imprecision in the
measurements, has also been reported when high salt concentrations and/or pH
adjustment are applied to the sample solution. This is not the case for LPME, where
the ionic strength of the sample solution and high or low pH values do not influence
the repeatability of the method or the condition of the hollow fiber. However, in
some cases, it appears that, in two-phase LPME, the presence of salt influences the
extractions kinetic and does not yield a significant increase in the response of the
analytical instrument seen with SPME.
2.4
Hollow-Fiber based Liquid-phase Microextraction
2.4.1 Hollow-Fiber Liquid-phase Microextraction Sampling Modes
There are two sampling modes that can be used with LPME: two-phase and
three-phase. In the two-phase LPME sampling mode, analyte i is extracted from an
aqueous sample (donor phase) through a water-immiscible solvent immobilized in
the pores of the hollow fiber into the same organic solvent (acceptor phase) present
inside the hollow fiber (Fig. 2.1a).
40
(a) Two-phase system
(b) Three-phase system
Figure 2.1 Cross section of the hollow fiber inside the aqueous sample during (a)
two-phase and (b) three-phase LPME
The extraction process of the two-phase LPME for analyte i may be illustrated as
follows:
id ↔ iorg
(2.1)
and is characterized by the partition ratio Korg/d, which is defined as the ratio of the
concentration of analyte i in the organic and donor phase at equilibrium condition.
Successful two-phase LPME requires large distribution ratios for target
analytes. Such Korg/d values correspond to moderately or highly hydrophobic
compounds containing acidic or basic groups, or neutral compounds of similar
hydrophobicity. It should be mentioned here that in two-phase LPME the final
extract is an organic phase, compatible with analytical techniques, such as GC or
HPLC.
In the three-phase LPME sampling mode, analytes i is extracted from an
aqueous solution (donor phase) through the organic solvent immobilized in the pores
of the hollow fiber (organic phase) into another aqueous phase (acceptor phase)
present inside the lumen of the hollow fiber (Fig. 2.1b). The organic phase in this
case serves as a barrier between the acceptor and the donor aqueous solutions,
preventing mixing of these two phases. The three-phase sampling mode is usually
combined with a HPLC or a capillary electrophoresis (CE) system, as the acceptor
phase is aqueous.
41
Overall, the three-phase LPME extraction process for analytes i may be
illustrated as follows:
id ↔ iorg ↔ ia
(2.2)
where i represents the target analytes in the donor phase, organic phase, and acceptor
phase respectively. The three-phase LPME process is characterized by Korg/d and
Ka/org, which are the partition ratios at equilibrium between the organic phase and the
donor phase, and the acceptor solution and the organic phase, respectively.
The overall partition ratio Ka/d between the acceptor and the donor phase can
be written as:
Ka/d = Korg/d . Ka/org
(2.3)
Adjustment of the composition of the donor and acceptor phase is critical for a
successful three-phase LPME. Large Ka/d (»1) can be achieved when the analytes in
the acceptor phase are converted by reaction, such as protonation or complexation, to
species that will have very small affinity for the organic phase. In this way, back
extraction of analytes from the acceptor to the donor solution is prevented. Hence,
the three-phase system will also extract acidic or basic compounds having a low
Korg/d, as long as the Ka/org value are high, thus preserving the requirement for the
large overall distribution ratio. This is very important, because the applicability
range of LPME may be extended to ionizable analytes having a low Korg/d by simply
changing from two-phase to three-phase LPME sampling mode.
2.4.2
Hollow Fiber Configurations
There are essentially two different configurations (Fig. 2.2 and Fig. 2.3) of
arranging the fiber during extraction with the two-phase or three-phase LPME
42
sampling modes. Despite the great potential capabilities of hollow fiber LPME, most
studies carried out have been manually performed in static mode, with several
improvements such as the use of “U-shaped” configuration, because of its benefits
with respect to reproducibility and recovery (Pezo et al., 2007). In Fig. 2.2, two
conventional medical syringe needles are inserted through a septum and the two ends
are connected with a piece of hollow fiber. The length of the fiber usually varies
from 4 cm to 8 cm. Longer fiber lengths (up to 27 cm) can be used and, in such
cases, the hollow fiber is folded around a solid support.
Figure 2.2 Schematic representation of the hollow-fiber configuration when two
syringe needles are connected to each other with a hollow fiber. Adapted from ref.
(Psillakis and Kalogerakis, 2003).
When using either the two-phase or the three-phase technique, the hollow
fiber is initially immersed for several seconds in the organic solvent in order to
immobilize the solvent in its pores. After impregnation, the hollow fiber is immersed
in the donor solution. The acceptor solution is then injected in the lumen of the
hollow fiber with the help of a microsyringe and, once extraction is completed, the
acceptor solution is collected in microvials by applying a small head pressure on the
inlet needle or withdrawn with the help of a microsyringe and submitted for analysis.
43
In the second configuration (Fig. 2.3), only one end of the hollow fiber is
used for injection or collection of the acceptor solution and the other one is left
suspended in the donor-sample solution. This is possible by using the tip of the
needle of a microsyringe as a support for the fiber. The microsyringe in this case
serves to support the hollow fiber, introducing or collecting the acceptor solution,
and as a sample-introduction device for subsequent analysis. In this case, a metal
tube is fitted in the center of the septum to help introduce/lace the fiber in the
septum.
Figure 2.3 Schematic representation of configuration where a microsyringe is used
for supporting the hollow-fiber, introducing or collecting the acceptor solution, and
serves as a sample introduction device for subsequent analysis. (Adapted from ref.
(Lee et al., 2001)).
Introduction or collection of the acceptor solution is affected with the help of
a microsyringe. In general, when configuration (Fig. 2.3) is used, the free end of the
hollow fiber can be flame sealed. Prior to immersion in the donor solution, the fiber
should be dipped in the organic solvent to impregnate it.
2.4 Analysis of Liquid-phase Microextraction Extracts
44
Liquid and solid samples extracted into organic solvent are usually analysed
by LC and GC. LC is selected if the analytes are polar, thermally labile or have high
molecular mass. Otherwise, GC is preferred because of better resolution. Detection
in GC can be achieved with Fourier transform infrared spectrometry (FT-IR), atomic
emission detection (AED) or the more common electron capture detector (ECD),
FID or MS. The most widely used detection methods for LC are UV and different
types of MS. LPME supported by two-phase system is compatible with GC analysis,
while three-phase system are suitable to HPLC due to the aqueous acceptor phase
(Basheer et al., 2004; Pedersen-Bjergaard et al., 2005).
2.5.1
Gas Chromatography
Gas chromatography (GC) is that form of chromatography in which a gas is the
moving phase. The important work was first published in 1952 when Martin and his
co-worker James acted on a suggestion made 11 years earlier by Martin himself in a
Nobel-prize winning paper on partition chromatography. It was discovered that GC
was simple, fast, and applicable to the separation of many volatile materials,
especially petrochemicals, for which distillation was the preferred method of
separation at that time. Simultaneously the demand for instruments gave rise to a
new industry that responded quickly by developing new gas chromatographs with
improved capabilities. Flame Ionization Detector (FID), Thermal Conductivity
Detector (TCD), and Electron Capture Detector (ECD) are the three most important
widely used detectors. TCD and ECD are concentration types and the FID is a mass
flow rate type.
The invention of the ECD (for GC) is generally attributed to Lovelock, based on
his publication in 1961 (McNair and Miller, 1997). It is a selective detector that
provides very high sensitivity for those compounds that “capture electrons”. These
compounds include halogenated materials like pesticide residues (Hercegová et al.,
45
2007). The carrier gas used for the ECD can be very pure nitrogen or a mixture of
5% methane in argon. When used with a capillary column some make-up gas is
usually needed, and it is convenient to use inexpensive nitrogen as make-up and
helium as the carrier gas. A schematic of a typical ECD is shown in Fig. 2.4.
Gas Outlet
Cathode
Radioactive
source (63Ni)
Anode
Gas Inlet
Figure 2.4 Schematic of an ECD
For successful LPME-GC, the acceptor solution in the LPME device should be
an organic solvent capable of direct injection. The solvent should be selected to
provide a high solubility for the analytes of interests (good extraction solvents). It is
also an additional advantage to utilize a solvent of relatively low volatility. When
pesticides containing electronegative functional groups, such as halogens,
phosphorus, and nitro groups pass by detector, they capture some of the electrons
and reduce the current measured between the electrodes.
46
For two-phase LPME, both the pores and the lumen of the hollow fibers were
filled with an organic solvent immiscible with water. The analytes were extracted
from the sample solution and into this solvent through the pores of the hollow fiber.
Normally, approximately 20 μL organic solvent inside was immobilized within the
pores of the hollow fiber, while the volume of organic solvent inside the lumen was
easily transferred to a micro insert application of a small head pressure on the inlet
tube for the hollow fiber. The microextract was directly compatible with GC.
Pan and Ho (2004), described about hollow fiber liquid-phase microextraction
(HF-LPME) and gas chromatographic-electron capture detection (GC-ECD) method
for the determination of six fungicides (chlorothalonil, hexaconazole, penconazole,
procymidone, tetraconazole, and vinclozolin) in 3 mL of water. The enrichment
factors of this method were from 135 to 213. Limits of detection were in the range of
0.004–0.025 µg/L. The relative standard deviations (RSDs) at 0.1 and 5 µg/L of
spiking levels were in the range 3–8%. Recoveries of six fungicides from farm water
at a spiking level of 0.5 µg/L were between 90.7 and 97.6%.
Huang and Huang (2006) studied the potential of dynamic HF-LPME technique
for the extraction of organochlorine pesticides (OCPs) in green tea leaves and in
commercially sold ready-to-drink tea prior to GC-ECD analysis. Good enrichments
were achieved (34-297) with this method, and good repeatibilities of extraction were
obtained with RSDs below 12.57%. Detection limits were much below 1 μg/L for
ready-to-drink tea and much below 1 μg/g for green tea leaves.
2.5.2
High Performance Liquid Chromatography
HPLC is the most powerful of all the chromatographic techniques. It can often
easily achieve separations and analyses that would be difficult or impossible using
other forms of chromatography. HPLC is a technique that has arisen from the
47
application to liquid chromatography (LC) of the theories and instrumentation that
were originally developed for GC (Meyer, 2004).
In analytical HPLC the mobile phase is normally pumped through the column at
a flow rate of 1-5 cm3min-1. If the composition of the mobile phase is constant, the
method is called ‘isocratic’ elution. Alternatively, the composition of the mobile
phase can be made to change in a predetermined way during the separation, which is
called ‘gradient’ elution (Lindsay, 1992).
For three-phase LPME, an aqueous phase was filled inside the lumen of the
hollow fiber. Thus, the analytes of interest were extracted from 1 to 4 mL aqueous
sample, where pH adjustment served to deionize the analytes, into the immobilized
organic solvent within the pores of the hollow fiber, and further into the aqueous
acceptor phase. The organic phase served as a barrier between the sample solution
and the acceptor phase. For the extraction of basic compounds, an aqueous solution
of an acid was utilized as acceptor phase, whereas an alkaline solution was used for
three-phase LPME of acidic components. Since the analytes became ionized
following extraction into the acceptor phase, they were prevented from re-entering
the organic solvent in the pores of the hollow fiber. The aqueous acceptor phase was
easily collected into a micro insert and analyzed directly by HPLC (Gjelstad et al.,
2006).
Huang and Lin (2008) developed a novel liquid-liquid-liquid microextraction
(LLLME) technique for the extraction of sulfonamides from aqueous systems; it
combined with HPLC-UV. Experiments were carried out where sulfonamides
extracted from a donor phase (water sample) into several microlitres of an organic
phase into an acceptor phase and then from the organic phase into an acceptor phase
by LLLME. Subsequent separation and quantitative analyses were performed using
HPLC/UV with 265 nm detection.
48
However, Ma et al. (2008) demonstrated the carrier mediated LPME, as a sample
pretreatment step prior to HPLC for the analysis of illicit drugs. By adding an
appropriate carrier in organic phase, simultaneous extraction and enrichment of
hydrophilic (morphine and ephedrine) and hydrophobic (pethidine) drugs were
achieved. The study has expanded the application of LPME to extract a hydrophilic
substance from complex biological matrices, which is very challenging for the
conventional LPME, with aid of the carrier-mediated transport.
Shariati et al. (2007) developed three-phase LPME technique coupled with
HPLC-UV to determine phenylacetic acid (PAA) and phenylpropionic acid (PPA) in
biological fluids. The proposed three-phase LPME technique is attractive enough
owing to its simplicity, analytical precision, low consumption of organic solvent,
low cost, and short sample preparation time. Compared to the traditional methods,
this method needs only one HPLC syringe for PAA and PPA determinations and the
current design employs small sample volume which is compatible with the biological
sample such as blood. Since a fresh acceptor phase is used for each extraction, there
is no memory effect. Three-phase LPME method has excellent clean-up and for this
method, an enrichment factor up to 110-folds was obtained. Low LOD makes the
three-phase LPME as a method of choice for the measurements of target analytes in
the complex matrices such as the body fluids.
2.6
Application of Two-phase Hollow-Fiber Liquid-Phase Microextraction
Ho et al. (2002) have compared LPME with LLE in terms of mathematical
descriptions of extraction recovery and enrichment. The LPME theoretical
calculations were verified by experimental determination of actual partition
coefficients and by data obtained with LPME in a robust hollow fiber formate. With
hollow fiber LPME operated in the two-phase mode, analytes were extracted from 1
to 4 mL aqueous samples into 25–50 mL of an organic solvent present in the pores
and in the lumen of the porous hollow fibers. Compared with conventional two-
49
phase LLE, two-phase HF-LPME provided substantially higher enrichments for
compounds with relatively large partition coefficients (Korg/d » 500). In contrast,
because of the large volume of organic solvent relative to the sample volume, LLE
provided high recovery and moderate enrichment even for compounds with
relatively low partition coefficients (Korg/d » 5). Thus, two-phase HF-LPME may be
used for substantially enhanced extraction selectivity and enrichment of relatively
hydrophobic analytes as compared with LLE whereas conventional two-phase LLE
is superior for more hydrophilic analytes.
In other work, Lambropoulou and Albanis (2004) developed a novel sample
pre-treatment technique for the determination of trace concentrations of some
organophosphorous insecticides namely dichlorvos, mevinphos-cis, ethoprophos,
chlorpyrifos methyl, phenthoate, methidathion and carbofenothion and one
carbamate (carbofuran) in aqueous samples and applied to the determination of the
selected analytes in environmental water samples. The extraction procedure was
based on coupling polypropylene HF-LPME with gas chromatography by flame
thermionic detection (GC-FTD). Several factors that influenced the efficiency of HFLPME were investigated and optimized including agitation, organic solvent, sample
volume, exposure time, salt additives and pH. The optimized methodology exhibited
good linearity with correlation coefficient = 0.990. The analytical precision for the
target analytes ranged from 4.3 to 11.1 for within-day variation and 4.6 to 12.0% for
between-day variation. The LODs for all analytes were found in the range from
0.001 to 0.072 g/L, well below the limits established by the EC DrinkingWater
Directive (EEC 80/778). Relative recoveries obtained by the proposed method from
drinking and river water samples ranged from 80 to 104% with coefficient of
variations ranging from 4.5 to 10.7%. The reported methodology was easy, rapid,
sensitive, and required small sample volumes to screen environmental water samples
for insecticide residues.
Pan and Ho (2004) applied the HF-LPME technique to the analysis of six
compounds in water samples by using GC-ECD. The method included the use of 3
μL of toluene as extraction solvent, 20 min of extraction time, a pH fixed at 4, a
stirring rate at 870 rpm, and no salt addition. The enrichment factor of this method
50
were from 135 to 213 and limits of detection (LODs) were in the range of 0.0040.025 μg/L. Relative standard deviations (RSDs) between 3 and 8% and recoveries
from farm water between 90.7 and 97.6% were obtained for different spiking levels.
Lambropoulou and Albanis (2004) demonstrated that a selective trace
enrichment of environmental andiandrogen vinclozolin from natural water samples
can be achieved by HF-LPME method. Vinclozolin acts as an antiandrogen by
altering the androgen binding to androgen receptor (AR) and subsequently also
altering the AR-dependent gene expression. The newly developed microextraction
technique has distinct advantages over conventional methods with respect to
extraction time and volume of solvents required, where a high level of precision and
detection limits are readily achieved, exceeding the requirement for vinclozolin
analysis in aqueous samples.
Huang and Chiang (2007) indicated that a selective trace enrichment of
carcinogenic haloethers from natural water samples can be achieved by a dynamic
HF-LPME method. Dynamic HF-LPME coupled with GC-FID and GC-ECD was
used for quantification of toxic haloethers in lake water. The analytes were extracted
from 5 mL of aqueous sample using 4 µL of organic solvent through a porous
polypropylene hollow fibre. The effects on extraction performance of solvent
selection, agitation rate, extraction time, extraction temperature, concentration of salt
added and volumes of solvent for extraction and injection were optimized. The
proposed method provided a good average enrichment factor of up to 231-fold,
reasonable reproducibility ranging from 9 to 12% (n = 3), and good linearity (R2 »
0.9973) for spiked water samples. Method detection limits (MDLs) ranged from 0.55
to 4.30 µg/L for FID and 0.11–0.34 µg/L for ECD (n = 7).
2.7
Application of Three-phase Hollow-Fiber Liquid-Phase Microextraction
Recently, three-phase microextraction was developed to extract ionizable and
chargeable compounds from different aqueous sample. Ebrahimzadeh et al. (2008)
51
used an unsupported liquid organic membrane to separate two aqueous phases, donor
phase and acceptor phase. The pH of donor phase was adjusted to basic and the
acceptor phase was acidic. An ionizable compound was extracted from the donor
phase to the organic phase, and then back extracted to the acceptor phase (Yamini et
al., 2007; Ugland et al., 2003). In the latter study, a hollow fiber was used to support
the liquid membrane. Hollow fiber membrane has proven to be useful for enrichment
of ionizable and charge species, giving a high degree of clean-up and enrichment of
various analytes in different sample (Berhanu et al., 2006).
Common to the published methods are that the analytes are relatively
hydrophobic. The extraction efficiency is governed by the partitioning of the analyte
between the sample matrices and the immobilized solvent and by the partitioning
between the acceptor and the immobilized solvent. Hydrophobic analytes are easily
extracted into organic solvents from aqueous sample solutions. In addition,
hydrophobic ionic analytes have large solubility differences in acidic and basic
aqueous solutions. On the other hand, polar analytes have low solubility in organic
solvents and small differences in their solubility in acidic and basic aqueous
solutions. Polar analytes are therefore difficult to extract by three-phase LPME. The
ability of LPME to exclude extraction of polar analytes contributes to the selectivity
of the method (Ho et al., 2003).
Ugland et al. (2003) demonstrated the behaviour of weak bases
(benzodiazepines and a non-benzodiazepine as a model of substances) in LPME with
varying physicochemical properties where not all experimental parameters can be
optimized. The extraction time, pH, and use of different organic phase were
parameters investigated for their effect on analytes behaviour in the LPME system.
Addition of methanol as a disruptor of the binding between protein and drug in
whole blood had minimal effect on the recovery. Dilution of the blood sample with
phosphate buffer gave satisfying results. This work shows that LPME can be used as
an extraction technique for weak bases.
Pedersen-Bjergaard et al. (2005) developed three-phase HF-LPME to report
the extraction recoveries for 49 new drugs. Based on the recovery data, the
52
application area of three-phase HF-LPME was defined in terms of analyte solubility
and log D (D = distribution coefficient between n-octanol and water) as calculated
by a commercial computer program. The basic drugs were extracted from 1.5 mL
water samples (pH 13) through approximately 15 µL of dodecyl acetate immobilized
within the pores of a porous polypropylene hollow fiber (organic phase), into 15 µL
of 10 mM HCl (acceptor solution) present inside the lumen of the hollow fiber. In
the solubility range 1-5 mg/mL, most drugs were effectively extracted (recovery >
30%), whereas drugs belonging to the solubility range 5-150 mg/mL were all
extracted with recoveries above 30% by three-phase system. The hydrophilic nature
of most drugs with solubilities above 150 mg/mL prevented them from entering the
organic phase, and only those with log D > 1.8 were effectively recovered by threephase HF-LPME.
Bårdstu et al. (2007) investigated three-phase LPME based on a supported
liquid membrane (SLM) sustained in the wall of a hollow fiber with special focus on
optimization of the experimental procedures in terms of recovery and repeatability.
Fig. 2.5 shows the schematic diagram of the setup for HF-LPME. Recovery data for
doxepin, amitriptyline, clomipramine, and mianserin were in the range of 67.879.8%. Within-day repeatability data for the four basic drugs were in the range of
4.1-7.7%. Bårdstu et al. (2007) also reported that, reuse of hollow fibers was found
to suffer from matrix effects due to built-up the analytes in the SLM, whereas
washing of the hollow fibers in acetone was beneficial in terms of recovery,
especially for the extraction of the most hydrophobic substance.
53
Figure 2.5 Schematic illustration of the setup for HF-LPME. Adapted from ref.
(Bårdstu et al., 2007)
A three-phase LPME technique with high selectivity for five aromatic
carboxylic acids and three phenolic compounds has been developed and optimized
by Rodríguez et al. (2008). The system consisted of an acidified donor phase, a thin
layer of solvent inside the wall pores of a hollow fiber, and an alkaline acceptor
phase located inside the hollow fiber. The analysis of the compounds was carried out
by CE with UV detection. Eugenol, thymol, and carvacrol were efficiently extracted
from the aqueous solution using chloropentane as organic phase, with recoveries
from 73.8% to 93.8%. However, using 2-octanone as the organic phase, the
recoveries for the aromatic carboxylic acid compounds ranged from 60.7% to 93.7%
whereas the phenols were not extracted. 2, 6- naphthalene dicarboxylic acid was
found to remain in the organic phase.
In the three-phase system, the consumption of organic solvent is lower than
in the two-phase mode (15-20 µL inside the pores of hollow fiber) and excellent
clean up have been observed even in biological samples, since this technique
includes two simultaneous extractions. In addition, the ionization of analytes in
acceptor solution prevents back-extraction into the organic phase, so high
enrichment factors are achievable (Bonato and Santana, 2008).
CHAPTER 3
EXPERIMENTAL
54
3.1
Materials and Chemicals
Analytes
Chemical structure
Molecular
Weight (g/mol)
log Ko/w
The Accurel Q3/2 polypropylene hollow fiber membrane (600 µm I. D., wall
thickness 200 µm, pore size 0.2 µm) was purchased from Membrana GmbH
(Wuppertal, Germany).
Hexaconazole, methidathion, quinalphos, and vinclozolin (used as internal
standard) were obtained from Dr. Ehrenstorfer GmbH (Augsburg, Germany). 2, 4dinitrophenol, 4-nitrophenol, 3-nitrophenol, and 2-nitrophenol were purchased from
Fluka (Buchs, Switzerland). Their molecular structures and molecular mass are
shown in the Table 3.1 and Table 3.2.
Acetone was from QRec (Germany), HPLC grade acetonitrile and n-hexane
were obtained from J. T. Baker (USA). Double-distilled deionized water of at least
18 MΩ was purified by Nano ultra pure water system (Barnstead, USA). Stock
standard solutions (1000 µg/mL) of the pesticides were prepared in acetonitrile and
were stored in the freezer at about -15 oC. An acetonitrile solution (1000 µg/mL) of
vinclozolin was prepared and used as the internal standard (I.S.). Reagent grade
sodium chloride (Showa Chemicals, Tokyo, Japan), isooctane (JT. Baker), toluene
(JT. Baker), and nonane (Merck, Darmstadt, Germany) were also used for this study.
Table 3.1: Properties of the target analytes
55
Methidathion
(C6H11N2O4PS3)
302.33
2.22
286.1
3.00
314.2
3.90
298.3
4.44
O
S
O
O
N
S
N
P
Vinclozolin
(C12H9Cl2O3)
O
S
O
Cl
N
O
Hexaconazole
(C14H17Cl2N3O)
Cl
O
Cl
OH
C
Cl
CH 2
N
N
N
Quinalphos
(C12H15N2O3PS)
N
O
O
N
P
S
O
Table 3.2: Properties of the nitrophenol compounds
Analytes
Chemical Structure
Molecular
pKa
log Kow
56
Weight (g/mol)
OH
2-nitrophenol
(C6H5N1O3)
O
139.11
7.22
1.79-1.91
139.11
8.30
1.90-2.00
184.11
4.89
1.67-1.73
N+
O-
2-nitrophenol
HO
3-nitrophenol
(C6H5N1O3)
O
N+
O-
3-nitrophenol
2,4-dinitrophenol
(C6H4N2O5)
3.2
Instrumentation
57
3.2.1
Gas Chromatography-Electron Captured Detector
A Perkin Elmer XL gas chromatography (GC) equipped with an electron
capture detector (ECD) (San Jose, United State) was employed for the analysis of
pesticides. The capillary column for GC determination was a non-polar HP-5
column, 30 m × 0.32 mm i.d., and 0.25 µm film thicknesses. Helium gas was used as
the carrier gas with a flow rate of 1.1 mL/min and nitrogen was used as the make-up
gas with a flow rate of 32.4 mL/min. The data were processed using Perkin Elmer
software Turbochrom Navigator version 4.1.
The GC oven temperature program used was as follows: 200 oC held for 1
min, 6 oC/min to 225 oC, held for 1 min, 4 oC/min to 240 oC, held for 1 min. The
injector temperature was 280 oC. The detector temperature was 300 oC. The carrier
gas was helium, maintained at a flow-rate of 1.0 mL/min. The pressure of the carrier
gas was 10.0 psi. Peak areas of the ECD chromatogram signals were used to
demonstrate the effect of parameters of the extraction and the efficiency of the
extraction.
3.2.2
High Performance Liquid Chromatography
The HPLC system series 200 Perkin Elmer, USA used for the analysis
consisted of Vacuum degasser, UV-Vis LC detector and pump series 200. The data
acquisition was controlled by the Chemstation 1100 series software (Agilent
Technologies, Heilbronn, Germany). A Nucleosil 100-5 C18, particle size of 5 μm
and dimensions of 250 mm × 4.6 mm was used as the analytical column.
Separations of prepared pesticides mixture were carried out utilizing mobile
phase made up of two parts, A:B (90:10). Part A consists of a water:acetonitrile
(60:40) mixture. Part B is a 75 mM phosphate buffer (pH 3)-acetonitrile (60:40)
mixture. The concentration of phosphate buffer solution was varied from 0 mM to 75
mM. Mixtures of acetonitrile-water were adjusted to pH 2.5, 3.0, 3.5, 4.0, and 5.0.
58
The water was filtered before mixing with acetonitrile and the mixture was degassed
prior to use. When a change was made to the mobile phase composition, the system
was flushed overnight at a flow rate 0.1 mL/min to condition the column. Before
injecting the analyte into the column, the whole system was stabilized at least 1 hour
to equilibrate the column by flowing the fresh eluent.
1 µL of prepared mixture was injected onto the column in triplicate. UV
detection of analytes was set up at 220 nm and at an eluent flow rate of 1 mL/min.
Retention factor, k; separation factor, α; theoretical number of separation
plates per column length, N/m; plate height, H, and the resolution, Rs were
calculated.
3.3
Sampling and Pretreatment of Samples
Stations for the collection of water samples included UTM farm water and
Danga Bay, Johor Bahru. 1 liter of water sample was taken from the upper 50 cm
surface of the water using polyethylene buckets that had been pre-cleaned and rinsed
with water samples. The samples were then transferred to polyethylene bottles.
Shortly after sampling, the water samples were filtered using 0.45 µm pore-size
membrane filters (NALG, Belgium). Filtered samples were collected in pre-cleaned,
hugh-density polyethylene bottles and acidified to pH 3 with concentrated H3PO4
(Merck, Belgium). Finally, the preserved samples were transported to the laboratory.
3.4
Hollow-Fiber Liquid-phase Microextraction Procedure
59
3.4.1 Extraction of Pesticides based on Two-phase Hollow-Fiber Liquid-phase
Microextraction
The experimental setup is illustrated in Fig. 3.1. Accurel Q3/2 polypropylene
tubular membranes were cut in pieces of 1.8 cm length for LPME experiments. Each
piece of fiber was employed only once to avoid any possibility of carryover. In the
optimized method, before use the hollow fiber was ultrasonically cleaned in acetone
for several minutes in order to remove any contaminants. A 4 μL volume of acceptor
phase (organic solvent) was withdrawn using the syringe. The needle tip was
inserted into the hollow fiber, and the assembly was immersed in the organic solvent
(organic phase) for around 10 s in order for the solvent to impregnate the pores of the
fiber wall.
Syringe and needle for
injection and collection of
acceptor solution
Screw top with
silicone septum
Vial
Acceptor
Porous hollow fiber
(Impregnated with
organic phase)
Sample solution
Stirring bar
Magnetic stirrer
(Cross section of the hollow-fiber inside the aqueous sample during LPME)
Figure 3.1 Basic setup for HF-LPME
Since the hollow fiber was hydrophobic, the fiber channel could be filled
with organic solvent. After this, the fiber was immersed into the sample (4 mL) in a
60
5 mL vial with screw tops and silicone septum by making sure that the whole fiber is
totally immersed in the water phase but above the magnetic stirrer bar, so it would
not be damaged during the stirring.
Finally, the organic solvent (4 μL) in the syringe was injected carefully and
completely into the hollow fiber. The sample was continuously stirred at room
temperature (25 oC) with a magnetic stirrer to facilitate the mass transfer process and
to decrease the time required for the equilibrium to be established. The stirring speed
was fixed at 850 rpm. After 40 min extraction, the analyte-enriched solvent (1 μL)
was withdrawn into the syringe, the fiber segment was removed and the organic
solvent was then injected into the heated injection port of the GC-ECD for further
analysis.
3.4.2
Extraction of Nitrophenol Compounds based on Three-phase HollowFiber Liquid-phase Microextraction
The device setup for three-phase hollow fiber LPME is the same with twophase HF-LPME and shown in Fig. 3.1. A polypropylene hollow fiber with an
internal diameter of 600 µm, with a 200 µm wall thickness, and with 0.2 µm pores
was used for the immobilization of the organic phase and for housing the acceptor
solution. The whole fiber was cut into small segments with length of 1.5 cm and one
end of each hollow fiber was heat-sealed using a sealer. A 10 µL syringe (Hamilton,
Bondaduz, Switzerland) was employed to introduce the acceptor phase, into the
lumen of the hollow fiber, and also to inject extracted analytes at the end of
extraction into the HPLC sample injection loop. Each piece of hollow fiber and tip
were used only once and discarded after extraction.
HF-LPME was performed according to the appropriate procedure: 2.5 mL
deionized water was spiked with 2-nitrophenol and 3-nitrophenol and placed in a 5
mL extraction vial. The sample was made acidic with 3 mL 50 mM phosphate buffer
(pH 4). 1-hexanol was immobilized in the wall of the hollow fiber, by dipping the
hollow fiber in the actual organic solvent (1-hexanol). The organic solvent filled the
61
pores of the hollow fiber wall. Subsequently, 5 µL of 0.1 M NaOH (pH 12) was
injected into the lumen of the hollow fiber with a microsyringe as acceptor solution.
The samples were agitated at 850 rpm for a period of 40 min during extraction. After
extraction, the syringe-fiber assembly was taken out of the solution. A 1 µL volume
of acceptor solution was withdrawn from the fiber, and then injected into the HPLC
system.
3.5
Method Optimization: Parameters affecting Hollow-Fiber Liquid-phase
Microextraction
3.5.1
Organic Solvent
Choosing the most suitable organic solvent is very important for achieving
good selectivity of the target compounds. The choosing of solvent should be based
on comparison of selectivity, extraction efficiency, and the level of toxicity. This
practical is very critical for both the two-phase and the three-phase HF-LPME. The
polarity of the organic phase should be similar to polypropylene fiber so that it can
be easily immobilized within the pores of the hollow fiber. Solvent impregnation
gives greatly affects to the performance of HF-LPME since extraction occurs on the
surface of the immobilised solvent (Psillakis and Kalogerakis, 2003; Yazdi and
Es’haghi, 2005).
In addition, it should have a low solubility in water so as to prevent
dissolution into the aqueous phase, and a low volatility, which will restrict solvent
evaporation during extraction. For two-phase HF-LPME, the organic solvent should
be selected in order to maximize the Korg/d partition coefficient, thus a very good
solvent for the analyte immiscible with water should be selected. The solvent also
provide high solubility for target analytes, and when coupled to a GC, it should have
a tremendous GC performance. Even though, in the three-phase HF-LPME, the
selected solvent should ensure high values for Korg/d and Ka/org in order to avoid
62
analyte trapping in the organic phase and consequently reduced analyte recovery. In
the present work, for the optimization of two-phase LPME, nonane, isooctane, and
toluene were chosen while for three-phase LPME, isooctane, toluene, nonane, nhexane, 1-octanol, and 1-hexanol were selected as organic solvent.
3.5.2
Agitation Rate
Stirring of donor phase accelerates the kinetics of extraction by decreasing
the thickness of the Nernst diffusion film around the interface between the phases.
This phenomenon enhances diffusion of analytes from donor to acceptor phase
(Shariati et al., 2007) and improves the repeatability of the extraction method. With
stirring, the analyte molecules are able to pass through the interfacial layer of the
hollow fiber more easily there by increasing the efficiency of extraction. In this
work, the analytes were agitated in the range of 340 rpm to 1190 rpm.
3.5.3
Extraction Time
Extraction efficiency depends on the period of extraction time. HF-LPME,
like SPME, is dependent on equilibrium rather than exhaustive extraction. The
amount of analyte transferred into the acceptor phase should reach its maximum
value when this equilibrium is established. The present study investigated the
extraction time for both of the methods in the range of 5 to 50 min.
3.5.4
Influence of Hollow Fiber Length
Fiber length is an important factor from the viewpoint of analyte recoveries
and sample preparation time. While the selection of longer fiber can be shorten
sample preparation time due to higher surface area and higher evaporation speed. On
the contrary, the use of a shorter fiber severely increases sample preparation time and
hence increases the sample loss or decomposition (Marsin et al., 2009). Thus, in the
present work, the length of the fiber chosen for optimization process was 0.7 cm, 1.0
cm, 1.5 cm, 1.8 cm, and 2.0 cm.
63
3.5.5
Volume of Donor and Acceptor Solutions
For two-phase and three-phase HF-LPME systems, the sensitivity of the
method can be improved by increasing the volume ratio of the donor-to-acceptor
phase. Thus, if, e.g., the volume of donor solution is 3 mL and the volume of
acceptor solution is 3 μL, analytes may be enriched by a factor up to 100. This is one
of the major factors making HF-LPME very attractive, especially for relatively small
sample volumes, as similar enrichment may not be obtained with SPE or LLE.
In the two-phase HF-LPME procedure, sample volumes in the range 3-7 μL
are usually immobilized in the lumen of the HF. However, in the three-phase LPME
procedures, higher volumes of acceptor phase, ranging between 10 and 25 μL, are
commonly used, for the sake of combination with the HPLC instruments.
3.5.6
Salt Addition
Depending on the nature of the target analytes, addition of salt to the sample
solution can decrease their solubility and enhance their partitioning into organic
solvent because of the salting-out-effect. In both the two-phase and three-phase HFLPME methods, the effect of adding salt to the donor solution prior to extraction has
been investigated. The results showed that, depending on the target analytes, an
increase in the ionic strength of the aqueous solution may have various effects upon
extraction: it may enhance, not influence, or even limit extraction (Psillakis and
Kalogerakis, 2003). Furthermore, caution should be taken when high salt
concentrations are used in the sample matrix, since under these conditions, in
combination with the agitation of the sample, the formation of air bubbles was
promoted, increasing the incidents of drop loss and/or dislodgement of organic
solvent (Lambropoulou and Albanis, 2007).
3.5.7
Adjustment of the pH
64
Adjustment of the pH can enhance extraction, as dissociation equilibrium is
affected together with the solubility of the acidic or basic target analytes. In twophase or three-phase HF-LPME, there are many reports where pH changes in the
donor aqueous solution resulted in higher analyte preconcentration (Psillakis and
Kalogerakis, 2003; Lambropoulou and Albanis, 2007).
In two-phase LPME, the adjustment of pH aims at the pH zone that promotes
the formation of the molecular state of the target compound and not the ionic one
(e.g., for acidic pesticides, the pH should differ at least 2-3 units from their pKa). The
donor and the acceptor phases’ pH play an important role in three-phase HF-LPME.
For ionizable analytes, protonation is the most common reaction utilized to enhance
Ka/d and to facilitate analyte extraction from donor to acceptor phase. The pH
difference between the donor and acceptor phases can promote the transfer of the
analytes from the donor to the acceptor phase. For practical applications, pH should
differ from the pKa values of the analytes by at least 2 units (Shariati et al., 2007).
For example, when investigating basic analytes, pH in the sample should be
high (preferably 3 units higher than the pKa value of the analytes), whereas the
acceptor phase should be acidic (preferably 3 units below the pKa value of the
analytes) and vice versa for acidic analytes.
3.6
Fundamental Theory
65
3.6.1
Determination of Partition Coefficient, Recovery, and Enrichment
Factor in Two-phase Hollow-Fiber Liquid-phase Microextraction
In two-phase HF-LPME system, the target analytes are extracted from the
aqueous sample (donor solution) and directly into the organic solvent (acceptor
solution) present both in the porous wall and inside the lumen of the hollow fiber.
This process may be illustrated with the following equation:
Asample ↔ Aacceptor
(3.1)
where A represents the target analytes in the sample (donor phase) and acceptor
respectively. The extraction process depends on the partition coefficient between the
acceptor solution and the donor solution (Ka/d) defined by Eq. (3.2):
Ka/d = Ceq,a / Ceq,d
(3.2)
where Ceq,a is the analyte concentration at equilibrium in the acceptor solution and
Ceq,a is the analyte concentration at equilibrium in the donor solution.
The extraction recovery, R, and the enrichment, E, in the two-phase HFLPME system can be calculated by Eqs. (3.3) and (3.4):
R = (100 . Ka/d . Va) / (Ka/d . Va + Vd)
(3.3)
E = (Vd . R) / (Va.100)
(3.4)
where Va is the volume of acceptor solution and Vd is the volume of donor solution.
3.6.2
Determination of Partition Coefficient, Recovery, and Enrichment
Factor in Three-phase Hollow-Fiber Liquid-phase Microextraction
66
In three-phase HF-LPME, the analytes are extracted from the aqueous
sample, through the organic solvent impregnated in the pores of the hollow fiber, and
further into the aqueous acceptor solution present inside the lumen of the hollow
fiber. This process can be illustrated by the following equation:
Asample ↔ Aorg ↔ Aacceptor
(3.5)
The total extraction process is affected by both the partition coefficient between the
organic phase and the donor phase (Korg/d) and between the acceptor solution and the
organic phase (Ka/org) defined by Eqs. (3.6) and (3.7):
Korg/d = Ceq,org / Ceq,d
(3.6)
Ka/org = Ceq,a / Ceq,org
(3.7)
where Ceq,org is the analyte concentration at equilibrium in the organic phase, Ceq,d is
the analyte concentration at equilibrium in the donor solution, and Ceq,a is the analyte
concentration at equilibrium in the acceptor solution. The partition coefficient
between the acceptor phase and the donor phase, which can be considered as the
overall driving force for the extraction, is calculated as the product of Korg/d and
Ka/org:
Ka/d = Ceq,a / Ceq,d = Korg/d / Ka/org
(3.8)
For acidic compounds, a high Ka/d is achieved by adjusting pH into acidic
region in the sample (typically pH 3 to deionize the analytes) and by alkaline the
acceptor phase (typically pH 12 to ionize the analytes).
The extraction recovery, R, in the three-phase HF-LPME system can be
calculated by Eq. (3.9):
67
R = (100 . Ka/d . Va) / (Ka/d . Va + Korg/d . Vorg + Vd)
(3.9)
where Va is the volume of acceptor phase, Vorg is the volume of organic phase
immobilized in the pores of the hollow fiber, and Vd is the volume of the donor
phase. Eq. (3.4) can also be used to calculate the enrichment factor for three-phase
HF-LPME.
From Eq. (3.4), it clearly shows the relation between analyte enrichment and
ratio of Vd/Va for both two-phase and three-phase HF-LPME. HF-LPME will
provide high analyte enrichments, since the ratio Vd/Va is normally high. For
example, the volume of donor phase (sample solution) is 2 mL and the volume of
acceptor solution is 25 μL, analytes may be enriched by factor up to 80. So, this is
the major factors making HF-LPME very attractive, especially for relatively small
sample volumes, as similar enrichment may not be obtained with SPE or LLE.
68
CHAPTER 4
APPLICATION OF DOUBLE-PHASE LIQUID PHASE
MICROEXTRACTION TO THE ANALYSIS OF PESTICIDES
4.1
Preliminary Investigation on the Separation of Pesticides Using GC-
ECD
Prior to optimization of analyte separation, standard solutions of individual
pesticide 20 ppm were injected into the GC-ECD system in triplicates to determine
their respective retention times before injecting them in their mixture form.
4.1.1 Conformation of Individual Peaks
The retention times of the pesticides obtained from the chromatograms are
presented in Table 4.1. All the four pesticides were eluted within 10 min and were
well resolved under the temperature program setting used.
Fig. 4.1 shows the GC chromatogram from direct injection of the standard
solution mixture. Based on the chromatogram obtained, it was noted that the elution
order for the pesticides was directly proportional to the molecular weight of the
analytes. Vinclozolin with the lowest molecular weight was first eluted across the
column followed by quinalphos, methidathion, and hexaconazole.
69
Table 4.1 Retention times of pesticides studied
Pesticides
Retention Time, tR (min)
Vinclozolin
5.7 ± 0.1
Quinalphos
7.6 ± 0.1
Methidathion
8.0 ± 0.1
Hexaconazole
8.6 ± 0.1
(1)
(2)
P
(5)
(4)
(3)
.
Figure 4.1 GC chromatogram of pesticides studied from direct injection of 20 ppm
solution. HP-5 column, 30 m × 0.32 mm i.d., and 0.25 µm film thickness. Detector:
Electron Capture Detection (ECD). Carrier gas: Helium at flow rate of 1.0 mL/min.
Injector temperature: 300 oC. Temperature programming: 200 oC held for 1 min, 6
o
C/min to 225 oC, held for 1 min, 4 oC/min to 240 oC, held for 1 min. Peaks
identification: (1) Solvent, (2) Vinclozolin (internal standard), (3) Quinalphos, (4)
Methidathion, and (5) Hexaconazole
4.2
Optimization of Hollow Fiber-Liquid Phase Microextraction in Aqueous
Samples
70
This work focused on the application of HF-LPME to the analysis of selected
pesticides namely quinalphos, methidathion, and hexaconazole. The effects of
different experimental parameters such as solvent type, extraction time, agitation
rate, volume of the sample, length of the fiber, salting-out effect, and pH on the yield
of the microextraction were studied in detail.
Measurements of analytes were carried out using the GC-ECD system. The
temperature program developed was capable of giving a good separation of the
analytes studied. In this stage, all analyses were programmed in triplicate and
internal standard was used. All optimization experiments were performed with the
water samples spiked with 50 µg/L of the target analytes.
4.2.1
Effect of Organic Solvent on Analytes Behavior in the Two-phase LPME
System
The type of organic solvent immobilized in the pores of the hollow fiber is an
essential consideration for efficient analyte preconcentration. As in LLE, the
principle “like dissolves like” also applies to LPME. The solvent should be of low
volatility to prevent evaporation, low viscosity to ensure rapid mass transfer, low
polarity to ensure compatibility with the hollow fiber (Bårdstu et al., 2007), and to
prevent leakage into the sample. Also, the solvent should provide high distribution
constants for the target analytes (Psillakis and Kalogerakis, 2003; Lambropoulou and
Albanis, 2007). High polarity solvents are not suitable for LPME due to solvent
leakage via the hollow fiber pores (Basheer et al., 2004).
Based on the preliminary experiments, three organic solvents namely nonane,
isooctane, and toluene were evaluated. Four mililiters of water samples were spiked
with all the pesticides at 50 μg/L, and the extraction time and agitation rate were 20
min and 680 rpm, respectively. It was found that, the extraction efficiency decreased
71
in order of toluene, isooctane, and nonane (Fig. 4.2). Toluene extracted the analytes
better than the other two solvents. Besides, toluene was easily immobilized onto the
fiber within seconds and its solubility in water is low (Basheer et al., 2004). Toluene
was thus utilized as the extraction solvent for the rest of the study.
Figure 4.2 Effect of extraction solvent on LPME efficiency
4.2.2 Effect of Agitation Rate on Analytes Behavior in the Two-Phase LPME
System
The extraction of HF-LPME can be enhanced by agitation of the sample
solution, thereby reducing the time required to attain thermodynamic equilibrium
especially for the higher molecular mass analytes (Lambropoulou and Albanis,
2005). For SDME, stirring speeds above 600 rpm often result in dislodgement of the
acceptor phase and difficulties in analyte quantification, especially with prolonged
exposure time. In membrane-based LPME, the organic solvent is sealed and
protected by the hydrophobic HF membrane, so it is easier to handle and it can
tolerate higher stirring speeds (Lambropoulou and Albanis, 2007).
Different stirring rates were tested to determine the efficiency of extraction of
the analytes. The experiments were carried out at stirring speeds ranging from 340 to
1190 rpm. Figure 4.3 indicates that partitioning of the analytes into organic solvents
generally increases with stirring speed from 340 to 850 rpm.
72
Figure 4.3 Effect of agitation rate on LPME efficiency
However, the values levelled off or slightly decreased at stirring speeds of
higher than 850 rpm. Higher spinning rate exceeding 1190 rpm were not evaluated
due to excessive air bubbles on the surface of the hollow fiber, which could lead to
poorer precision and possible experimental failure. Subsequent experiments were
therefore performed with a stirring speed of 850 rpm.
4.2.3
Effect of Extraction Time Profile on Analytes Behavior in the Two-Phase
LPME System
Mass-transfer is a time-dependent process and its rate is reduced the closer
the system reaches equilibrium conditions. Whether extraction is exhaustive or
works as a pre-concentration technique, equilibrium is attained only after exposing
the acceptor solution to the sample for a long period of time. For method
optimization, it is therefore important to establish the extraction-time profiles of
target compounds so as to configure the time after which equilibrium is attained in
practice. Although longer exposure times of the acceptor solution generally result in
increased extraction efficiency, it is not always practical to apply extended extraction
times (Psillakis et al., 2003).
73
In this work, a series of exposure times were investigated by extracting
aqueous solution containing 50 ppb of each analytes. It was found that the analytical
signals increased dramatically within 50 min of extraction time (Fig. 4.4). However,
it is not normally considered practicable to maintain a long extraction time for
equilibrium to be attained (Chen et al., 2006), because the potential for solvent loss
due to dissolution increases with time. Thus, the extraction time for all subsequent
experiments was standardized at 20 min.
Figure 4.4 Effect of extraction time on LPME efficiency
4.2.4
Effect of Sample Volume on Analytes Behavior in the Two-Phase LPME
System
The influence of sample volume on the peak was studied in the range of 3-5
mL. The experimental was carried out using three different volumes due to the
limited sample (donor phase). The results shown in Fig. 4.5 indicate that the
analytical signal for most of the target analytes virtually increases with sample
volume in the range of 3-4 mL and decreases after 4 mL. Hence, a sample volume of
4 mL was applied for subsequent experiments.
74
Figure 4.5: Effect of sample volume on LPME efficiency
4.2.5
Effect of Length of Hollow Fiber on Analytes Behavior in the Two-Phase
LPME System
Fiber length is an important factor from the viewpoint of analytes recoveries
and sample preparation time. The selection of longer fiber can shorten sample
preparation time due to higher surface area and higher evaporation speed. On the
other hand, the use of a shorter fiber severely increases sample preparation time and
hence increases the sample loss or decomposition (Liu et al., 2007).
The effect of hollow fiber length on the HF-LPME extraction efficiency was
studied by exposing the hollow fiber of different lengths impregnated with 3 µL of
toluene to the sample solution for 10 min. The sample stirring rate was maintained at
850 rpm. The results (Fig. 4.6) indicate that the peak area increased with increasing
in hollow fiber length from a value of 0.7 to 1.8 cm. Reduction in extraction
efficiency was observed when using 2.0 cm hollow fiber. On the basis of these
results, 1.8 cm length hollow fibers were used for the optimization of the extraction
parameters.
75
Figure 4.6 Effect of hollow fiber length on LPME efficiency
4.2.6 Effect of Salt Addition on Analytes Behavior in the Two-Phase LPME
System
In two-phase LPME method, the effect of salt addition to the donor solution
prior to extraction has been widely investigated. Depending on the target analytes, an
increase in the ionic strength of aqueous solution may have various effects on
extraction: it may enhance, not influence or limit an extraction (Psillakis and
Kalogerakis, 2003).
To investigate the salt effect on the HF-LPME, the extraction was performed
with 4 mL sample solution containing 0, 1.5, 2.5, 3.5, and 5% (w/v) NaCl as shown
in Fig. 4.7. The extraction efficiency increased for all the analytes at salt
concentration below 1.5% (w/v), resulting, presumably, from the salting-out effect.
The extraction efficiency decreased for solution with 2.5% (w/v) salt. The
anomalous effect of NaCl on the extraction of quinalphos, methidathion, and
hexaconazole is probably due to two factors. The first is the salting-out effect, which
decreases the solubility of the analytes and thus increases their partitioning into
organic solvents. Secondly, salt dissolved in the solutions may vary the physical
properties of the Nerst diffusion layer on the hollow fiber, and thereby reducing the
diffusion rate of the analyte through the static layer to the thin organic layer on the
fiber (Chiang and Huang, 2007; Liu et al., 2007).
76
Furthermore, caution should be taken when high salt concentrations are used
in the sample matrix, since under these conditions; the formation of air bubbles is
promoted, increasing the incidents of drop loss or dislodgement of organic solvent.
Since the extraction efficiency for most of the analytes decreased beyond 1.5% NaCl
(w/v), all subsequent experiments were conducted at this concentration.
Figure 4.7 Effect of salt on LPME efficiency
4.2.7
Effect of pH on Analytes Behavior in the Two-Phase LPME System
It is common practice to acidify natural aqueous samples shortly after
collection in order to limit both abiotic and biotic degradation of organic
contaminants. However, changing the pH will change the ionization form of certain
analytes and will thereby affect their water solubility and extractability. Thus, pH
also appears to be an important factor for the LPME extraction of pesticides from
water samples.
There are many reports where, in two-phase or three-phase LPME, pH
changes in the donor aqueous solution results in higher analyte pre-concentration. In
two-phase LPME, the adjustment of pH aims at the pH zone that promotes the
formation of the molecular state of the target compound and not the ionic one (e.g.,
77
for acidic pesticides, the pH should differ by at least 2–3 units from their pKa)
(Lambropoulou and Albanis, 2007).
In the present study, the effect of pH by additioning of H3PO4 upon
pesticides extractability with a HF-LPME was also investigated by varying the pH
values from 2 to 8 (Fig. 4.8). The extractions of quinalphos and methidathion were
slightly increased when pH value was increased from 2 to 3, decreased gradually
from 3 to 5, and then decreased to almost zero at pH 6. The extraction of
hexaconazole was affected obviously from pH 2 to 3, decreased sharply at pH 3 to 5,
and then increased gradually from pH 5 to 7. At pH 8, the extraction yield of all the
target analytes decreased slightly. This effect may be attributed to the higher
hydrolysis rate of pesticides in alkaline media (Lambropoulou and Albanis, 2004).
Good extraction efficiencies for all of the pesticides was observed at pH 3. Thus, pH
3 was chosen as the optimum pH.
Figure 4.8 Effect of pH on LPME efficiency
4.3
Performance of HF-LPME Procedure
On the basis of the experiments discussed above, the following conditions
have been selected to evaluate the performance of the method: toluene as organic
78
solvent, 4 mL water samples, 20 min extraction time, 850 rpm stirring rate, 1.8 cm
HF length, 1.5 % (w/v) NaCl content, and pH 3.
The analytical performance of the proposed method HF-LPME was validated
through the determination of linearity, limits of detection, and precision with
vinclozolin as internal standard (Table 4.2).
Table 4.2 Summary of the performance of the developed methods
Analytes
Linear range
R2
RSDa
(µg/L)
a
LOD
Recovery
(ng/L)b
(%)
Methidathion
10-50
0.996
5.71
2.57
107.33
Quinalphos
10-50
0.997
7.02
2.55
103.55
Hexaconazole
10-50
0.996
4.46
2.80
98.15
Repeatability was investigated at concentration-quinalphos: 40.00 µg/L,
methidathion: 30.00 µg/L, hexaconazole: 50.00 µg/L
b
LODs were calculated as three times the standard deviation of three replicate runs
of spiked sample
4.3.1
Precision
The repeatability, expressed as relative standard deviations (RSDs) for three
replicate analyses, spiked at the following level of the target compounds: 50.00 µg/L
hexaconazole, 40.00 µg/L quinalphos, and 30.00 µg/L methidathion, ranged from
4.46% to 7.02%. Good repeatability was obtained for each of the analytes using HFLPME method.
4.3.2
Linearity
The square root of the peak areas was used as the basis for the calculations to
compensate for the quadratic response of the ECD detector. The peak area of
79
pesticides relative to the area of internal standard was plotted against the
concentration of the pesticides to generate calibration curves.
Viclozolin was added as internal standard into each of the mixture prepared
to maintain the consistency of the integrated data. Vinclozolin was selected as an
internal standard because it was well resolved and eluted near the standard mixture
of pesticides studied which is important to minimize the error in measurement the
peak area.
The calibration graphs for pesticides studied (hexaconazole, methidathion,
and quinalphos) are shown in Fig. 4.9, Fig. 4.10, and Fig. 4.11, respectively. The
linearity of the method was tested over a range of 10.00-50.00 µg/L using five
concentration levels and analyzing each level in triplicate. The correlation coefficient
for the three target pesticides varied from 0.996 to 0.997. All the calibration graphs
showed good linearity correlation coefficients of greater than 0.99.
Figure 4.9 Calibration graph of hexaconazole using HF-LPME
80
Figure 4.10 Calibration graph of methidathion using HF-LPME
Figure 4.11 Calibration graph of quinalphos using HF-LPME
4.3.3 Detection Limits
Limit of detection (LODs) were calculated as three times the standard
deviation of three replicate runs of spiked samples at lowest concentration of the
analytes. The LODs of the pesticides were 0.00257 µg/L for methidathion, 0.00255
µg/L for quinalphos, 0.00280 µg/L for hexaconazole.
Hexaconazole with higher LOD is the compound with the higher response in
the GC system but the higher detection limit can be attributed to lower enrichment
factor compared to methidathion and quinalphos. Nevertheless, the analytical
method for all target compounds meets the EU regulatory levels for drinking water
of 0.01 µg/L (Lambropoulou and Albanis, 2005).
81
Overall, the detection limits achieved by the proposed method were better or
comparable to other published extraction techniques for pesticides. Pan and Ho
applied the HF-LPME technique to the analysis of six compounds of pesticides in
water sample by using GC/ECD and found that the LODs were in the range of
0.004-0.025 µg/L (Pan and Ho, 2004).
4.4
Partition Coefficient and Enrichment Factor
At equilibrium, the partition coefficient for the analytes in the two-phase
system is as in Eq. 3.2 or can be rewritten as
K = Co,eq / Ca,eq
(4.1)
where Co.eq is the equilibrium concentration of analytes in the organic phase and
Ca.eq is the equilibrium concentration of analytes in the aqueous phase. According to
the mass balance relationship
Ct.Va = Co.eq.Vo + Ca.eq.Va
(4.2)
where Ct is the original concentration of analytes, Vo is the volume of the organic
solvent, and Va is the volume of the aqueous sample. Thus, LPME is an equilibrium
process and can be very effective for analyte enrichment because of the increase in
the volume ratio of donor solution and acceptor phase.
In an analysis, it is desired to evaluate the initial concentration of analyte in
an aqueous sample solution based upon the measured value of the analyte
concentration in the organic phase at sampling time. Similar to SPME, an exhaustive
extraction does not occur in LPME. Instead the analyte is partitioned between the
bulk aqueous solution and the organic acceptor phase (Lambropoulou and Albanis,
2004).
82
The enrichment factor, EF, can be defined as the ratio of the equilibrium
concentration of analytes in the organic phase to the original concentration of
analytes in the aqueous phase (Lee and Chung, 2008). EF can be calculated from the
Eqs. (4.1) and (4.2).
EF = 1/(Vo/Va + 1/K)
(4.3)
Eq. (4.3) shows that in order to obtain high EF, a low Vo/Va ratio and high
partition coefficient are required. LPME may be applied to medium-polarity and
non-polar analytes and to those whose polarities can be reduced before extraction.
The major factor affecting EF was the partition coefficient. The EF was found to be
directly related to the Korg/d.
In the present study, the amount of quinalphos, methidathion, and
hexaconazole in the hollow fiber following extraction was calculated using the peak
area ratio measurements and the calibration curves of the standards only. Data of the
enrichment factor for the target analytes were determined and are presented in Table
4.3 which indicates the correlation between the enrichment factor and partition
coefficient, Korg/d. The enrichment factor of hexaconazole was slightly lower
compared to methidathion and quinalphos. This could be due to the higher solubility
of hexaconazole in the donor phase thus reducing its partition into the organic phase.
Table 4.3
Analytical performance in terms of enrichment factor and partition
coefficient
a
Enrichment was determined at concentration all the analytes: 50.00 µg/L.
4.5
Application of HF-LPME in Real Water Samples
Pesticides
Methidathion
Quinalphos
Hexaconazole
Enrichment factor, EFa
343
331
314
Korg/d
605
584
554
83
In order to investigate the applicability of the proposed trace enrichment
microextraction method, two water samples of different origins were studied. The
optimum conditions obtained in the optimization of two-phase HF-LPME was then
applied to the determination of pesticides in selected water samples. Performance of
the overall method for a drinking (underground water) sample and farm water
sample (UTM farm, Skudai, Johor) was compared with that for distilled water.
Common components of farm water samples, such as humic acid and inorganic salts,
could reduce the applicability of the method in analysis by decreasing the recovery
(Pan and Ho, 2004).
The analysis of blank using deionized water was carried out with the
optimum conditions using two-phase HF-LPME. Fig. 4.12 shows an example of
chromatogram of blank. The smooth baseline of chromatogram shows that the
hollow fiber was clean and free from any chemical contaminants or memory effects.
In all quantitative applications of HF-LPME, memory effects must be considered,
i.e., incomplete transfer of analyte from the membrane to the acceptor phase. The
memory effect not only causes a reduction in the recovery but also leads to carryover effects in sequential extractions in automated system (Jönsson and Mathiasson,
1999).
The HF-LPME-GC-ECD method was applied to the analysis of pesticides in
water samples. The identification of pesticides from the water samples was based on
their retention time compared to the standard pesticides studied. However, none of
the target analytes were found despite the low detection limits obtainable with the
HF-LPME-GC-ECD method. This suggests that sampled waters were not polluted
by the pesticides studied. Figure 4.13 and Figure 4.14 show chromatograms for farm
water sample and drinking water sample respectively, after being extracted using
HF-LPME-GC-ECD method.
mV
84
Figure 4.12 Example of GC chromatogram of blank. GC conditions: HP-5 column,
30 m × 0.32 mm i.d., and 0.25 µm film thickness. Detector: Electron Capture
Detection (ECD). Carrier gas: Helium at flow rate of 1.0 mL/min. Injector
temperature: 300 oC. Temperature programming: 200 oC held for 1 min, 6 oC/min to
225 oC, held for 1 min, 4 oC/min to 240 oC, held for 1 min
(1)
(2)
Figure 4.13 GC chromatogram from farm water extracts using HF-LPME. HP-5
column, 30 m × 0.32 mm i.d., and 0.25 µm film thickness. Detector: Electron
Capture Detection (ECD). Carrier gas: Helium at flow rate of 1.0 mL/min. Injector
temperature: 300 oC. Temperature programming: 200 oC held for 1 min, 6 oC/min to
225 oC, held for 1 min, 4 oC/min to 240 oC, held for 1 min. Peaks identification: (1)
Solvent and (2) Vinclozolin (internal standard)
(1)
(2)
85
Figure 4.14 GC chromatogram from drinking water extracts using HF-LPME. HP-5
column, 30 m × 0.32 mm i.d., and 0.25 µm film thickness. Detector: Electron
Capture Detection (ECD). Carrier gas: Helium at flow rate of 1.0 mL/min. Injector
temperature: 300 oC. Temperature programming: 200 oC held for 1 min, 6 oC/min to
225 oC, held for 1 min, 4 oC/min to 240 oC, held for 1 min. Peaks identification: (1)
Solvent and (2) Vinclozolin (internal standard)
To demonstrate the applicability of the method for water analysis, farm water
sample and drinking water sample were spiked with the studied analytes. The water
samples were spiked with 20 ppb pesticides and analyzed in triplicate by the
proposed method by adjusting the pH of the samples to pH 3 and the representative
chromatograms can be seen in Figure 4.15 and Figure 4.16, respectively. The
recoveries were calculated for triplicate samples. Table 4.4 and Table 4.5 show the
relative recoveries of the spiked samples respectively. In general, the relative
recovery ranged between 96-102% with RSD of less than 7%.
mV
(1)
(5)
(2)
(4)
(3)
86
Figure 4.15
GC chromatogram of spiked standard pesticides from farm water
sample concentration level of 20.00 µg/L. HP-5 column, 30 m × 0.32 mm i.d., and
0.25 µm film thickness. Detector: Electron Capture Detection (ECD). Carrier gas:
Helium at flow rate of 1.0 mL/min. Injector temperature: 300 oC. Temperature
programming: 200 oC held for 1 min, 6 oC/min to 225 oC, held for 1 min, 4 oC/min to
240 oC, held for 1 min. Peaks identification: (1) Solvent, (2) Vinclozolin (internal
standard), (3) Quinalphos, (4) Methidathion, and (5) Hexaconazole
mV
(1)
(5)
(2)
(4)
(3)
Time (min)
Figure 4.16 GC chromatogram of spiked standard pesticides from drinking water
sample concentration level of 20.00 µg/L. HP-5 column, 30 m × 0.32 mm i.d., and
0.25 µm film thickness. Detector: Electron Capture Detection (ECD). Carrier gas:
Helium at flow rate of 1.0 mL/min. Injector temperature: 300 oC. Temperature
programming: 200 oC held for 1 min, 6 oC/min to 225 oC, held for 1 min, 4 oC/min to
240 oC, held for 1 min. Peaks identification: (1) Solvent, (2) Vinclozolin (internal
standard), (3) Quinalphos, (4) Methidathion, and (5) Hexaconazole
Table 4.4 Application performance of HF-LPME in farm water samples
Pesticides
Linear dynamic
Detected
range
amount
UTM Farm water
87
(µg/L)
2
µg/L
R
Relative recoveries %
RSD %
Quinalphos
10-60
0.9975
19.4
99.7
3.1
Methidathion
10-60
1.0000
20.1
101.5
2.4
Hexaconazole
10-60
0.9970
19.4
96.7
3.2
Table 4.5 Application performance of HF-LPME in drinking water samples
Pesticides
Linear dynamic
Detected
range
amount
Drinking water
(µg/L)
µg/L
R2
Quinalphos
10-60
0.9978
Methidathion
10-60
Hexaconazole
10-60
Relative recoveries %
RSD %
19.2
96.8
2.9
0.9998
19.1
96.2
2.0
0.9998
19.8
99.7
4.5
In spiked technique, high recoveries in analysis of samples usually
correspond to high accuracy. The relative recovery is calculated from percentage of
standard pesticides spiked expected area by comparison with standard solution of
pesticides at the same level. More than 96% relative recoveries were obtained for
most of the analytes in the two samples indicating that there was only a minor
influence of sample matrix on the extraction since there was not much difference in
the relative recoveries between deionized water, farm water, and drinking water.
Minor matrix effect on the LPME extraction is probably attributed to the selectivity
of the hollow fiber because of the pores in its wall. The pores of the hollow fiber
function as a filter in “dirty” sample, since large molecules, which can also be
soluble in the organic solvent, will not be extracted (Lambropoulou and Albanis,
2005).
88
CHAPTER 5
APPLICATION OF THREE-PHASE LIQUID PHASE MICROEXTRACTION
TO THE ANALYSIS OF NITROPHENOLS
5.1
Preliminary Investigation on the Separation of Nitrophenols using
HPLC-UV
Prior to optimization of analyte separation, solutions of 2,4-dinitrophenol
(IS), 3-nitrophenol, and 2-nitrophenol were injected individually at 10 ppm each to
determine their respective retention times before injecting them in the mixture form.
5.1.1
Conformation of Individual Peaks
Separation of mixture of nitrophenols analyzed using HPLC at a flow rate of
1.0 mL/min on C18 column is illustrated in Fig. 5.1. All the analytes had a longer
retention time compared to the solvent peak (acetonitrile) which appeared before 2,4dinitrophenol (IS), 3-nitrophenol, and 2-nitrophenol (Fig. 5.1). This was due to
acetonitrile having a smaller molecular weight compared to the target analytes and a
higher polarity and solubility in water.
Separation here is due to differences in the partition coefficients of solutes
between the stationary liquid and the liquid mobile phase. It was noted that the
elution order for the nitrophenols were based on log Kow value. Generally,
compounds with high molecular weight were eluted later than compounds with low
molecular weight. 2,4-dinitrophenol has the highest molecular weight compared to 2nitrophenol and 3-nitrophenol (see Fig. 3.2).
89
However, 2,4-dinitrophenol was eluted first among the nitrophenols studied.
This phenomenon can be explained by comparing their log Kow value that represents
their relative polarity, which means compound with lower log Kow value is more
polar than compound with higher log Kow value. Elution follows log Kow value which
is more polar compound eluted first, least polar compound will be eluted last because
of the interaction with the column. Compounds with the same polarity have higher
column affinity and thus, will be more retained on the column.
UV abs
Figure 5.1 Separations of nitrophenols at 220 nm using phosphate buffer (pHmin4, 50
mM)-acetonitrile 60:40 (v/v) as eluent. Separation conditions: C18 column (100 mm
× 4.6 mm I. D.); flow rate: 1.0 mL/min; peaks identification: (1) 2,4-dinitrophenol
(IS), (2) 3-nitrophenol, (3) 2-nitrophenol; injection volume: 1 µL
5.1.2
Evaluation of Flow Rate on the Separation Efficiency of Nitrophenols
The criteria used for the optimization of the separation of the analytes were
the baseline resolution of each individual analyte together with no peak splitting in
the shortest possible time. The retention time, tR, and retention factor, k, are
summarized in Table 5.1. Retention factor is degree of interaction of a solute with
the stationary phase in a given chromatographic system. It was observed that the
retention factor for the analytes decreased significantly with increasing flow rate
from 0.7 to 1.5 mL/min. For example, the retention factor of 3-nitrophenol was 3.23
at 0.7 mL/min and it decreased to 2.39 at 1.5 mL/min. In HPLC separations,
retention factors of solutes were kept between 1 and 10, since if k values are too low
it is likely that the solutes may not be adequately resolved, and for high k values the
analysis time may be too long.
90
Table 5.1 Effect of different flow rates on the retention time and retention factor of
the nitrophenols
Compounds
0.7
2,4-dinitrophenol
3-nitrophenol
2-nitrophenol
3.20
4.25
5.51
2,4-dinitrophenol
3-nitrophenol
2-nitrophenol
2.40
3.23
4.13
Flow Rate, mL/min
1.0
Retention Time, tR
2.28
3.27
4.01
Retention Factor, k
2.00
2.85
3.58
1.5
1.52
1.93
2.38
1.98
2.39
3.00
From Fig. 5.2, flow rate of 1.5 mL/min gave the shortest time for eluting all
the analytes but the 2 peaks of 2,4-dinitrophenol and 3-nitrophenol were not well
separated. At 0.7 mL/min, the symmetry of all the peaks worsened and broadening
were obvious. In isocratic separations, the peak area is strongly influenced by the
flow rate of the mobile phase. If the analyte flows slowly through the detector cell, it
will be broad but its height is only determined by the concentration at the peak
maximum; therefore its area will be large. If the peak is running fast, its area will be
small. Good peak shapes were observed and the separations improved considerably
at the flow rate of 1.0 mL/min.
1
2
3
1
(a ) 1.5
2
3
(b ) 1.0
91
Figure 5.2 Separations of nitrophenols at different flow rate (a) 1.5 mL/min; (b) 1.0
mL/min; (c) 0.7 mL/min without adding phosphate buffer. Separation conditions:
C18 column (100 mm × 4.6 mm I. D.); mobile phase: acetonitrile-water 60:40 (v/v);
peaks identification: (1) 2,4-dinitrophenol (IS), (2) 3-nitrophenol, (3) 2-nitrophenol;
UV absorbance at 220 nm; injection volume:1 µL
5.1.3 Evaluation of Mobile Phase Compositions on the Separation Efficiency
of Nitrophenols
When HPLC method is developed, an acceptable degree of separation for all
the components of interest in a sample is desirable in a reasonable time. The
purposes of the separation is to separate all the components with some defined
minimum resolution, or to achieve the maximum resolution in a certain time, or may
only be interested in resolving one or two components from the others, and so on. To
some extent, the separation can be influenced by operating parameters such as
column, temperature, and flow rate, but, by far the most important factor that
controls the separation is the composition of the mobile phase.
The variation of separation resolution and separation factor with mobile
phase compositions were also studied in this work. The resolution, Rs, and separation
factor, α, are summarized in Table 5.2. For each compound studied, the Rs and α
value were inversely proportional to the percentages of organic modifier used in the
mobile phase. For example, with acetonitrile-buffer 70:30 (v/v) the resolution was
1.00 and increased to 2.17 with acetonitrile-buffer 30:70 (v/v). From the results of
92
this study, all the compositions of mobile phases gave good resolutions based on the
theoretical aspects; resolution of Rs = 1.0 represents an overlap of 2% (or 98%
separation) and a resolution of Rs = 1.25 represents 99.4% separation or almost
complete separation. However, the overall results showed that the resolution for each
pair of components were mostly greater than 1.25 and hence, sufficient for accurate
quantification.
Table
5.2
Effect
of
different
mobile
phase
compositions
(phosphate
buffer:acetonitrile) on the resolutions and separation factor of the nitrophenols
Nitrophenols
Mobile Phase Composition:Phosphate buffer (75 mM, pH 4)-Acetonitrile (v/v)
30:70
40:60
50:50
60:40
70:30
Resolution, Rs
R1,2
1.00
1.33
1.36
1.75
2.17
R2,3
0.89
1.30
2.00
2.85
3.65
Separation Factor, α
α1,2
1.57
1.56
1.64
1.66
1.69
α2,3
1.41
1.43
1.51
1.58
1.62
From the chromatogram in Fig. 5.3, it was observed that the retention time
for the nitrophenols increased significantly with decreasing percentage of organic
modifier used in the mobile phase ranging from 70 to 30 (v/v). The test compounds
were eluted within 5 min. The optimized composition of mobile phase was observed
when using buffer-acetonitrile 60:40 (v/v). The results show the peaks obtained were
sharp and no overlapping peak.
UV abs
2
(a) Buffer - Acetonitrile 70:30
1
3
2
1
3
93
Figure 5.3
Separations of nitrophenols at different mobile phase compositions
(phosphate buffer:acetonitrile). Separation conditions: C18 column (100 mm × 4.6
mm I. D.) mobile phase: phosphate buffer:acetonitrile (a) 70:30; (b) 60:40; (c) 50:50;
(d) 40:60; and (e) 30:70 (v/v); flow rate: 1.0 mL/min; peaks identification: (1) 2,4dinitrophenol (IS), (2) 3-nitrophenol, (3) 2-nitrophenol; detection: UV absorbance at
220 nm; injection volume: 1 µL
5.1.4
Evaluation of Concentration and pH of Phosphate Buffer on the
Separation Efficiency of Nitrophenols
Buffers are required in ion-exchange chromatography and frequently also in
reversed-phase chromatography. If ionic or ionizable compounds need to be
separated, it is often, although not always, an imperative necessity to run the
chromatography at a well-defined pH. The analytes can be forced either into the nondissociated or the ionized form, depending on the mode of separation. The chosen
94
pH of the buffer must be two units apart from the pKa of the compounds of interest in
order to get the vast majority of the molecules in one single form (PedersenBjergaard and Rasmussen, 2005).
For acidic analytes, buffer pH 2 units lower than pKa gives non-dissociated
species and buffer pH 2 units higher than pKa gives ionized species (anions). On the
other hand, buffer pH 2 units higher than pKa gives non-dissociated species and
buffer pH 2 units lower than pKa gives ionized species (cations).
The effect of concentration of the buffer on the separation is shown in Fig.
5.4. With no buffer in the mobile phase, the peaks appeared to give the worst
performance because of the high degree of broad broadening. Too low an ionic
strength has no effect; i. e. the buffer capacity is low. Concerning the ionic strength
of a buffer was initially prepared a 25 mM solution. The results demonstrated that
the performance of the peaks improved compared to without using buffer. As a
whole, as the concentration of buffer in mobile phase was increased, the resolutions
of the nitrophenols improved. However, the use of high ionic strength (75 mM) can
cause the buffer solution to remain within the HPLC system and will damage the
column. Thus, 50 mM buffer-acetonitrile (60:40) (v/v) was used as the mobile phase.
2
UV abs
1
3
(a) 75 mM
1
2
3
95
Figure 5.4 Separations of nitrophenols at different buffer concentrations (a) 75 mM;
(b) 50 mM; (c) 25 mM; (d) 0 mM using phosphate buffer-acetonitrile 60:40 (v/v) as
eluent. Separation conditions: C18 column (100 mm × 4.6 mm I. D.); flow rate: 1.0
mL/min; peaks identification: (1) 2,4-dinitrophenol (IS), (2) 3-nitrophenol, (3) 2nitrophenol; detection: UV absorbance at 220 nm; injection volume: 1 µL
The pH value of a buffer is clearly defined in water. When the mobile phase
is prepared it is necessary to adjust the pH of the aqueous solution correctly. At an
acidic pH, weak acids are not dissociated. In their undissociated form, an interaction
with a C18 phase is possible and the retention time is large. However, bases are
protonated in their ionic form, an interaction with the hydrophobic surface is
unlikely, and they elute early in the chromatogram. The situation is reversed in the
alkaline range; bases are not dissociated and acids are ionized. This is the reason
why the selectivity of acids and bases can be manipulated by the pH of the mobile
phase and the results are illustrated in Fig. 5.5.
1
2
3
UV abs
(a) pH 5.0
1
2
3
(b) pH 4.5
2
1
3
96
Figure 5.5 Separations of nitrophenols at different pH of mobile phase (a) pH 5.0;
(b) pH 4.5; (c) pH 4.0; (d) pH 3.5; (e) pH 3.0 using phosphate buffer-acetonitrile
60:40 (v/v) as eluent. Separation conditions: C18 column (100 mm × 4.6 mm I. D.);
flow rate: 1.0 mL/min; peaks identification: (1) 2,4-dinitrophenol (IS), (2) 3nitrophenol, (3) 2-nitrophenol; detection: UV absorbance at 220 nm; injection
volume: 1 µL
The acceptor solution was made up of NaOH at varying concentrations. The
nitrophenols exist mainly as ions in the basic matrix. In order to address this issue,
different pH of mobile phase was made up. The buffer in the mobile phase
neutralizes the NaOH at the injection point due to its buffering capacity and the ionic
nitrophenols in the extract revert to the neutral state (Zhu et al., 2001). From the
chromatogram, pH 4.0 gave the high enrichment factor. As a result, 50 mM bufferacetonitrile (60:40) (v/v) (pH 4.0) was used as the mobile phase and applied to the
further experiment.
5.1.5
Evaluation of Wavelength Absorbance on the Separation Efficiency of
Nitrophenols
97
Ultraviolet-visible spectrophotometer (UV-VIS) detector is the most
commonly used type of detector as it can be rather sensitive, has a wide linear range,
is relatively unaffected by temperature fluctuations and is also suitable for gradient
elution. It records compounds that absorb ultraviolet or visible light. Such
compounds include alkenes, aromatics, and compounds having multiple bonds
between C and O, N or S.
UV detects the difference in the amount of light absorbed by an analyte
transiting the detector relative to the amount of light absorbed by the solvent.
However, from the chromatogram below (Fig. 5.6), the best wavelength was at
absorbance 220 nm.
UV abs
1
2
1
2
3
(a)
254 nm
3
(b) 235 nm
2
1
3
(c) 220 nm
min
Figure 5.6 Separations of nitrophenols at different wavelength absorbance (a) 254
nm; (b) 235 nm; (c) 220 nm using phosphate buffer (pH 4, 50 mM)-acetonitrile
60:40 (v/v) as eluent. Separation conditions: C18 column (100 mm × 4.6 mm I.D.);
flow rate: 1.0 mL/min; peaks identification: (1) 2,4-dinitrophenol (IS), (2) 3nitrophenol, (3) 2-nitrophenol; injection volume:1 µL
5.2
Basic Principle of Extraction using Three-Phase LPME System
A liquid-liquid-liquid phase microextraction (LLLME) technique involves
three phases. Analyte is extracted from an aqueous solution donor (donor phase)
through the organic solvent immobilized in the pores of the hollow fiber (organic
phase) into another aqueous phase (acceptor phase) present inside the lumen of the
hollow fiber (Huang and Lin, 2008). The organic phase in this case serves as a
barrier between the donor and the acceptor aqueous solutions, preventing mixing of
these two phases. The three-phase sampling mode is usually combined with HPLC or
CE, as the acceptor phase is aqueous (Pedersen-Bjergaard and Rasmussen, 2005).
As an example for an analyte such as A, the extraction process is:
Adonor phase ↔ Aorganic phase ↔ Aacceptor phase
(5.1)
And the initial amount of analyte n, is equal to the sum of individual amounts of
analyte present in the all phases during the whole extraction process.
ni = nd + norg + na
(5.2)
where nd, is the amount of analyte in the donor phase (sample), norg, is the amount of
analyte in the organic phase and na, is the amount of analyte in the acceptor phase,
respectively.
At equilibrium, Eq. (5.2) can be written as:
CiVd = Ceq,dVd + Ceq,orgVorg + Ceq,aVa
(5.3)
where Ci, is the initial concentration of analyte, Ceq,d, Ceq,org, and Ceq,a are analyte
concentration in the donor, organic, and acceptor phase at equilibrium condition,
2
respectively. Vd, Vorg, and Va, are the volume of the donor, organic, and acceptor
phase, respectively.
In the three-phase LPME system, partition coefficients between the phases
are:
Korg/d = Ceq,org / Ceq,d
(5.4)
Ka/org = Ceq,a / Ceq,org
(5.5)
Ka/d = Ceq,a / Ceq,d = Ka/org / Korg/d
(5.6)
and
By rearrangement the Eq. (5.2) at equilibrium, can be written as:
neq,a = (Ka/dVaCiVd)/(Ka/dVa + Ka/dVa + Vd)
(5.7)
The relative recovery can be expressed as:
R = 100neq,a / CiVd = 100 Ka/dVa / (Ka/d + Korg/dVorg + Vd)
(5.8)
The enrichment factor (EF), can be calculated as follows:
EF= Ca / Ci = VdR / 100Va
(5.9)
or
EF = 1/[(Ka/org / Korg/d) + (Korg/aVorg)/Vd +(Va / Vd)
(5.10)
3
In the three-phase LPME system, the volume of the organic solvent immobilized in
the pores of the hollow fibre (Vorg) is small, and Eq. 5.10 simplified to:
EF = 1 / (1/Ka/d + Va / Vd)
5.3
(5.11)
Optimization of Three-phase LPME in Aqueous Samples
5.3.1 Effect of Organic Solvent on Analytes Behavior in the Three-Phase
LPME System
The choice of proper organic solvent in the three-phase HF-LPME is of great
importance. The selection criteria for a suitable organic solvent are as follows: (1)
solubility of the analytes in the organic solvent should be higher than in the donor
phase but lower than in the acceptor phase so that the analytes could be transferred
from the donor into the acceptor phase with high extraction efficiency (Chen et al.,
2006), in order to achieve a high degree of recovery of analytes in the acceptor phase
(Ugland et al., 2003); (2) it should be of low volatility to prevent solvent loss and be
immiscible with water to avoid dissolution during the extraction and serve as a
barrier between the donor and acceptor phases; (3) solvents with low viscosity are
preferred due to larger diffusion coefficient of the analytes; and finally (4) the
solvent should have no toxicity.
Preliminary trials showed that the use of non polar solvents such as isooctane,
toluene, nonane and n-hexane considerably decreased the extraction efficiency. It
was seen that by increasing the solvent polarity, the extraction efficiency was
improved and in the presence of alcohols, the best extraction efficiencies were
obtained. The reason was that the analytes had relatively high polarity and could be
extracted into the solvents with sufficient polarity. Therefore, the solvents with
higher polarity were examined as organic phase. The solvents such as 1-octanol and
4
1-hexanol were examined. Fig. 5.7 shows that 1-hexanol and 1-octanol provided the
best extraction efficiencies for the target analytes. 1-hexanol was therefore selected
as the extraction solvent due to better chromatographic behavior compared to 1octanol.
Figure 5.7 Effect of the organic solvent on the extraction of analytes. Extraction
conditions: 50 μg/L of analytes; donor phase: 2.5 mL of solution containing 0.1 M
H3PO4 (pH 3); donor phase temperature: 25oC; acceptor phase: 5 μL of the aqueous
solution containing 0.01 M NaOH (pH 10); extraction time: 20 min
5.3.2
Effect of Composition of Donor and Acceptor Phases on Analytes
Behavior in the Three-Phase LPME System
The donor and the acceptor phases’ pH play an important role in three-phase
HF-LPME. For ionizable analytes, protonation is the most common reaction utilized
to enhance Ka/d and to facilitate analyte extraction from donor to acceptor phase. The
extraction efficiency is governed by the partitioning of the analyte between the
sample matrices and the immobilized solvent and by the partitioning between the
acceptor phase and the immobilized solvent (Ho et al., 2003). The pH difference
between the donor and acceptor phases is one of the major parameters that promote
the transfer of the analytes from the donor to the acceptor phase. For practical
applications, pH should differ from the pKa values of the analytes by at least 2 units.
5
Since our analytes were weak acidic compounds (pKa of 2-nitrophenol: 7.22 and pKa
of 3-nitrophenol: 8.36), the donor solution should be sufficiently acidic to maintain
the neutrality of the analytes and consequently reduce their solubility within the
donor phase. Also, the acceptor phase should be alkaline in order to promote
dissolution of the acidic analytes.
Experiments were conducted to optimize the pH of both donor and acceptor
solutions. First, the effect of the donor phase’s pH on extraction efficiency was
investigated. The concentration of H3PO4 in the donor phase was varied between
0.01 and 1.0 M and at the same time, the concentration of NaOH was also varied
between 0.01 and 1.0 M. In the donor phase, when using the highest pH, the
extraction efficiency decreased because protonation reaction was not complete and a
large portion of the analytes existed in ionic form. Thus, in further experiments, pH
of the donor phase was adjusted in acidic medium. Like the pH of donor phase, the
pH of the acceptor phase can affect extraction efficiency as well. Acceptor pH should
be adjusted to pH values where the analytes are highly charged (Rodríguez et. al.,
2008) and thereby fully trapped in the acceptor phase (Ugland et. al., 2003).
Zhu and co-workers (2001) reported on their research that 0.5 M of NaOH in
the acceptor phase gave the highest enrichment factor for nitrophenols. Generally,
the use of higher concentration of NaOH is too basic for the C18 column and will
damage the column. However, in the present investigation for the separation of target
compounds potassium dihydrogen phosphate buffer (pH 4) was used that will
neutralize the NaOH at the injection point due to its buffering capacity.
The effect of compositions of donor and acceptor phases is given in Table
5.3. The results show that 1.0 M H3PO4 in the donor phase and 0.1 M NaOH in the
acceptor phase provided the highest enrichment factor for 2-nitrophenol and 3nitrophenol and was chosen as optimum concentration for the subsequent
experiments.
6
Table 5.3 Effect of compositions of donor and acceptor phases on the enrichment
factor
Enrichment Factor
0.01 M H3PO4
0.1 M H3PO4
0.5 M H3PO4
1.0 M H3PO4
5.3.3
0.01 M
0.1 M
0.5 M
1.0 M
NaOH
NaOH
NaOH
NaOH
3-nitrophenol
10.93
176.84
9.37
40.86
2-nitrophenol
17.38
229.10
8.36
66.73
3-nitrophenol
15.02
148.74
63.08
41.91
2-nitrophenol
7.95
227.09
109.92
84.67
3-nitrophenol
35.81
145.67
110.86
63.24
2-nitrophenol
30.43
142.11
125.01
86.71
3-nitrophenol
26.51
190.66
124.07
70.68
2-nitrophenol
30.77
243.74
131.48
67.74
Effect of Extraction Time on Analytes Behavior in the Three-Phase
LPME System
Mass transfer is a time-dependent process, so the function of extraction time
was studied here. However, when choosing an extraction time profile, precise timing
becomes essential for good precision.
The extraction experiment was performed on a standard mixture solution in
1.0 M H3PO4 (50 μg/L of each nitrophenol). 5.0 µL of the acceptor phase was 0.1 M
NaOH (pH 12), the impregnation solvent was 1-hexanol, and the stirrer speed was
fixed at 510 rpm. Fig. 5.8 shows the effect of extraction time on the enrichment
factor. The enrichment factor increased rapidly with the extraction time to reach a
maximum at 15 min before decreasing slowly. Thus, 15 min was selected as the
optimum extraction time.
7
Figure 5.8 Effect of the extraction time on the extraction of analytes. Extraction
conditions: 50 μg/L of analytes; donor phase: 2.5 mL of solution containing 1.0 M
H3PO4 (pH 3); donor phase temperature: 25 oC; organic phase: 20 μL of 1-hexanol;
acceptor phase: 5 μL of the aqueous solution containing 0.1 M NaOH (pH 12);
extraction time: 20 min
5.3.4
Effect of Salt Addition on Analytes Behavior in the Three-Phase LPME
System
Depending on the nature of the target analytes, addition of salt to the sample
solution can decrease their solubility and therefore enhance extraction because of the
salting-out effect. The present study investigated the effect of adding salt to the
donor solution prior to extraction using three-phase HF-LPME method.
The results showed that, depending on the target analytes, an increase in the
ionic strength of the aqueous solution may have various effects upon extraction: it
may enhance, not influence, or even limit extraction. NaCl was added to the donor
solution to study the possibility of salting-out effect. No significant increase in
enrichment was achieved when 1%, 5%, 10%, 15%, 20%, and 30% NaCl was added.
8
Figure 5.9 Effect of the NaCl on the extraction of analytes. Extraction conditions:
50 μg/L of analytes; donor phase: 2.5 mL of solution containing 1.0 M H3PO4 (pH
3); donor phase temperature: 25 oC; organic phase: 20 μL of 1-hexanol; acceptor
phase: 5 μL of the aqueous solution containing 0.1 M NaOH (pH 12); extraction
time: 15 min
5.3.5
Effect of Agitation Rate on Analytes Behavior in the Three-Phase LPME
System
The importance of agitation of the whole extraction assembly in LPME has
been highlighted in several publications (Gjelstad et al., 2007). Stirring of donor
phase accelerates the kinetics of extraction by decreasing the thickness of the Nernst
diffusion film around the interface between the phases. This phenomenon enhances
diffusion of analytes from donor to acceptor phase and improves the repeatability of
the extraction method. Agitation also reduces the time required to reach
thermodynamic equilibrium and induces convection in membrane phase. Using
larger magnets for sample agitation increases extraction but they are not suitable due
to the production of vortex flow in the membrane phase that can reduce stability of
the acceptor microdrop.
9
Application of the maximum stirring speed was found to promote the
formation of air bubbles that tended to adhere to the surface of the hollow fiber, thus
accelerating solvent evaporation and introducing imprecision in the measurements.
Some studies have shown that extraction efficiency improves by increasing the
stirring speed up to 1020 rpm, which is the highest speed that could be achieved by
magnetic stirrer. Fig. 5.10 illustrates the effect of extraction stirring speed on the
enrichment factor. The enrichment factor increased with stirring speed up to 1020
rpm and decreased thereafter. Thus, stirring speed of 1020 rpm was applied in the
subsequent experiments.
Figure 5.10 Effect of the stirring rate on the extraction of analytes. Extraction
conditions: 50 μg/L of analytes; donor phase: 2.5 mL of solution containing 1.0 M
H3PO4 (pH 3); donor phase temperature: 25 oC; organic phase: 20 μL of 1-hexanol;
acceptor phase: 5 μL of the aqueous solution containing 0.1 M NaOH (pH 12);
extraction time: 15 min
5.3.6
Effect of Acceptor Volume on Analytes Behavior in the Three-Phase
LPME System
In three-phase LPME, higher enrichment factors can be resulted by
decreasing the volume ratio of acceptor to donor phase (Yazdi and Es’haghi, 2005).
10
In the present work, the phase volume of acceptor solutions was optimized by
changing the volume of the acceptor phase in the range of 2.0-6.0 μL, whereas the
volume of the donor phase was kept constant at 2.5 mL.
Experiments were carried out to determine the recoveries for all the analytes
by changing the Vd/Va values (Table 5.4). Vd/Va ratio of 500 was considered the best
recovery for all of analyses and thus used for the further experiment.
Table 5.4 Optimization of phase volume on LLLMEa
Volume of acceptor phase (µL)
a
Vd/Vab
Relative Recovery
2-nitrophenol
3-nitrophenol
98.1
95.2
6.0
417
5.0
500
99.5
96.9
4.0
3.0
2.0
625
833
1250
81.6
64.3
50.7
87.3
60.9
48.2
Extraction conditions: volume of donor phase, 2500 μL. Initial concentration of
each, 50 µg/L. Extraction time: 15 min. Stirring speed: 1020 rpm. All extractions
were performed in triplicate
b
Phase volumes ratio, Vd is volume of the donor phase, and Va is volume of the
acceptor phase
5.4
Analytical Performance of LPME
Repeatability,
linearity,
correlation
coefficient,
detection
limit
and
enrichment factors were investigated under chosen experimental conditions. The
analytical performance of the developed method is tabulated in Table 5.5. It can be
seen that, the RSD values are smaller than 6.0% based on the peak areas for five
replicated runs. The linearity of this method for analyzing standard solution has been
investigated between the ranges 0.6-200 µg/L. Each analyte exhibited good linearity
with correlation coefficient, r2 > 0.9994. The limit of detection values were
calculated as three times the standard deviation of five replicate runs. Also, the
11
enrichment factors (EFs) were defined as the ratio of the analyte concentration in the
acceptor phase to its initial concentration in the donor phase.
Zhu et al. (2001) reported that the LOD for 2-nitrophenol and 3-nitrophenol
were 1.0 ppb and 0.5 ppb respectively. However, it is worth noting that LOD levels
are lower than the values established three-phase LPME system for this type of
analytes.
Table 5.5 Performance of LLLMEa
Compound
2-nitrophenol
3-nitrophenol
a
Enrichment
factor
RSD (%)
(n = 5)
569
502
4.04
3.27
Linear
range
(μg/L)
Correlation
coefficient
(r2)
0.8-200
0.6-200
0.9994
0.9997
LOD
(μg/L)
(n = 5)
0.6
0.3
Relative
recovery
(%)b
90.6
89.2
Conditions: 50 μg/L of analytes; donor phase: 2.5 mL of solution containing 1.0 M
H3PO4 (pH 3); donor phase temperature: 25 oC; organic phase: 20 μL of 1-hexanol;
acceptor phase: 5 μL of the aqueous solution containing 0.1 M NaOH (pH 12);
extraction time: 15 min; agitation rate: 1020 rpm.
b
Recovery values were calculated as the ratio of the amount of analyte in the
acceptor phase to its initial amount in the donor phase.
5.5
Determination
of
Partition
Coefficients
and
Verification
of
Experimental Results
The experimental values were used to calculate theoretical recoveries, and
these were compared with experimental values obtained by three-phase LPME. As
illustrated in Table 5.6, the Ka/d values were high for both of nitrophenols and the
individual partition coefficients (Korg/d and Ka/org) promoted efficient simultaneous
extraction from the donor through the organic phase and further into the acceptor
phase. Based on the individual partition coefficients, theoretical recoveries were
calculated in five replicates and these were in acceptable agreement with the values
determined experimentally by three-phase LPME of both nitrophenols dissolved at
50 µg/L level in deionized water.
12
Table 5.6 Experimental partition coefficients, theoretical recovery, and experimental
recovery of 2-nitrophenol and 3-nitrophenol for three-phase LPME
Compound
Partition coefficients
Recovery (%)
Experimental
*
Korg/d
Ka/org
Ka/d
Theoretical*
Experimental*
2-nitrophenol
90
95
8550 (R.S.D = 5.1%)
90.9
90.6 (R.S.D = 4.2%)
3-nitrophenol
75
99
7425 (R.S.D = 4.8%)
90.3
89.2 (R.S.D = 3.3%)
Va = 5 μL, Vorg = 20 μL, Vd = 2.5 mL
5.6
Application of Three-Phase LPME in Sea Water Samples
In order to determine nitrophenol in real sample, the developed method was
applied to the analysis of sea water. The overall concentration of H3PO4 in the sea
water was made to 1.0 M using concentrated H3PO4. A 2.5 mL volume of this
acidified sample was placed in the sample vial and was extracted using the general
procedure. The acceptor phase was then injected into HPLC. No target compounds
were found in the tested sea water sample as illustrated in Fig. 5.11 (B).
13
UV abs
(A)
3-nitrophenol
2-nitrophenol
2,4-dinitrophenol
(B)
(C)
3-nitrophenol
2,4-dinitrophenol
1
2
2-nitrophenol
3
4
5
min
Figure 5.11 The chromatograms of nitrophenols: (A) 5 μg/L standard solution; (B)
sea water blank sample; (C) 5 μg/L spiked sea water sample
For evaluating the precision and accuracy of the present method, relative
recovery were conducted by spiking known amounts of 2-nitrophenol and 3nitrophenol, each at 1.0, 5.0, and 10.0 µg/L, into sea water and the results are
summarized in Table 5.7. It shows that the relative recoveries and precision are in the
ranges of 90.1-105.3% and 1.9-7.5% (R.S.D) in intra-day measurements (n = 3) and
91.3-101.7% and 3.4-8.0% (R.S.D) in inter-day measurements (5 consecutive days),
respectively, indicating that LLLME-HPLC method is applicable to real water
sample.
14
Table 5.7 Results of the relative recovery for determining 2-nitrophenol and 3nitrophenol in the concentrations of spiked sea water sample (1.0, 5.0, and 10.0
µg/L) in intra-day (n = 3) and inter-day measurements (n = 5)
Relative recovery (%) mean ± S.D.
Analyte
Intra-day
Inter-day
1.0 µg/L
5.0 µg/L
10.0 µg/L
1.0 µg/L
5.0 µg/L
10.0µg/L
2-nitrophenol
90.9 ± 6.6
105.3 ± 2.4
99.9 ± 3.4
91.3 ± 3.5
99.0 ± 5.2
98.4 ± 7.2
3-nitrophenol
90.1 ± 5.4
101.7 ± 1.9
98.0 ± 7.5
101.7 ± 4.1
98.5 ± 8.0
97.1 ± 4.3
15
CHAPTER 6
CONCLUSIONS AND SUGGESTIONS FOR FURTHER STUDY
6.1
Conclusions
The extractions of pesticides and nitrophenols based on polypropylene hollow
fiber liquid-phase microextraction (HF-LPME) with two-phase and three-phase
mode system have been examined successfully. In general, two-phase HF-LPME is
applicable for the analytes with high solubility in non-polar organic solvents and 3phase HF-LPME is applicable for basic and acidic analytes with ionizable
functionalities. The determination of analytes partition coefficients in two-phase and
three-phase HF-LPME has also been successfully achieved. Extraction parameters
were successfully optimized and provided high enrichment factor.
A simple and solvent minimized sample preparation technique based on twophase hollow fiber-protected liquid-phase microextraction has been developed and
applied to the analysis of three pesticides namely hexaconazole, quinalphos, and
methidathion. Separation of the analytes was successfully achieved within 10 min by
using GC-ECD. The analytes were extracted from a donor phase through 3 µL of an
organic solvent immobilized in the pores of a porous polypropylene hollow fiber and
then into the acceptor phase present inside the hollow fiber. Extraction conditions
such as solvent identity, salt concentration, stirring speed, extraction time, length of
the hollow fiber, volume of donor phase, and pH value were optimized. The
following conditions have been selected as the optimum extraction conditions:
toluene as organic solvent, 4 mL water samples, 20 min extraction time, 850 rpm
stirring rate, 1.8 cm HF length, 1.5% (w/v) NaCl content, and pH 3.
16
The repeatability of the proposed method expressed as relative standard
deviations (RSD) and was found to range between 4% and 7% indicating that good
repeatability was obtained for each of the analytes using HF-LPME method. Each
analyte exhibited good linearity with correlation coefficient, r2 > 0.9974. Limit of
detection (LODs), were calculated as three times the standard deviation of three
replicated runs of spiked samples at lowest concentration of the analytes. The LODs
ranged from 0.0025 to 0.0028 µg/L. These values were found to be better than those
reported elsewhere that utilized LPME method for extracting pesticides from this
type of matrix (Pan and Ho, 2004).
LPME is an equilibrium process that can be very effective for analyte
enrichment because of the increase in the volume ratio of donor solution and
acceptor phase. The proposed method provided good average enrichment factors of
up to 350-fold and successfully determined the partition coefficient for the selected
analytes that were found to be directly correlated to the enrichment factor. The
enrichment factor of hexaconazole was slightly lower compared to methidathion and
quinalphos. This is due to the higher solubility of hexaconazole in the donor phase
thus reducing its partition into the organic phase.
The HF-LPME method was applied to the analysis of pesticides in drinking
water and farm water samples. However, none of the target analytes were found
despite the low detection limits obtainable with HF-LPME-GC-ECD method. The
waters were evidently not polluted beyond the LODs. The relative recoveries of the
spiked samples ranged between 96-102% with RSD of less than 7% for three
replicates. More than 96% relative recoveries were obtained for the analytes in the
two samples indicating that there was only a minor influence of sample matrix on the
extraction since there was not much difference in the relative recoveries between
spiked deionized water, spiked farm water, and spiked drinking water. Minor matrix
effect on the LPME extraction was probably attributed to the selectivity of the
hollow fiber because of the pores in its wall.
This study has successfully demonstrated for the first time the development
of three-phase HF-LPME methods coupled with HPLC to extract some selected
17
nitrophenols from sea water samples with a high efficiency. All the analytes were
successfully separated by using HPLC at flow rate of 1.0 mL/min on C18 column
within 5 min after achieving the optimum conditions of separation.
The compounds were extracted from 2.5 mL aqueous solution with the
adjustment of pH at 3.0 (donor solution) into an organic phase (1-hexanol)
immobilized in the pores of the hollow fiber and finally back-extracted into 5.0 µL of
the acceptor microdrop (pH 12.0) located at the end of the microsyringe needle. After
a prescribed back-extraction time, the acceptor microdrop was withdrawn into the
microsyringe and directly injected into the HPLC system under the optimum
conditions (donor solution: 1.0 M H3PO4, pH 3.0; organic solvent: 1-hexanol;
acceptor solution: 5 µL of 0.1 M NaOH, pH 12.0; agitation rate: 1050 rpm;
extraction time: 15 min). The use of 1-hexanol as organic phases provided high
selectivity for phenols. This solvent is a very stable organic phase which provides
very low relative standard deviations.
An enrichment factor up to 500-fold was obtained for three phase LPME
method. This is higher than the enrichment factor obtained by other researchers that
utilized the same type of approaches. In, addition, low limits of detection for 3nitrophenol and 2-nitrophenol which is 0.3 µg/L and 0.6 µg/L respectively make
three-phase LPME as a method of choice for the measurement of target analytes in
the environment samples such as sea water. The calibration curve for these analytes
was linear in the range of 0.6-200 µg/L with r2 > 0.9994 and the RSD values of less
than 6.0%. The amount of nitrophenols in real sea water samples was below than the
detection limit.
This study also found that the partition coefficient (Ka/d) values were high for
2-nitrophenol and 3-nitrophenol and the individual partition coefficients (Korg/d and
Ka/org) promoted efficient simultaneous extraction from the donor through the organic
phase and further into the acceptor phase. Analytes with high Ka/d values are
typically moderately or highly hydrophobic compounds containing acidic or basic
functionalities besides neutral compounds are not or very poorly extracted into the
acceptor phase in three-phase HF-LPME.
18
It was shown that two-phase and three-phase HF-LPME provided high
recovery only for compounds with high overall partition coefficients (Ka/d and Korg/d
respectively), whereas extraction was inefficient for compounds with lower partition
coefficients. Enrichment factor of three-phase HF-LPME were higher than with
hollow fiber based LPME performed in the two-phase mode because the volume of
the acceptor phase in three-phase mode was smaller. In addition, three-phase HFLPME provided higher selectivity because neutral compounds were not extracted
with this technique.
6.2
Suggestions for Further Study
Although satisfactory results were obtained using two-phase and three-phase
HF-LPME in this study, there are some other aspects that can be considered in the
future study. In optimization of the parameter affecting the HF-LPME, another
parameter could be added. The temperature effect study is valuable because the
temperature could obviously affect the extraction efficiency in two ways. On one
hand, temperature could affect the mass transfer rates of analytes, while on the other
hand, temperature could affect the partition coefficients between the organic and
aqueous phase. When the temperature is increased, the mass transfer rates of analytes
are increased, but the partition coefficients are decreased.
Only a few papers have been published for HF-LPME of food and beverages.
This is surprising as the technique is very well suited for extraction of pesticides and
other toxic compounds, as shown by the numerous methods published for these
substances in the environmental field. To our knowledge, no systematically work has
been done in order to study the behavior of three-phase LPME for studying the
migration of food packaging components into different aqueous matrices used as
food simulants. Thus, this method can be applied in the future to demonstrate that the
three-phase LPME technique provide reliable data for a broad range of analytical
applications.
19
The present work has successfully developed methods for the extraction of
analytes using a short piece of a porous hollow fiber which is a rod configuration
with a closed bottom. This concept is also much compatible with modern autosamplers. However, some works have been reported utilizing U-shape configuration
where both ends of the hollow fiber are connected to guiding tubes. But, procedure
for transferring of the acceptor solution was somewhat cumbersome and difficult to
automate. Therefore, more studies are required regarding the applications for the
determination of pesticides and nitrophenols on water using this technique.
The proposed methods are simple and the use of the disposable hollow fiber
completely eliminated carry-over effects. However, for three-phase HF-LPME, it
remains as a challenge to automate it without substantial expenditure on an autosampler system whose software also needs to be programmed to handle the
operations involved in the extraction process.
20
REFERENCES
Bårdstu, K. F., Ho, T. S., Rasmussen, K. E., Pedersen-Bjergaard, S., and Jönsson, J.
Ǻ. (2007). Supported Liquid Membranes in Hollow Fiber Liquid-phase
Microextraction (LPME)- Practical Considerations in Three-phase Mode. J.
Sep. Sci. 30, 1364-1370.
Basheer, C., Lee, H. K., and Obbard, J. P. (2004). Application of Liquid-phase
Microextraction and Gas Chromatography-Mass Spectrometry for the
Determination of Polychlorinated Biphenyls in Blood Plasma. J. Chromatogr.
A. 1022, 161-169.
Beceiro-Gonzalez, E., Concha-Granha, E., Guimaraes, A., Goncalves, C.,
Muniatagui-Lorenzo, S., and Alpendurada, M. F. (2007). Optimization and
Validation of a Solid-phase Microextraction Method for Simultaneous
Determination of Different Types of Pesticides in Water by Gas
Chromatography–Mass Spectrometry. J. Chromatogr. A. 1141, 165-173.
Berhanu, T., Liu, J-F., Romero, R., Megersa, N., Jonsson, J. A. (2006).
Determination of Trace Levels of Dinitrophenolic Compounds in
Environmental Water Samples using Hollow Fiber Supported Liquid
Membrane Extraction and High Performance Liquid Chromatography. J.
Chromatogr. A. 1103, 1-8.
Billings. S. A. (1980). Identification of Nonlinear Systems: A survey. Proc. IEE,
Part D. 127(6), 272-284.
21
Bishop, E. J. and Mitra, S. (2007). Measurement of Nitrophenols in Air Samples by
Impinger Sampling and Supported Liquid Membrane Micro-extraction. Anal.
Chim. Acta. 583, 10-14.
Bonato, P. S. and Santana, F. J. M. (2008). Enantioselective Analysis of Mirtazapine
and
its
Two
Major
Chromatography-Mass
Metabolites
Spectrometry
in
Human
after
Plasma
Three-phase
by
Liquid
Liquid-phase
Microextraction. Anal. Chim. Acta. 606, 80-91.
Chen, C-C., Melwanki, M. B., and Huang, S-D. (2006). Liquid-liquid-liquid
Microextraction with Automated Movement of the Acceptor and the Donor
Phase for the Extraction of Phenoxyacetic Acids prior to Liquid
Chromatography Detection. J. Chromatogr. A. 1104, 33-39.
Chiang, J.-S. and Huang, S.-D. (2007). Determination of Haloethers in Water with
Dynamic Hollow Fiber Liquid-phase Microextraction using GC-FID and GCECD. Talanta. 71, 882–886.
Chimuka, L., Cukrowska, E., and Jönsson, J. Å. (2004). Why Liquid Membrane
Extraction is an Attractive Alternative in Sample Preparation. Pure Appl.
Chem. 76(4), 707-722.
Ebrahemzadeh, H., Yamini, Y., Sedighi, A., and Rouini, M. R. (2008).
Determination of Tramadol in Human Plasma and Urine Samples using
Liquid Phase Microextraction with Back Extraction combined with High
Performance Liquid Chromatography. J. Chromatogr. B. 863, 229-234.
Gallardo, E., Barroso, M., Margalho, C., Cruz, A., Vieira, D. N., and LopezRivadulla, M. (2006). Determination of Quinalphos in Blood and Urine by
Direct Solid-phase Microextraction combined with Gas ChromatographyMass Spectrometry. J. Chromatogr. B. 832, 162-168.
22
Gjelstad, A., Andersen, T. M., Rasmussen, K. E., and Pedersen-Bjergaard, S. (2007).
Microextraction Across Supported Liquid Membranes Forced by pH
Gradients and Electrical Fields. J. Chromatogr. A. 1157, 38-45.
Gjelstad, A., Rasmussen, K. E., and Pedersen-Bjergaard, S. (2006). Electrokinetic
Migration across Artificial Liquid Membranes Tuning the Membrane
Chemistry to different Types of Drugs Substances. J. Chromatogr. A. 1124,
29-34.
Hercegová, A., Dömötorová, M., and Matisová, E. (2007). Sample Preparation
Methods in the Analysis of Pesticide Residues in Baby Food with Subsequent
Chromatographic Determination. J. Chromatogr. A. 1153, 54-73.
Ho, T. S., Pedersen-Bjergaard, S., and Rasmussen, K. E. (2002). Recovery,
Enrichment and Selectivity in Liquid-phase Microextraction Comparison
with Conventional Liquid–liquid Extraction. J. Chromatogr. A. 963, 3–17.
Ho, T. S., Halvorsen, T. G., Pedersen-Bjergaard, S., and Rasmussen, K. E. (2003).
Liquid-phase Microextraction of Hydrophilic Drugs by Carrier-mediated
Tarnsport. J. Chromatogr. A. 998, 61-72.
Hou, L., Shen, G., and Hian, K. L. (2003). Automated Hollow Fiber-protected Liquid
Phase Microextraction of Pesticides for Gas Chromatography-Mass
Spectrometric Analysis. J. Chromatogr. A. 985, 107-116.
Hu, B., and Xiong, J. (2008). Comparison of Hollow Fiber Liquid Phase
Microextraction and Dispersive Liquid-liquid Microextraction for the
Determination of Organosulfur Pesticides in Environmental and Beverage
Samples by Gas Chromatography with Flame Photometric Detection. J.
Chromatogr. A. 1193, 7-18.
23
Huang, S.-P and Huang, S.-D. (2006). Dynamic Hollow Fiber Protected Liquid
Phase Microextraction and Quantification using Gas Chromatography
combined with Electron Capture Detection of Organochloride Pesticides in
Green Tea Leaves and Ready-to-drink Tea. J. Chromatogr. A. 1135, 6-11.
Huang, S.-P and Chiang, J.-S. (2007). Determination of Haloethers in Water with
Dinamic HF-LPME using GC-FID and GC-ECD. Talanta. 71, 882-886.
Huang, S-D. and Lin, C-Y. (2008). Application of Liquid-liquid-liquid
Microextraction and High Performance Liquid Chromatography for the
Determination of Sulfonamides in Water. Anal. Chim. Acta. 612, 37-43.
Jönsson, J. Ǻ. and Mathiasson, L. (1999). Liquid Membrane Extraction in Analytical
Sample Preparation. Trends in Anal. Chem. 18 (5), 318-334.
Kati Lüthje. On-line Coupling of Pressurised Hot Water Extraction and Microporous
Membrane Liquid-Liquid Extraction with Chromatography in Analysis of
Enviromental Samples. Academic Dissertation. University of Helsinki
Finland; 2004.
Lambropoulou, D. A. and Albanis, T. A. (2004). Sensitive Trace Enrichment of
Environmental Andiandrogen Vinclozolin from Natural Waters and Sediment
Samples using Hollow-Fiber Liquid-phase Microextraction. J. Chromatogr.
A. 1061, 11–18.
Lambropoulou, D. A. and Albanis, T. A. (2005). Application of Hollow Fiber Liquid
Phase Microextraction for the Determination of Insecticides in Water. J.
Chromatogr. A. 1072, 55–61.
24
Lambropoulou, D. A. and Albanis, T. A. (2007). Liquid Phase Microextraction
Techniques in Pesticides Residue. J. Biochem. Biophys. Methods. 70, 195228.
Lee, H. K., Basheer, C., Alnedhary, A. A., and Rao, B. S. M. (2007). Determination
of Organophosphorous Pesticides in Wastewater Samples using Binarysolvent Liquid-phase Microextraction and Solid-phase Microextraction: A
Comparative Study. Anal. Chim. Acta. 605, 147-152.
Lee, H. K., Lee, L., Rasmussen, K. E., and Pedersen-Bjergaard, S. (2008).
Environmental and Bioanalytical Applications of Hollow Fiber Membrane
Liquid-phase Microextraction: A Review. Anal. Chim. Acta. 624, 253-268.
Lee, M-W. and Chung, L-W. (2008) Evaluation of Liquid-phase Microextraction
Conditions for Determination of Chlorophenols in Environmental Samples
using Gas Chromatography-Mass Spectrometry without Derivatization.
Talanta. 76, 154-160.
Lee, H. K. and Zhu, L., Zhu, L. (2001). Liquid-liquid-liquid Microextraction of
Nitrophenols with a Hollow Fiber Membrane prior to Capillary Liquid
Chromatography. J. Chromatogr. A. 924, 407-414.
Lindsay, S. (1992). High Performance Liquid Chromatography. (2nd ed.) Thames
Polytechnic, London: Wiley.
Liu, X., Chen. X., Yang, S. and Wang, X. (2007). Comparison of Continuous-flow
Microextraction
and
Static
Liquid-phase
Microextraction
for
the
Determination of p-toluidine in Chlamydomonas reinhardtii. J. Sep. Sci. 30,
2506-2512.
25
Ma, M., Zhang, Z., Zhang, C., Su, X., Chen, B., and Yao, S. (2008). Carriermediated Liquid Phase Microextraction coupled with High Performance
Liquid Chromatography for Determination of Illicit Drugs in Human Urine.
Anal. Chim. Acta. 621, 185-192.
McNair, H. M., and Miller, J. M. (1997). Basic Gas Chromatography. US: Wiley.
Meyer, V. R. (2004). Practical High Performance Liquid Chromatography. (4th ed.)
Switzerland: Wiley.
Pan, H-J., and Ho, W-H. (2004). Determination of Fungicides in Water using Liquidphase Microextraction and Gas Chromatography with Electron Capture
Detection. Anal. Chim. Acta. 527, 61-67.
Pan, J., Xia, X-X., and Liang, J. (2008). Analysis of Pesticide Multi-Residues in
Leavy
Vegetables
by
Ultrasonic
Solvent
Extraction
and
Liquid
Chromatography-Tandem Mass Spectrometry. Ultrasonic Sonochemistry. 15,
25-32.
Pedersen-Bjergaard, S. and Rasmussen, K. E. (2007). Liquid-phase Microextraction
with Porous Hollow Fibers, a Miniaturized and Highly Flexible Format for
Liquid-liquid Extraction. J. Chromatogr. A. 113, 67-72.
Pedersen-Bjergaard, S. and Rasmussen, K. E. (2005). Bioanalysis of Drugs by
Liquid-phase Microextraction Coupled to Separation Techniques. J.
Chromatogr. B. 817, 3-12.
Pedersen-Bjergaard, S., Rasmussen, K. E., Brekke, A., Ho, T. S., and Halvorsen, T.
G. (2005). Liquid-phase Microextraction of Basic Drugs-Selection of
Extraction Mode based on Computer Calculated Solubilty Data. J. Sep. Sci.
28, 1195-1203.
26
Pezo, D., Salafranca, J., and Nerín, C. (2007). Development of an Automatic
Multiple Dynamic Hollow Fiber Liquid-phase Microextraction Procedure for
Specific Migration Analysis of New Active Food Packagings containing
Essential Oils. J. Chromatogr. A. 1174, 85-94.
Psillakis, E., Kalogerakis, N., Ratola, N., and Alves, A. (2008). Hollow-Fiber LiquidPhase Microextraction: A Simple and Fast Cleanup Step used for PAHs
Determination in Pine Needles. Anal. Chim. Acta. 618, 70-78.
Psillakis, E., and Kalogerakis, N. (2003). Development in Liquid-phase
Microextraction. Trends in Anal. Chem. 22(10), 565-574.
Psillakis, E., and Kalogerakis, N. (2003). Hollow-fibre Liquid-phase Microextraction
of Phthalate Esters from Water. J. Chromatogr. A. 999, 145-153.
Rasmussen, K. E., Pedersen-Bjergaard, Krogh, M., Ugland, H. G. and Grønhaug, T.
(2000). Development of a Simple in-vial Liquid-phase Microextraction
device for Drug Analysis Compatible with Capillary Gas Chromatography,
Capillary Electrophoresis and High-performance Liquid Chromatography. J.
Chromatogr. A. 873, 3-11.
Rasmussen, K. E. and Pedersen-Bjergaard, S. (2004). Development in Hollow Fiberbased, Liquid-phase Microextraction. Trends in Analytical Chemistry. 23(1),
1-10.
Rodríguez, A., Pedersen-Bjergaard, S., Rasmussen, K. E., and Nerín, C. (2008).
Selective Three-phase Liquid-phase Microextraction of Acidic Compounds
from Foodstuff Simulants. J. Chromatogr. A. 1198-1199, 38-44.
Sanagi, M. M., See, H.H., Ibrahim, W. A. W., and Naim, A. A. (2007).
Determination of Pesticides in Water by Cone-shaped Membrane Protected
27
Liquid Phase Microextraction prior to Micro-liquid Chromatography. J.
Chromatogr. A. 1152, 215–219.
Sanagi, M. M., Miskam, M., Ibrahim, W. A. W., and Hermawan, D. (2009).
Determination of Partition Coefficients for Nitrophenols using Three Phase
Liquid Phase Microextraction prior to Micro-liquid Chromatography.
Proceedings of ICORAFSS 2009. 164-168.
Ugland, H. G., Krogh, M., and Reubsaet, L. (2003). Three-phase Liquid-phase
Microextraction of Weakly Basic Drugs from Whole Blood. J. Chromatogr.
B. 798, 127-135.
Xiao, Qin., Hu, B., Duan, J., He, Man., and Zu, W. (2007). Analysis of PBDEs in
Soil, Dust, Spiked Lake Water, and Human Serum Samples by Hollow FiberLiquid Phase Microextraction Combined with GC-ICP-MS. J. Am. Mass
Spectrom. 18, 1740-1748.
Xiong, J. and Hu, B. (2008). Comparison of Hollow Fiber Liquid Phase
Microextraction and Dispersive Liquid–liquid Microextraction for the
Determination of Organosulfur Pesticides in Environmental and Beverage
Samples by Gas Chromatography with Flame Photometric Detection. J.
Chromatogr. A. 1193, 7–18.
Xu, H., Liao, Y. and Yao, J. (2007). Development of a Novel Ultrasound-assisted
Headspace Liquid-phase Microextraction and its Application to the Analysis
of Chlorophenols in Real Aqueous Samples. J. Chromatogr. A, 1167, 1-8.
Yamini, Y., Shariati, S., Darabi, M., and Amini, M. (2007). Three Phase Liquid
Phase Microextraction of Phenylacetic Acid and Phenylpropionic Acid from
Biological Fluids. J. Chromatogr. B. 855, 228-235.
28
Yazdi, A. S., and Es’haghi, Z. (2005). Two-step Hollow Fiber-based, Liquid-phase
Microextraction combined with High-Performance Liquid Chromatography:
A New Approach to Determination of Aromatic Amines in Water. J.
Chromatogr. A. 1082, 136-142.
Zhang, J., Su, T., and Lee, H. K. (2006). Development and Application of
Microporous Hollow Fiber Protected Liquid-phase Microextraction via
Gaseous Diffusion to the Determination of Phenols in Water. J. Chromatogr.
A. 1121, 10-15.
Zhao, L. and Lee, H. K. (2001). Determination of Phenols in Water using Liquid
Phase Microextraction with Back Extraction combined with HighPerformance Liquid Chromatography. J. Chromatogr. A. 931, 95-105.
Zhao, L. and Hian, K. L. (2002). Liquid-Phase Microextraction Combined with
Hollow Fiber as a Sample Preparation Technique Prior to Gas
Chromatography/Mass Spectrometry. Anal. Chem. 74, 2486-2492.