close

Вход

Забыли?

вход по аккаунту

?

Improvements in sample preparation and introduction for inductively coupled plasma-mass spectrometry incorporating microwave energy

код для вставкиСкачать
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type o f computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely afreet reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI
A Bell & Howell Information Company
300 North Zed> Road, Ann Arbor MI 48106-1346 USA
313/761-4700 800/521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IMPROVEMENTS IN SAMPLE PREPARATION AND INTRODUCTION FOR
INDUCTIVELY COUPLED PLASMA-MASS SPECTROMETRY
INCORPORATING MICROWAVE ENERGY
A Dissertation Presented
by
NEIL FITZGERALD
Submitted to the Graduate School of the
University of Massachusetts Amherst in partial fulfillment
of the requirements for the degree of
DOCTOR OF PHILOSOPHY
February 1999
Department of Chemistry
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 9920601
UMI Microform 9920601
Copyright 1999, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MI 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
© Copyright by Neil Fitzgerald 1999
All Rights Reserved
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IMPROVEMENTS IN SAMPLE PREPARATION AND INTRODUCTION FOR
INDUCTIVELY COUPLED PLASMA-MASS SPECTROMETRY
INCORPORATING MICROWAVE ENERGY
A Dissertation Presented
by
NEIL FITZGERALD
Approved as/to style and content by:
Julian P.
, Chair
Ramon M. Bames, Member
"7
William C. Conner,
Lila M. Gierasch, department Head
Department of Chemistry
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGMENTS
The author wishes to acknowledge Prof. Tyson for his guidance during
this work and the members of the research group including Paul, Pete, Chris,
Robert, Cesar, Hakan, Zikri, Dave and Emily for their many insights and
observations. The author would also like to acknowledge Richard and Lee for
many useful and informative discussions (scientific and otherwise). The
author would also like to express his gratitude to Prof. Yngvesson for helpful
discussions on the theoretical work, Lois Jassie and CEM for partial financial
support and helpful discussions and David Leighty and Permapure for the
provision of equipment and information on Nation dryers. The author wishes
to express his gratitude to his family for their support and to Tiffany whose love
and encouragement has made this work possible.
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
IMPROVEMENTS IN SAMPLE PREPARATION AND INTRODUCTION FOR
INDUCTIVELY COUPLED PLASMA-MASS SPECTROMETRY
INCORPORATING MICROWAVE ENERGY
FEBRUARY 1999
NEIL FITZGERALD, B.Sc., UNIVERSITY OF KENT AT CANTERBURY
M.Sc., LOUGHBOROUGH UNIVERSITY OF TECHNOLOGY
Ph.D., UNIVERSITY OF MASSACHUSETTS AMHERST
Directed by: Professor Julian F. Tyson
The area of atomic spectrometry has long suffered from inefficient
sample preparation and introduction systems. Microwave heating can be a
powerful technique for improving these systems but has tending not to be well
understood and used in analytical chemistry. The purpose of this work has
been to use microwave energy in order to improve sample preparation and
introduction for plasma source spectrometry. An objective of this work has
been to develop an on-line pressurized microwave digestion system. This
system was developed to be capable of digesting difficulty soluble organic
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
materials with nitric acid alone rapidly while reducing possibilities for
contamination and providing the possibility for automation.
The use of microwave energy has been considered as a technique for
desolvating an aqueous aerosol prior to solvent removal using a Nation dryer.
The purpose of this system was to increase the efficiency of sample
introduction to a plasma source mass spectrometer.
In order to accomplish
this some fundamental studies of microwave interaction with water droplets
were performed and a theoretical background was developed.
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS.......................................................................................... iv
ABSTRACT................................................................................................................ v
LIST OF TABLES.................................................................................................... xii
LIST OF FIGURES..................................................................................................xiv
Chapter
1.
INTRODUCTION TO SAMPLE PREPARATION AND INTRODUCTION
TECHNIQUES FOR ATOMIC SPECTROMETRY............................1
1.1
1.2
1.3
1.4
1.5
2.
Introduction...........................................................................................1
Conventional Sample Preparation Techniques..............................2
Sample Introduction Techniques....................................................... 5
The Application of Microwave Radiation to Analytical
Chemistry................................................................................9
References......................................................................................... 12
DEVELOPMENT OF AN ON-LINE PRESSURIZED MICROWAVE
DIGESTION SYSTEM..................................................................... 14
2.1
2.2
Introduction........................................................................................ 14
On-line Digestion and Analysis of Milk Powder.............................18
2.2.1 Experimental........................................................................... 18
2.2.1.1 Development of a Pressurized On-Line
Digestion System.......................................... 18
2.2.1.2 Development of a Pressure Transducer.............. 19
2.2.1.3 Development of the Microwave Program............. 20
2.2.1.4 Reagents.................................................................. 20
2.2.1.5 Spectrometer........................................................... 21
2.2.1.6 Procedure................................................................ 21
2.2.2 Results and Discussion.......................:.................................. 23
2.3
On-line Digestion and Analysis of Bovine Liver............................ 24
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3.1 Experimental.......................................................................... 24
2.3.1.1 Modification of the Pressurized On-Line
Digestion System.......................................... 24
2.3.1.2 Reagents..................................................................24
2.3.1.3 Spectrometer........................................................... 25
2.3.1.4 Procedure................................................................25
2.3.2 Results and Discussion........................................................ 26
2.4
2.5
3.
Conclusions......................................................................................27
References........................................................................................28
REDUCTION OF WATER LOADING EFFECT IN ICP-MS USING
NAFION MEMBRANE MATERIAL..................................................42
3.1
3.2
Introduction.......................................................................................42
Experimental.....................................................................................46
3.2.1
3.2.2
3.2.3
3.2.4
3.2.5
3.2.6
3.2.7
3.3
Results and Discussion...................................................................50
3.3.1
3.3.2
3.3.3
3.3.4
3.3.5
3.3.6
3.3.7
3.4
3.5
4.
Instrumentation.......................................................................46
Reagents................................................................................ 47
System Optimization............................................................. 47
Dryer Efficiency......................................................................48
Detection Limit for Iron.......................................................... 48
Addition of Nitrogen to the Plasma.......................................49
Heated Spray Chamber....................................................... 50
Development of Dryer System.............................................50
Efficiency for Water Removal................................................51
System Optimization............................................................. 52
Comparison with Conventional Sample Introduction
53
Detection Limits for Iron........................................................ 54
Addition of Nitrogen to the Plasma.......................................54
Heated Spray Chamber....................................................... 55
Conclusions......................................................................................56
References........................................................................................57
INVESTIGATION OF MICROWAVE DESOLVATION OF WATER
AEROSOLS..................................................................................... 72
4.1
4.2
Introduction.......................................................................................72
Experiments with a Multi-mode Cavity...........................................76
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.1 Experimental......................................................................... 76
4.2.1.1 Apparatus Development....................................... 76
4.2.1.2 Determination of Maximum Available
Microwave Power........................................ 78
4.2.1.3 Spectrometer and Reagents................................. 78
4.2.1.4 Trapping and Analysis of Exiting Material........... 79
4.2.1.5 Destination of Sample Material............................ 79
4.2.1.6 Calculation of System Efficiency.......................... 80
4.2.1.7 Temperature Measurements................................. 81
4.2.1.8 Effect of Microwave Energy on the System
Efficiency...................................................... 82
4.2.2 Results and Discussion....................................................... 83
4.2.2.1 Determination of Maximum Available
Microwave Power........................................ 83
4.2.2.2 Trapping and Analysis of Exiting Material........... 83
4.2.2.3 Destination of Sample Material............................ 84
4.2.2.4 Calculation of System Efficiency...........................84
4.2.2.5 Temperature Measurement................................... 86
4.2.2.6 Effect of Microwave Energy on the System
Efficiency...................................................... 86
4.3
Experiments with a Single-Mode Cavity....................................... 87
4.3.1 Experimental..........................................................................87
4.3.1.1 Cavity Design......................................................... 87
4.3.1.2 Calibration for Microwave Power..........................88
4.3.1.3 Experiments of Aerosol Heating............................88
4.3.2 Results and Discussion........................................................89
4.4
4.5
5.
Conclusions......................................................................................90
References........................................................................................91
THEORY OF MICROWAVE HEATING OF WATER DROPLETS
5.1
100
Introduction.....................................................................................100
5.1.1 Microwave Heating of Water.............................................. 101
5.1.2 Frequency Effects............................................................... 104
5.2
Mathematical Models.................................................................... 107
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2.1 Field Strength Model........................................................... 107
5.2.2 Results and Discussion...................................................... 109
5.2.3 Field Strength Model Incorporating
Penetration Depth...................................................110
5.2.4 Results and Discussion.......................................................112
5.2.5 Absorbed Power Model...................................................... 113
5.2.6 Results and Discussion...................................................... 115
5.3
5.4
6.
Conclusions.................................................................................... 116
References.......................................................................................117
DEVELOPMENT OF A DESOLVATION SYSTEM FOR ICP-MS
UTILIZING MICROWAVE ENERGY............................................. 125
6.1
6.2
Introduction..................................................................................... 125
Experimental....................................................................................131
6.2.1 Development of a Microwave Thermospray
Desolvation Device................................................ 131
6.2.2 Development of a Microwave Heated Nebulizer
Desolvation System...............................................133
6.2.3 Development and Testing of a Microwave Heated
Cyclone Spray Chamber Desolvation System ...134
6.3
Results............................................................................................. 135
6.3.1 Development of a Microwave Thermospray
Desolvation Device................................................ 135
6.3.2 Development of a Microwave Heated Nebulizer
Desolvation System............................................... 138
6.3.3 Development and Testing of a Microwave Heated
Cyclone Spray Chamber Desolvation System ...139
6.4
6.5
7.
Conclusions.................................................................................... 141
References...................................................................................... 142
CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK.............152
7.1
7.2
7.3
7.4
7.5
On-line Microwave Digestion........................................................152
Reduction of Water Loading Effects in ICP-MS Using a
Nation Dryer........................................................................155
Microwave Desolvation................................................................. 156
Conclusions.................................................................................... 158
References...................................................................................... 158
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDICES
A.
B.
C.
PROGRAM FOR FIELD STRENGTH MODEL......................................... 159
PROGRAM FOR FIELD STRENGTH MODEL INCORPORATING
PENETRATION DEPTH............................................................... 161
PROGRAM FOR ABSORBED POWER MODEL..................................... 163
BIBLIOGRAPHY...................................................................................................164
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table
Page
2.1
Instrument parameters for the determination of milk powder
by flame atomic absorption spectrometry......................... 31
2.2
Microwave digestion program for milk powder.............................32
2.3
Results for the digestion of SRM A-11 milk powder slurries
by flame atomic absorption spectrometry......................... 33
2.4
Instrument parameters for the determination of bovine
liver by ICP-MS using a Nafion dryer sample
introduction system..................................................34
2.5
Microwave digestion program for bovine liver............................. 35
2.6
Results for the digestion of SRM 1577b bovine
liver by ICP-MS................................................................... 36
2.7
Results for the digestion of SRM 1577b bovine liver by
ICP-MS Using a Nafion Dryer........................................... 37
2.8
Results for the determination of a 6 blank solutions by
ICP-M S................................................................................ 38
3.1
Operating conditions for ICP-MS and dryer system.....................59
3.2
FIAS program for iron determination............................................. 60
3.3
Efficiency of the dryer system for water removal...........................61
3.4
Comparison of oxide ratios and ArO+ intensities for
conventional sample introduction and the
dryer system............................................................ 62
3.5
Limits of detection for iron-56 and iron-57.................................... 63
3.6
Limits of detection (pg dm'3) for iron-56 with the addition of
acetonitrile............................................................................64
4.1
Operating parameters of integrator and spectrometer for
multi-mode experiments.....................................................92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2
Destination of sodium after collection of exitingmaterial by
bubbling through water.......................................................93
4.3
Destination of sample material in the system................................94
5.1
Some physical constants and definitions.................................... 118
6.1
Optimized parameters for microwave heated cyclone
spray chamber system......................................................145
xiii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figure
Page
2.1
Schematic diagram of the on-line pressurized microwave
digestion system used for the digestion
of milk powder.......................................................... 39
2.2
Schematic diagram of pressure transducer.................................. 40
2.3
Schematic diagram of the on-line pressurized microwave
digestion system used for the digestion
of bovine liver........................................................... 41
3.1
Chemical structure of Nafion material............................................ 65
3.2
Schematic of dryer system.............................................................. 66
3.3
Effect of dryer temperature on oxide ratios and ArO+................... 67
3.4
Effect of purge gas flow rate on oxide ratios
and ArO+ intensity................................................................ 68
3.5
Flow injection peak for a 10 |ig dm"3 iron solution with and
without the dryer system......................................................69
3.6
Effect of flow rate on intensity for iron-56 for the addition of
nitrogen gas to the plasma..................................................70
3.7
Flow injection peak for 10 pg dm'3 iron-56 using a heated
spray chamber..................................................................... 71
4.1
Schematic diagram of the multi-mode cavity cross flow
nebulizer test system............................................................95
4.2
Schematic diagram of the multi-mode cavity Meinhard
nebulizer test system............................................................96
4.3
Schematic diagram of the spectrometer sample introduction
system modified to allow measurement of material
from the test system..................................................97
4.4
Temperature measurements in the test system with
microwaves on and off after sample injection.................. 98
xiv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.5
Schematic of the single mode microwave system........................ 99
5.1
Variation of penetration depth with frequency for water............ 119
5.2
Absorption coefficient vs. frequency for a cloud..........................120
5.3
Schematic diagram of mathematical model............................... 121
5.4
Variation of field strength necessary to evaporate a 10 pm
water droplet with applied frequency according to
the field strength model.........................................122
5.5
Variation of field strength necessary to evaporate a 10 pm
water droplet with applied frequency according to
the field strength incorporating
penetration depth model...........................123
5.6
Effect of applied frequency on the power absorbed by
a 10 pm water droplet according to the
absorbed power model.......................................... 124
6.1
Schematic diagram of the microwave thermospray
desolvation device.............................................................146
6.2
Schematic diagram of the adapted cross flow interface............ 147
6.3
Schematic diagram of the microwave heated nebulizer
device.................................................................................. 148
6.4
Schematic diagram of the microwave heated cyclone spray
chamber desolvation system............................................149
6.5
Normalized intensity against microwave power for a
microwave desolvation system with an unheated
transfer line............................................................150
6.6
Normalized intensity against microwave power for a
microwave desolvation system with a heated
transfer line.............................................................151
XV
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1
INTRODUCTION TO SAMPLE PREPARATION AND INTRODUCTION
TECHNIQUES FOR ATOMIC SPECTROMETRY
1.1 Introduction
Sample preparation and introduction techniques have tended to lag
behind the recent advances in atomic spectrometry instrumentation and are
still the predominant time consuming and inefficient processes in the analytical
determination.1 The purpose of any sample preparation and introduction
procedure for plasma spectrometry is to produce atoms of the analyte in the
plasma source. The preparation of a sample solution breaks most of the
bonds that are present in the sample material and produces a homogeneous
and easily transported liquid. The kinetics of the dissolution stage is often
slow and requires heating the sample in concentrated acids. Microwave
energy can be used to increase the speed of the dissolution stage. Open
vessel dissolutions are prone to contamination from the surroundings and can
lose analyte species. The development of a closed vessel, rapid dissolution
stage would offer many advantages over existing systems. The introduction of
a sample solution into a plasma source requires the desolvation of sample
droplets in the plasma prior to the formation of atoms which leads to the
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
absorption of a significant amount of energy. This desolvation process can
lead to plasma cooling and extinction. The use of a desolvation system prior
to sample introduction can reduce these loading effects by removal of the
solvent and introduction of dry sample particles into the plasma. This allows
more of the plasma energy to be available for producing atoms.
The following is a brief overview of sample preparation and sample
introduction techniques for plasma spectrometry as well as a discussion of the
current use of microwave energy in the field of analytical chemistry.
1.2 Conventional Sample Preparation Techniques
In order to analyze a sample by atomic spectrometry, a sample
preparation procedure is normally required. Typically a liquid sample is
preferred in order to achieve homogeneity of the sample and to ease transport
to the atomization source. Ideally this preparation step should convert all
samples to aqueous solutions, destroy all organic matter, retain all analytes of
interest in solution at detectable concentrations, not add any interfering
species, and adjust sample viscosity to the optimum for the analytical
technique.2
The most common sample dissolution step involves the use of an acid
which is normally heated with the sample in either an open or closed
container. The use of a heated open vessel digestion (normally achieved
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
using a glass or PTFE container on a hot plate) is a well accepted technique.
This technique has the advantages that there is little limitation on the sample
size that can be dissolved, it is simple, inexpensive, easy to add sample and
reagents to and is less likely to leach trace metals from the container than
pressurized digestion systems.2 The disadvantages of this technique are that
it is slow, allows the loss of volatile elements and can add impurities from the
large quantities of acid used. It also needs to be closely monitored, can be
contaminated easily (e.g. by airborne particles), can be hazardous (e.g.
production of acid fumes) and often requires large quantities of acid.2 Closed
vessel techniques allow for an increase in pressure which has the effect of
reducing the digestion time by increasing the reaction temperature. This
technique started to be developed more successfully after the introduction of
PTFE vessels that are chemically and thermally resistant.3 The main
advantages of elevated pressure and temperature digestions in closed
systems are the ability to decompose resistant solids which are not digestible
in open systems and the shortened digestion time. There is also little loss of
volatile species, low volumes of acids needed, reduction of contamination and
a substantial decrease in the blank value.3 Disadvantages of this method are
the need to monitor and control the pressure in the system, possible hazards
due to the use of hot acids under pressure and the increased possibility of
leaching from the container.
Other digestion techniques are based on dry methods, examples of
these include, dry ashing and fusion. Dry ashing is a useful technique for
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
organic samples which cannot be easily digested or fused. This method
involves the heating of a sample under high temperature in air to oxidize the
material to an ash followed by the dissolution of the ash with an acid. This is a
simple method which requires little monitoring and has little limitation of
sample size. However it is slow, can lose volatile species, requires expensive
apparatus and has substantial contamination problems.2 Fusion techniques
require the use of a flux material that reacts with the sample at high
temperatures to form a substance which can be readily dissolved in an
appropriate reagent. This technique is typically used for materials that are
insoluble in acids. This method is rapid, simple and requires little monitoring
but suffers from contamination from the flux. The technique requires a sample
size of typically 0.1 to 0.2 g and the use of expensive apparatus.2
The future trend in sample preparation methods is likely to involve the
use of high pressure digestions with pure acids. This would allow for the
digestion of difficulty soluble samples rapidly with little or no contamination,
while being simple to use and automated to reduced the need for trained
personnel. Efficient heating of these systems can be provide by the use of
microwave energy. Recent advances in the use of microwave digestion
procedures for environmental analysis has been reviewed by Lamble and
Hill.4 The development of new high pressure digestion vessels and high
power microwave ovens has led to the development of static high pressure
commercial systems such as the MARS system (CEM). This system is capable
of pressures up to 1500 psig for the digestion of difficulty soluble materials.5
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
This system requires the use of trained personnel in order to load the sample
and acids into the vessels and a cooling down time is necessary before the
vessels can be opened. On-line digestions systems operating at atmospheric
or elevated pressures including the Spectroprep (CEM) and Q-Prep
(Questron) are also available commercially. These systems tend to be more
suitable for easily digestible materials and require the introduction of a slurry
which can cause blockages. One method for overcoming the need for a slurry
has been suggested by Legere and Salin6 who introduced the solid sample
enclosed in a polyacrylamide capsule into an U-shaped digestion vessel. This
system suffers from a temperature limitation due to the melting point of the
PTFE digestion vessel and requires a manual stage in order to remove the
digested material from the vessel. These limitations could be overcome by the
development of an automated on-line system that allows for rapid digestion
and collection of a difficulty soluble sample without the necessity to introduce
the sample as a slurry.
1.3 Sample Introduction Techniques
In order to determine the concentration of a species by atomic
spectrometry, the sample must be introduced into the atomization source. This
may be achieved using samples in the solid, liquid or gaseous phase,
however, in most cases the liquid form is the most convenient. Liquid samples
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
have the advantages of homogeneity and ease of transport (e.g. via tubing);
although, bulk liquids are not directly compatible with plasma instruments.
Methods for the introduction of liquid samples to plasma spectrometers have
been reviewed by a number of authors.1,7,8 The most common method for
introducing a liquid sample into a plasma spectrometer is in the form of an
aerosol, which allows liquids to be introduced as small drops which can be
desolvated in the plasma without significant energy loss or extinction. The
production of a fine aerosol is generally achieved with the use of a nebulizer,
to produce the aerosol, and a spray chamber which allows only the small
droplets to pass to the plasma. The use of nebulizers and spray chambers has
been reviewed comprehensively by Sharp.9,10 Pneumatic nebulizers are the
most common type of nebulizer due to their high sample throughput and
robustness, however, other nebulizers such as ultrasonic, thermospray and
hydraulic high pressure nebulizers tend to be more efficient in terms of
producing a greater amount of small droplets.10 For these more efficient
systems, it is often necessary to reduce the amount of solvent vapor that enters
the plasma by the use of a condenser, membrane dryer or by cooling the spray
chamber.10 In order to overcome the inefficiency problems of common
nebulizer sample introduction systems, desolvation techniques have been
used to evaporate solvent from the aerosol droplets and remove the excess
solvent vapor so that greater amounts of sample can be introduced to the
source without significantly increasing the amount of solvent loading.10
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The introduction of solids into a plasma source has been accomplished
in a number of ways.1 Laser ablation is a method for introducing a sample as
a fine dust into an atomic spectrometer and has been reviewed by Darke and
Tyson.11 Other solid sampling techniques have included electrothermal
vaporization, direct insertion and slurry introduction. The introduction of solid
particles into an ICP often leads to precision problems due to incomplete
vaporization in the plasma and inhomegeneity of the sample.1,12
The introduction of a gaseous sample has many benefits for plasma
spectrometry. A gas can be transported efficiently to the plasma, can allow the
element of interest to be separated from the matrix and offers the possibility of
preconcentration.8 The main disadvantage is the need to carry out the
chemistry necessary to produce the gaseous form of the species of interest.
While some elements can form an elemental vapor under the appropriate
conditions (e.g. Hg and Cd) often a volatile species of an element needs to be
formed (e.g. stable hydrides of As, Se, Sb, Sn, Bi, Pb and Te). Usually the
experimental conditions for the formation of gaseous species of different
elements are not the same. This makes simultaneous detection of different
elements difficult. Other disadvantages are the interelement interferences
from other hydride forming elements and the effect of the produced hydrogen
gas in the plasma.
Liquid sample introduction would appear to be the best method for the
introduction of samples for multi-element determinations with good accuracy
and precision, however, the low efficiency of liquid sample introduction
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
systems and solvent loading effects in the plasma are a substantial limitation.
Recent advances in the sample introduction to plasma spectrometry have
included the development of efficient nebulizers coupled to membrane
dryers,13*16 although these systems often suffer from sample loss during the
dryer stage. Advances in instrument design have overcome some
interferences effects caused by solvent loading.17 These have included the
use of a cool plasma,18 formed using a lower radio frequency power and
increased nebulizer gas flow rate, which has the effect of reducing
interferences for a limited number of elements. The use of ICP-MS
instruments with high resolution mass spectrometers can be used to
discriminate an element signal from a interfering signal but tend to be more
expensive than the more common quadrupole instruments. More recently the
use of a hexapole ion lens positioned behind the skimmer cone containing a
small amount of helium gas has been used to break down interfering
molecular species and significantly reduce interferences.19 The most
promising method for overcoming plasma cooling and reducing polyatomic
interferences caused by solvent loading would appear to be the use of
desolvation systems although the development of a system that can efficiently
vaporize a sample aerosol and remove the produced solvent vapor has still to
be achieved.
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.4 The Application of Microwave Radiation to Analytical Chemistry
Microwave radiation was first developed in 1933 as means to
communicate between France and England. It later found applications in
radar during the second world war and has been used for other
communications applications ever since.20 In 1945 the Raytheon Company
(Waltham, Massachusetts) patented microwaves as a way to cook food. The
manufacture of ovens for microwave cooking has since become a massive
industry, however, the use of microwave energy in chemistry laboratories has
proceeded at a much slower pace.
The mechanism of the microwave heating effect has been described by
a number of authors21'23and is generally considered to be a combination of
ionic conduction and dipole rotation. Ionic conduction is the migration of ions
in the applied field. This migration of ions is a flow of charge and causes the
formation of heat due to the resistance in the material in much the same way
as heat is produced in the wire when a electric current flows through it. Dipole
rotation refers to the alignment of molecules, due to the applied electric field, of
a sample that contains permanent or induced dipoles. When a electric field is
applied to molecule with a permanent or induced dipole, the molecule will
align itself to the field. When the electric field is reduced, the molecules will
return to disorder. The molecules in a material will take a certain time to return
to their original disorder, the relaxation time. At low frequencies these
rotations synchronize with the field but as the frequency increases, the inertia
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of the molecules causes the rotation to lag behind the field. The phase lag
between the rotation and the applied electric field leads to an absorption of
energy and Joule heating.
The main application of microwave energy in the laboratory is the area
sample digestion. This was initially accomplished using domestic ovens and
was first reported in 1975 by Abu-Samara et al.24 More recently microwave
digestion systems have gained in popularity leading to the development of
laboratory microwave ovens incorporating pressurized vessels, exhaust fans
to remove acid fumes, corrosion resistant cavities, improved power control and
temperature and pressure monitoring. A good review of the status of
microwave sample dissolution and decomposition for elemental analysis was
published in 198921 and discusses open and closed batch digestion systems
as well as microscale digestion and commercial microwave digestion systems.
Another good source of information can be found in the book edited by
Kingston and Jassie in 1988.22 More recently some interest has centered on
the development of on-line microwave systems which offers the advantages
over off-line methods of reduced digestion times. These advantages include
the ability to digest troublesome matrices, and dissolution in an enclosed
environment (reducing volatile losses, atmospheric contamination and
increases personal safety).23 The development of a commercial high pressure
on-line microwave digestion system that can provide rapid, complete
digestions of difficulty soluble sample matrices without suffering from
blockages has yet to be accomplished.
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The application of microwave energy to analytical and environmental
chemistry has been reviewed by Zlotorzynski25 who also discusses the theory
of microwave radiation in some detail. As well as sample digestion,
microwave radiation can be used in the extraction of various species from
many different types of samples. The extraction is typically accomplished by
mixing the ground sample with an appropriate solvent and irradiating with
microwaves. For flammable solvents, the necessity of a spark free cavity and
the effective exhaust of solvent fumes becomes very important. Other
applications of microwave energy in the chemistry laboratory have included
speeding up preconcentration steps,25 microwave catalysis25 microwave
induced plasmas1 and speeding up of organic synthesis.26 The theory of
microwave interaction with matter, however, is still not well understood and is
an area of debate. For instance a controversy still centers on the existence of
a ‘microwave effect’. This dispute questions whether the ability of microwaves
to increase reaction rates of some organic syntheses is due thermal effects,
such as hot spots, superheated solvents or temperature gradients or if there is
some other energetic effect that is caused by the microwave field directly.27
Attempts to use microwave energy to dry an aerosol for use in a
desolvation system have also been reported with limited success.28,29 By
increasing the knowledge of microwave theory and the effect of microwave
frequency on its ability to interact with matter it is hoped that the development
of a microwave desolvation system may be more successful and that
microwave energy in general can be used in a more effective manner as a tool
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in the chemistry laboratory for sample introduction and preparation prior to
analysis by atomic spectrometry.
1.5
References
1.
Montaser, A., and Golightly, D. W., “Inductively Coupled Plasmas in
Analytical Atomic Spectrometry”, 1987, VCH publishers (New York).
2.
“Sample Preparation Techniques", 1996, short course notes, University
of Massachussetts.
3.
Sulcek, Z., and Povondra, P., “Methods of Decomposition in Inorganic
Analysis”, 1989, CRC Press (Boca Raton, Florida).
4.
Lamble, K. J., and Hill, S. J., Analyst, 1998,123, 779.
5.
CEM Product literature, CEM Corporation, Matthews, NC.
6.
Legere, G., and Salin, E. D., Appl. Spectrosc., 1995,
7.
Koropchack, J. A., Spectroscopy, 1993, 8, 20.
8.
Browner, R. F., and Boom, A. W., Anal. Chem., 1984, 56, 788A.
9.
Sharp, B. L., J. Anal. At. Spectrom., 1988, 3, 613.
10.
Sharp, B. L., J. Anal. At. Spectrom., 1988.3, 939.
11.
Darke, S. A., and Tyson, J. F., J. Anal. At. Spectrom., 1993, 8, 145.
12.
Stewart, I., and Olesik, J. W., FACSS Conference, 1997, Abstract 462,
13.
Botto, R. I., and Zhu, J. J., J. Anal. At. Spectrom., 1994, 9, 777
14.
Brenner, I. B., Zhu, J., and Zander, A., Fresenius J. Anal. Chem., 1996,
335, 774.
15.
Tao, H., and Miyazaki, A., J. Anal. At. Spectrom., 1995,10, 1.
16.
Yang, J., Conver, T. S., Koropchak, J. A., and Leighty, D. A.,
Spectrochim. Acta, 1996, 51B, 1491.
49,
14A.
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17.
Evans, H. E., and Giglio, J. J., J. Anal. A t Spectrom., 1993, 8 ,1 .
18.
Niu, H., and Houke, R. S., Spectrchim. Acta, 1996,
19.
Product Literature, Micromass, http://www.micromass.co.uk, accessed 9
September 1998.
20.
Copson, D. A., “Microwave Heating”, 1962, AVI Publishing Company
(Westport, C l).
21.
Matusiewicz, H., and Sturgeon, R. E., Prg. Analyt. Spectrosc., 1 9 89 ,12,
21.
22.
Kingston, H. M., and Jassie, L. B., “Introduction to Microwave Sample
Preparation", 1988, American Chemical Society (Washington DC).
23.
Zhi, Z., Rios, A., and Valcarcel, M., Crft. Rev. Anal. Chem., 1996,
239.
24.
Abu-Samra, A., Morris, J. S., and Koirtyohann, S. R., Anal. Chem., 1975,
47, 1475.
25.
Zlotorzynski, A., Crit. Rev. Anal. Chem., 1995,
26.
Michael, D., Mingos, P., and Baghurst, D., Chem. Soc. Rev., 1991.
1.
27.
Dangani, R., C&EN, 1997,
28.
Gras, L., Mora, J., Todoli, J. L., Hemandis, V., and Canals, A.,
Spectrochim. Acta, 1997, 52B, 1200.
29.
Mora, J., Canals, A., Hemandis, V., van Veen, E. H., and de LoosVollebregt, M. T. C., J. Anal. At. Spectrom., 1998,13, 175.
February 10,
51B, 779.
26,
25, 43.
31.
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20,
CHAPTER 2
DEVELOPMENT OF AN ON-LINE PRESSURIZED MICROWAVE DIGESTION
SYSTEM
2.1 Introduction
Atomic spectroscopy techniques commonly require the conversion of a
solid sample material to a homogeneous liquid. This can often be the most
time-consuming and error-prone procedure in the analytical method. For
biological samples the oxidation of organic materials to carbon dioxide and
water can usually be accomplished with nitric acid given sufficient digestion
time and temperature.1 Microwave heating has recently become a more
popular technique than hot plate methods for the digestion of biological
samples due to its ability to achieve high temperatures rapidly.2 Pressurized
microwave digestion methods can reduce the digestion time even further by
allowing higher temperatures to be achieved.2 On-line digestion procedures
have been developed in order to reduce contamination and human error in
sample preparation while accomplishing digestions of samples quickly and
easily. Incorporating a microwave oven into a flow system has the advantages
of reducing digestion times, allowing the digestion of some troublesome
matrices and ensuring dissolution in an enclosed system which reduces
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
contamination from atmospheric sources, analyte losses and hazards due to
acid fumes.3 Initial on-line microwave digestion systems operated at
atmospheric pressure by mixing samples with acid in a coiled tube situated
inside a microwave field. The temperatures generated in these systems were
not high enough to digest many biological materials fully although analyses
have been performed on whole blood for the determination of copper, zinc and
iron by flame atomic absorption spectroscopy,4 and with in vivo sample uptake
for the determination of cobalt by electrothermal atomic absorption
spectrometry.5 The same authors also investigated the digestion of human
kidney and liver tissue using a system which incorporated six glass test tubes
located in a Pyrex jar inside a microwave oven. The samples were
simultaneously heated in acid at atmospheric pressure and then pumped into
a flow system and analyzed by flame atomic absorption spectrometry.6
Sewage sludge has been studied in a number of papers after an on-line acid
leaching step with microwave heating and subsequent analysis of the
resulting slurry by atomic spectroscopy.7'9 Water samples have also been
investigated following a microwave treatment step for the determination of total
phosphorus,10,11 and metals.12 Other authors have determined aluminum13 and
selenium14 in shellfish samples, lead in beverages and fruit slurries using an
on-line digestion hydride generation technique,15 and a number of metals in
biological materials.16,17 Similar systems have incorporated a single-mode
microwave cavity to provide more efficient heating of the sample and have
been used successfully for the analysis of water and urine for the
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
determination of mercury, arsenic, lead and tin using microwave sample pre­
treatment with on-line chemical vapor generation techniques and atomic
absorption spectrometry.18*20 Other samples which have been analyzed after
pretreatment in a single-mode microwave system include whole blood for the
determination of mercury by cold vapor atomic absorption spectrometry;21 milk,
blood and urine using ionically coupled plasma atomic emission
spectrometry;22 environmental materials for the determination of mercury by
atomic fluorescence spectrometry23 and waste waters for the determination of
total phosphate using colorimetric detection.24 In order to increase digestion
temperatures for on-line digestions pressurized system have been developed.
Haswell and Barclay25 used at restrictor at the outlet of the digestion tubing to
increase the digestion pressure. A similar method was used to increase the
digestion temperature using a single mode microwave.26 This concept has
been developed into commercial systems and have been used to digest
various biological samples.27*29 Karanassios et al.30 used a stopped-flow
method in order to increase digestion time and pressure for the analysis of
biological materials. These pressurized techniques have been successful for
a variety of samples but are limited in terms of the pressure and efficiency of
mixing between the sample and acid. Gluodenis and Tyson31 overcame some
of these limitations by the development of a stopped-flow digestion system,
incorporating a vertically mounted glass digestion tube, which was capable of
completely digesting biological materials at pressures up to 400 psig. This
system had the limitation of requiring solid samples to be made into a stable
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
slurry and was prone to contamination due to acid attack of the digestion tube
end fittings.32 Legere and Salin33 suggested the use of polyacrylamide
capsules as a means of introducing solid samples into a pressurized on-line
digestion system. This system incorporated a PFA U-tube that allowed a
working temperature of 200°C and showed good mixing due to boiling
processes inside the tube.
In this chapter further developments of the glass tube digestion system,
based on the design by Gloudenis and Tyson,31 are described. A pressurized
stopped-flowed microwave digestion system has been developed
incorporating a glass U-tube which permits high pressure operation and good
mixing of the digesting material while reducing the amount of contact between
the acid and the end fittings in order to reduce contamination. The system can
be used in either slurry sample introduction or direct solid sample introduction
modes and can achieve digestion of biological samples rapidly using nitric
acid alone.
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.2 On-line Digestion and Analysis of Milk Powder
2.2.1 Experimental
2.2.1.1 Development of a Pressurized On-Line Digestion System
The purpose of this work was to develop an on-line pressurized
microwave digestion system based on the system previously developed by
Gluodenis32 that could be used to digest biological matrices quickly without the
need for a filtration step while avoiding sources of contamination and allowing
the system pressure to be measured accurately. To remove sources of
contamination in the system, a U-tube was constructed from a 10 cm long 1 cm
i.d borosilicate glass tube (Omnifit) so that the amount of contact between the
hot acid and the PTFE end-fittings was reduced. In order to remove the
sample after digestion, a length of PTFE tubing was inserted into the U-tube
which could be used to pump out the digestion mixture. The system
developed for the digestion of milk powder samples is shown in Figure 2.1. A
peristaltic pump (Ismatec) with Tygon pump tubing (Cole Palmer) was used to
pump acid reagent through PTFE tubing to the digestion vessel which was
positioned inside a CEM MDS-81 microwave oven. For milk powder
digestions, a 4-way PTFE slider valve with flow channels drilled to 1.5 mm i.d
and a 0.2 ml injection loop was used to introduce sample in the form of a
slurry. All tubing was made of 0.8 mm i.d. PTFE except for the tubing from the
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
slider valve, V7, to V3 and between V3 and the digestion vessel which was
1.5 mm i.d to reduce blockages in the system. All manifold connections were
made with high-pressure gripper fittings (Omnifit).
2.2.1.2 Development of a Pressure Transducer
In order to achieve an accurate measurement of pressure during the
digestion procedure, an electronic pressure transducer was connected to the
pressure line of the digestion system. The pressure in the system was
measured with a Sensotec A-105 pressure sensor and Sensotec model GM
digital readout unit connected to the pressure line via a pressure transducer
housing constructed in-house as shown in Figure 2.2 which is based on a
previous design.34 The pressure transducer housing was constructed from a 1
inch diameter cylinder of Kel-F material. The cylinder was drilled such that the
pressure sensor could be screwed into the top and fitted tightly against a disk
constructed of 0.025 inch thick PTFE which sits on top of a shallow well. The
well was connected to the pressure line by two PTFE pressure fittings which
were screwed into the bottom of the block.
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.2.1.3 Development of the Microwave Program
The microwave program was designed to heat the sample to the point
where the organic matrix would begin to breakdown and form carbon dioxide
gas. In order to prevent a pressure ‘spike’ which occurs when the pressure
rises too quickly and is no longer under control, a second stage was added
with 0% power during which time the pressure increased to a maximum and
began to decrease due to cooling. A final microwave stage was then
conducted to remove any remaining organic material. The pressure ranges
indicated in Table 2.2 were the final pressures seen on the completion of each
stage and vary according to the actual mass of organic material in the vessel.
It is also worth noting that the oven needed to be run for a short period before
a digestion in order to warm up the magnetron to achieve reproducible
pressures in the system.
2.2.1.4 Reagents
Certified reference material Milk Powder A11 was obtained from the
International Atomic Energy Agency (IAEA). Milk powder slurries were
prepared in deionised water and digested with reagent grade nitric acid
(Fisher). All subsequent dilutions were prepared with deionised water.
Standard solutions for flame atomic absorption determinations were prepared
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
by the dilution of 1000 mg I'1 stock solutions of calcium, zinc, sodium and
magnesium standard solutions (Fisher) with deionised water and the
appropriate amount of nitric acid to achieve acid matching with the digested
solutions.
2.2.1.5 Spectrometer
For milk powder digests, the analysis was performed with a PerkinElmer 1100B flame atomic absorption spectrometer using parameters shown
in Table 2.1.
2.2.1.6 Procedure
Milk powder samples were introduced to the digestion system in the
form of a slurry. Slurries were prepared by mixing approximately 1.00 g of
powder with 9.00 g of deionised water and 0.4 ml of slurry was then introduced
to the digestion vessel using two repeat injections through the slider valve, V7,
by pumping the earner line with air with valve V3 open to the digestion vessel
and V4 closed. The retractable tubing was removed from the digestion vessel
during this stage by unscrewing the PTFE connector from the end-fitting and
sliding the tubing out of the vessel. Nitric acid (1.0 ml) was then pumped
through the earner line to ensure that all of the sample was transferred to the
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
digestion vessel. V3 was then closed to the carrier line and opened to the acid
line. With V1 open and V2 open to V3 and closed to the waste line the acid
line was filled with nitric acid. When the acid line was full, the retractable
tubing was reconnected to the vessel and V1 closed so that any pressure rise
in the system would cause the acid in the pressure transducer to push against
the membrane and allow for pressure measurement in the system by the
pressure sensor. The system was then heated using the microwave program
shown in Table 2.2. After heating, the system was then allowed to cool until
the pressure had reduced to less than 100 psig. The pressure was then
relieved by opening V4 with V5 open to the pressurization system and V6
closed. The pressure could then be safely reduced to atmospheric pressure
by slowly opening V6. The digested material was collected by pumping nitric
acid through the acid line with V1 open until the liquid had reached a known
level marked on the first depressurization tube, then switching V5 such that it
was closed to the digestion vessel and open to the collection vessel. The
liquid was then pumped through the collection line into a 10 ml volumetric
flask. V5 was then opened to the digestion vessel and more acid pumped
through the acid line up to the mark on the depressurization tube. This
assured complete transfer of the digested material. The acid was collected as
before into the same collection vessel by switching V5 and pumping the
collection line. The collected material was made up the mark with deionised
water. A clear, colorless or slightly yellow solution of approximately 50% nitric
acid by volume was obtained. It was not found necessary to filter any of the
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
digested solutions. The digestion system was cleaned by pumping deionised
water. The digested material was diluted 10-fold with deionised water and
analyzed by flame atomic absorption spectrometry against acid matched
standards.
2.2.2 Results and Discussion
The results for the analysis of digested SRM milk powder A-11 samples
by flame atomic absorption spectrometry are shown in Table 2.3. The results
are found to be in good agreement with the certified values for calcium,
magnesium and sodium with good precision for replicate digestions. The
results for zinc are significantly higher than the certified value. An
investigation was conducted into possible sources of contamination in the
system. Significant concentrations of zinc were found by pumping acid
through a new piece of Tygon pump tubing, diluting to 5% acid with deionised
water and analyzing by flame atomic absorption spectrometry. The amount of
zinc found was more than enough to account for the higher zinc concentration
in the results and indicates that zinc may be leached from new pump tubing by
concentrated nitric acid. In subsequent digestions the pump tubing was
washed through with concentrated acid before use.
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3 On-Line Digestion and Analysis of Bovine Liver
2.3.1 Experimental
2.3.1.1 Modification of the Pressurized On-Line Digestion System
In the case of bovine liver samples, it was impossible to produce a
stable slurry. The bovine liver particles tended to aggregate in solution
causing a blockage in the slider valve. It was therefore found necessary to
develop a procedure for the direct addition of solid samples into the digestion
vessel. This could easily be achieved by removal of the U-tube from the
system and weighing the sample directly into it. For this procedure the
apparatus was simplified slightly as shown in Figure 2.3 to remove the slider
valve and tubing that were necessary for slurry introduction.
2.3.1.2 Reagents
Digestions of bovine liver were achieved with sub-boiled nitric acid and
diluted with deionised water. Standard solutions for determination by ICP-MS
were prepared by diluting standard solutions of silver (Spectrum), cobalt (Alfa),
strontium (Spectrum), lead (Johnson Matthey), cadmium (Aldrich), manganese
(Fisher), rubidium (Spex), calcium (Fisher), iron (Johnson Matthey) and zinc
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Fisher) with the appropriate amount of deionised water and sub boiled nitric
acid to achieve acid matching with the digested samples.
2.3.1.3 Spectrometer
ICP-MS analysis of bovine liver was performed with an Elan 5000 ICPMS instrument using parameters shown in Table 2.4. The Nation dryer device
(Permapure) described previously35 was used to reduce possible polyatomic
interferences.
2.3.1.4 Procedure
Bovine liver samples were added to the digestion vessel directly by
unscrewing the end fittings and weighing approximately 30 mg the bovine liver
powder into the digestion tube. The digestion tube could then be reattached to
the system and the digestion earned out using a similar method as that for milk
powder. Acid was added to the bovine liver sample through the acid line with
V1 open, V2 open to V3 and V3 open to the digestion tube. The retractable
tubing was withdrawn with V4 closed during the acid addition stage. Sub­
boiled nitric acid was pumped through the acid line for 75 s at a rate of 2
ml/min in order to fill the acid line and add sufficient acid (approx. 1.5 ml) to the
sample. V1 was then closed, the retractable tubing replaced and the system
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
heated using the microwave program shown in Table 2.5. After allowing the
system to cool, the digested material was collected in the same manner as for
the milk powder digestion, resulting in a colorless or slightly yellow clear
solution of digested bovine liver in 50% acid solution. It was not found
necessary to filter any of the digested solutions. The digests were diluted 10fold with deionised water and analyzed by ICP-MS against acid matched
standards.
2.3.2 Results and Discussion
Results for bovine liver digestions measured in the absence of the dryer
are shown in Table 2.6. Results for silver, strontium and lead are not shown as
the concentrations measured for these elements were found to be below the
detection limits of the instrument. Due the high values and poor precision
obtained for some elements, the measurements were repeated using the
Nafion dryer device and the results shown in Table 2.7. The use of the dryer
allowed the measurement of Fe-56 due to the reduction of the ArO+
interference and gave an improvement in terms of accuracy and precision for
some elements, however, values obtained for Fe, Ca, Mn and Rb were still
found to be significantly higher than the certified values. Results for the
analysis of a blank solutions obtained by repeating the digestion procedure in
the absence of any sample material are shown in Table 2.8 (measured without
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the Nation dryer) and indicate significant contamination for iron and calcium
and a lesser contamination of rubidium and manganese. The concentrations
of these values in the blank solutions are low due to the lower pressures that
can be achieved in the absence of a sample material. The concentrations of
the elements found in the blank did not show any noticeable trend that might
suggest a memory effect in the system. The main source of contamination in
the system is likely to be from the glass digestion vessel which is the only
material in direct contact with the digestion mixture during the heating stage
and is known to contain significant concentrations of calcium as well as iron,
lead, manganese, rubidium and strontium.36 This can account for the high
values of calcium, iron, manganese and rubidium found in the sample
solutions. The reason for the poor precision found for cadmium is unknown.
2.4 Conclusions
An on-line microwave digestion system has been developed which is
capable of digesting biological samples at elevated pressures with nitric acid
alone. The design of the system reduces the contact between the digestion
mixture and the end-fittings reducing the possibility for contamination and
possible deterioration of the end-fittings caused by acid attack. The new
design is also thought to increase mixing of the system to improve the
digestion process. The pressure measurement system allows the pressure to
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
be measured with precision and accuracy. The system produced clear
solutions for digested milk powder and bovine liver samples in a short period
of time giving acceptable results for most elements. Contamination in the
system from the glass digestion vessel was not a problem for the milk powder
samples due to the higher concentrations of elements in the analyzed
solutions however it seems evident that the use of glass as a vessel for
performing digestion prior to trace element determinations leads to poor
accuracy and precision for some elements.
2.5
References
1.
Moede-Wedler, T., MS. thesis, 1994, University of Massachusetts.
2.
Kingston, H. M., and Haswell, S., “Microwave Enhanced Chemistry”,
1997, American Chemical Society (Washington DC).
3.
Zhi, Z., Rios, A., and Valcarcel, M., Crit. Rev. Anal. Chem., 1996, 26,
239.
4.
Burguera, M., and Burguera, J. L., and Alarcon, O. M., Anal. Chim. Acta.,
1986, 179, 351.
5.
Burguera, M., Burguera, J. L., Rondon, C., Rivas, C., Carrera, P.,
Gallignani, M., and Brunetto, M . R., J. Anal. At. Spectrom., 1995,10,
343.
6.
Burguera, M., Burguera, J. L., and Alarcon, O. M., Anal. Chim. Acta.,
1988, 214, 421.
7.
Carbonell, V., de la Guardia, M., Salvador, A., Burguera, M., and
Burguera, J. L., Anal. Chim. Acta., 1990, 238, 417.
8.
Bordera, L., Hemandis, V., Mora, J., and Canals, A., Fresenius J. Anal.
Chem., 1996, 355, 112.
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9.
de la Guardia, M., Carbonell, A., Morales-Rubio, A., and Salvador, A.,
Talanta, 1993, 40, 1609.
236,
10.
Hinkamp, S., and Schwedt, G., Anal. Chim. Acta., 1990,
11.
Benson, R. L., McKelvie, I. D., Hart, B. T., and Hamilton, I. C., Anal. Chim.
Acta., 1994, 291, 233.
12.
Maksimova, I. M., Morosanova, E. I., Kuz’min, N. M., and Zolotov, Y. A.,
Fresenius J. Anal. Chem., 1997, 357, 946.
13.
Arruda, M. A. Z., Mercedes, G., and Valcarcel, M., J. Anal. At. Spectrom.,
1995, 10, 501.
14.
Arruda, M. A. Z., Mercedes, G., and Valcarcel, M., J. Anal. At. Spectrom.,
1996, 11, 169.
15.
Cabrera, C., Madrid, Y., and Camara, C., J. Anal. At. Spectrom., 1994,
1423.
16.
Carbonell, V., Morales-Rubio, A., Salvador, A., de la Guardia, M.,
Burguera, J. L., and Burguera, M., J. Anal. At. Spectrom., 1992, 7, 1085.
17.
Burguera, M., and Burguera, J. L., J. Anal. At. Spectrom., 1993, 8, 235.
18.
Tsalev, D. L., Sperling, M., Welz, B., Analyst, 1992,117, 1729.
19.
Tsalev, D. L., Sperling, M., Welz, B., Analyst, 19 92 ,117, 1735.
20.
Welz, B., Tsalev, D. L., Sperling, M., Anal. Chim. Acta., 1992,
21.
Guo, T., and Baasner, J., Talanta, 1993,
22.
Martines Stewart, L. J., and Barnes, R. M., Analyst, 1994,119, 1003.
23.
Morales-Rubio, A., Mena, M. L., McLeod, C. W., Anal. Chim. Acta., 1995,
308, 364.
24.
Cerda, A., Oms, M. T., Forteza, R., and Cerda, V., Anal. Chim. Acta.,
1997, 351, 273.
25.
Haswell, S. J., and Barclay, D., Analyst, 1 9 9 2 ,117, 117.
26.
Welz, B., He, Y., and Sperling, M., Talanta, 1993,
27.
Williams, K. E., Haswell, S. J., Barclay, D. A., and Preston, G.,Analyst,
1993, 118, 245.
40,
345.
9,
261, 91.
1927.
40,
1917.
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28.
Sturgeon, R. E., Willie, S. N., Methven, B. A., Lam, J. W. H., and
Matusiewicz, H., J. Anal. At. Spectrom., 1995,10, 981.
29.
Beary, E. S., Paulsen, P. J., Jassie, L. B., and Fassett, J. D., Anal.
Chem., 1997, 69, 758.
30.
Karanassios, V., Li, F. H., and Salin, E. D., J. Anal. At. Spectrom., 1991,
6, 457.
31.
Gluodenis, Jr, T. J., and Tyson, J. F., J. Anal. At. Spectrom., 1993, 8,
697.
32.
Gluodenis, Jr, T. J., PhD. Dissertation, 1993, University of
Massachusetts.
33.
Legere, G., and Salin, E. D., Appl. Spectrosc., 1995, 49, 14A.
34.
Legere, G., Personal communication, 1994.
35.
Fitzgerald, N„ Tyson, J. F., and Leighty, D. A., J. Anal. At. Spectrom.,
1998, 13, 13.
36.
SRM 611, NIST Catalogue, 1998, NIST (Gaithersburg, MD).
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.1. Instrument parameters for the determination of milk powder by
flame atomic absorption spectrometry.
Parameter
Calcium
Magnesium
Sodium
Zinc
Wavelength/nm
422.7
285.2
589.0
213.9
Lamp current/mA
10
4
6
10
Bandpass/nm
0.7
0.7
0.2
0.7
Acetylene flow rate/l min'1
2.0
2.0
2.0
2.0
Air flow rate/l min'1
8.0
8.0
8.0
8.0
Number of replicates
3
3
3
3
Integration Time/s
3
3
3
3
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.2. Microwave digestion program for milk powder.
Program stage
Power/%
Time/min
Pressure
range/psig
1
50
1
160-280
2
0
1
0-130
3
45
1
50-300
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.3. Results for the digestion of SRM A-11 milk powder slurries by flame
atomic absorption spectrometry.
Element
Concentration found
/gkg"1
Certified concentration
/gkg'1
Ca
12.3 (±1.2)
12.9 (±0.8)
Mg
1.17 (±0.09)
1.10 (±0.08)
Na
4.17 (±0.62)
4.42 (±0.33)
Zn
0.0698 (±0.0043)*
0.0389 (±0.0023)
values in parentheses indicate 95% confidence intervals based on 5 replicate
digestions
* based on 4 replicate digestions
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.4. Instrument parameters for the determination of bovine liver by ICPMS using a Nafion dryer sample introduction system.
ICP-MS Parameter
Power/W
1100
Nebulizer gas flow/l min'1
0.95
Plasma gas flow/l min'1
15.0
Auxiliary gas flow/l min'1
0.90
Number of replicate readings
15
Dwell time per reading/ms
200
Nafion dyer parameters
Dryer temperature/°C
58
Purge gas flow/l min'1
15
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.5. Microwave digestion program for bovine liver.
Program stage
Power/%
Time/min
Pressure
range/psig
1
30
1
120-350
2
0
1
160-230
3
10
1
240-270
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.6. Results for the digestion of SRM 1577b bovine liver by ICP-MS.
Concentration found in
Certified concentration
solid /mgkg'1
/mg kg'1
Mo-98
2.7(±0.3)*
3.5(±0.3)
Co-59
0.28**
0.25*
Cd-114
0.42(±0.34)
0.50(±0.03)
Mn-55
23.0(±1.1)*
10.5(±1.7)
Ca-44
404(±194)
116(±4)
Fe-57
247(±106)*
184(±15)
Zn-64
91 (±14)*
127(±16)
Rb-85
25.3(±5.7)
13.7(±1.1)
Element
Values in parentheses indicate 95% confidence intervals based on 4 replicate
digestions
* based on 3 replicate digestions
** based on 2 replicate digestions
* information value
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.7. Results for the digestion of SRM 1577b bovine liver by ICP-MS
Using a Nafion Dryer.
Element
Concentration found in
Certified concentration
solid /mgkg'1
/mg kg'1
Mo-98
3.8(±0.3)*
3.5(±0.3)
Co-59
0.68(±0.15)*
0.25#
Cd-114
0.50(±0.29)
0.50(±0.03)
Mn-55
19.9(±4.9)
10.5(±1.7)
Ca-44
586(±42)
116(±4)
Fe-56
279(±146)*
184(±15)
Zn-64
91 (±10)
127(±16)
Rb-85
23.5(±4.3)
13.7(±1.1)
Values in parentheses indicate 95% confidence intervals based on 4 replicate
digestions
* based on 3 replicate digestions
* information value
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.8. Results for the determination of 6 blank solutions by ICP-MS.
Isotope determined
Mean concentration found in solution
/p.g dm'3
Pb-208
0
Mo-98
0
Ag-107
0
Co-59
0
Sr-88
0
Cd-114
0
Mn-55
0.1
Rb-85
0.1
Ca-44
4.1
Fe-57
1.2
Zn-64
0
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to exhaust
pressure
transducer
microwave cavity
v6
acid
v2
earner
2-stage
depressuization
system
waste
retractable
tubing
v5
collection
vessel
digestion
tube
Figure 2.1. Schematic diagram of the on-line pressurized microwave
digestion system used for the digestion of milk powder.
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pressure sensor
PTFE membrane
pressure line
pressure line
Figure 2.2. Schematic diagram of pressure transducer.
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to exhaust
pressure
transducer
microwave cavity
v6
2-stage
depressurization
system
waste
acid
retractable
tubing
v4
v5
collection
vessel
digestion
tube
Figure 2.3. Schematic diagram of the on-line pressurized microwave
digestion system used for the digestion of bovine liver.
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3
REDUCTION OF WATER LOADING EFFECTS IN ICP-MS USING NAFION
MEMBRANE MATERIAL
3.1 Introduction
Inductively coupled plasma source mass spectrometry is a modem
technique which has gained in popularity due to its low detection limits and
ability to achieve multi-element determinations almost simultaneously. The
instrument’s performance is often critically affected by the sample introduction
system. Aqueous samples are usually preferred due to their homogeneity and
ease of handling; however, problems are encountered due to the solvent load
in the plasma. Hutton and Eaton1 have shown that water from an aerosol can
reduce the efficiency of the plasma for ion production and produce interfering
polyatomic and doubly charged species which are problematic for the
determination of some elements. Desolvation systems have been developed
to improve the sample introduction procedure by evaporating the solvent from
the aerosol and removing the vapor produced.2 This condensation step has
been found to reduce oxide and double charged species but is unable to
eradicate water loading completely. This step also provides a mechanism for
sample loss and memory effects by condensation of the solvent back onto the
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
sample particles or loss of particles onto the cold surfaces. Alves and
coworkers3 used a cryogenic desolvation system which allowed repetitive
heating and cooling of an aerosol inside copper coils as a means of removing
organic solvents. This system has also been found to reduce chloride species
which is particularly advantageous for the determination of arsenic which often
suffers from an ArCI+ interference.4 Removal of solvent vapor via a membrane
separator is an alternative to condensation and was first considered for atomic
spectrometry by Gustavsson and coworkers.5-6 They used a silicone polymer
membrane for the removal of organic solvents such as Freon and chloroform
with efficiencies of 80 to 100%. A semi-permeable PTFE membrane has also
been used successfully for solvent reduction prior to introduction of a sample
into a plasma.7,8 This system while being chemically resistant and efficient for
solvent reduction is not specific for the solvent and may allow sample loss
through the permeable membrane material. A non-porous polyimide
membrane separator has been used as an alternative to porous membrane
dryers.9 An aerosol produced by a cross flow nebulizer and double pass
spray chamber was passed through the polyimide dryer into a ICP-MS
instrument. The high permeation rate for water vapor with this system led to
much reduced oxide intensities. Problems were encountered due to
electrostatic charge build up in the transfer tubing of the system and long times
were required for the analyte signal to reach a steady state.
Nafion has many properties which make it suitable for use as a
membrane separator for aqueous samples. It is a non-porous ion exchange
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
material which allows small polar molecules to pass through while sample
particles and non-polar molecules are unaffected. The chemical structure of
Nafion, shown in Figure 3.1, has a polymer backbone which is similar to PTFE
making it very resistant to chemical attack. The sulfonic acid group side chains
have the ability to attract small polar compounds which can then pass through
the material and be removed by a dry purge gas. These properties have been
successfully utilized by researchers for reducing water vapor in the
determination of mercury.10 Nafion has also been used to remove water vapor
for the determination of arsenic and selenium11 by chemical vapor generation
atomic fluorescence spectrometry and in chemical vapor generation atomic
absorption spectrometry.12 Recently the use of a multi-strand Nafion dryer for
aerosol desolvation for sample introduction to inductively couple plasmaatomic emission spectrometry has been investigated.13,14 An aerosol from a
thermospray nebulizer was introduced into a heated spray chamber. The bulk
of the solvent vapor was then removed with a Friedrichs condenser and the
moisture content further decreased with a 200-strand Nafion dryer. This
system showed a significant reduction in detection limits for a number of
elements with improved precision compared to the thermospray system in the
absence of the dryer and with conventional pneumatic nebulization sample
introduction.
The addition of a molecular gas to the plasma has been proposed as a
method for increasing sensitivity in ICP-MS.15,16 The mechanism for this
improvement in sensitivity is thought to be due to a greater heat transfer to the
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
analyte particles.17 Akino et al. found an Improvement In sensitivity with an
addition of nitrogen to an argon plasma with a membrane desolvation system.
They suggested that the reason for this improvement is that hydrogen is need
for the heat transport properties in the plasma. Desolvation effectively reduces
the amount of hydrogen present by reducing the amount of hydrogen and
oxygen which are introduced to the plasma in the form of water.18 Addition of
nitrogen to the plasma in the presence of a desolvation system has also been
studied by Lam etal.19 who did not observe an increase in sensitivity for
uranium.
This chapter shows the results of an investigation into the ability of a
Nafion dryer to reduce the water load in ICP-MS for a conventional nebulizer
and spray chamber sample introduction system. The reduced water load
leads to a decrease in the signal of the oxide species and improvement in the
detection limits for iron at m/z 56 and 57 which suffer from isobaric overlap
from the 40Ar16O and 40Ar16O1H species. The effect of heating the spray
chamber and addition of nitrogen in the presence of the dryer system were
also investigated.
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2 Experimental
3.2.1 Instmmentation
A Perkin-Elmer (Norwalk, CT) Elan 5000 ICP-MS instrument was used
with the operating conditions shown in Table 3.1. The dryer system, shown in
Figure 3.2, contained a Perma Pure (Toms River, NJ) PD-060 dryer consisting
of 56 strands of Nafion material, inside a metal box which was heated in one
half in order to achieve a temperature gradient across the dryer. A
conventional cross-flow nebulizer and double pass spray chamber were
attached to the heated end of the dryer while the other end was attached to the
plasma torch via a Teflon connector and approximately 1 m of PTFE tubing.
An aerosol was generated in an argon gas stream and passed through the
Nafion strands which allow small polar molecules, such as water vapor, to
pass through and be removed by a purge gas stream. The temperature of the
exiting purge gas stream was monitored and could be controlled. The
resulting sample was then introduced into the plasma allowing the
measurement of analyte with a reduction of oxide interferences.
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2.2 Reagents
Iron standards were prepared by the dilution by weight of 1000 mg dm'3
ICP Fe standard solution (Johnson Matthey) in 1% nitric acid (sub boiled
analytical grade) with deionized water. A 10 jig dm'3 test solution was
produced by a similar dilution in 1% acid of a 100 mg dm'3 test solution
prepared from 1000 mg dm'3 barium, rhodium, magnesium, cerium and lead
ICP standard solutions (Johnson Matthey).
3.2.3 System Optimization
ICP-MS operating conditions were optimized for forward power, plasma
gas flow, auxiliary gas flow, nebulizer gas flow and sample flow by a univariate
method for the dryer system and for a normal spray chamber sample
introduction system. Optimized conditions for the conventional sample
introduction system (Table 3.1) were found to be very similar to the conditions
for the dryer system suggesting that the use of the dryer does not require
extensive reoptimization of the ICP-MS operating conditions.
The dryer system was optimized by a univariate method for nature of the
purge gas, purge gas flow rate and exiting purge gas temperature using a 10
|ig dm'3 test solution of cerium and barium for low CeO+/Ce+, Ba07Ba+, and
Ba2+/Ba+ratios and ArO+ intensity. Dry nitrogen was used as the purge gas in
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
order to prevent any possibility of oxygen entering the sample gas stream at
higher temperatures.
3.2.4 Dryer Efficiency
The efficiency of the dryer system to remove water vapor from a spray
exiting a conventional spray chamber was determined with the use of drierite
tubes (Fisher). Weighed drierite tubes were attached at the exiting gas stream
and the exiting sample stream. An aerosol of deionized water was introduced
into the system at 1.2 cm3min*1 using 0.95 dm3min*1 of argon as the nebulizer
gas. The increases in mass of the tubes after the system was run for 20
minutes was assumed to be due to the total mass of water introduced and the
efficiency of transfer across the membrane could then be calculated.
3.2.5 Detection Limit for Iron
In order to demonstrate the advantages of reduced water loading in
ICP-MS determinations the detection limits for Fe-56 and Fe-57 isotopes were
measured with the dryer sample introduction system and compared to the
detection limits with conventional sample introduction under optimized
conditions. Flow injection experiments were conducted using a Perkin-Elmer
FIAS 200 unit using the program shown in Table 3.2. Sample was introduced
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
through a 380 |iL sample loop. Straight line calibration curves were produced
by flow injection analysis of 10, 20 and 50 (ig dm'3 standards in 1% nitric acid.
Data points were collected with a dwell time of 50 ms. Data was then exported
to Microsoft Excel for data treatment to allow a more flexible treatment of data
and prevent overloading the available memory of the Elan 5000 software. The
detection limits were calculated using the flow injection peak for a 10 (ig dm'3
standard. The first 50 data points in each experiment were taken to be the
blank signal and the detection limits calculated as 3 times the standard
deviation of the blank divided by the sensitivity based on peak height.
3.2.6 Addition of Nitrogen to the Plasma
Nitrogen was introduced into the plasma with the dryer present and the
intensities of iron in a test solution monitored in order to investigate the
possibility of increasing the efficiency of the plasma by the addition of nitrogen
gas to the plasma. Initial attempts to add nitrogen gas to the nebulizer gas
before the nebulizer via a PTFE T-piece were unsuccessful due the back
pressure produced by the nebulizer which forced argon gas back through the
nitrogen gas line. In order to overcome this problem, nitrogen gas was
introduced via the oxygen mass flow controller of the ICP-MS to a T-piece
which was positioned in the line between the dryer and the plasma torch. The
intensity of a flow injection peaks of 10 ng I*1 of iron in a 1% nitric acid solution
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
was measured at various nitrogen flow rates with the argon kept at a constant
flow rate of 900 ml min'1.
Nitrogen was also added to the plasma by the addition of acetonitrile to
sample solutions. 0%, 5% and 10% solutions of acetonitrile containing 10 pg I'
1 iron were introduced using a flow injection system and the limit of detection
based on peak height for four replicate determinations calculated.
3.2.7 Heated Spray Chamber
In order to increase the degree of desolvation of the system the spray
chamber was wrapped with heating tape (Bamstead Thermalyne briskheat)
with the control box (Geo. Ulanet Co. model 515) set to 10 and left for 10
minutes to heat up. A 10 pg I'1iron solution was then introduced via a FIAS
200 unit.
3.3 Results and Discussion
3.3.1 Development of Drver System
Initial experiments were carried out with a 200 strand dryer but the back
pressure did not allow sample to reach the plasma source. The back pressure
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
built up in the spray chamber and pushed sample through the drain line.
Therefore, a 56 strand dryer was used with wider bore PTFE tubing of 3.2 mm
id and a PTFE fitting to the plasma torch constructed with a 6.4 mm id bore and
a screw thread in order to fix it securely to the plasma source inlet. The dryer
device was designed so it could be heated at the inlet end to provide a
temperature gradient along the length of the strands. This temperature
gradient is required in order for the dryer to operate at its greatest efficiency.20
The aerosol flow rate was kept constant at the optimum for the spectrometer
performance at 0.9 dm3 min'1 as this has a critical impact on the performance of
the plasma source mass spectrometer.
3.3.2 Efficiency for Water Removal
The efficiency of the dryer for water removal as a function of purge gas
temperature using drierite tubes is shown in Table 3.3. The system reached a
maximum efficiency for the removal of water vapor for a spray in an argon gas
stream at about 60IC with a maximum efficiency for this system of 97% (± 1%).
The precision of these results was determined by the replicate (n=5)
determination of the efficiency at 58 °C.
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.3.3 System Optimization
The effect of dryer temperature on the Ce07Ce+, BaO+/B a \ and Ba27Ba+
ratios and ArO* intensity is shown in Figure 3.3. The results indicate that an
optimum temperature of about 601C is necessary to achieve low ratios for
cerium and barium and a low argon oxide intensity. This result is in
agreement with the results for dryer efficiency determined by using drierite
tubes and suggests that the dryer removes water more efficiently at about
60°C which leads to lower concentrations of oxygen in the plasma and,
therefore, lower concentration of oxides. The effect is particularly apparent for
CeO+ and ArO+ intensities with a smaller variation in intensity found for BaO+
and Ba2+. The degree of dependence of oxide intensity on the water content in
the plasma tends to be due to the strength of the oxide bond with the strongly
bound species showing greater dependence on the water content in the
plasma.21 The optimum temperature for water removal of 60°C was an
unexpected result as it was thought that efficient removal of solvent would be
more likely at temperatures above the dew point of water when a larger
amount of water vapor would be present and subsequently removed from the
system more efficiently by the membrane. Similar results were obtained for
the nitrogen and air purge gases indicating that the effect of purge gas on the
oxide ratios was negligible. Nitrogen was used as the purge gas in all
subsequent experiments.
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The effect of purge gas flow rate on oxide ratios is shown in Figure 3.4
and indicates little effect for BaO+/Ba+ and Ba27Ba+ ratios while the CeOVCe*
ratio shows a minimum between 5 and 15 dm3min'1. The intensity of ArO+
showed little or no dependence on the purge gas flow rate. A flow rate of 10
dm3min'1 was chosen for all further experiments.
3.3.4 Comparison with Conventional Sample Introduction
A comparison of oxide ratios and intensities for conventional sample
introduction with those obtained when the dryer device was used is shown in
Table 3.4 for the optimized conditions in Table 3.1. The background intensity
obtained when the sample uptake to the nebulizer was closed was found to be
around 10,000 counts s'1 and is thought to be due to the presence of oxide
forming gases in the plasma from the surrounding atmosphere or in the argon
gas supply. It is evident from these results that the dryer system is removing a
large proportion of the water from the sample aerosol. The signal intensities of
the test elements for the conventional sample introduction system and the
dryer system were found to be similar which suggests that little or no sample
loss is occurring due to the addition of the dryer device.
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.3.5 Detection Limits for Iron
Flow injection peaks for Fe-56 in the absence and presence of the dryer
are shown in Figure 3.5. The reduction in water loading with the dryer present
leads to a reduction in ArO+ and ArOH+ intensities at m/z ratios of 56 and 57
respectively such that the background intensities are significantly reduced.
This reduction in background intensity leads to a reduction in the standard
deviation of the background and, therefore, a reduction in the detection limits
as shown in Table 3.5. It can also be seen from the relative widths of the flow
injection peaks that the addition of the dryer system has little effect on the
dispersion of the sample.
3.3.6 Addition of Nitrogen to the Plasma
The addition of low flow rates of nitrogen gas into a T-piece in the
nebulizer gas line led to a decrease in intensity in the signal for iron compared
to that for an experiment without the nitrogen gas added. This might suggest
that some sample is being lost at the T-piece. As the nitrogen gas flow was
increased the intensity for iron was also increased as shown in Figure 3.6. It
can also be seen that the background intensity rose as the nitrogen flow rate
was increased. At nitrogen flow rates of greater than 100 ml min'1, the plasma
became unstable and more likely to be extinguished.
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The addition of acetonitrile to the sample solutions was found to give a
significant improvement in detection limits for iron-56 as seen in Table 3.6.
This increase in detection limit may be due to the addition of carbon and
nitrogen into the plasma from the acetonitrile or may be caused by an
improvement in nebulization efficiency due to the more volatile solvent.
3.3.7 Heated Spray Chamber
Heating the spray chamber has the effect of increasing the evaporation
of droplets in the aerosol allowing more sample to pass to the plasma. This
increase in evaporation would also be expected to increase the amount of
water vapor which passes to the dryer. A typical flow injection peak for a 10 jig
dm'3 iron solution is shown in Figure 3.7. The intensity of the peak was slightly
lower than the case for the cool spray chamber, however, the width of the flow
injection peak was seen to increase indicating the greater amount of iron
entering the plasma and an increase in dispersion in the system compared to
a unheated spray chamber. The noise on the signal was also seen to
increase with the background intensity becoming less stable however the
average intensity of the background remained low suggesting that the dryer is
capable of removing most of the extra water vapor that is transferred with a
heated spray chamber.
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.4 Conclusions
A multi-tube Nafion dryer system reduced the water vapor introduced
into a plasma source spectrometer from a conventional sample introduction
system. The optimization of the system has indicated that water can be
removed with little adjustment to the ICP-MS operating conditions and the
system provides a simple, inexpensive and rugged method for the reduction of
water-loading effects in ICP-MS determinations. This reduction in waterloading in the plasma reduced formation of interfering oxide species. In
particular, the decrease in intensity of the ArO+ ion species reduced the
interference in iron 56 and significantly improved the detection limit for this
isotope. The addition of nitrogen gas to the nebulizer gas stream did not give
much improvement in the detection limit for iron, due to the instability of the
plasma at greater nitrogen gas flows. The addition of acetonitrile to iron
solutions, however, did give a significant improvement in detection limit
although the process responsible for this improvement is unclear. Heating the
spray chamber increased the amount of sample entering the plasma with little
increase in background intensity although the instability of the signal and
increased noise made this an ineffective method for increasing the detection
limits of the system.
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.5 References
1.
Hutton, R. C., and Eaton, A. N., J. Anal. A t Spectrom., 1987, 2, 595.
2.
Tsukahara, R., and Kubota, M., Spectrochim. Acta, 1990,
3.
Alves, L. C., Minnich, M. G., Wiederin, D. R., and Houk, R. S., J. Anal. At.
Spectrom., 1994, 9, 399.
4.
Minnich, M. G., Houk, R. S., Woodin, M. A., and Christiani, D. C., J. Anal.
At. Spectrom., 1997,12, 1345.
5.
Gustavsson, A., Spectrochim. Acta, 1988,
6.
Backstrom, K., Gustavsson, A., and Hietala, P., Spectrochim. Acta, 1989,
44B, 1041.
7.
Botto, R. I., and Zhu, J. J., J. Anal. At. Spectrom., 1994, 9, 777.
8.
Brenner, I. B., Zhu, J. and Zander, A., Fresenius J. Anal. Chem., 1996,
355, 774.
9.
Tao, H., and Miyazaki, A., J. Anal. At. Spectrom., 1995,10, 1.
10.
Corns, W. T., Ebdon, L., Hill, S. J., and Stockwell, P. B., Analyst, 1992,
117, 717.
11.
Corns, W. T., Stockwell, P. B., Ebdon, L., and Hill, S. J., J. Anal. At.
Spectrom., 1993, 8, 71.
12.
Sundin, N. G., Tyson, J. F., Hanna, C. P., and McIntosh, S. A.,
Spectrochim. Acta, 1995, 50B, 369.
13.
Yang, J., Conver, T. S., Koropchak, J. A., Leighty, D. A., Spectrochim.
Acta, 1996, 51B, 1491.
14.
Conver, T. S., Yang, J., Koropchak, J. A., Shkolnik, G., and FlajnikRivera, C., Appl. Spectrosc., 1997, 51, 68.
15.
McLaren, J. W., Lam, J. W., and Gustavsson, A., Spectrochim. Acta.,
1990, 45B, 1091.
16.
Evans, H. E., and Giglio, J. J., J. Anal. At. Spectrom., 1993, 8,1.
43B,
45B, 581.
917.
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17.
Montaser, A.t and Golightly, D. W., “Inductively Coupled Plasmas in
Analytical Atomic Spectrometry”, 1987, VCH Publishers (New York).
18.
Akino, T. A., Carnahan, J. W., 1995, FACSS Abstracts, paper 856.
19.
Lam, J. W., and McLaren, J. W., J. Anal. At. Spectrom., 1990, 5, 419.
20.
Leighty, D. A., personal communication.
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3.1. Operating conditions for ICP-MS and dryer system.
ICP-MS
Power/W
1200
Plasma gas flow/dm3min'1
15
Auxiliary gas flow/dm3min'1
0.8
Nebulizer gas flow/dm3min'1
0.9
Sample flow rate/cm3min'1
1.2
Dryer
Purge gas temperature/°C
60
Purge gas flow rate/dm3min*1
10
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3.2. FIAS program for iron determination.
time/s
pump 1
pump 2
speed/rpm
speed/rpm
valve position
Pre-run
20
120
0
load
1
15
50
80
load
2
30
0
80
inject (read)
-
40
40
load
post run
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3.3. Efficiency of the dryer system for water removal.
Temperature
/°C
Mass of water
Mass of water
collected from collected from
sample gas/g
purge gas/g
Total mass of
Percentage
water
water
collected/g
removed by
dryer
25
0.026
0.652
0.678
96.2
45
0.021
0.514
0.535
96.1
58
0.020
0.560
0.580
96.6
80
0.033
0.541
0.574
94.3
87
0.026
0.515
0.541
95.2
98
0.040
0.582.
0.622
93.6
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3.4. Comparison of oxide ratios and ArO+ intensities for conventional
sample introduction and the dryer system.
Conventional
Dryer system
sample introduction
CeO+:Ce+
1.58 x 10*
2.08 x 10'3
BaCT.Ba*
1.59 x 10*
5.78 x 10-4
ArO+ /counts s'1
90 000
10 000
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3.5. Limits of detection for iron-56 and iron-57
Detection limit without
Detection limit with dryer
dryer (jig dm'3)
(jig dm'3)
Fe-56
1.8
0.37
Fe-57
1.4
0.93
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3.6. Limits of detection (fig dm'3) for iron-56 with the addition of
acetonitrile.
Replicate number
0% acetonitrile
5% acetonitrile
10% acetonitrile
1
2.4
2.0
1.0
2
1.8
1.9
0.8
3
1.9
1.4
1.0
4
6.5'
1.7
1.0
Mean (r.s.d)
2.0(14% )
1.8(15% )
1.0(12% )
’ value rejected by Q-test at 95% confidence,
(limits of detection calculated at 3s).
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
[(CF,— CF2)
2 'm
-C F -
•CF,]
2-In
- I
O
I
c f2
I
2
CF— -CF,
-
I
o
I
c f9
I
2
c f9
I
2
s o 3h . [ h 2o i
Figure 3.1. Chemical structure of Nation material (m= 5 to 13.5, n= ca. 1000,
z= 1,2,3..., x= 1 to 13)
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
heated compartment
control compartment
Nation membrane
dryer
os
OS
dryer heater
sample
gas
inlet
sample
gas
outlet
0
temperature
controller
flow
meter
purge gas exhaust
Figure 3.2. Schematic of dryer system.
purge
needle
valve
inlet
pressure
regulator
Temperature/ deg. C
I
40
70000
60
80
1< 0
■
«
♦ ■ 40000
(0
0c
« Ba2+/Ba+
■ CeO+/Ce+
0.00
30000
4 BaO+/Ba+
x ArO+
20000
1 0000
0
0.00
Figure 3.3. Effect of dryer temperature on oxide ratios and ArO+
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ArO+
50000
0.01
Intensity/ counts
s-1
60000
CeO+/Ce+
BaO+/Ba+
Ba++/Ba+
Ratio
.Oil
.001
.0001
20
flow rate/dm3 min-1
Figure 3.4. Effect of purge gas flow rate on oxide ratios and ArO+ intensity.
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
300000
250000
!%■■
■ ■
■■■"‘
’ v W ■ ■p ■■* '> w■ ,■ ■ ■■
Without Dryer
m 200000
5 150000
~ 100000
50000
With Dryer
0
2
4
6
8
10
12
14
16
18
20
Time/s
Figure 3.5. Flow injection peak for a 10 |ig dm'3 iron solution with and without
the dryer system.
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40000
■ ■
■ ■
„
30000
!
R?-5A
Blank 56
c
5
20000
•
•
•
10000
•
100
Nitrogen
Flow (ml/min)
Figure 3.6. Effect of flow rate on intensity for iron-56 for the addition of
nitrogen gas to the plasma.
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
200
80000
s-1
50000
Intensity/
60000
counts
70000
40000
30000
20000
*
0
2
4
6
....................................
8
10
12
14
16
18
Tim e/ s
Figure 3.7. Flow injection peak for 10 |ig dm'3 iron-56 using a heated spray
chamber.
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
CHAPTER 4
INVESTIGATION OF MICROWAVE DESOLVATION OF WATER AEROSOLS
4.1 Introduction
A major limitation of plasma spectroscopy for aqueous samples is the
inability of the plasma source to deal with significant amounts of water vapor.
The introduction of water into the plasma can cause cooling and ultimately
lead to extinction of the plasma source. This limitation has led to sample
introduction systems which typically introduce only 1 or 2% of the sample in
the form of small droplets in order to limit the amount of water vapor that is
transported to the plasma. These sample introduction system also have the
effect of reducing the rate of introduction of sample to the plasma and therefore
reducing the analytical signal. Desolvation is a method for increasing the
analytical signal. This is achieved by evaporating solvent from the droplets
followed by subsequent removal of the solvent vapor in order to allow a
greater rate of sample introduction to the plasma. The evaporation of water
from the droplets is normally achieved by thermally heating the walls of a
spray chamber to increase the temperature of the droplets and increase
evaporation. The process of heat transfer from the chamber walls to the
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
droplets in these systems is inefficient and necessitates a temperature of well
in excess of the solvent boiling point1 which causes aerosol droplets to deposit
on the spray chamber walls leading to gradual build-up of salt and subsequent
memory effects.
Microwave energy is known to be an efficient method for heating some
materials. Water can absorb microwave energy very efficiently due to the
dipole of the water molecule which can undergo dipole rotation in a
microwave field as discussed in chapter 5. The rapid movement of the water
molecules due to the application of a microwave field leads to rapid heating
effects.2 It is also known that some materials are transparent to a microwave
field, allowing heating to be achieved inside the sample without the need to
heat the containing vessel. This concept has successfully been applied to the
digestion of aqueous based sample solutions contained within microwave
transparent materials such as PTFE.2
The application of microwave energy for the heating of an aqueous
aerosol contained inside a microwave transparent chamber would have
advantages over conventional desolvation systems. The direct heating of
aerosol droplets contained inside a microwave transparent chamber would
remove memory effects which occur in thermally desolvation systems as
droplets come into contact with hot spray chamber walls and evaporate on the
surface allowing salt particles to be deposited. A microwave desolvation
system might also be expected to produce a more even heating effect and
allow greater control of the input energy.
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Christiansen and Unruh3 investigated the possibility of heating an
aqueous aerosol in order to develop a spray drying process for the
preparation of homogeneous powders of complex metal oxide systems. They
used a signal mode cavity of 2.45 GHz and up to 2.2 kW to attempt to heat a
stream of droplets of approximately 30 |im in diameter. The authors, however,
found little evidence for coupling of the microwaves with the droplets and
suggested that the inefficient coupling was due to unfavorable dielectric and
geometric effects. It was not possible to raise the interior temperature of the
individual droplets significantly and therefore thermal conduction through the
droplet surface to the surrounding gas could remove the energy from
microwave heating nearly as rapidly as it is absorbed. It was concluded that a
dramatic improvement in the calculated evaporation rates of small droplets
could be achieved by increasing the microwave frequency.
Radiation in the microwave region has been investigated for microwave
remote sensing and is known to be absorbed by clouds4 and fog5 containing
water droplets of similar dimensions to aerosols produced by a pneumatic
nebulizer. Tsang et al.4 calculated the absorption for a cloud of 0.8 g/m3 water
density and a mode droplet diameter of 20 pm at varying frequencies and
showed that the calculated absorption is very small at lower frequencies such
as the frequency used for most microwave ovens (2.45 GHz). Gras et al.,6
however, claimed that microwave radiation is absorbed at a frequency of 2.45
GHz and 890 W by water droplets inside a multi-mode microwave cavity. They
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
positioned a concentric pneumatic nebulizer and Pyrex glass spray chamber
inside a domestic microwave oven and passed the exiting material through
two condensers via a length of silicone rubber tubing before introducing it into
the plasma source of an atomic emission spectrometer. This system was
compared with a similar system heated to 130°C by wrapping the spray
chamber with heating tape. Although some differences were seen between
the two systems, the authors failed to prove that this was due to direct
absorption of the microwave energy by the droplets. A possible explanation
for the effects observed in this system is that thermal desolvation occurred due
to absorption, and subsequent heating, of various system components in the
microwave field. These components may include the Pyrex glass spray
chamber (which is not completely microwave transparent), silicone tubing and
bulk liquid produced by aerosol striking the walls of the chamber and transfer
tubing. It may also be possible that sample was heated as a liquid stream in
the nebulizer and connecting tube to the nebulizer.
In this work the ability of a microwave field at 2.45 GHz to heat an
aqueous aerosol by direct absorption of the microwave energy by the droplets
has been investigated.
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2 Experiments with a Multi-mode Cavity
4.2.1 Experimental
4.2.1.1 Apparatus Development
A CEM MDS-81 microwave oven containing a 1 liter glass vessel
(approx. 15 cm long x 11 cm diameter) that fits tightly to the top of the oven with
3 holes of 3.5 cm diameter in the top casing of the oven was used for all of the
experiments. The glass container was used as a spray chamber inside the
oven in order to investigate the effect of microwave energy on an aerosol.
Initial experiments in which a cross flow nebulizer was used as shown in
Figure 4.1 were concerned with investigating the destination of sample
material and trapping of exiting material in water. The cross flow nebulizer
was housed on the outside of the oven as it contained metal parts which may
cause sparking in the microwave field. The nebulizer was tightly connected to
a glass tube (1.6 cm i.d., 20 cm long) in order to increase the residence time of
the aerosol in the microwave field. Liquid was transferred to the nebulizer
using a peristaltic pump.
Subsequent experiments were conducted using a quartz concentric
nebulizer (Meinhard) which could be placed inside the glass container (Figure
4.2). The liquid stream to the nebulizer was inserted through the first of the 3
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
holes in the oven through a length of 0.8 mm i.d. PTFE tubing. The gas stream
was provided through the central hole from a compressed air cylinder at
approximately 1 L/min via a length of Tygon tubing. These 2 holes were
sealed by rubber bungs to prevent loss of the aerosol. The liquid stream was
pumped by a peristaltic pump (Ismatec) at 1.0 ml/min.
In order to measure the exiting material in both systems a glass to metal
connector constructed in house was inserted into the third hole allowing a
heated transfer line to be firmly attached at the other end via a length of metal
tubing. In this way the aerosol could be created inside the glass vessel and
the resulting material passed through a heated transfer line to the
spectrometer. The heated transfer line was modified from a transfer line used
in a gas chromatography system. The transfer line consisted of a length of
copper tubing (120 cm long by 1.6 mm id) wrapped with wire which provided
resistive heating contained inside an insulating sleeve. The heat was
obtained by passing a current through the wire which could be controlled with
a potentiostat. The other end of the transfer line was modified so that the
original fitting was removed and replaced by a length of metal tubing which
was wide enough (3.2 mm o.d.) to fit tightly into the spray chamber of a flame
atomic absorption spectrometer (IL Video 22) via the hole which originally
housed the impact bead, Figure 4.3. In this way the exiting material from the
transfer line was directly introduced into the instrument spray chamber. This
experimental setup allowed for the measurement of the efficiency of the spray
chamber system under various conditions for analyte transport by the
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
introduction of solutions. Sodium was chosen as a test element due to its
strong emission intensity in an air/acetylene flame and was introduced to the
sample stream by injection of 0.5 ml of sodium chlooride solution into a water
earner stream via a 6 port rotary valve.
4.2.1.2 Determination of Maximum Available Microwave Power
To determine the maximum power available for heating a sample in the
OEM oven, 1 L of distilled water was placed in a glass beaker in the center of
the microwave oven and heated for 5 minutes at 100% power. The initial and
final temperatures of the water were measured with a mercury in glass
thermometer and the power calculated as 35 times the temperature change.7
4.2.1.3 Spectrometer and Reagents
The spectrometer used was a IL video 22 flame spectrometer
connected to an integrator, HP 3394A, in order to obtain a print out of signal
produced. When signals were being collected the spectrometer and integrator
parameters were set as shown in Table 4.1 using sodium as a test element.
Deionised water was used as the carrier stream. Sodium solutions were
prepared by dissolving the appropriate mass of sodium chloride (Fisher) in
deionised water.
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.1.4 Trapping and Analysis of Exiting Material
In order to measure the amount of sample exiting the oven during a
typical run, the outlet from the oven was bubbled through water and analyzed
by flame spectrometry. The cross flow nebulizer system was used with the end
of the transfer line disconnected from the spectrometer and placed into a
beaker containing 20 ml of deionised water. A sample of sodium solution (0.5
ml of 1000 mg/L) was injected into a earner stream of deionised water and
aspirated at approx. 0.5 ml/min with compressed air approx. 1L/min) into a
preheated glass vessel (5 minutes at 50% power). The power was continued
at 70% for a further 5 minutes. After each run the beaker of water was made
up to 100 ml with deionised water and analyzed by flow injection-flame
spectrometry for the determination of sodium. The vessel was washed out with
deionised water which was collected, made up to 100 ml and analyzed in the
same manner.
4.2.1.5 Destination of Sample Material
In order to determine the destination of sample material in the system,
the cross flow nebulizer system (Figure 4.1) was used with the outlet
connected to the flame atomic absorption spectrometer via a heated transfer
line. Sodium solution, 0.5 ml of 1000 mg L '\ was injected into a carrier stream
of deionised water at 0.5 ml/min and aspirated into a preheated spray
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
chamber (5 minutes at 50% power) with compressed air. The system was
heated for a further 5 minutes at 50% power. After each run the glass vessel in
the oven, instrument spray chamber, instrument waste line and transfer line
were washed out with deionised water, made up to 100 ml and analyzed by
flame spectrometry using flow injection sample introduction.
4.2.1.6 Calculation of System Efficiency
The efficiency of the system was determined by injecting 0.5 ml of a 10
mg L*1 sodium solution into the spray chamber with the microwave power set at
50% after having heated the system for 5 min at 70% power. The peak area
was compared to a calibration curve for sodium constructed by injecting 0.5 ml
of 0.5,1.0, 5.0 and 10.0 mg L'1 sodium into a water earner stream connected to
the instrument nebulizer which was earned out immediately before running the
microwave system. It should be noted that this calibration could be performed
directly before the desolvation system was run without extinguishing the flame
and that the impact bead was removed in order to accommodate the end of the
transfer line. The efficiency was then calculated by determining the mass of
sodium chloride entering the flame and dividing by the total mass of sodium
chloride introduced into the system. In order to determine the mass of sodium
entering the flame the efficiency of the instrument nebulizer and spray
chamber system must be known. This was done by introducing a known mass
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of sodium chloride, 0.005 mg, into the spectrometer via a flow injection
manifold. The liquid in the waste line, which had previously been cleaned and
filled with deionised water, was then analyzed after dilution to 50 ml with
deionised water. The mean value of 4 replicate runs was determined and the
efficiency of the sample introduction system calculated for these experimental
conditions.
4.2.1.7 Temperature Measurements
The temperature inside the glass vessel was measured using a fiber
optic probe which had been previously calibrated in a water bath against a
mercury in glass thermometer. The temperature was measured every 30
seconds during a run in which a sample of 0.5 ml of 100 mg L‘1 sodium was
added after 2 minutes at 50 % microwave power and heating continued at
70% power for another 4 minutes. Another set of temperature measurements
were taken after heating for 2 minutes at 50% power, injection of 0.5 ml of 100
mg L‘1 sodium into the earner stream and then switching off the microwave
power.
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.1.8 Effect of Microwave Energy on the System Efficiency
In order to identify the role of microwave energy in the desolvation
system, a series of experiments were conducted with a sample injected into a
hot spray chamber with the microwave energy on and off alternately, in order
to discover if the desolvation process is due to the thermal effect of heating the
system by the microwaves or by direct interaction of the microwaves with the
aerosol droplets. The experiments were conducted by heating the system for
2 minutes at 70% power with a water aerosol and injecting 0.5 ml of 100 mg L'1
sodium into the water stream with the power at 70% for a further 5 minutes.
This experiment was then repeated with the power switched off after the
sample was injected. In this way the temperature of the system was similar for
both experiments at the time of injection so that the degree of desolvation due
to the temperature of the spray chamber should be similar. Any increase in
efficiency for the system when the microwave power was continued after the
sample was injected was assumed to be due to an interaction between the
microwave energy and the aerosol droplets. A statistical analysis, to
determine any significant difference between the two sets of experiments was
made.
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.2 Results and Discussion
4.2.2.1 Determination of Maximum Available Microwave Power
The temperature for 1 liter of water was found to rise from 19.6°C to
39.7°C for 5 minutes at 100% power corresponding to a maximum absorbed
power of 703.5 W. A simple calculation indicates that aproximately 2.6 kJ is
necessary to heat 1 ml of water to the boiling point and convert it to steam
which corresponds to a power requirement of 43 W for water flowing at 1
ml/min. The oven would, therefore, appear to have a more than adequate
power output to desolvate an aerosol which is produced at 1 ml/min into a 1
liter vessel in a gas stream flowing at 1 l/min.
4.2.2.2 Trapping and Analysis of Exiting Material
The results for 6 runs are shown in Table 4.2 and show a good total
recovery of sodium injected into the system. The amount of sample exiting the
system and being collected was low and suggests a low mass transport
efficiency for this experiment. In this case the low efficiency maybe due to the
increased back pressure which is produced when the exiting material is
bubbled through a beaker of water but also indicates an intrinsically inefficient
design for the spray chamber.
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.2.3 Destination of Sample Material
The amounts of sodium found in the glass vessel, instrument spray
chamber, instrument waste line and transfer line for 4 replicate experiments
expressed as a percentage of the total amount of sodium introduced are
shown in Table 4.3. These results suggest that most of the sample is retained
in the glass vessel with small amounts collecting in the instrument spray
chamber, waste line and the transfer line. Approximately 5% of the sample
that was introduced was not accounted for and was assumed to have exited
the spray chamber through the burner. Sample found in the transfer line was
probably caused by to the deposition of sample droplets and subsequent
drying on the heated walls of the tube. The sample found in the instrument
spray chamber and waste line was thought to be deposited by condensation of
water vapor onto the sample particles or droplets, creating larger droplets
which could not pass through to the burner.
4.2.2.4 Calculation of System Efficiency
In order to calculate the efficiency of the desolvation system, the
efficiency of the spectrometer sample introduction system needed to be
determined. This was estimated by determination of the concentration of
sodium in the waste line after addition of a known amount of sodium by a flow
injection method. 83% of the sample was found in the waste line which
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
indicated that about 17% of the sample was transported to the flame. The
efficiency of the desolvation system could then be calculated by comparing the
signal from the system with a calibration curve obtained using the instrument
nebulizer and spray chamber. The mean of 5 replicate measurements was
used to calculate the efficiency for this system which was calculated to be 2%
for sample transport to the flame. This is lower than the efficiency of the cross
flow system which was determined by washing out the various components of
the system, analyzing the resulting solutions and subtracting from the total
mass of sample introduced. This discrepancy could be due to the presence of
a greater amount of water vapor transferred to the flame compared to the
standard solutions which may lower the flame temperature, suppressing the
emission signal. The low efficiency of the system is probably due to the size
and shape of the spray chamber and transfer line which were not the optimum
designs for mass transport to the spectrometer. When the system was run
without microwave heating a signal could not be distinguished from the
background noise which suggests that desolvation of the sample is occurring
in the presence of the microwave field.
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.2.5 Temperature Measurement
The temperature of the system with microwave heating is shown in
Figure 4.4. It can be seen from this graph that the temperature of the system
reaches about 55°C in the spray chamber at the time of sample injection. It is
worth noting that this is the temperature at the center of the spray chamber
which is cooled by the nebulizer gas and water droplets and not at the walls of
the spray chamber which may be at a significantly higher temperature. The
temperature of the system then continues to heat to nearly 95°C after 6
minutes of heating. When the microwave power is switched off after injection
of the sample, the temperature decreases slowly to about 50°C after 6
minutes. At the time of sample injection the temperature for the 2 experiments
was almost identical.
4.2.2.6 Effect of Microwave Energy on the System Efficiency
Statistical treatment of the results from a series of experiments
conducted by alternating between the experiment with the microwave field
switched off after injection of the sample and the microwave field left on after
injection using a paired t-test at 95% confidence (n=11) did not show a
significant difference between the experiments. This results suggests that any
desolvation of the aerosol in the presence of the microwave field is
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
predominantly due to heating of the system and not due to direct interaction
between the field and the droplets.
4.3 Experiments with a Sinale-Mode Cavity
4.3.1 Experimental
4.3.1.1 Cavity Design
A single mode microwave cavity system was designed and built by
CEM and is shown schematically in Figure 4.5. The system consists of an
aluminum cavity (46.5 cm long, 4.5 cm wide, 9.0 cm high) with a magnetron
(Hitachi 2M214, 300W output power, 2.45 GHz) attached to the side at one
end such that the antenna entered the box approx. 5 cm from one end. A
small fan (ETRI141LS) was mounted on top of the magnetron in order to
provide cooling. The cavity was mounted in a metal frame to keep it in
position. In order to conduct experiments with this apparatus, holes were
drilled in the ends of the cavity to allow a quartz tube (2 cm o.d. approx. 60 cm
long) to pass through. Two lengths of steel tube (2.1 cm i.d., 6.5 cm long) were
pushed tightly into the holes in the end pieces to act as chokes and prevent
microwaves from leaking out the cavity. The cavity was monitored for
microwave leaks using a microwave survey meter (Holaday model 1501).
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
One end of the quartz tube was narrowed so that a cross flow nebulizer and
double pass spray chamber could be fitted to It to allow aerosol droplets to
pass through the cavity.
4.3.1.2 Calibration for Microwave Power
In order to determine the microwave power available for heating in the
cavity, 50 ml of deionised water was placed into the quartz tube and was
Pwjition to be totally inside the cavity. The temperature of the water was
monitored using a fiber optic temperature probe inside the cavity. The
temperature rise of the water after a known period of time (30 s) was measured
and absorbed power calculated from the mean of 4 runs.
4.3.1.3 Experiments of Aerosol Heating
In order to determine the degree of heating inside the single-mode
cavity due to direct absorption of the microwave energy by an aerosol an
experiment was performed with an aerosol introduced into a quartz tube inside
the cavity. The aerosol was produced by a cross flow nebulizer and double
pass spray chamber placed at one end of the tube. Deionised water was
introduced to the nebulizer by free uptake (2 to 3 ml/min) using a flow of air at
16 psi (approx. 1 L/min) from a pressurized cylinder. The aerosol could then
88
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pass through the cavity and the temperature measured using a fiber optic
probe placed inside the cavity. As quartz is transparent to microwave energy
any temperature rise in the system should be due to heating of the nebulizer
gas or the aerosol droplets. A series of experiments were run with the
nebulizer gas only and with the aerosol present to determine any absorbance
of microwave energy by the water droplets.
4.3.2 Results and Discussion
Calibration of the single mode cavity gave a mean value of 245 W of
absorbed power. Heating of the system with nebulizer gas only without an
aerosol present gave a small temperature rise of 2 to 10°C after 5 minutes of
microwave heating. Similar temperature rises were seen with the aerosol
present. For a series of experiments alternating between an experiment with
the aerosol present and without an aerosol present, no significant difference
was observed between the two sets of results (95% confidence). If the aerosol
was being directly heated by the microwave energy it might be expected that a
greater temperature rise would be observed by the presence of water vapor in
the tube or by deposition of heated water droplets onto the fiber optic
temperature probe. As this was not seen it can be assumed that direct
absorption of microwave energy is minimal and therefore any temperature
increase is due to absorption of microwave energy by the nebulizer gas or
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
heating of the system due to conduction to the cavity walls from the magnetron
which became warm during operation.
4.4 Conclusions
A test system has been constructed incorporating a multi-mode
microwave cavity which shows an increase in sample transport efficiency
when it is heated in a microwave field. The spray chamber system heats
significantly in the presence of a microwave energy. The efficiency for the
heated system has been estimated as approximately 5% for the cross flow
nebulizer system using a mass balance experiment and approximately 2% for
the Meinhard nebulizer system determined by comparison of the emission
peak compared to stardard solutions. A sample emission signal could not be
detected for an unheated system. The increase in sample transport appears to
be primarily due to heating of the apparatus or bulk water inside the
microwave field. A single-mode cavity system has also been used to test the
possibility of microwave absorption by water droplets by ensuring that the
vessel walls remain unheated and determining the temperature rise with and
without an aerosol present. There is no evidence for absorption of microwave
energy by water droplets in these experiments.
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.5 References
1.
Sharp, B. L., J. Anal. A t Spectrom., 1988, 3, 939.
2.
Kingston, H. M., and Jassie, L. B., “Introduction to Microwave Sample
Preparation”, 1988, American Chemical Society (Washington DC).
3.
Christiansen, D. E., and Unruh, W. P., Ceram. Trans., 1991, 21, 597.
4.
Tsang, L., Kong, J., and Shin, R., “Theory of Microwave Remote
Sensing”, 1985, J. Wiley and Sons (New York).
5.
Collin, R. E., “Antennas and Radiowave Propagation”, 1985, McGrawHill (New York).
6.
Gras, L., Mora, J., Todoli, J. L., Hemandis, V., and Canals, A.,
Spectrochim. Acta, 1997, 52B, 1200.
7.
CEM MDS-81D Operation Manual, 1989, CEM Corporation (Matthews,
NC).
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4.1. Operating parameters of integrator and spectrometer for multimode experiments.
Integrator Parameters
Attenuation
2
Chart Speed
0.1 cm s'1
Peak Width
4 min
Threshold
2
Area Rejected
0
Spectrometer Parameters
Number of Runs
99
Lamp Current
8 mA
Gain
0.1
Wavelength (emission)
589 nm
Slit Width
0.3 nm
Acetylene Flow
2.0 L min'1
Air Flow
8.0 L min'1
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4.2. Destination of sodium after collection of exiting material by bubbling
through water.
Experiment
mass of
mass of
amount of
Total sodium
number
sodium in
sodium
sodium
recovery/ %
vessel/ mg
collected/ mg
collected as
percentage
1
0.4862
0.0035
0.71
98
2
0.5085
0.0054
1.1
103
3
0.4830
0.0240
4.7
101
4
0.5370
nd
0.0
107
5
0.5010
nd
0.0
100
6
0.4850
nd
0.0
97
nd= not determined (below detection limit).
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4.3. Destination of sample material in the system.
Percentage of total sample found
1
2
3
4
Mean
Glass vessel
107.4
88.2
88.4
88.0
88.3'
Spray chamber
4.0
3.2
1.6
2.4
2.8
Waste line
1.0
3.6
2.0
2.2
2.2
Transfer line
2.4
2.5
1.2
2.4
2.1
Sample not found
-14.8
2.5
6.8
5.0
4.8'
' Trial 1 reject by Q-test (95% confidence)
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
glass to metal connector
gas inlet
to transfer
line
liquid inlet
nebulizer
glass
tube
Glass
Vessel
Figure 4.1. Schematic diagram of the multi-mode cavity cross flow nebulizer
test system.
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
glass to metal connector
liquid inlet
Jto transfer
line
gas inlet
-CD
OD-
nebulizer
Glass
Vessel
Figure 4.2. Schematic diagram of the multi-mode cavity Meinhard nebulizer
test system.
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
burner head
instrument
nebulizer
I rupture
I disk
transfer line
drain
Figure 4.3. Schematic diagram of the spectrometer sample introduction system
modified to allow measurement of material from the test system.
100
90
O 80
O)
V
■o
70
Microwaves on
microwaves offj
®
W
3
«
w
60
©
O.
E
I-
50
40
30
0
60
120
180
240
300
360
Time/ s
Figure 4.4. Temperature measurements in the test system with microwaves
on and off after sample injection.
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
>.
>
CO
o
0)
Figure 4.5. Schematic of the single mode microwave system.
>
CO
5
o
o
E
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5
THEORY OF MICROWAVE HEATING OF WATER DROPLETS
5.1 Introduction
Microwave energy has become a common method for heating materials
in analytical laboratories particularly for the digestion of samples prior to
analysis by atomic spectrometric techniques. The first reported use of
microwave energy for this purpose was in 1975,1 subsequently the
acceptance of microwave energy as a heating method for sample dissolution
has become widespread. The advantages of microwaves over conventional
methods include their ability to heat aqueous solutions rapidly, the greater
control of input power and its even heating due to the ability of microwaves to
pass through some container walls and heat the materials to be digested
directly. Despite the common use of microwave energy in analytical chemistry,
the theory of microwave heating is often not well understood by many
analytical chemists
In this chapter the theory of microwave interaction with water and the
effects of the applied frequency on this interaction will be discussed followed
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
by the description of three mathematical models that have been developed in
order to gain an understanding of the effect of varying the applied frequency
on the evaporation of a droplet of water. The mathematical symbols used in
these models are summarized in Table 5.1 or are defined within the text. The
results of these models will be discussed along with the apparent strength and
weakness of the different approaches.
5.1.1 Microwave Heating of Water
Much of the success of microwave heating for digestion techniques is
due to the ability of microwaves to heat aqueous samples. Microwave energy
has a frequency range of 0.3 to 300 GHz and is non-ionizing radiation. The
radiation does not directly cause any changes in molecular structure but can
produce heating by ionic migration and dielectric polarization.2 In microwave
spectroscopy samples are generally in the gas phase where the molecules
can rotate freely giving rise to sharp absorption bands. In the case of
microwave heating for sample preparation, the samples tend to be in the solid
or liquid phase where the molecules are restricted in movement due to the
presence of neighboring molecules. These materials, therefore, absorb
microwave radiation over a range of frequencies and do not produce sharp
absorption bands in the microwave region.
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Conduction heating occurs in a sample which contains dissolved ions
that have the ability to move in an applied field. The movement of ions
occurring in a sample gives rise to heat produced by the resistance to the flow
causing a temperature rise in the sample. In the case of pure water this
heating effect is minimal and heating is predominantly caused by polarization.
The total polarization (separation of charge),p, of a material is the sum of a
number of individual components:3
P ~ P'+Pa+Pd+Pi
1
where, pe, is the electronic polarization which arises from the
realignment of electrons around a specific nuclei: pa> is the atomic polarization
caused by the relative displacement of the nuclei due to unequal distribution of
charge within the molecule: pd, is the dipolar polarization resulting from the
orientation of permanent dipoles by the electric field: and p„ is the interfacial
polarization which occurs when there is a build up of charge at an interface
(e.g. between a solid and a liquid). The effect of each term in a material that is
subject to an alternating field is dependent upon the time scale for the
polarization and depolarization. The rate of growth and decay of polarization
is characterized by the relaxation time which can be described in terms of the
frictional forces between molecules in the medium. For pe and pa the
polarization and depolarization are much faster than the time scale of
microwave frequencies and therefore these parameters do not significantly
influence the heating effect. In the absence of interfacial polarization of
comparable time scale, the heating effect in water is predominately caused by
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
polarization associated with the orientation of permanent dipole moments in
the water molecules which have a similar time scale to microwave
frequencies. Dipole polarization in water occurs due to the permanent dipole
of the water molecule which is formed due to the differing electronegativities of
the oxygen and hydrogen atoms. In the presence of an oscillating field, the
water molecule will move in order to align itself with the field. At low oscillation
frequencies, the molecules move in phase with the applied field and very little
heating occurs due to the realignment. If the frequency is high, the molecule
cannot move fast enough to follow the applied field so that rotation does not
occur and no heating is observed. At frequencies in the microwave region, the
response time of the dipoles is similar to the time that the field changes such
that the molecule will experience a force and move in order to try and follow
the field. As the molecule rotates in order to align with the field, however, the
field will start to weaken and switch direction allowing the molecules to relax
back to a lower energy position such that the molecule always lags behind the
applied field. The relaxation of the molecules gives rise to a release of energy
in the form of heat producing a temperature rise in the liquid. This is the
predominant mechanism for microwave heating in pure liquid water.
In order to describe the dielectric properties of a material, two
parameters are usually used, the dielectric constant, e', and the dielectric loss,
e". The dielectric constant describes the ability of a material to be polarized by
an electric field, while the dielectric loss indicates the ability of the material, to
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
convert this energy absorbed from the field to heat. At low frequencies, the
dielectric constant (s') will reach a maximum as the material stores the
maximum amount of energy that it can from the applied field. Using these two
parameters the dielectric loss tangent (or dissipation factor) can be defined as:
e"
tand = —
s'
This parameter indicates how well a material will heat in a microwave
field and is a function of the material’s ability to absorb microwave energy (s')
and its ability to dissipate that energy as heat (e").
5.1.2 Frequency Effects
Both the dielectric constant and dielectric loss are strongly frequency
dependent and therefore the efficiency of heating a material is also greatly
dependent on the frequency of the applied field. It is worth noting that in the
case of bulk liquid water, the greatest heating effect occurs around 20 GHz3
while domestic and laboratory microwaves ovens operate at 2.45 GHz. The
reason for the use of this less efficient frequency is due to the frequency
dependence of penetration depth. The penetration depth of microwave
radiation is defined as the depth at which the field strength has fallen to 1/e
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.
(0.368) of its original value at the surface of the material. Penetration depth
can be calculated using the equation:4
3.
where ?iis the wavelength of the microwave radiation and k' is the
relative dielectric constant, as defined in Table 5.1. In order to heat samples
efficiently in domestic or laboratory ovens, it is necessary for the microwaves to
penetrate far enough into the sample to ensure heating throughout the whole
sample. The variation of penetration depth with frequency for water at 25°C is
shown in Figure 5.1 and shows a penetration depth of about 2 cm at 2.45 GHz
while a frequency of 20 GHz would only heat the outer 2 mm of the sample.
For a small sample such as a water droplet produced by a typical pneumatic
nebulizer the diameter of the droplet is much less than the penetration depth of
the microwave power at 2.45 GHz such that a higher frequency becomes more
favorable in terms of efficient dielectric loss and penetration depth of the
microwave field. Indeed it could be hypothesized that a smaller penetration
depth would allow more of the applied energy to be absorbed by a small
droplet while microwaves with a larger penetration depth would tend to pass
through with little absorption of the microwave field by the droplet. The
absorbed power per unit volume, Pv is also frequency dependent and is given
by the equation:5
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Pv = jtfe"E2
4.
where f is the applied frequency and E is electric field strength inside
the sample.
From microwave remote sensing of clouds and fog which contain water
droplets of a similar size to those produced by a pneumatic nebulizer,
microwaves are seen to absorb with much greater efficiency at frequencies
above 2.45 GHz.6,7 This behavior is shown in Figure 5.2 for a cloud of 0.8 g/m3
water density and a mode droplet size of 20 pm diameter. The absorption
coefficient is given in terms of the absorbance per centimeter pathlength for an
assumed concentration resulting in units of inverse centimeters. Christiansen
and Unruh8 used a single mode microwave cavity at 2.45 GHz with powers of
up to 2.2 kW to attempt to dry a stream of 30 pm diameter droplets of an
aqueous solutions of metal oxide systems. Only partial drying of the droplets
was observed due to the “unfavorable dielectric properties of water in
conjunction with spherical geometry." The spherical shape of the droplets was
thought to reduce the internal electric field by a factor of 3/( k/+2) from a
calculation of the boundary conditions. It was concluded that for micrometer
sized aqueous droplets the rate of coupling of microwave energy at 2.45 GHz
becomes comparable to the rate at which thermal energy is conducted to the
surroundings and that “increasing the applied microwave frequency, to take
advantage of the more favorable dielectric parameters , makes a dramatic
improvement in the calculated evaporation rates of very small droplets.”
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2 Mathematical Models
5.2.1 Field Strength Model
In order to determine the effect of frequency on the evaporation time of a
small (10 pm diameter) water droplet, a model has been developed and is
shown schematically in Figure 5.3. In this simple case a droplet of water of
known diameter passes into a microwave cavity. In this radiation zone the
droplet absorbs microwave energy which is used to raise the temperature of
the drop. It is assumed that this is the only effect occurring during this stage
(i.e. there is assumed to be no mass or energy losses from any other
mechanism such as evaporation or conduction). After a given period of time
the droplet emerges from the radiation zone as a hot droplet of the same
diameter as the initial droplet. It then undergoes a period of evaporation, due
to the difference in temperature between the droplet and the surrounding gas,
during which its diameter is reduced to zero, again assuming no other mass or
energy losses occur.
By assuming a time for evaporation (10 ms) it is possible to calculate
the temperature difference between the droplet and surrounding gas that is
necessary for complete evaporation to occur. The temperature rise due to
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
microwave absorption can then be calculated (assuming that the droplet is
initially at the same temperature as the surrounding gas) from the equation:9
s
where T is the temperature difference between the droplet and the
surrounding gas (K), D0 is the initial droplet diameter (m), D is the final droplet
diameter (zero in this case), 1^ is the latent heat of vaporization at 298K (J/g), d
is the density of the droplet (g/L), K„ is the thermal conductivity of the
surrounding gas (W/Km) and t is the time for the evaporation (s). The energy
that would need to be absorbed in order to raise the temperature of the droplet
can then be calculated from the equation:
Energy = mCT
which gives energy (J) from the mass of the droplet, m (g), the
temperature rise, T (K), and the heat capacity of water at 298K, C (J/gK).
According to the model, this energy must come from the microwave field such
that the internal field strength, E^,, necessary to heat the droplet can be
calculated by the rearrangement of equation 4 substituting Pv=Energy/Vt to
give:5
f Energy
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
where tabs is the time that the droplet spends in the microwave field.
Using boundary conditions the external applied electric field strength can be
calculated by:
£ _ Eia(K' + 2)
j
3
In this way the electric field strength necessary to evaporate a droplet
can be calculated at different frequencies using the dielectric data for water
from conductivity measurements given by Von Hippel.10
5.2.2 Results and Discussion
A plot of the applied field strength that would be necessary to evaporate
a 10 pm diameter water droplet in an evaporation time of 10 ms against
applied frequency is shown in Figure 5.4. The values on this graph were
calculated using the computer program (Mathcad Plus, Mathsoft) shown in
appendix A. The appendix shows the values and equations used for the
calculation of applied field strength necessary to evaporate a droplet at 3.0
GHz and suggests that a field strength of 5.8x104 V/m would be necessary at
this frequency. This field strength should be achievable in a microwave cavity
(the breakdown point of air is approx. 1x106V/m)8 however the model is limited
in that it ignores the loss of energy and mass from the droplet due to other heat
transfer effects during the time that the droplet is in the radiation zone. In
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reality it is likely that the droplet will evaporate and lose heat as it absorbs
microwave energy to the point where is may lose energy as quickly as it is
absorbed from the microwave field as suggested by Christiansen and Unruh.8
The model does, however, strongly suggest that higher frequencies would
allow for more efficient absorption of microwave energy and would give a
greater possibility for total evaporation to occur. The evaporation would also
be favored by the increased residence time of the droplet inside the radiation
zone.
5.2.3 Field Strength Model Incorporating Penetration Depth
The penetration depth in water for microwave energy at 2.45 GHz
(approx. 2 cm) is much larger than the droplet size produced by a typical
pneumatic nebulizer. It can be suggested that only a small fraction of the
energy at this frequency is likely to be absorbed by a single droplet. At higher
frequencies the penetration depth of microwave radiation reduces significantly
(Figure 5.1) and the microwave energy will dissipate over a shorter distance.
Therefore a small droplet would be expected to absorb more microwave
energy at a higher frequency. In order to see the effect of penetration depth on
the evaporation of a water droplet, a penetration depth term was incorporated
into the model described previously.
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The penetration depth, Dp (the depth at which the field strength fall to
1/e of its initial value), is given in equation 3. The attention factor,a, can be
calculated as the inverse of the penetration depth:4
The ratio of the applied field strength to the field strength after passing
through the droplet (E/E0) can then be calculated from the attention factor and
initial droplet diameter (D0) by the rearrangement of the defining equation for
the attenuation factor:4
From the inverse of this value the fraction of the microwave field
strength that is not absorbed by the droplet can be found and, therefore, the
fraction of the microwave field that is absorbed by the droplet, E ^ , can be
calculated.
This value can then be used to find the total applied field strength, Etotal,
necessary to evaporate a droplet by dividing the applied field strength
calculated from the previous model, E, by the absorbed fraction of the field
strength, Eabs.
Ill
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2.4 Results and Discussion
The field strength with penetration model modifies the field strength
model by taking into account the possibility that only a small fraction of the
electric field strength is absorbed by the droplet due to the fact that the
penetration depth is large in comparison to the diameter of the droplet. The
model can be used to calculate the applied field strength need to evaporate a
10 (im droplet in 10 ms using the computer program (Mathcad Plus, Mathsoft)
shown in appendix B. This program is modified from the program in appendix
A to allow the effect of penetration depth to be included. The effect of
penetration depth is seen to dominate the model indicating that a field strength
of 1.4x108 V/m would be necessary to achieve evaporation, in 10 ms, of a
droplet at 3.0 GHz. This field strength would be impossible to achieve in a
microwave cavity due to the breakdown of air under normal conditions. The
dependence of the field strength required for evaporation as a function of
applied frequency is shown in Figure 5.5 which was constructed from data
obtained using the program in appendix B and shows that evaporation may be
achieved at a frequency of approximately 10 to 25 GHz which corresponds to
the smallest penetration depth of microwaves in water.
11 2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2.5 Absorbed Power Model
A model for the determination of microwave power absorbed by a water
droplet as a function of applied frequency has been developed by Prof.
Yngvesson (University of Massachusetts, department of electrical and
computer engineering). A good source of background information used in the
development of this model can be found in Von HippePs book on “Dielectrics
and waves”.4
In this model a cavity size of A/4 by A/2 by
10A. is assumed where Ais
the applied wavelength for a single mode cavity. The volume of this cavity is
then:
1.25A.3 = 1.25-4
f
12 .
where c is the speed of light.
The quality factor (Q) of the cavity which is defined by the equation:
2ic(storedenergy)f
13.
where Ps is the dissipated power from the microwave source and the
stored energy is given by:
storedenergy = A —£0.E'2Vrcuv
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14.
where A is a geometric factor and close to one and Vav is the volume of
the cavity. Therefore, we can combine these equations to get:
e 1=
u
1.257ve0c
The electric field strength that the droplet is exposed to, E, gives rise to
an internal electric field, E,,,,, which is smaller. The field strength, E, is given by:
16 .
where D is the dielectric flux density or displacement which is defined
as:
17.
Taking the droplet shape into account (assumed to be spherical) we
can calculate the internal electric field using the equation:
3E
iat ~
k
'+ 2 ~
3E
k
'
The power, P, absorbed by the droplet is given by:
P = 7fe"EltldV
19.
where dV is the change in volume of the drop. Substituting E^ and E
into this equation now gives:
20.
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
which can be used to find the power absorbed by a droplet given the
source power, Q factor, dielectric constant and loss, applied frequency and
volume change of the droplet and assumes that the cavity size is scaled as the
frequency changes. The dielectric constant and loss can be found from the
real and imaginary parts of the complex formula:
a
e = ----------- j ------
21.
I + l ( 1.7xIO ,o)
where k' is taken as the value at low frequency (approx. 80).
5.2.6 Results and Discussion
The absorption of microwaves in a water droplet model (developed by
Prof. Yngvesson) describes the power that can absorbed by a 10 pm diameter
droplet at varying frequencies and is shown in Figure 5.6. This graph was
constructed using the computer program (Mathcad Plus, Mathsoft) shown in
appendix C which is based on equations 20 and 21. This model shows
increasing power absorbance as the applied frequency is increased which
suggests that a higher frequency would produce a greater degree of heating in
the droplet and would be more likely to cause evaporation of an aerosol. The
power absorbed by the droplet calculated by this model is very low at lower
frequencies and suggests that the droplet is essentially absorbing little or no
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
microwave power at frequencies around 2.45 GHz or below. The absorbed
power at these frequencies may not be sufficient to desolvate the droplet
allowing for other heat transfer effects.
5.3 Conclusions
The mathematical models described, although simplistic, give an
indication of the possibility of utilizing microwave energy for the desolvation of
an aqueous aerosol. The models describe the field strength necessary to
provide enough energy for the droplet to evaporate and the frequency which
would provide the best transfer of microwave energy to a droplet. The models
could be modified to be more realistic by incorporating other heat and mass
transfer effects. Improvements could also be made by the development of a
dynamic model which would allow for the study of the effect of changing
droplet size on the ability of the microwave to absorb microwaves. Other
limitations of the models are the absence of any surface energy or resonance
terms. The models show that the commonly used frequency of 2.45 GHz is not
the most efficient frequency for aerosol desolvation and may explain the
difficulties in developing an aerosol desolvation system at this frequency. All
of the models indicate that a higher frequency (about 10 GHz) would be more
efficient for this purpose.
116
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.4 References
1.
Abu-Samra, A., Morris, J., and Koirtyohann, S., Anal. Chem., 1975, 49,
1475.
2.
Kingston, H. M., and Jassie, L. B., “Introduction to Microwave Sample
Preparation-Theory and Practice”, 1988, American Chemical Society
(Washington DC).
3.
Michael, D., Mingos, P., and Baghurst, D., Chem. Soc. Rev., 1991, 20,
1.
4.
Von Hippel, A. R., “Dielectrics and Waves", 1954, M.I.T. Press (New
York).
5.
Perkin, R., J. Heat Mass Transfer, 1980, 23, 687.
6.
Tsang, L., Kong, J., and Shin, R., “Theory of Microwave Remote
Sensing”, 1985, J. Wiley and Sons (New York).
7.
Collin, R. E., “Antennas and Radiowave Propagation”, 1985, McGrawHill (New York).
8.
Christiansen, D. E., Unruh, W. P., Ceram. Trans, 1991, 21, 597.
9.
Masters, K., “Spray Drying Handbook”, 1985, J. Wiley and Sons (New
York).
10.
Von Hippel, A. R., “Dielectric Materials and Applications”, 1954, page
361, M.I.T. Press (New York).
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5.1. Some physical constants and definitions.
Quantity
Symbol
Equation
Units
angular frequency
attenuation constant
(0
area
complex dielectric constant
A
CO=27Cf
■«(1/k)ln(5/E )
A=x2
rad s'1
m'1
m2
£
£=£'-]£"
farad m'1
conductivity
a=J/E
lll=dQ/dt
mho/m
current
a
1
current density
J
IJI=dl/dA
amp
am p/m2
dielectric conductivity
dielectric constant
G
a=co£"
mho/m2
£'
e'=IDI/IEI
8.854E-12
farad m'1
dielectric constant of free space
oc
farad m'1
dissipation factor (loss tangent)
distance
£o
tan5
X
electric charge
Q'
electric dipole moment
P
E
p.=Q'x
E=F/Q'
coul m
D
D=£'E
e"
e"=J1
oss/ coE
P
F
P=D-£oE
d(mv)/dt
coul m'2
farad m*1
coul m'2
newton
electric field strength
electric flux density (displacement)
electric loss factor
electric polarization
force
frequency
tanSse'Ve7
m
coulomb
mass
f
m
power
P
P=dW/dt
Q of dielectric (quality factor)
resistivity
Q
Q=1/tan8
P
k"
P = 1 /g
relative dielectric loss factor
relative dielectric constant
time
velocity
work
volt m'1
S'1
kg
watt
ohm m
K"=8,7e0
K7
t
K,=£7e0
V
w
v=x/t
s
W=Fx
118
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
m s'1
newton m
Log(penetration
depth/m)
3
-
1
log
(frequency/Hz)
Figure 5.1. Variation of penetration depth with frequency for water (dielectric
data from Von Hippel10).
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10-4 -
5
cm ’ 1
10
Frequency in GHz
100 200 300
Fig 5.2. Absorption coefficient vs. frequency for a cloud (0.8 g/m3, 20 pm mode
droplet diameter). Redrawn from Tsang, L., Kong, J., and Shin, R., Theory of
Microwave Remote Sensing", J. Wiley and Sons, 1985.
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Water
Droplet
Hot Droplet
Microwave Cavity
6 -------------------
► •*
Radiation Zone
Figure 5.3. Schematic diagram of mathematical model.
Evaporation
Zone
7
6 .5
~ 4 .5
3 .5
8
10
Log frequency (Hz)
Figure 5.4. Variation of field strength necessary to evaporate a 10 pm water
droplet with applied frequency according to the field strength model.
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
15
14
13
1 1
9
O
9
-J
8
7
6
5
6
7
8
9
10
Log Fraquency (Hz)
Figure 5.5. Variation of field strength necessary to evaporate a 10 pm water
droplet with applied frequency according to the field strength incorporating
penetration depth model.
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
Log Frequency (Hz)
0
8
9 .0 0
.5 0
10 .0 0
1 0 .5 0
2
4
6
8
-I
10
12
-1 4
Figure 5.6. Effect of applied frequency on the power absorbed by a 10
water droplet according to the absorbed power model.
124
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 DO
CHAPTER 6
DEVELOPMENT OF A DESOLVATION SYSTEM FOR ICP-MS UTILIZING
MICROWAVE ENERGY
6.1 Introduction
The sample introduction system for atomic spectroscopy has been
described as the Achilles’ heel of atomic spectroscopy.1 The restrictions of
sample introduction to flame and plasma spectrometers are well known. In
general liquid samples are preferred in analytical experiments due to their
homogeneous, well mixed nature and ease of transport to the instrument, e.g.,
by pumping through tubing. The predominant problem with the introduction of
liquid samples into a plasma source is the amount of energy required by the
plasma to desolvate the liquid. This can cause cooling and extinction of the
plasma and lead to the formation of interfering species due to the presence of
the solvent. The ultimate consequence of these effects has been the necessity
to reduce the amount of solvent, and therefore the amount of analyte entering
the source. Conventional sample introduction systems consist of nebulizer
and spray chamber combinations which are designed to introduce small
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
droplets of the sample into the plasma and are typically 1 to 2% efficient in
terms of analyte mass transport.
One of the most effective means to reduce the sample introduction
limitations in ICP-MS and ICP-AES is to decrease the degree of desolvation
that is necessary within the plasma. This can be achieved by adding a
desolvation system prior to introduction of the sample to the plasma source.
The desolvation is usually achieved by heating the spray chamber to
evaporate solvent from the sample particles followed by a solvent vapor
removal step by condensation or with the use of a membrane dryer. Simple
desolvation systems for increasing the sensitivity of flame spectrometers have
been reported as long ago as 1952.2 This system produced a fine spray into a
heated single pass spray chamber and introduced the resulting material to the
flame without removing the solvent vapor. A desolvation system for
introducing sample to a r.f. plasma was described by Veillon and Margoshes
in 19683 who used a heated spray chamber connected to a Freidrichs
condenser in order to desolvate an aerosol and allow greater sample transport
to the plasma. This system was found to have an overall efficiency of 35%,
defined as the fraction of aspirated analyte entering a flame, and produced a
10-fold increase in sensitivity compared with that for an unheated spray
chamber for an argon plasma instrument. Many similar systems have been
reported for ICP-AES, e.g. for the determination of trace metals in freshwaters4
and the analysis of blood samples.5*7 A desolvation system has also been
described for use with a capacitively coupled plasma atomic emission
126
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
spectrometer.8 Gustavsson investigated the use of a jet separator9 and
membrane dryers10"13 in place of the condensation stage in a desoivation
system for ICP-AES. Desolvation systems incorporating heated spray
chambers and condensers have also been reported for ICP-MS to increase
sample transport and reduce polyatomic interferences.14'16 Peltier coolers have
also been used for the condensation stage in desolvation systems.17,18
Cryogenic desolvation is a method for removing solvent from a sample
by passing an aerosol through a length of coiled tubing mounted
horizontally.19 The top portion of the coil was heated while the lower portion
was cooled such that the sample aerosol goes through an evaporation and
condensation stages for each turn of the coil. This has the effect of removing
much of the solvent from the aerosol while reducing the amount of
condensation back onto the sample particles. Cryogenic desolvation has
been shown to significantly reduce interfering oxide and hydride interferences
in ICP-MS arising from the use of organic solvents20 although substantial
memory effects were observed for several metal complexes. This system has
also been proposed in a method for screening urine samples for vanadium by
ICP-MS21 due to its ability to reduce the interfering CIO+ species. In a
comparison of cryogenic desolvation with membrane desolvation22 for the
attenuation of oxide, hydride, hydroxide ions and chlorine containing ions,
cryogenic desolvation was found to give a greater reduction in intensity for
LaO+, ThO+ and ThOH+. Membrane desolvation achieved a lower detection
limit for vanadium in urine samples.
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Alternative methods for evaporating solvent from an aerosol have been
considered. The use of infrared radiation as a method for desolvating an
aerosol without the necessity to heat the spray chamber walls was first
reported by Hell et. al. in 196S.23 Eastgate et. al.24 compared a spray chamber
embedded in an electrically heated copper block to a spray chamber of similar
dimensions placed at the focus of a radiant heater. The radiation system was
found to give better short term stability than the conduction system while
sharply reducing wash-out times and carryover effects. This work was the
basis of the Mistral system (Fisons Instruments) which was evaluated by
Schron and Muller who found a clear improvement in detection limits (by a
factor of between 1.5 and 6) for water samples compared to conventional
sample introduction.25
Efficient nebulizers that produce a greater quantity of small droplets
usually require desolvation methods to prevent large amounts of solvent from
entering the plasma. The ultrasonic nebulizer has become popular in plasma
spectroscopy due to its high efficiency. Initially desolvation systems
incorporating a thermally heated stage and a condensation stage were used
to reduce the solvent loading produced by these nebulizers.26,27 Significant
analyte losses have, however, been reported in ultrasonic nebulizer systems
both in the spray chamber and the condenser.28 A desolvation system for an
ultrasonic nebulizer has also been successfully developed incorporating a
porous PTFE membrane in order to remove solvent vapor. This system has
been evaluated for the analysis of volatile solvents by ICP-AES29,30and ICP-
128
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MS.31 Other high efficiency nebulizers have also been combined with
desolvation systems for plasma spectrometry including the thermospray32-33
and microconcentric nebulizers.34
The use of microwave energy for the desolvation of a sample has been
investigated by Canals et al. using two different approaches. In one approach
microwave energy was used to produce a thermospray for liquid sample
introduction in ICP-AES.35 In this work a sample stream was pumped by an
HPLC pump through a length of PTFE tubing which was passed through a
focused microwave oven. The tubing then emerged from the bottom of the
oven where a short piece of narrower silica or PTFE tubing was attached. A
hot aerosol could be produced at the exit of the capillary in this manner. This
aerosol was transported to the plasma after passing through a non­
thermostated, single-pass spray chamber and two Leibig condensers. The
results from this system were compared with those obtained with a Meinhard
nebulizer and double pass spray chamber without a desolvation unit.
Although the results showed an increase in intensity of 1.0 to 2.1 (counts per
second for the desolvation system divided by counts per second for
conventional sample introduction) this improvement is relatively small in
comparison with that obtained with other desolvation systems. Another
approach considered by this group was to produce an aerosol in a spray
chamber inside a microwave oven in order to evaporate the solvent from the
sample droplets prior to solvent vapor removal by condensation.36,37 In this
work an aerosol was produced by a Meinhard nebulizer in a single pass glass
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
spray chamber which is positioned inside a domestic microwave oven. The
resulting heated aerosol was then passed through a length of Tygon tubing to
two Leibig condensers to remove the solvent vapor. For ICP-MS
determinations, the results were compared with those obtained for
conventional sample introduction (Meinhard nebulizer and double pass spray
chamber) and showed significant improvements with 2 to 14 times higher
intensities and lower Ce07Ce+ and Ba27Ba+ ratios. Despite these
improvements, the mechanism for the desolvation remains unclear. Although
it may be possible that the aerosol droplets are being heated directly by the
microwaves under these conditions (900W, 2.45 GHz) there are also other
possible mechanisms for the absorption of heat by the droplets from the
surroundings. Both the Pyrex glass spray chamber and Tygon transport tubing
are known to absorb microwave energy to some degree and may be heating
the sample aerosol. The system described is mounted vertically with the
nebulizer at the bottom, this suggests that any droplets that strike the spray
chamber and tubing walls are being heated either immediately or at the
bottom of the spray chamber by the microwaves which would have the effect of
filling the system with water vapor and raising the temperature of the aerosol.
In the work described in this chapter an attempt has been made to
develop a desolvation system using microwave energy and couple it to a
membrane dryer device to improve the sensitivity of an ICP-MS instrument
without significantly increasing water loading effects.
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6.2 Experimental
6.2.1 Development of a Microwave Thermosprav Desolvation Device
An attempt was made to develop a microwave thermospray type device
using the single mode microwave cavity described previously (chapter 4). An
HPLC pump (Waters, model M-6000) was used to pump water through approx.
10 m of PTFE tubing (0.8 mm, i.d.) placed inside the microwave cavity. The
PTFE tubing was connected to a length of fused silica capillary (15 to 20 m,
320 pm) via a PEEK sleeve and union (Upchurch Scientific). The capillary
was coiled to fit inside the cavity so that the end just exited the choke at the
end of the cavity, as shown in Figure 6.1. The water was then pumped at
varying flow rates with different lengths of capillary tubing (between 15 and 20
m). This was done in an attempt to boil the water at the tip of the capillary and
produce a thermospray effect which occurs when a hot stream of liquid is
converting into an aerosol by water vapor produced in the capillary. The
capillary tip diameter was varied by placing the capillary in a Bunsen flame
and drawing it out so that the end could be cut off to produce a smaller
diameter tip in order to increase the pressure in the system. The degree of
heating was observed by touch or measured with a mercury in glass
thermometer.
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A desolvation system was developed using the same equipment
housed in a CEM MDS-81 multi mode microwave oven. Water was pumped
into the oven through a hole in the top of the oven to the PTFE and fused silica
coils, the end of the silica capillary was narrowed in a Bunsen flame and
exited the oven through a separate hole in the top. The exiting material was
then coupled to the Nation dryer device described in chapter 3 by a cross-flow
type interface, Figure 6.2a. The interface consisted of a PTFE T-piece, one
side of which was connected to a compressed air cylinder via a length of PTFE
tubing while the other end was connected to the inlet of the Nation dryer by a
short length of Tygon tubing. The bottom arm of the T-piece was sealed with a
piece of parafilm so that the end of the silica capillary could be pushed through
the film and held in place in the gas stream (greater than 1 l/min). Nitrogen at
approx. 10 l/min was used as the purge gas the dryer which was heated to
60°C. In this way a spray could be produced by the compressed air flowing
over the tip of the capillary. In subsequent experiments the capillary was
introduced into the gas stream prior to the T-piece through a small hole drilled
in the PTFE tubing that was covered with parafilm to reduce gas leakage,
Figure 6.2b. This configuration allowed the bottom arm of the T-piece to be
used as a drain and prevent build up of water in the tubing. Positive pressure
was applied to the drain by use of a 0.8 mm i.d PTFE U-tube attached to the
lower arm of the T-piece. The efficiencies of these systems were measured by
the mass increase of drierite tubes placed at the outlet of the dryer and the
outlet of the purge gas stream.
132
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6.2.2 Development of a Microwave Heated Nebulizer Desolvation System
In order to produce a heated aerosol, a Meinhard nebulizer was placed
at the exit of the single mode microwave cavity with the tip of the nebulizer
held in place by a piece of polystyrene, Figure 6.3. Water was pumped to the
nebulizer by a HPLC pump through a length of PTFE coupled to a length of
silica capillary tubing located inside the cavity. Compressed air was supplied
to the nebulizer via a length of PTFE tubing which was passed through the
cavity. The temperature of the exiting aerosol was measured by a mercury in
glass thermometer. This device was coupled to the Nation dryer system
described in chapter 3 via a length of quartz tubing (11 cm long, 1.8 cm i.d.)
which fitted into the end of the choke. The end of the quartz tube was
narrowed to (0.8 cm i.d) and connected to the dryer with a short length of PTFE
tubing (30 cm long, 0.4 cm i.d.). Transport and dryer efficiencies were
measured by drierite tubes which were placed at the exit of the sample gas
and purge gas streams of the Nafion dryer. The increase in mass of the
drierite tubes after a known length of time was assumed to be due to water
absorbed.
133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6.2.3 Development and Testing of a Microwave Heated Cvclone Sprav
Chamber Desolvation System
A Meinhard nebulizer attached to a glass cyclone spray chamber was
placed inside a CEM MDS-81 multi mode microwave oven, Figure 6.4. The
spray chamber was mounted onto a polystyrene block and placed in the
center of the cavity. Liquid was transported to the nebulizer by a peristaltic
pump through 0.8 mm i.d PTFE tubing, argon gas was supplied through the
ICP-MS mass flow controller via a length of Tygon tubing, PTFE tubing (0.8
mm i.d.) was used to pump waste from the spray chamber using a peristaltic
pump. The outlet of the spray chamber was connected to the Nafion dryer
device described earlier (chapter 3) with a length of PTFE tubing (approx. 50
cm long, 0.5 cm i.d.). This tubing was fixed tightly into the outlet arm of the
spray chamber using 2 O-rings to prevent gas leakage. The Nafion dryer was
setup and operated using the optimum conditions described in chapter 3. The
ICP-MS operating conditions and liquid flow were optimized for high signal
intensities for the three test elements by an alternating variable search
method. The optimized conditions are shown in Table 6.1. Heating of the
transfer line between the exit of the microwave oven and the Nafion dryer was
accomplished using 2 lengths of heating tape (Bamstead, Briskheat)
controlled independently by 2 control boxes (Geo. Ulanet Co., Robotemp).
Testing of this system was achieved using a combined standard solution of 10
134
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ng/ml magnesium, rhodium and lead in a solution of 1% nitric acid in
deionised water.
6.3 Results
6.3.1
Development of a Microwave Thermosprav Desolvation Device
The purpose of this work was to produce a thermospray nebulizer using
microwave energy by heating a liquid stream to a temperature where water
vapor would be produced forming a spray at the tip of a capillary tubing. A
system was set up without coupling to the dryer so that the conditions could be
varied and the effect on the exiting material observed. Initially a 17m length of
silica was used, tightly coiled inside the microwave cavity and the flow rate
varied. The optimum flow rate for heating was estimated to be approx. 4
ml/min which gave the greatest degree of heating, measured by touch, and
produced some water vapor along with the liquid stream. The increase in
temperature of the liquid stream was difficult to measure using a mercury in
glass thermometer due to the low volume of water exiting the capillary which
was not sufficient to completely cover the thermometer bulb. The experiment
was repeated for a 20 m capillary which appeared to heat well at 5 ml/min. In
this experiment, placing large lengths (greater than about 10 meters) of silica
capillary into the cavity without breaking was a considerable difficulty. To
135
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
produce a greater degree of heating in a shorter length of capillary tubing, the
diameter of the capillary tip was reduced using a Bunsen flame. For a 15m
length of capillary this seemed to produce a greater degree of heating at lower
flow rates (1 to 2 ml/min) although it was still not possible to produce an
aerosol from the capillary tip.
In work by Bordera et al.35 a thermospray type effect was observed at
the tip of a capillary in a microwave field. In this work a focused microwave
oven was used which may allow a greater degree of microwave energy to be
absorbed in a shorter length of capillary tubing. It should also be noted that
the spray produced in this technique may be a type of heated hydraulic
nebulizer rather than a thermospray. The difference between these effects is
discussed by Bemdt and Yanez38 who describe a thermospray as a technique
that involves the external heating of a liquid stream to a temperature where
partial vaporization of the solution occurs creating an intense vapor jet at
supersonic velocity from the capillary tube. A hydraulic nebulizer is
characterized by the applied pressure and nozzle type and produces a highly
turbulent flow inside the nozzle to produce a spray of the solution. By heating
this system, energy can be made available at the exit of the nozzle to vaporize
some of the solution. The Bordera system did not have a choke allowing the
capillary tip to be positioned at the exit of the cavity. In our work a choke was
thought to be necessary for safety purposes in order to prevent microwave
leakage. Due to the inability to see inside the cavity or choke, it was found
necessary to place the capillary tip external to the choke. The length of
136
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
capillary outside of the microwave field may have been long enough to allow
the liquid stream to cool. This may prevent the water from reaching a high
enough temperature at the capillary tip to produce the amount of water vapor
necessary for a thermospray type effect to occur.
Although a spray was not produced, a significant increase in
temperature was observed such that if an aerosol could be produced from this
heated stream it may produce heated droplets that would desolvate in the gas
stream. In order to test this theory, the capillary tip (using a multi mode cavity)
was interfaced to a Nafion dryer device using a cross flow type interface
(Figure 6.2a). The transport efficiency of this system was determined using
preweighed drierite tubes attached to the exit of the dryer and exit of the purge
gas stream. The transport efficiency was found to be 5% at 80% microwave
power for the first run, however, after this run a build up at water was observed
in the dryer. In order to prevent this build up of water, the interface was
adapted, Figure 6.2b, to allow excess water to drain out. Using this system,
the aerosol transport rate was found to be between 5 and 8%. Attempts to
couple this desolvation system to an ICP-MS instrument were unsuccessful
due to the inability to produce an aerosol at the normal flow rates used for ICPMS and the ease of breakage of the capillary tip.
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6.3.2 Development of a Microwave Heated Nebulizer Desolvation System
The aerosol transport efficiency for the Meinhard nebulizer system
connected to the Nafion dryer (Figure 6.3) was evaluated with drierite tubes at
the exit of the purge gas and sample gas streams. The total transport
efficiency of water to the dryer was calculated to be only 0.5% with 97% of this
water removed by the purge gas of the dryer. In the absence of the dryer and
quartz tube interface, the temperature of the aerosol produced could be
measured using a mercury in glass thermometer positioned in the aerosol
stream to increase by approximately 2°C when the microwave energy was
applied. One possible reason for the low transport efficiency was the
backpressure produced by the dryer. Therefore, an experiment was
conducted with the dryer removed from the system and replaced by a single
drierite tube with the microwave power off. In this way, the total amount of
water transported from the system could be measured as the increase in
weight of the tube. The transport efficiency was found to be between 0.4 and
0.7% with a build up of water observed in the quartz tube connector. This
transport efficiency remained the same even after the addition of a small glass
drain arm on the bottom of the quartz tube to remove excess build up of water.
The low efficiency is probably due to a large proportion of the aerosol striking
the walls of the quartz tube. In order to increase the efficiency of this system a
larger diameter connector would be needed, however, the diameter of the
connector in this system is restricted by the cavity dimensions.
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6.3.3 Development and Testing of a Microwave Heated Cvclone Spray
Chamber Desolvation System
The effect of microwave power on the signal intensities produced for
Mg-24, Pb-208, Rh-103 and ArO-56 was determined and the results shown in
Figure 6.5 as the ratio of the signal intensity found divided by the signal found
at 0% microwave power without heating the transfer line. Although an
increase in signal intensity is noted for all of the test species, this increase was
small. The overall intensity of the unheated system was found to be lower than
the signals produced by a conventional cross flow nebulizer and double pass
spray chamber by approximately a half to a third for all of the species studied.
It was observed during the microwave heating stages that condensation of the
aerosol produced from the spray chamber was occurring on the walls of the
transfer line between the microwave oven and the dryer. This became
particularly apparent at 100% microwave power when a plug of liquid was
observed in the transfer line. This build up of liquid in the transfer line may be
the cause of the low transport efficiency of the system particularly at 100%
power when the signal was seen to decrease significantly. In order to avoid
condensation in the transfer tubing, heating tape was wrapped around it
between the exit of the oven and the inlet of the dryer. At settings above 2 on
the heating tape control boxes the PTFE tubing was found to soften and melt,
therefore, the control boxes were set to ‘lo’ in all subsequent experiments. The
addition of the heating tape led to an increase in signal intensity compared to
the system without a heated transfer line, Figure 6.6. The heated transfer line
139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
also allowed for a greater increase in signal intensities with an increase in
microwave power up to a maximum of almost twice the signal obtained at 0%
power with an unheated transfer line. The results for both the heated and
unheated transfer line systems indicate that the ArO+ intensity tends to follow
the same trends as the intensities for the other test elements. This suggests
that an increase in water vapor accompanies an increase in sample species in
the plasma. The intensity for the ArO+ signal, however, was significantly lower
than conventional sample introduction using a cross flow nebulizer and
double pass spray chamber in all conditions and was always below 10 000
counts s'1.
Although these results show that signal improvements are possible by
placing a spray chamber and nebulizer in a microwave field, the signal
improvements are low. The overall intensities of the signals for the test
elements were not significantly higher than the intensities achieved using a
conventional sample introduction system. Mora et al.37 described a
desolvation system incorporating a Meinhard nebulizer and single pass spray
chamber positioned inside a domestic microwave oven, coupled to a pair of
condensers and reported significant signal intensity improvements (up to 10
fold) over conventional sample introduction. The difference in performance
between this reported system and the one described here may in part be
explained by the single pass spray chamber used. The reported system
probably has a much higher sample transport efficiency compared to the
cyclone spray chamber used in this study. The signal intensities for the system
140
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
without microwave heating could be a useful piece of information in order to
make a comparison between the two systems but was not reported in the Mora
paper. The higher maximum power of the domestic oven used (890W)
compared to the CEM oven used in this study (700W) may also be significant
for increasing the signal intensities.
Although these results do show that microwave radiation can be used to
desolvate a sample aerosol, the effects seen might be expected to be much
higher if the microwaves are efficiently coupling directly to the aerosol
droplets. It is therefore assumed that most of the desolvation effect observed is
due to thermal heating of the droplets by microwave absorption of the
surroundings and that significant coupling between the droplets and the
microwave field does not occur under these conditions. These results are
supported by the theory of microwave interaction with water droplets at 2.45
GHz as discussed in chapter 5.
6.4 Conclusions
The use of the single-mode microwave cavity apparatus was not able to
produce a thermospray at the tip of a silica capillary. Coupling a heated water
stream produced in a silica capillary by a multi-mode cavity to the Nafion dryer
apparatus was found to be possible using a cross-flow type interface. This
system was determined to be between 5 and 8% efficient in terms of water
141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
transfer to the dryer. It was not possible to successfully couple this system to
an ICP-MS instrument. Positioning of a Meinhard nebulizer and capillary
tubing inside a single mode microwave oven to produce a heated spray was
possible. The dimensions of the system did not allow for efficient transfer of the
resultant material to the Nafion dryer device. A desolvation system has been
developed incorporating a Meinhard nebulizer and cyclone spray chamber
placed in a microwave field and coupled to a Nafion dryer device. This system
can increase signal intensities by approximately 2 fold using a heated transfer
line at 100% microwave power compared to the same system at 0%
microwave power without heating the transfer line. It is important for the
transfer line to be heated to prevent condensation of the heated aerosol on the
walls of the transfer line. The signal intensities found using this desolvation
system do not show significant improvement compared to a conventional
sample introduction system. This increase in signal intensity is lower than
might be expected with efficient coupling of the microwave energy with the
aerosol droplets.
6.5 References
1.
Browner, R. F., and Boom, A. W., Anal. Chem., 1984, 56, 786A.
2.
Dubbs, C. A., Anal. Chem., 1952, 24, 1654.
3.
Veillon, C., and Margoshes, M., Spectrochim. Acta, 1968, 23B, 553.
4.
Goulden, P. D., and Anthony, D. H. J., Anal. Chem., 1982, 54, 1678.
142
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.
Marinov, M. I., Spectrosc. Lett., 1984,17,1.
6.
Marinov, M. I., Spectrosc. Lett., 1984,17, 33.
7.
Marinov, M. I., Spectrosc. Lett., 1984,17,151.
8.
Uchida, H.t Masamba, W. R., Uchida, T., Smith, B. W., and Winefordner,
J. D., Appl. Spectrosc., 1989, 43, 425.
9.
Gustavsson, A., Spectrochim. Acta, 1987, 42B, 111.
10.
Gustavsson, A., and Nygren, O., Spectrochim. Acta, 1987, 42B, 883.
11.
Gustavsson, A., and Hietala, P., Spectrochim. Acta, 1990, 45B, 1103.
12.
Gustavsson, A., Spectrochim. Acta, 1988, 43B, 917.
13.
Backstrom, K., Gustavsson, A., and Hietala, P., Spectrochim. Acta, 1989,
44B, 1041.
14.
Tsukahara, R., and Kubota, M., Spectrochim. Acta, 1990, 45B, 581.
15.
Lam, J. W., and McLaren, J. W., J. Anal. At. Spectrom., 1990, 5, 419.
16.
Jakubowski, N., Feldmann, I., and Stuewer, D., Spectrochim. Acta,
1992, 47B, 107.
17.
Hill, S. J., Hartley, J., and Ebdon, L., J. Anal. A t Spectrom., 1992, 7, 23.
18.
Weir, D. G. J., and Blades, M. W., Spectrochim. Acta, 1990, 45B, 615.
19.
Alves, L. C., Weiderin, D. R., and Houke, R. S., Anal. Chem., 1992, 64,
1164.
20.
Alves, L. C., Minnich, M. G., Weiderin, D. R., and Houk, R. S., J. Anal. At.
Spectrom., 1994, 9, 399.
21.
Minnich, M. G., Houk, R. S., Woodin, M. A., and Christiani, D. C., J. Anal.
At. Spectrom., 1997,12, 1345.
22.
Minnich, M. G., and Houke, R. S., J. Anal. At. Spectrom., 1998,13,167.
23.
Hell, A., Ulrich, W. F., Shifrin, N., and Ramirez-Munoz, J., Appl.
Opt.,1968, 7, 1317.
24.
Eastgate, A. R., Fry, R., C., and Gower, G. H., J. Anal. At. Spectrom.,
1993, 8, 305.
25.
Schron, W., and Muller, U., Fresenius. J. Anal. Chem., 1997, 357, 22.
143
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26.
Fassel, F. A., and Bear, B. R., Spectrochim. Acta, 1986, 41B, 1089.
27.
Castillano, T. M., Vela, N. P., Caruso, J. A., and Story, W. C., J. Anal. At.
Spectrom., 1992, 7, 807.
28.
Tarr, M. A., Zhu, G., and Browner, R. F., J. Anal. At. Spectrom., 1992, 7,
813.
29.
Botto, R. I., and Zhu, J. J., J. Anal. At. Spectrom., 1994, 9, 905.
30.
Brenner, I. B., Zhu, J., and Zander, A., Fresenius J. Anal. Chem., 1996,
355, 774.
31.
Brenner, I. B., Zander, A., Plantz, M., and Zhu, J., J. Anal. At. Spectrom.,
1997, 12, 273.
32.
Koropchak, J. A., and Winn, D. H., Anal. Chem., 1986, 58, 2558.
33.
Yang, J., Conver, T. S., Koropchak, J. A., and Leighty, D. A.,
Spectrochim. Acta, 1996, 51B, 1491.
34.
CETAC Technologies Product Literature, Omaha, Nebraska.
35.
Bordera, L., Todoli, J. L., Mora, J., Canals, A., and Hemandis, V., Anal.
Chem, 1997, 69, 3578.
36.
Gras, L., Mora, J., Todoli, J. L., Hemandis, V., and Canals, A.,
Spectrochim. Acta, 1997, 52B, 1200.
37.
Mora, J., Canals, A., Hemandis, V., van Veen, E. H., and de LoosVollebregt, M. T. C., J. Anal. At. Spectrom., 1998,13,175.
38.
Bemdt, H., and Yanez, J., J. Anal. At. Spectrom., 1996,11, 703.
144
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 6.1. Optimized parameters for microwave heated cyclone spray
chamber system.
Heated Transfer
Unheated
Line
Transfer Line
Power / W
995
1005
Nebulizer Flow Rate/ L min'1
0.898
0.691
Plasma Flow Rate/ L min*1
14.5
15.0
Auxiliary Flow Rate/ L min*1
0.966
0.898
Sample Flow Rate/rev min*1
24
24
145
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Magnetron
Microwave Cavity
Antenna
PEEK
Union
Pump
Water
Reservoir
PTFE
Coil
Silica
Capillary
Coil
Figure 6.1. Schematic diagram of the microwave thermospray desolvation device.
Compressed
A ir
FTFE T-Piece
------------------------------
Silica
C a p illa ry
To Dryer
‘P a ra f ilm
From
Oven
a. Original design
P a raf ilm
Compressed
A ir
_
1
/
Silica
C apillary
_ .
From
Oven
T
- R
e
c e
To Dryer
to U tube
Drain
b. Adapted design
Figure 6.2. Schematic diagram of the adapted cross flow interface.
147
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Magnetron
Antenna
zr
Microwave Cavity
Meinhard Nebulizer
PEEK
Union
Quartz Tube
00
Water F eservo r
Silica
Capillary
Coil
From
Gas Cylinder
Figure 6.3. Schematic diagram of the microwave heated nebulizer device.
polystyrene
block
to dryer
to waste
sample
O-ring
gas
cyclone
spray chamber
Meinhard
nebulizer
microwave cavity
Figure 6.4. Schematic diagram of the microwave heated cyclone spray
chamber desolvation system.
149
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.2
In ten sity
1.1
1
Noramlzed
0.9
M g-24
R h-103
P b-208
0.8
A rO -56
0.7
0.6
20
40
60
80
100
Power/%
Figure 6.5. Normalized intensity against microwave power for a microwave
desolvation system with an unheated transfer line.
150
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Normalized
Inten sity
2 .2
1.8
M g-24 :
Rh-103!
1.6
Pb-208l
ArO-56!
1.4
1.2
0
20
40
60
80
100
Pow er/ %
Figure 6.6. Normalized intensity against microwave power for a microwave
desolvation system with a heated transfer line.
151
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 7
CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK
7.1 On-line Microwave Digestion
The development of an on-line microwave digestion system that is
capable of the dissolution of difficulty soluble materials rapidly while being
automated and easy to use has long been a goal of many researchers. In this
work a system has been developed that is capable of digesting organic
materials in a short period of time with nitric acid in an on-line system. The
system incorporates a pressure transducer to allow for continuous monitoring
of the system pressure. It has been demonstrated that milk powder can be
introduced into this system as a slurry and can be digested such that a number
of minor elements can be determined by flame atomic absorption
spectrometry. By adapting the system, a small amount of sample can be
weighed directly into the digestion vessel allowing materials that are difficult to
make into a stable slurry to be digested. This overcomes many of the
problems of blocking of the valves and tubing which occurs in most slurry
based on-line digestion systems. By using a stopped flow system it is possible
to achieve pressures in excess of 300 psig while the U-shaped digestion
152
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
vessel has the advantages of allowing effective mixing of the digesting
material (by motion of bubbles from one side of the tube to the other) and not
allowing the digesting material to come into direct contact with the end fittings.
In this way the contamination due to acid attack of the PTFE end fittings and
Viton O-rings is reduced compared to that of a previous design.1 The
incorporation of a 2-stage depressurization system reduces the overall time for
the digestion procedure by allowing sample to removed from the vessel soon
after the digestion has been accomplished. The short digestion time and
depressurization system allow for the complete digestion and collection of a
sample in as little as 10 minutes. This is high sample throughput in
comparison with other pressurized digestion techniques which require waiting
for the pressure to reduce, in the case of a static system, or the production of a
stable slurry, for most on-line systems.
The digestion of bovine samples has been achieved followed by
determination of a number of trace elements by ICP-MS. Values were
acceptable for many elements compared to the reference concentrations
although many elements were observed to have high values and poor
precision. Contamination from the glass digestion vessel is thought to be the
most likely source of the high values found. The use of a quartz vessel in
place of the borosilicate glass should eliminate this problem and allow these
elements to be determined. In this work it was hoped to obtain a quartz vessel
of the same dimensions as the glass vessel. A glass vessel was sent to
Precision Glassblowing in order to make a quartz copy, however, due to the
153
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
non standard threads used they were not able to buy quartz threads of the
same dimensions and the threads could not be machined well enough to
produce a pressure tight connection to the end fittings.2 It would be possible to
attach glass end-fittings to a quartz vessel, however, this would involve greatly
increasing the size of the vessel and may be not completely eliminate the
contamination from glass. Future work may concern the adaptation of this
design in order to use a quartz tube that can be attached to the PTFE tubing
with pressure tight fittings while not requiring the manufacture of screw threads
from quartz. The application of this system for other organic based materials
should be performed in order to test the ability of the system to digest a wide
variety of sample materials. Measurement of the residual carbon content of
the digested material would also be a useful parameter to determine the
completeness of digestion.
The developed system appears to be promising as a basis for a
commercial digestion system provided that the contamination problems can be
overcome. In producing a commercial system, the automation of the values
and pumps of the digestion system would be advantageous in order to simplify
the use of the digestion system and allow for computer control. The computer
control of the microwave power by monitoring the digestion pressure and
temperature would also be advantageous in order to provide effective
digestion and allow the pressure to be controlled to prevent possible failure of
the vessel or fittings. Directly coupling the digestion system to an atomic
spectrometer could also be investigated.
154
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.2 Reduction of Water Loading Effects in ICP-MS Using a Nafion Drver
The use of a Nafion dryer can significantly reduce the amount of water
vapor that is introduced to a plasma source by a conventional sample
introduction system. This reduction of water can lower some polyatomic
interferences as has been demonstrated for the ArO+ interference on Fe-56 as
an example. The effect of dryer temperature and gas flow rate have been
investigated and shown to have an effect on the efficiency of the dryer for
water removal. Studies of other possible inteferences in real samples could
also be investigated such as the interference of CIO+ on V and the interference
of KO+on Mn. The effect of the dryer may be more prominent with the
reduction of oxygen entering the plasma by air entrapment or via the argon
gas supply. This may be achieved by the use of a different model of ICP-MS
that is better designed to reduce oxygen entrapment, drying the argon gas
prior to use or using a purer argon gas supply.
The dryer appears to be capable of removing large amounts of water
vapor efficiently and could be useful as a desolvation system for more efficient
sample introduction systems such as an ultrasonic nebulizer. The dryer shows
much promise as a commercial desolvation device due its simple operation,
relatively low cost and rugged design although a external control of the purge
gas supply and a more convenient size and shape would be advantageous.
155
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.3 Microwave Desolvation
The development of a microwave desolvation system is problematic
due to the ineffective heating of aerosol droplets by microwave energy at 2.45
GHz. The heating of a liquid stream however is more efficient at this frequency
such that it may be possible to produce a hot aerosol by heating the sample
stream before nebulization. This has been attempted for a cross flow type
interface which showed some promise but needs to be further developed in
order to be more rugged and fix the liquid and gas stream in position. The
possibility of producing a thermospray type effect by heating a liquid stream
has been reported by Bordera et al.3 This effect could be the basis of further
investigation with the use of a higher pressure system, microwave absorbent
tube, focused microwave energy source or high power microwave source.
A microwave heated desolvation system has been demonstrated as a
method for increasing signals in ICP-MS and could be further improved by the
use of a sample introduction system with a higher sample throughput
efficiency. However the signal improvement of such as system at 2.45 GHz
seems unlikely to be dramatic due to the inefficient coupling between the
droplets and the applied radiation. A similar system has been reported by
Mora et al.4 however, these workers fail to give any proof of direct interaction of
the microwaves with the aerosol droplets and demonstrate little understanding
of the theory of their system. With the use of simple mathematical models it
has been demonstrated that 2.45 GHz would be expected to give little
156
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
interaction with a small water droplet. This effect is due to the frequency
dependence of the dielectric constant and dielectric loss, the effect of the
droplet shape on the internal electric field strength and may also be related to
the frequency dependence of the microwave penetration depth. All of the
models discussed suggest that a higher frequency would allow a substantially
improved heating effect. The models described could be expanded further
with the incorporation of other heat and mass transfer mechanisms. Further
improvements could be achieved by the use of a more dynamic approach
allowing for the changes in droplet temperature and mass on the efficiency of
microwave absorption to be incorporated.
The use of microwave energy could still present a substantial
improvement in a desolvation system if a microwave source can be obtained
which is capable of causing substantial heating in a microwave droplet without
significantly heating the walls of a spray chamber. In this way evaporation of
the droplet should be possible while reducing the sample loss and memory
effects caused by droplets drying on the spray chamber walls which occurs in
thermal desolvation systems. A simple experiment could be performed using
a higher frequency source to determine the ability of such a source to dry an
aqueous aerosol and the possible use of such a source for the development
of an efficient microwave desolvation system. This type of experiment could
also be used to validate or refine the theoretical models. A Nafion dryer
appears to be an efficient way to remove the water vapor that would be
produced in such a system without the loss of sample particles.
157
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.4 Conclusion
It can be concluded that with a better understanding of the
characteristics of microwave energy among analytical chemists the uses of
microwave energy could offer substantial benefits in the future for analytical
determinations.
7.5 References
1.
Gluodenis, Jr. T. J.t PhD. Dissertation, 1993, University of
Massachusetts.
2.
Lato, B., Precision Glassblowing, Personal Communication, 1998.
3.
Bordera, L., Todoli, J. L„ Mora, J., Canals, A., and Hemandis, V., Anal.
Chem., 1997, 69, 3578.
4.
Mora, J., Canals, A., Hemandis, V., van Veen, E. H., and de LoosVollebregt, M. T. C., J. Anal. At. Spectrom., 1998,13, 175.
158
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX A
PROGRAM FOR FIELD STRENGTH MODEL
n ..
_ c tn -6
Radius - 5-10
m
r
= 0.0238 watt • ( K ) " 1-m’ 1
Kd
D
Droplet radius
= 0-m
Thermal conductivity of air at 298K
Final droplet diam eter
DO = 2 -Radius
Initial droplet diameter
Lv = 2283.333 joule-gm * 1
d = 1000 gm - liter" 1
t = 0.1 sec
Latent heat of vaporization for w ater
Density of w ater
Evaporation time
T ._ [ L v d - ( D 0 2 - D 2) ]
(8-Kd-t)
T = 11.992’ K
V
=
Tem perature increase
I -4 it \ -Radius3
\3
I
mass = V -d
C
= 4 .1 8 0 joule-gm " 1-K " 1
Specific heat capacity of w ater
Energy := mass -C T
f = 3 109 H z
tabs = 10-sec
Microwave frequency
Tim e that droplet absorbs microwaves
e0
- 8.854-10 I2 farad m 1
K'
= 76.7
tan6 := 1570-10 4
tan delta=loss factor
159
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
k ” - tan 5 k '
k" = 12.042
e"
=
K ” -e O
c
Eint =
Energy
>0.5
i ceM- f V t a b s
Eint = 2.234* 103 *kg*m*sec 2 *coul 1
Internal field strength
E int-(K ’ + 2)
E = ---------------------
Correction for shape
E = 5.859*104 *kg*m*sec 2 *cou! 1
Applied field strength
160
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX B
PROGRAM FOR FIELD STRENGTH MODEL INCORPORATING
PENETRATION DEPTH
_ ,.
_ . , rt-6
Radius = 5 10 -m
Droplet radius
v
Kd = 0.0238 watt • ( K ) ’ 1•m 1
D = 0-m
Thermal conductivity o f air at 298K
Final droplet diam eter
DO = 2 -Radius
Initial droplet diam eter
Lv := 2283.333 joule-gm" 1
Latent heat o f vaporization for water
d = 1000 gm-liter 1
t = 0.1 -sec
Density of water
Evaporation time
[ L v d - ( D 0 2 - D2)]
(8-Kd-t)
T = 11.992* K
V
/4
\
Tem perature increase
- I — k | • Radius
3
Droplet volume
mass - V -d
C
= 4. 180 joule-gm 1-K 1
Specific heat capacity of water
Energy := mass -C T
f = 3 -109 -H z
tabs = 10-sec
Microwave frequency
Time that droplet absorbs microwaves
£0 = 8.854-10 12 farad-m 1
K’ = 76.7
tan 8 = 1570 -10 4
uin delta=loss factor
161
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
K" = tan6 k ’
tc” = 12.042
t"
- tc"-e0
Energy
Eini =
,0.5
7t-e” f V t a b s
Eint = 2.234* 103 *kg*m*sec 2 *coul 1
E in t(K ‘ + 2)
Internal field strength
Correction for shape
E :
E = 5.859* 104 *kg*m*sec 2 *coul 1
Applied field strength
, . _ ( 3 - 1 0 8)
-i
A . ----------------- msec
f
0.5
Dp =
a
-
6.28
Penetration Depth
K ' - [ [ l + (tan8)2]
-
1
1
Dp
Fr = exp( DO- a)
Fr is the applied field strength/ the transmitted
field strength
Eabs = I — —
Fr
Eabs is the fraction of field strength
absorbed by the droplet
Etot -
Eabs
Etot = 1.362* 108 *kg*m*sec 2 *coul 1
T otal R eid strength necessary
16 2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX C
PROGRAM FOR ABSORBED POWER MODEL
f
- 3-10
E
=
Applied frequency (Hz)
80
Complex dielectric constant
f
1+ i
1.7-1010]
eO = 8 .8 5 -1 0 '12
Q
= 100
Quality factor
Ps = 500
Source power (W )
c - 3 . jo 8
Speed of light m/s
radius = 5 - 10'6
dV
I
4 \
- |3.14 -
j -radius3
Droplet Volume
pabs __ ~ [ ? . 2 - ( f 3) ( l m ( e ) ) - d V - Q - P s ]
[[(Re(e))2]-(c3)]
Pabs = 4.285* 10 10
Pow er absorbed by droplet (W)
163
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
BIBLIOGRAPHY
Abu-Samra, A., Morris, J. S., and Koirtyohann, S. R., Anal. Chem., 1975, 47,
1475.
Aklno, T. A., Carnahan, J. W.f 1995, FACSS Abstracts, paper 856.
Alves, L. C., Minnlch, M. G., Wiederin, D. R., and Houk, R. S., J. Anal. At.
Spectrom., 1994, 9, 399.
Alves, L. C., Weiderin, D. R., and Houke, R. S., Anal. Chem., 1992, 64, 1164.
Arruda, M. A. Z., Mercedes, G., and Valcarcel, M., J. Anal. At. Spectrom.,
1995.10, 501.
Arruda, M. A. Z., Mercedes, G., and Valcarcel, M., J. Anal. At. Spectrom.,
1996.11, 169.
Backstrom, K., Gustavsson, A., and Hietala, P., Spectrochlm. Acta, 1989,
44B, 1041.
Benson, R. L., McKelvie, I. D., Hart, B. T., and Hamilton, I. C., Anal. Chim.
Acta., 1994, 291, 233.
Bemdt, H., and Yanez, J., J. Anal. At. Spectrom., 1996,11, 703.
Bordera, L., Hemandis, V., and Canals, A., Fresenius J. Anal. Chem., 1996,
355,112.
Bordera, L., Todoli, J. L., Mora, J., Canals, A., and Hemandis, V., Anal.
Chem, 1997, 69, 3578.
Botto, R. I., and Zhu, J. J., J. Anal. At. Spectrom., 1994, 9, 777
Botto, R. I., and Zhu, J. J., J. Anal. At. Spectrom., 1994, 9, 905.
Brenner, I. B., Zander, A., Plantz, M., and Zhu, J., J. Anal. At. Spectrom.,
1997,12,273.
Brenner, I. B., Zhu, J., and Zander, A., Fresenius J. Anal. Chem., 1996, 355,
774.
Browner, R. F., and Boom, A. W., Anal. Chem., 1984, 56, 788A.
Burguera, M., and Burguera, J. L., and Alarcon, O. M., Anal. Chim. Acta.,
1986, 179, 351.
Burguera, M., and Burguera, J. L„ J. Anal. At. Spectrom., 1993, 8, 235.
164
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Burguera, M., Burguera, J. L., and Alarcon, O. M., Anal. Chim. Acta., 1988,
214,421.
Burguera, M., Burguera, J. L., Rondon, C., Rivas, C., Carrera, P., Gallignani,
M., and Brunetto, M. R., J. Anal. At. Spectrom., 1995,10, 343.
Cabrera, C., Madrid, Y., and Camara, C., J. Anal. At. Spectrom., 1994, 9,
1423.
Carbonell, V., de la Guardia, M., Salvador, A., Burguera, M., and Burguera, J.
L., Anal. Chim. Acta., 1990, 238, 417.
Carbonell, V., Morales-Rubio, A., Salvador, A., de la Guardia, M., Burguera,
J. L., and Burguera, M., J. Anal. At. Spectrom., 1992, 7, 1085.
Castillano, T. M„ Vela, N. P., Caruso, J. A., and Story, W. C., J. Anal. At.
Spectrom., 1992, 7, 807.
CEM Product literature, 1996, CEM Corporation (Matthews, NC).
CEM-81D Operation Manual, 1989, CEM Corporation (Matthews, NC).
Cerda, A., Oms, M. T., Forteza, R., and Cerda, V., Anal. Chim. Acta., 1997,
351,273.
CETAC Technologies Product Literature, (Omaha, Nebraska).
Christiansen, D. E., and Unruh, W. P., Ceram. Trans., 1991, 21, 597.
Collin, R. E., “Antennas and Radiowave Propagation”, 1985, McGraw-Hill
(New York).
Conver, T. S., Yang, J., Koropchak, J. A., Shkolnik, G., and Flajnik-Rivera, C.,
Appl. Spectrosc., 1997, 51, 68.
Copson, D. A., “Microwave Heating”, 1962, AVI Publishing Company
(Westport, CT).
Corns, W. T., Ebdon, L., Hill, S. J., and Stockwell, P. B., Analyst, 1992, 117,
Corns, W. T., Stockwell, P. B., Ebdon, L., and Hill, S. J., J. Anal. At.
Spectrom., 1993, 8, 71.
Dangani, R., C&EN, 1997, February 10, 31.
Darke, S. A., and Tyson, J. F., J. Anal. At. Spectrom., 1993, 8, 145.
de la Guardia, M., Carbonell, A., Morales-Rubio, A., and Salvador, A.,
Talanta, 1993, 40, 1609.
Dubbs, C. A., Anal. Chem., 1952, 24, 1654.
Eastgate, A. R., Fry, R., C .f and Gower, G. H., J. Anal. At. Spectrom., 1993,
8, 305.
165
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Evans, H. E., and Giglio, J. J., J. Anal. At. Spectrom., 1993, 8,1.
Fassel, F. A., and Bear, B. R., Spectrochim. Acta, 1986, 41B, 1089.
Fitzgerald, N., Tyson, J. F., and Leighty, D. A., J. Anal. At. Spectrom., 1998,
13,13.
Gluodenis, Jr, T. J., and Tyson, J. F., J. Anal. At. Spectrom., 1993, 8, 697.
Gluodenis, Jr, T. J., PhD. Dissertation, 1993, University of Massachusetts.
Goulden, P. D., and Anthony, D. H. J., Anal. Chem., 1982, 54, 1678.
Gras, L., Mora, J., Todoli, J. L., Hemandis, V., and Canals, A., Spectrochim.
Acta, 1997, 52B, 1200.
Guo, T., and Baasner, J., Talanta, 1993,40,1927.
Gustavsson, A., and Hietala, P., Spectrochim. Acta, 1990, 45B, 1103.
Gustavsson, A., and Nygren, O., Spectrochim. Acta, 1987, 42B, 883.
Gustavsson, A., Spectrochim. Acta, 1987, 42B, 111.
Gustavsson, A., Spectrochim. Acta, 1988, 43B, 917.
Haswell, S. J., and Barclay, D., Analyst, 1992,117,117.
Hell, A., Ulrich, W. F., Shifrin, N., and Ramirez-Munoz, J., Appl. O pt,1968, 7,
1317.
Hill, S. J., Hartley, J., and Ebdon, L., J. Anal. At. Spectrom., 1992, 7, 23.
Hinkamp, S., and Schwedt, G., Anal. Chim. Acta., 1990, 236, 345.
Hutton, R. C., and Eaton, A. N., J. Anal. At. Spectrom., 1987, 2, 595.
Jakubowski, N., Feldmann, I., and Stuewer, D., Spectrochim. Acta, 1992,
47B, 107.
Karanassios, V., Li, F. H., and Salin, E. D., J. Anal. At. Spectrom., 1991, 6,
457.
Kingston, H. M., and Haswell, S., “Microwave Enhanced Chemistry”, 1997,
American Chemical Society (Washington DC).
Kingston, H. M ., and Jassie, L. B., “Introduction to Microwave Sample
Preparation , 1988, American Chemical Society (Washington DC).
Koropchack, J. A., Spectroscopy, 1993, 8, 20.
Koropchak, J. A., and Winn, D. H., Anal. Chem., 1986,58,2558.
Lam, J. W., and McLaren, J. W., J. Anal. At. Spectrom., 1990, 5, 419.
166
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Lamble, K. J., and Hill, S. J.t Analyst, 1998,123, 779.
Legere, G., and Salin, E. D., Appl. Spectrosc., 1995, 49, 14A.
Maksimova, I. M., Morosanova, E. I., Kuz’min, N. M.t and Zolotov, Y. A.,
Fresenius J. Anal. Chem., 1997, 357, 946.
Marinov, M. I., Spectrosc. Lett., 1984,17, 1.
Marinov, M. I., Spectrosc. Lett., 1984,17, 151.
Marinov, M. I., Spectrosc. Lett., 1984,17, 33.
Marlines Stewart, L. J., and Barnes, R. M., Analyst, 1994,119,1003.
Masters, K., “Spray Drying Handbook”, 1985, J. Wiley and Sons (New York).
Matusiewicz, H., and Sturgeon, R. E., Prg. Analyt. Spectrosc., 1989, 12, 21.
McLaren, J. W., Lam, J. W., and Gustavsson, A., Spectrochim. Acta., 1990,
45B, 1091.
Michael, D., Mingos, P., and Baghurst, D., Chem. Soc.Rev., 1991,20,1.
Minnich, M. G., and Houke, R. S., J. Anal. At. Spectrom.,1998,13, 167.
Minnich, M. G., Houk, R. S., Woodin, M. A., and Christiani, D. C., J. Anal. At.
Spectrom., 1997,12, 1345.
Moede-Wedler, T., MS. thesis, 1994, University of Massachusetts.
Montaser, A., and Golightly, D. W ., “Inductively Coupled Plasmas in Analytical
Atomic Spectrometry^ 1987, VCH publishers (New York).
Mora, J., Canals, A., Hemandis, V., van Veen, E. H., and de Loos-Vollebregt,
M. T. C., J. Anal. At. Spectrom., 1998,13, 175.
Morales-Rubio, A., Mena, M. L., McLeod, C. W., Anal. Chim. Acta., 1995,
308, 364.
NIST catalogue, SRM 611, NIST (Washington DC).
Niu, H., and Houke, R. S., Spectrochim. Acta, 1996, 51B, 779.
Perkin, R., J. Heat Mass Transfer, 1980, 23, 687.
Product Literature, Micromass, http://www.micromass.co.uk, accessed 9
September 1998.
“Sample Preparation Techniques”, 1996, short course notes, University of
Massachusetts.
Schron, W., and Muller, U., Fresenius. J. Anal. Chem., 1997, 357, 22.
167
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Sharp, B. L., J. Anal. At. Spectrom., 1988, 3, 613.
Sharp, B. L., J. Anal. At. Spectrom., 1988, 3, 939.
Stewart, I., and Olesik, J. W., FACSS Conference, 1997, Abstract 462,
Sturgeon, R. E., Willie, S. N., Methven, B. A., Lam, J. W. H., and
Matusiewicz, H., J. Anal. At. Spectrom., 1995,10, 981.
Sulcek, Z., and Povondra, P., “Methods of Decomposition in Inorganic
Analysis”, 1989, CRC Press (Boca Raton, FL).
Sundin, N. G., Tyson, J. F., Hanna, C. P., and McIntosh, S. A., Spectrochim.
Acta, 1995, 50B, 369.
Tao, H., and Miyazaki, A., J. Ahal. At. Spectrom., 1995,10, 1.
Tarr, M. A., Zhu, G., and Browner, R. F.f J. Anal. At. Spectrom., 1992, 7, 813.
Tsalev, D. L., Sperling, M., Welz, B., Analyst, 1992,117,1735.
Tsalev, D. L., Sperling, M., Welz, B., Analyst, 1992,117,1729.
Tsang, L., Kong, J., and Shin, R., “Theory of Microwave Remote Sensing”,
1985, J. Wiley and Sons.
Tsukahara, R., and Kubota, M., Spectrochim. Acta, 1990, 45B, 581.
Uchida, H., Masamba, W. R., Uchida, T., Smith, B. W., and Winefordner, J.
D., Appl. Spectrosc., 1989, 43, 425.
Veillon, C., and Margoshes, M., Spectrochim. Acta, 1968, 23B, 553.
Von Hippel, A. R., “Dielectric* Materials and Applications”, 1954, page 361,
M.I.T. Press.
’
Von Hippel, A. R., “Dielectrics and Waves", M.I.T. Press, 1954.
Weir, D. G. J., and Blades, M. W., Spectrochim. Acta, 1990, 45B, 615.
Welz, B., He, Y., and Sperling, M., Talanta, 1993, 40,1917.
Welz, B., Tsalev, D. L., Sperling, M., Anal. Chim. Acta., 1992, 261, 91.
Williams, K. E., Haswell, S. J., Barclay, D. A., and Preston, G., Analyst, 1993,
118,245.
Yang, J., Conver, T. S., Koropchak, J. A., and Leighty, D. A., Spectrochim.
Acta, 1996, 51B, 1491.
Zhi, Z., Rios, A., and Valcarcel, M., Crit. Rev. Anal. Chem., 1996, 26, 239.
Zlotorzynski, A., Crit. Rev. Anal. Chem., 1995, 25,43.
168
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IMAGE EVALUATION
TEST TARGET (Q A -3 )
w
%
150mm
IIW 4 G E . In c
1653 E ast M ain S treet
R ochester. NY 14609 USA
P hone: 716/482-0300
Fax: 716/288-5989
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 795 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа