close

Вход

Забыли?

вход по аккаунту

?

Microwave dissolution: Development and application of a new sample preparation technique

код для вставкиСкачать
INFORMATION TO USERS
The most advanced technology has been used to photo­
graph and reproduce this manuscript from the microfilm
master. UMI films th e text directly from th e original or
copy submitted. Thus, some thesis and dissertation copies
are in typewriter face, while others may be from any type
of computer printer.
The quality of th is 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 affect reproduction.
t
In the unlikely event th at the author did not send UMI a
complete manuscript and there are missing pages, these
will be noted. Also, if unauthorized copyright m aterial
had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are re­
produced by sectioning the original, beginning a t the
upper left-hand comer and continuing from left to right in
equal sections w ith small overlaps. Each original is also
photographed in one exposure and is included in reduced
form at the back of the book. These are also available as
one exposure on a standard 35mm slide or as a 17" x 23"
black and w hite photographic p rin t for an additional
charge.
Photographs included in the original m anuscript 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.
University Microfilms International
A Bell & Howell Information C om pany
3 00 North Z ee b R oad, Ann Arbor, Ml 4 8 1 0 6 -1 3 4 6 USA
3 1 3 /7 6 1 -4 7 0 0 8 0 0 /5 2 1 -0 6 0 0
Order Number 9014653
M icrowave dissolution: D evelopm ent and application of a new
sam ple preparation technique
Jassie, Lois B., Ph.D.
The American University, 1989
C opyright © 1989 by Jassie, Lois B . A ll rights reserved.
UMI
300 N. Zeeb Rd.
Ann Arbor, MI 48106
MICROWAVE DISSOLUTION: DEVELOPMENT AND APPLICATION
OF A NEW SAMPLE PREPARATION TECHNIQUE
*»
Lois B. Jassie
submitted to the
Faculty of the College of Arts and Sciences
of The American University
in Partial Fulfillment of
The Requirements for the Degree
of
Doctor of Philosophy
in
Chemistry
Signatures of^bmm^tee^
Chair:
A t A
.
G r a u L s (& .
Amz
'Dean of the College
6 December 1989
Date
1989
The American University
Washington, D.C. 20016
THE AMERICAN UNIVERSITY LIBRARY
Qj/ oI bAjU
© COPYRIGHT
by
LOIS B. JASSIE
1989
ALL RIGHTS RESERVED
To the Memory of David Smith-Magowan
MICROWAVE DISSOLUTION: DEVELOPMENT AND
APPLICATION OF A NEW SAMPLE
PREPARATION TECHNIQUE
BY
Lois B. Jassie
ABSTRACT
Results presented here establish and quantify the theoretical basis for
understanding the technique of microwave sample dissolution. Obstacles associated with
temperature and pressure measurement in closed-vessel microwave digestions were
overcome and studied in a systematic fashion. By measuring these parameters
fundamental principles have been investigated. A complete microwave digestion system
using 2450 MHz is described and mechanisms of microwave heating are discussed.
Since microwave energy couples directly with the solvent medium, the behavior
of mineral acids used for dissolution has been investigated and characterized. Equations
for estimating the power absorptions of nitric, hydrochloric, hydrofluoric and sulfuric
acids from 25- lOOOg are given. Procedures for determining the maximum power output
of a microwave unit and a calibration scheme for determining the most useful linear
region are documented. This information permits the transfer of analytical procedures
from one microwave unit to another.
Continuous, real-time monitoring of temperature provides new insights into
matrix component decompositions. This capability allows the mechanism of
decomposition to be observed and the procedures to be generalized to a variety of
samples. Analysis of the decomposition behavior of the three principal classes of
biological materials in nitric acid ranks carbohydrates, proteins, and lipids according to
their decomposition temperatures. Optimal conditions for digestions now can be tailored
specifically to the matrix. Reproducibility o f thermal conditions o f similar samples
prepared by microwave dissolution is achieved by monitoring the temperature. The
concept of modeling dissolution on the behavior of the acid is presented as an alternative
to real-time measurements.
Studies of the microwave interaction with mineral acids include 1) a comparison
of acid absorption efficiencies and their relationship with mass and proportional power 2)
the prediction of the power absorption of mixtures by treating them additively 3) the
nature of heat loss from the vessels, power rejection, and their influence on measurement
and prediction.
Results for the analysis of selenium and mercury in sediment suggest that volatile
analytes are retained in solution; these elements may now be determined directly and
simultaneously with other trace metals. Practical applications for the microwave
dissolution of soils, sediments, inorganic and biological matrices are presented along with
a recommended procedure for high-particulate water.
ACKNOWLEDGEMENTS
Foremost, I am indebted to the members of my committee who have been,
throughout the course of this endeavor, inspirational and dedicated teachers, true mentors
and collaborators. In particular, Dr. James F. Girard, whose compassion and
understanding of the joys and conflicts of the role of a "Woman-in-Science", has always
been a source of encouragement and comfort. He is ever responsive to the plight of the
graduate student and more than once, has come to the Author's rescue. Dr. Paul Waters
has kept the work honest and I am grateful for his thoughtful suggestions for
improvement of the manuscript. My mentor and colleague at the National Institute of
Standards and Technology, formerly the National Bureau of Standards, Dr. H. M.
(Skip) Kingston, with whom I have had the pleasure of working these last 6 years,
deserves special mention. Without his visionary drive this entire "microwave project"
might not have come to pass. I am grateful to him for selecting the Author as a Research
Associate.
It has been a distinct pleasure working with everyone at CEM Corp. (Matthews,
NC) but I would especially like to acknowledge the technical and materiel assistance of
Judy Gilbert, Will Grooms, Wyatt Hargett, Elaine Hasty, Bob Revesz, Sara Littau,
Dennis Manchester and Vickie Serrett. If Mike Collins, the President of CEM
Corporation (manufacturer of our equipment) did not share the vision and had not
willingly committed CEM's resources and expertise to the project, the field of microwave
sample preparation would probably still be in the Dark Ages. His enthusiasm for the
project and dedication to good science helped give the project its "raison d'etre". I am
especially grateful to Dr. Ed Neas, Principal Scientist at CEM, for his patient and
persevering perusal of the manuscript, for the suggestions about absorption efficiency,
and for the many instructive conversations about microwave heating. Steve Smith has
been the creative force responsible for producing much of the artwork. His contribution
to the professional look of the manuscript is a genuine source of pride. And finally,
CEM's financial support in the form of tuition assistance and research grants is most
appreciated.
Over the years, I have come to know many wonderful people associated with the
Chemistry Department at The American University. They include Drs. Mary Aldridge,
Nell Buell, Tom Cantrell, Nancy (Rowan) Gordon, Steve Grebe (Biology), Lou Hughes
and Nina Roscher. And I haven't forgotten Marie Matheny. Yes, there have been role
models that I have looked to over the years; I hope that in some way I may prove equal to
the task.
Many individuals at NIST have been generous with their time and suggestions
when I have sought their advice and counsel. My thanks to Drs. Yung C. Wu and
Kenneth W. Pratt of the Electroanalytical Group especially concerning technical matters
relating to heat loss; to Bill Bowman and Bill King for engineering assistance and
shopwork connected with the project, and to Susannah Schiller of the Statistical
Engineering Group for helping me to express the results of data more precisely, within
the 95% confidence interval. A special debt of gratitude is owed to Janiel Reed who
patiently learned the "ins and outs" of Microsoft Word® while preparing this manuscript.
I am ever in her debt for helping to keep the project manageable and the Author sane.
v
I am grateful to the American Chemical Society for permission to use a number of
figures and tables that first appeared in Introduction to Microwave Sample Preparation, to
the International Microwave Power Institute for permission to use the generation of
2450 MHz waves figure, and to Elaine Hastey and Bob Revesz of CEM Corporation for
permission to use unpublished data and the hydrochloric acid heating rate figure.
Last, but certainly not least, this undertaking would not have been at all possible
or worthwhile without the continuing support, confidence, love and devotion of the first
doctor in our family, Newton Jassie, MD.
TABLE OF CONTENTS
ABSTRACT..............................................................................................................
ii
ACKNOWLEDGEMENT.....................................................................................
iv
LIST OF TABLES..................................................................................................... viii
LIST OF FIGURES...................................................................................................
x
Chapter
1.
INTRODUCTION.........................................................................................
1
2.
BACKGROUND........................................................................................
7
3.
INSTRUMENTATION AND THEORY OF MICROWAVE
HEATING................................
14
4.
EXPERIMENTAL DETAILS, RESULTS AND DISCUSSION................... 58
5.
PRACTICAL APPLICATIONS.............................................................. 143
6.
CONCLUSIONS........................................................................................ 179
APPENDICES.......................................................................................................... 200
BIBLIOGRAPHY..............................................
v ii
227
LIST OF TABLES
Table
Page
1. Thermal and Microwave Characteristics of Laboratory Container
Materials.................................................................................................
27
2. Heat Capacity of Mineral Acids and Solutions
57
............................
3. Maximum Microwave Power Output Measured in Multiple Containers
63
4. Maximum Power Output Measured in Glass and Plastic Vessels...................
64
5. Effect of Insulated Containers on the Measurement of Power Absorption
in Liquids at 2450 MHz...........................................................................
67
6. Measurement of Maximum Microwave Power Output as a Function of
Time........................................................................................................
71
7. Microwave Power Absorbed by Small Volumes of Mineral Acids and
Water.......................................................................................................
82
8. Average Percent Error for the First- Order and Fourth-Order Models
87
9. Actual Time Required for a Single 250 mg Sample of Biological Material
in 5 mL of Concentrated Nitric Acid to Reach a Specific
Temperature ......................................................................................
89
10. Predicted and Actual Time for Urine Sample to Reach a Target
Temperature............................................................................................
92
11. Microwave Power Absorbed by One-Liter Mixtures of Hydrofluoric
and Nitric Acids....................................................................................... 101
12. Microwave Power Absorbed by 200 mL Mixtures of Hydrofluoric and
Nitric Acids..............................
101
13. Microwave Power Absorbed by Mixtures of Hydrofluoric and Nitric
Acids Estimated from Quartic Equations and by Linear Combination
102
14. Microwave Absorption Efficiency of One-Liter Acid Solutions at
Full Input Pow er.................................................................................... 105
15. Microwave Absorption Efficiency of Individual and Mixed Acids as a
Function of Vessel Volume.....................................................................
Ill
16. Microwave Absorption Efficiency of Mineral Acids and Their Mixtures
as a Function of Proportional Power....................................................... 113
17. Microwave Absorption Efficiency of Mineral Acid and Their Mixtures
as a Function of Mass and Power Input.........................................
114
18. Microwave Power Output Requirements for the Maintenance of Target
Temperatures in Mineral Acids................................................................
115
19. Predicted Values for the Transition Temperature(Xo) of Model
Biological Components and Real Samples......................................
138
20. Elemental Analysis of Standard Reference Materials by Laser
Enhanced Ionization (LEI)....................................................................... 151
21. Reproducibility of Temperature and Pressure in Digestion of Buffalo
River Sediment SRM 2704.....................................................................
162
22. Elemental Analysis of Buffalo River Sediment SRM 2704 by ICP-MS
andGFAAS...............................
163
23.
Elemental Analysis by ICP of Glass Frit Used for the Vitrification of
Simulated Nuclear Waste........................................................................ 169
24. Comparison of Traditional Digestions with Microwave Sample
Preparation Methods......................................................................... 198
25. Transducer Calibrations................................................................................. 207
26. Coefficients for the First-Order Model.................................................. 224
27.
Coefficients for the Fourth-Order Model........................................................ 224
ix
LIST OF FIGURES
Figure
Page
1. Schematic of a Mode Pattern in a Microwave Cavity.................................... 17
2. Equipment Setup for Microwave Dissolution...............................................
21
3. Generation of 2450 MHz Microwave in a Magnetron, (a) Wind-Like Path
of Electrons in Tube, (b) Vane Containing Bunched Electrons................ 24
4. Moderate-Pressure 120 mL Teflon® Vessel with Temperature and
Pressure Probes................................ :.................................................... 31
5. Expanded View of 120 mL Moderate-Pressure Vessel with
CEM Relief Disk......................... i .......................................................... 32
6. Cross-section of Release Mechanism in Disk Valve..................................... 34
7. Parr High-Temperature and High-Pressure Vessel
..................... 35
8. Electromagnetic Spectrum Wavelengths and Energy States.......................... 44
9. Dipole Rotation - Microwave Electric Field Interaction with
Water Molecule, (a) Aligned With the Field, (b) Rotation to Follow
Field, (q) Rotation in Opposite Direction to Keep Up with Field
48
10. Ionic Conduction- (a) Asymmetric Effect (b) Electrophoretic Effect .......... 52
11. Calibration of a Microwave Unit Showing Linearity Over Entire
Proportional Range................................................................................. 68
12. Calibration of a Microwave Unit Showing Bias at Upper and Lower
Proportional Levels................................................................................. 70
13. Maximum Microwave Power Output as a Function of Irradiation Time
72
14. Microwave Digestion Profile Showing a Continuously Increasing Partial
Power Program-300 mg Bovine Liver in 5 mL of Nitric Acid ............... 76
15. Microwave Digestion Profile Showing a Rapid Initial Partial Power
Program Followed by Decreased Power-Five 45 mL Samples of
Water each with 5 mL of Nitric A cid....................................................... 78
16.
Microwave Heating Profile of 5 mL of Tetrafluoioboric Acid...................... 81
17.
Absorbed Power Estimated from the Exponential Fit of Power to Mass
with 95% Confidence Bands for Prediction IncludedCubic Model for Water. .....................................................................
83
Absorbed Power Estimated from the Exponential Fit of Power to Mass
with 95% Confidence Bands for Prediction IncludedQuartic Model for Nitric Acid, 16M........................................................
84
18.
19.
Microwave Heating Profile of 5 mL of Nitric Acid....................................... 90
20.
Microwave Heating Profile of 5 mL of Nitric Acid with 300 mg of
Bovine Liver...................................................................................... 91
21.
Microwave Heating Profile of 16 mL of Nitric and Hydrofluoric
Acids (5:3v/v).......................................................................................... 98
22.
Temperature and Pressure Profile of 16 mL of Nitric and Hydrofluoric
Acids (5:3v/v).......................................................................................... 99
23.
Heating Rates of 10 mL Samples of Hydrochloric Acid in Single and
M ultiple Vessels.............................
106
24.
Absorption Efficiency of Individual Acids as a Function of the Number
of Vessels Containing 10 mL......................................................... 108
25.
Absorption Efficiency of Mixed Acids as a Function of the Number
of Vessels Containing 10 mL......................................................... 109
26.
Newton Cooling of 120 mL Teflon Vessel with 10 g of Distilled Water
Heated to 180 °C in a Microwave Cavity........................................ 119
27.
Heating Rate of the Dummy Load as a Function of Sample Mass............
122
28.
Microwave Heating Profile of 1 g of Wheat Flour SRM 1577a
in 10 mL of Nitric Acid ...............................................................
124
29.
Temperature and Pressure Profile of 1 g of Wheat Flour SRM 1577a
Decomposition in 10 mL of Nitric Acid................................................... 125
30.
Temperature and Pressure Profile of 240 mg of Soluble Starch
Decomposition in 5 mL of Nitric A cid....................................................
127
31.
Temperature and Pressure Profile of 250 mg of Mixed Starch
Polymer Decomposition in 5 mL of Nitric Acid............................... 128
32.
Temperature and Pressure Profile of 240 mg of Glucose Monomer
Decomposition in 5 mL of Nitric A cid....................................................
xi
129
33. Temperature and Pressure Profile of Albumin (Bovine Serum)
Decomposition in 5 mL of Nitric Acid....................................................
34.
Temperature and Pressure Profile of Albumin (Bovine Serum)
Decomposition in 4 mL of Nitric and Phosphoric Acids (3:1 v/v)
130
131
35.
Comparison of the Temperature and Pressure Profiles of the First and
Second Digestions of Albumin (Bovine Serum) in 5 mL of Nitric Acid.. 132
36.
Temperature and Pressure Profile of 220 mg of Tristearin
Decomposition in 6 mL Nitric Acid...............................................
134
37.
Sample Logistic Curve......................................................................... 135
38.
Predicted Decomposition Curves Showing Transition Temperatures
for Model Biological Components.........................................................
137
Predicted Decomposition Curves Showing Transition Temperatures
for Real Materials..............................................................
139
39.
40.
Temperature and Pressure Profile of Bovine Liver Decomposition in
Nitric Acid Compared with Protein Model...................................... 141
41.
Temperature and Pressure Profile of Bovine Liver Decomposition in
Nitric Acid Compared with Lipid M odel................................................
142
42.
Microwave Heating Profile of the First Digestion of Total Diet SRM 1548.
Nine Samples of 250 mg Each in 10 g of Nitric Acid.............................. 145
43.
Temperature and Pressure Profile of the First Digestion of Total Diet
SRM 1548 in 10 g of Nitric Acid............................................................
147
Microwave Heating Profile of the Second Digestion of Six Samples of
Total Diet SRM 1548 in 10 g of Nitric Acid............................................
148
Microwave Heating Profile of the Second Digestion of Nine Samples of
Total Diet SRM 1548 in 10 g of Nitric Acid............................................
149
44.
45.
46.
Temperature and Pressure Profile of the Second Digestion of Total Diet
SRM 1548 in 10 g of Nitric Acid............................................................ 150
47.
Microwave Heating Profile of the First Digestion of Apple and Peach Leaves
SRMs 1515 and 1547. Nine Samples of 250 mg Each in 9.5 g of
Nitric and 0.5 g Hydrofluoric Acids.............................................. 154
48.
Microwave Heating Profiles of Two Replicates of the First Digestion
of 9 Diet Samples in Nitric Acid.............................................................. 155
49. Microwave Heating Profiles of Two Replicates of the Second Digestion
of 6 Leaf Samples in Nitric Acid.............................................................
156
50. Microwave Heating Profiles of the Decomposition Conditions for 2 and
6 Samples of Peruvian Soil SRM 4355 in 10 m L of Nitric Acid
158
51. Microwave Heating Profiles of 13 mL of Nitric and Hydrofluoric Acid
(9:4 v/v) Model and with 0.5 g River Sediment SRM 2704....................
161
52.
Microwave Heating Profile of 250 mg of Simulated Nuclear Waste in
10 mL of Nitric and Hydrofluoric Acids (1:1 v/v)................................... 168
5 3. Microwave Heating Profile of 276 mg of Alpha-Alumina Digested in
6 mL of Sulfuric and Phosphoric Acids (1:1 v/v)............................
54.
Superposition of the Microwave Heating Profiles of Four Different
Batches of Five 45 mL Water Samples Each Digested with 5 mL Nitric
A cid...................................
171
174
55.
Effect of Power Changes on the Heating Program of Five 45 mL Water
Samples with 5 mL of Nitric Acid........................................................... 175
56.
Superposition of the Microwave Heating Profiles of Four Different
Batches of Five 45 mL Water Samples Each Digested with 5 mL of Nitric
and Hydrochloric Acids (1:1 v/v)............................................................ 177
57.
Effect of Power Changes on the Heating Program of Five 45 mL Water
Samples with 5 mL of Nitric and Hydrochloric Acids (1:1 v/v).............. 178
58.
Torquing Device........................................................................................... 201
59.
One-Piece Thermocouple Type Temperature Probe...................................... 204
60.
Absorbed Power Estimated from the Exponential Fit of Power to Mass.
with 95% Confidence Bands for Prediction Included-Quartic Model
for Nitric Acid, 1M
..................................................................... 210
61.
Absorbed Power Estimated from the Exponential Fit of Power to Mass.
with 95% Confidence Bands for Prediction Included-Quartic Model
for Hydrofluoric Acid, 29M.................................................................... 211
62.
Absorbed Power Estimated from the Exponential Fit of Power to Mass.
with 95% Confidence Bands for Prediction Included-Quartic Model
for Sulfuric Acid, 18M............................................................................ 212
63.
Absorbed Power Estimated from the Exponential Fit of Power to Mass.
with 95% Confidence Bands for Prediction Included-Quartic Model
for Hydrochloric Acid, 12M........................................................... 213
64. Absorbed Power Estimated from the Exponential Fit of Power to Mass.
with 95% Confidence Bands for Prediction Included-Quartic Model
for Hydrochloric Acid, 6M...................................................................... 214
65. Absorbed Power Estimated from the Exponential Fit of Power to Mass.
with 95% Confidence Bands for Prediction Included-Quartic Model
for Hydrochloric Acid, 1M...................................................................... 215
66. Ln Power Estimated as Linear and Quartic Functions of Ln Mass for Water95% Confidence Bands for Prediction Included...................................... 216
67. Ln Power Estimated as Linear and Quartic Functions of Ln Mass for Nitric
Acid, 16M-95% Confidence Bands for Prediction Included................ 217
68. Ln Power Estimated as Linear and Quartic Functions of Ln Mass for Nitric
Acid, lM-95% Confidence Bands for Prediction Included....................
218
69. Ln Power Estimated as Linear and Quartic Functions of Ln Mass for
Hydrofluoric Acid, 29M-95% Confidence Bands for Prediction
Included................................................................................................... 219
70. Ln Power Estimated as Linear and Quartic Functions of Ln Mass for
Sulfuric Acid, 18M-95% Confidence Bands for Prediction Included
220
71. Ln Power Estimated as Linear and Quartic Functions of Ln Mass for
Hydrochloric Acid, 12M-95% Confidence Bands for Prediction
Included................................................................................................... 221
72. Ln Power Estimated as Linear and Quartic Functions of Ln Mass for
Hydrochloric Acid, 6M-95% Confidence Bands for Prediction
Included................................................................................................... 222
73. Ln Power Estimated as Linear and Quartic Functions of Ln Mass for
Hydrochloric Acid, lM-95% Confidence Bands for Prediction
Included................................................................................................... 223
xiv
CHAPTER 1
INTRODUCTION
Statement of Problem
The last twenty years have witnessed the growth and development of numerous
instrumental techniques with multi-element capability for the analysis of metals. These
analytical methods have proven themselves to be both rapid as in the case of ICP-ES and
extremely sensitive as in the case of ICP-MS. The art of sample preparation has
unfortunately not kept pace; classical methods of preparing materials for elemental
analysis are traditionally slow, labor-intensive processes that can be both variable and
occasionally dangerous. Conventional open-vessel digestions are slow because the
maximum average temperature in solution is limited to the boiling point of the reagent.
Digestions in Teflon containers are further retarded because of the excellent thermal
insulating capability of Teflon and the difficulty of transferring heat to the solution inside.
Preparations of nearly all biological, botanical, metallic, geological and
environmental matrices for analysis of their trace element content invariably require
dissolution into a suitable liquid form for quantitative determination by instrumental
analysis. These digestion processes normally involve the wet-ashing of material in
strong mineral acids in closed, semi-closed or open containers using ovens and hot plates
in processes frequently lasting hours or days. Such processes are always tedious and
frequently risk contamination of the analyte by exposure to laboratory environment and
some, such as perchloric acid digestions and Carius tube work, are dangerous. Many
1
2
processes also risk losing volatile elements of interest when completely contained
decompositions are not performed.
Control of the reaction is obtained by maintaining control of the decomposition
conditions so that uniformly reproducible temperatures are attained in the reaction vessel.
This is especially difficult when numerous samples are digested simultaneously on a
hotplate; heating blocks suffer from similar problems. Variable analytical data can result
because the samples are at different stages of decomposition as the result of the different
hotplate temperatures
Microwave assisted wet ashing was first attempted in 1974 (1) using commercial
appliances and the preliminary efforts attempted to harness the microwave energy as a
heat source to drive the reactions. Decomposition conditions were determined by simply
transferring existing methods to the new system. Until very recently, the relationship
between acid and microwave interaction resulting in sample digestions has remained
purely empirical. Once the obstacles associated with temperature and pressure
measurement in closed-vessel microwave digestions could be overcome, these parameters
could then be studied in a systematic fashion and the fundamental principles involved in
this process could begin to be investigated. Although early researchers felt the
microwave technique showed promise as a simpler, faster and cleaner process, no
research effort was expended to establish the fundamentals of sample preparation in
mineral acids with microwave heating. Nor was there any attempt to provide
instrumentation that addressed the specific requirements of the many and varied materials
requiring sample preparation using specific reagents.
Proposed Solution
Reducing sample preparation time can be accomplished by speeding up the
dissolution process. This may be achieved with closed-system acid decompositions
which have many advantages over atmospheric decompositions. The elevated
temperatures produce significant increases in oxidation potential and the formation of
intermediates, such as free radicals, that facilitate the chemical attack of compounds (2).
Another significant advantage is the ability of acid combinations such as nitric and
hydrofluoric acids to decompose materials that would not react at atmospheric pressures
and temperatures. In addition to faster reaction rates, the closed system also minimizes
contamination from laboratory air and reduces the amount of acid necessary in the
decomposition, thus reducing the analytical blank associated with the sample preparation.
Volatile trace elements that would normally be lost in open systems are retained using
closed digestion systems. Levels of many otherwise difficult elemental determinations
become feasible when these decompositions are performed in closed Teflon PFA vessels.
In order to gain some understanding of microwave-assisted acid decompositions
it is necessary to elicit the fundamental concepts governing the interactions o f microwave
energy and dissolution reagents and the influence of samples on the reaction. This will
be accomplished by studying the microwave absorption characteristics of mineral acids
and their temperature and pressure dependence in closed containers. The presence of
organic and inorganic samples in solution will be studied to determine their influence on
temperature and pressure parameters. By monitoring the conditions in the vessels control
of decomposition reactions should be feasible
This dissertation will document the development of the microwave decomposition
process from its inception as an innovative sample preparation technique through the
establishment of the technique's ability to model the decomposition of specific matrices.
This can be attained through a thorough understanding of the microwave absorption
behavior of the wet-ashing reagents coupled with the essential role of monitoring
temperature and pressure in closed containers. Understanding the processes that occur
during microwave-heated acid decomposition will be assisted by the ability to measure
temperature and pressure in-situ. Control of the temperature and pressure during closed
vessel work is critical to the efficiency, reproducibility, and above all, safety of the
procedure (3). Continuous, real-time monitoring of temperature provides new insights
into matrix component decompositions (4-6). This capability allows the mechanism of
decomposition to be observed and the procedures to be generalized to a variety of sample
types. Because microwaves couple directly with the solvent heat is a by-product as
compared with conventional conduction heating techniques. Microwave radiation can be
used as an energy source to heat digestion reagents without the limitations normally
encountered in conventional thermal ovens. New methods are therefore required to
properly apply this technology to chemical analyses. From the outset, temperature
measurement has been considered an essential parameter for controlling the reaction. For
that reason, temperature sensors have played an important role in the growth,
development and sophistication of this technique. Although optical fiber technology for
temperature sensing is now widely used, a non-traditional metallic thermocouple was
originally developed for early use as a sensor in the microwave environment.
Refinements in manufacturing now make thermocouples a true, low-cost alternative to the
expensive fiberoptic system. Fundamental design principles for the use of thermocouple
sensors are documented in this work.
5
A study of microwave energy interactions with mineral acids was undertaken to
develop a method for estimating the power absorption of reagents for microwave
dissolution. This information assists in predicting the temperature attained at the end of a
fixed interval during the inductive heating portion of the cycle and can, as well, predict
the time necessary to achieve that specific temperature. The absorption efficiency of
several mineral acids and their combinations can be determined from this data. Such
information may be used to approximate the real heat lost during dissolution at various
target temperatures. Monitoring the temperature and pressure will permit control of
digestion reactions so that specific decomposition temperatures of biological components
can be determined and then sample materials can be targeted to specific temperatures
based on composition. The absorption behavior of acid mixtures will be studied to see if
they can be treated as simply additive. When the behavior of a system is well
documented it can serve as a model for unknown materials. To demonstrate the
feasibility of modeling, examples of real sample decompositions will be examined.
Model systems will be examined for their uniformity of temperature and pressure
parameters and reproducibility of decomposition conditions. Several applications of
microwave sample preparation for the analysis of trace elements in different matrices will
also be covered. These examples will be used to demonstrate how closed vessel
microwave sample preparation has not only improved the quality of samples prepared
using this technique but has resulted in substantial time saved.
Several factors have contributed significantly to the increasing acceptance of
microwave dissolution as an effective sample preparation technique and to the research
efforts aimed at providing the basis for understanding how it works. The development of
specially designed equipment for analytical chemical use has been a most important step.
Introduction of a commercial analytical microwave system addressed, for the first time,
problems such as acid fumes, small sample power reflection, field inhomogeneity, and
long duty cycles, that were encountered by analysts trying to modify home appliances.
Until a strong, inert, and microwave-transparent container for acid dissolution was
fabricated from Teflon PFA, closed-vessel experimentation was dangerous and limited.
Finally, routine measurement of the elevated dissolution temperatures and pressures
during microwave exposure was difficult until the equipment modifications, Teflon
vessels, and new fluoroptic thermometry were applied simultaneously. The evolution of
microwave systems for the laboratory has provided researchers in microwave sample
dissolution with new and better tools to irradiate, contain, and measure the process in a
safe and reliable fashion.
CHAPTER 2
BACKGROUND
Historical Perspective
Sample dissolution procedures practiced in analytical laboratories the world over
have changed very little in the past hundred years. Heating open beakers over flames is
an ancient art and has evolved as heat sources have become more sophisticated. Hot
plates, water baths, and reflux apparatuses, are still open vessel techniques and the
advent of automation has served to relieve the analyst from baby-sitting but has not
seriously reduced the time factor for samples that must be completely decomposed prior
to trace element analysis. It is well accepted that the experimental conditions that are
obtained in open-beaker work are empirically based.
With the advent of the Carius tube in 1860, digestions could be performed at
higher temperatures as the result of the closed container resulting in more efficient and in
some cases more rapid dissolutions. Carius tubes are inherently dangerous and such
digestions are not routinely practised in analytical labs because of the care and skill
required in sealing and unsealing the tubes. There is always the danger of explosion
upon unsealing glass tubes that are substantially above atmospheric pressure and it must
be performed behind a glass shield. Other dissolutions are carried out on materials that
are fused at very high temperatures and then dissolved in an appropriate reagent for
analysis of the metal analyte.
Traditional digestion techniques comprise both open and closed-vessel
arrangements whereby thermal energy is transmitted to the sample in acid medium by
conduction across an air gap, in water or through thick-walled glass or plastic containers.
7
In conventional closed-vessel techniques heat-up and cool-down times can be substantial
as the time required to conduct heat through steel jackets and Teflon inner vessel walls is
substantial. Ceramic jackets, like Teflonware are insulating and thus further aggravate
the time factor. Such closed systems are normally used for veiy difficult to digest
materials requiring temperatures of 240 °C as might be obtained in the Carius tube using
nitric acid or 200-240 °C in Parr bombs where the elevated temperature is provided by an
external heat source such as a convection oven.
Traditional open vessel decomposition techniques such as hotplate digestion or
acid refluxing in a Kudema-Danish type apparatus normally requires from 1-2 h to
several days of digestion depending on the sample matrix and acid medium. The
increased time demanded by such techniques is a consequence of the maximum
temperature limitation imposed by being open to the atmosphere. In such containers, a
drastic temperature gradient is established throughout the liquid contents of the flask due
to cooling that occurs at the surface as the result of evaporation and localized superheating
that may occur at the surface in contact with the heating source. The overall average
temperature inside the vessel can never be more than the atmospheric boiling point of the
reagent. Another factor that contributes to the inefficiency of such heating techniques is
that massive bombs and thick containers required for strength and safety are generally
insulating material and thus are inherently poor conductors of heat.
At elevated temperatures mineral acids are far more efficient in promoting
decomposition (2) than at lower temperatures and closed containers permit the attainment
of temperatures in excess of the normal boiling points expected in conventional open
beaker dissolutions. At the elevated temperatures possible in closed containers the energy
of the acid is a function only of the temperature, according to Jule's Law, and the
oxidation potential of the acid may be increased several-fold over that at the boiling point
(7). In addition, reaction rates increase with increasing temperature; for every 10 °C rise
in temperature a doubling of the rate can be expected (ibid pgl24). When a nitric acid
digestion can be conducted at 60 °C above the traditional boiling point, marked reductions
in the time required for dissolution o f even the most stubborn and tenacious materials are
expected.
In current techniques for the determination of metals, contamination is a major
impediment to successful low level analysis. This includes contamination from the
environment in which the analysis is performed, from the vessels in which the
decomposition is carried out, from the reagents used in the process, and from the analyst
himself. These sources of interference are largely eliminated when samples are prepared
in a clean environment and the digestions are conducted in closed Teflon containers (8).
Review of Literature
The use of microwave energy as a heat source in wet ashing procedures was first
demonstrated in 1975 (1), and most o f the early papers described specific applications
using open or covered vessels (at atmospheric pressure) for the acid dissolution of bone
(9), biological tissue (10-12), and botanical matrices (13). Several studies compared the
technique with different digestion procedures (14-16) and examples are given where the
technique has been applied successfully to biological samples suitable for atomic
absorption and emission spectroscopy (17-19). Early researchers realized that openvessel work involved the risk of environmental contamination as well as mechanical or
volatile loss of the analyte. Such conditions also limited the maximum sample
temperature to the boiling point of the acids. In an attempt to deal with these problems,
investigators turned to closed polycarbonate bottles (20,21) and Teflon PFA digestion
10
vessels to obtain the elevated temperatures and pressures needed for the digestion of
steels (22), geologic species such as ores, zircons and other rocks (23-26) including soils
(27), environmental samples (28), and biologicals (3,29,30).
Not only were significant.ieductions in sample preparation time realized as the
result of the high temperatures and pressures that could be obtained in 2-3 min, but new
applications for microwave heating in closed containers became apparent. For instance,
elevated pressure in closed reaction vessels heated by microwaves was shown to increase
not only the reaction rate in classical organic syntheses of esters, hydrolysis reactions and
oxidations, but to increase the product yield as well (31-33). Not only were routine
syntheses amenable to the microwave environment, but Diels-Alder, Claisen and "ene"
rearrangements have been conducted at known temperatures that similarly resulted in
improved yields in a fraction of the usual time (34). Decreased reaction times were found
when synthesizing radio-labelled (18F, 180131I and H Q pharmaceuticals which
improved production and lead to more versatile and better controlled studies (35).
Kjeldahl nitrogen determinations using a microwave system have successfully reduced
the normal 2 h average run to about 8 min (36). Such a dramatic decrease in preparation
time is possible because microwave energy couples directly with the sulfuric acid reagent
and heating is controlled at the optimum temperature, 400 °C, where decomposition is
most efficient (ibid)
Early work in acid dissolution using microwaves as a heat source was
accomplished using home appliances that were often extensively modified to enable
researchers to utilize the unit safely. Reaction vessels were placed inside evacuated
desiccators or large plastic jars as precautionary measures aimed at containment of the
acid vapors and at reducing corrosion and the hazards of an explosion. When researchers
11
realized that small sample volumes in the cavity did not absorb all the magnetron power,
additional loads were placed in the cavity to reduce reflected radiation, which damages the
magnetron and alters its power output. These auxiliary loads reduce microwave power
that would be absorbed by samples and acid; consequently the samples receive neither
constant nor reproducible power, a situation that fiequently produces incomplete
dissolutions.
Although wet ashing of samples using microwaves was described over a decade
ago, the method has remained something of a curiosity. Since 1984, however, there has
been renewed interest in microwave based sample preparation for analytical chemistry.
The results of research on the extraction of metals from sediments (37) and research on
trace elements in biological tissue (38-40) have been presented at various regional and
national conferences. The first conference session devoted solely to presentations in the
field of sample preparation using microwave dissolution took place at the 1986 Eastern
Analytical Symposium in New York City. In 1986 four articles in Analytical Chemistry
were devoted to the subject of microwave dissolution and its inherent suitability for steels
(22), geologic materials (23-25), and biological matrices (3). Keen interest has continued
with conference presentations (41-43) and journal articles on recovery studies in
environmentally important matrices (44-46), biological (47-50) and botanical tissue (48,
51), pharmaceuticals (52), the chemistry of decomposition in biological matrices (4,53,
54), and mineralization of blood in a flow injection system (55). Many researchers are
actively comparing the completeness of open-vessel vs. closed-vessel digestion (46,56,
57) and high pressure decompositions of environmental materials in various bomb type
containers and fusions with high-temperature and high-pressure microwave digestions
(45,54,58-59). The determination of environmentally significant elements in marine
12
samples (49), geologic materials (60,61) and biologicals (62) by Inductively Coupled
Plasma (ICP), Graphite Furnace Atomic Absorption Spectrometry (GFAAS) and cold
vapor AAS (47) from samples prepared by microwave dissolution continues to
supplement our knowledge of the safe as well as toxic levels of nutritionally important
elements. Renal calculi and artificial stone mixtures were prepared for determination of
Ca, Mg and P as well as minor and trace elements by ICP-Atomic Emission Spectrometry
(AES) analysis using microwave-assisted digestions (63). In 1988, the American
Chemical Society published the first text on microwave sample preparation (64) in its
Professional Reference Book Series.
All efforts demonstrated that this dissolution technique gives results comparable
with those obtained by classic methods. When combined with clean room techniques,
closed vessel microwave digestions in Teflon PFA containers have been able to reduce
analysis blanks and the limits of detection in GFAAS at the same time as tightening the
range of blanks for aluminum in biological materials (65). When combined with
robotics, flexible automation of the microwave dissolution process can improve the
overall efficiency of mineral analyses (66). For the most part, little attention has been
paid to the importance of temperature or pressure conditions during digestion until the
first serious attempts were begun to investigate and understand the various parameters
that influence acid dissolution in microwave systems (3,6).
In the area of materials research, extensive guidelines for the preparation of
geologic and metallic samples using microwave techniques have been detailed by Matthes
(67). Chemical dissolutions of similar materials in radioactive environments have been
implemented by remote automation of the dissolution (68). The Bureau of Mines has
been evaluating the potential of rapid microwave heating for mineral processing (69-71)
13
and for the Department of Energy, G. Fanslow has been perfecting the desulphurization
of coal by dielectric heating using 2450 MHz radiation (72). Super- conducting ceramics
of uniformly high purity and crystallinity have been produced by irradiation at 2450 MHz
in a fraction of the time of comparable conventional conductive heating schemes (73).
The theoretical basis for understanding microwave dissolution lies in an
examination of related topics in classic physical chemistry, thermodynamics,
electromagnetic radiation, and dielectric materials. The ability to apply simplified
thermodynamic relationships to actual measurements enables the analyst to gain the
practical understanding that allows the technique to be generalized to a particular sample
matrix. It is this ability to generalize from concepts to real samples that makes this
technique useful to the majority of the analytical community preparing samples for
today's modem instrumentation.
CHAPTER 3
INSTRUMENTATION AND THEORY OF MICROWAVE HEATING
Design and Use of Microwave Equipment
Instrument Design
As described by Neas (36) the typical analytical microwave instrument used for
heating samples consists of six major components: the magnetron or microwave
generator, the waveguide, the microwave cavity or applicator, the mode stirrer, a
circulator and a turntable. Microwave energy produced by the magnetron is propagated
down the waveguide and injected directly into the microwave cavity where the mode
stirrer distributes the energy in various directions.
The Magnetron
In the generation of microwave energy the magnetron may be thought of as an
electron tube or diode containing a central cathode surrounded by a cylindrical multicavity
anode. Mutually coupled resonant cavities are arrayed around the cathode and connected
to the cathode-anode space by slots. When a potential of several thousand volts is
reached across the diode, the electrons in transit from the cathode to the anode are
deflected by the field of a powerful permanent magnet superimposed on the diode. The
microwaves that are generated are roughly of the same order of magnitude as the
magnetron device.
14
15
Cycling the magnetron
To obtain an average power level, the power output of the magnetron to the
cavity, in most microwave systems, is controlled by cycling. This duty cycle is the
fraction of the time base that the magnetron is on. For example, if the time base is 1 s
and the time on is .5 s, the duty cycle is one-half. For the magnetron in the microwave
system used in this research the time base is 1 s. Thus, to generate one-half the rated
output of a 600 W tube, the magnetron would have to be on for 1/2 s and off for 1/2 s.
Most commercial and appliance grade microwave units have a 10 s time base which is not
terribly efficient for heating small analytical samples because of the substantial amount of
heat lost during the long time-off portion- 5 s in a 0.5 duty cycle.
The Waveguide
Waveguides are generally made of metal so that the microwaves emitted from the
magnetron can be reflected off the walls and channeled in the direction of the
cavity(applicator). Dimensions are critical to both waveguides and cavities to avoid
mismatching. Effective transmission of microwaves is accomplished in guides that are
less than one free-space wavelength (X) in width and greater than A/2. For the 2450
MHz microwave the width of waveguides is > 6 but < 12.25 cm. Although the energy
propagates down a waveguide with several distinct distributions of the electric and
magnetic fields, a single dominant mode results from this configuration.
The Circulator
A circulator is a device that is designed to permit the magnetron to maintain
constant power output by protecting it from reflected energy when mismatching occurs
(74). Mismatching occurs when some portion of the electromagnetic energy directed into
the cavity is not absorbed by the sample but is reflected. Normally, the microwaves are
transferred from the magnetron down the waveguide and into the cavity without
reflectance; such a system is said to be perfectly matched. Energy injected into the cavity
that is not absorbed is eventually reflected back down the waveguide to the magnetron
where it may cause overheating, a loss in power output and eventually, could destroy the
magnetron tube. This situation is frequently the cause of magnetron failure in commercial
appliance-grade units because they are not normally equipped with an isolating device.
The microwave system used in this research incorporates a terminal circulator (isolator)
consisting of a ferrite device that allows microwaves to pass in one direction onlyforward, but diverts the backstreaming waves away from the source of the incident wave
(75) to a dummy load where the energy is dissipated harmlessly as heat. Digestions of
small samples in the microwave cavity are situations that can produce reflected
microwaves and result in a mismatch between the cavity and the magnetron leading to
excessive heating that can change the output of the magnetron. When more than 0.5 Kg
(500 g) of an aqueous solution or other good dielectric material with a high dissipation
factor is heated in a microwave cavity, the behavior approaches that of a perfectly
matched system by absorbing nearly all of the input power (36). Microwave dissolutions
however, are frequently conducted with much smaller quantities of reagent, usually
between 25 and 200 g, depending on the individual sample size and number of
containers. In this situation, the total dissipation factor is substantially reduced and can
result in reflected power that in commercial units would ordinarily return to the
magnetron causing overheating which changes its output A microwave unit that has its
magnetron protected from reflected microwave energy can be expected to continue
producing constant power over many years of operation. For example, the magnetron in
the equipment used in this study is protected from reflected microwave energy by an
waveguide
U H ^ w j y i f c l l ^ ’i w w j v n w w w v w w
..
;! 1 I 1.1i. . Mi
i !
:
•
w v n fm w w i i m h
:j | t
•:
i
•
. I
!11 ! 11 I
I
S
i
!
S
I I
i is
s !l s
! 1i 5 I
imtwvw^uMtyyvvmw vwvwww|u*utwwwfrw>wrwvifl»w»
M
s is
Ii sI :!
microwave pattern
!
T]p
si |
s I
I 3
sI ||:! ss }11
s
I
II il I
microwave cavity
Figure 1. Schematic of a Mode Pattern in a Microwave Cavity.
(Reprinted with permission from reference 101, copyright 1988 American Chemical
Society)
18
isolator and has maintained a constant 574 ± 7 W for over 4 y of frequent operation. As
will be seen later on, stable power output is very important to temperature reproducibility
and quality control through reproducible digestion conditions.
The Micro wave Cavity
Microwave energy generated by the magnetron and conducted through the
waveguide is propagated into the (metal) cavity applicator where the waves are reflected
from surface to surface and interfere with one another producing a pattern of well defined
and recognizable modes. These numerous modes arise because the cavity is large
(41x38x26 cm) compared to the size of the free-space wavelength making oscillations
possible in many different modes within the given 2450 MHz frequency. This is the
"multimode" cavity to distinguish it from single-mode cavities where only a single wave
has been transmitted and results in a focusing effect, e.g the electric field strength is
maximized to a given spot. Of necessity, such cavities are much smaller.
Once inside the cavity, the microwaves that are reflected off all of the cavity walls
form a complex pattern of standing waves. As seen in a conceptual drawing in Figure 1,
such a pattern is characterized by regions of high and low field intensity. The presence of
these hot (high field intensity) and cold (low field intensity) spots can severely affect the
final temperatures realized by a sample or multiple samples distributed throughout the
cavity. If one of the vessels is in a hot spot where it is exposed to greater intensity it will
heat more than another vessel in a cold spot. The result is nonuniform heating of
samples. Devices that circulate the sample vessels through the field or that move the
pattern around in the cavity generally help produce more uniform heating.
19
The Mode Stirrer
A mode stirrer is used to reflect and mix the energy entering the microwave cavity
from the waveguide. When installed in box-type cavities such devices have been found
to better distribute the energy more uniformly within the cavity (76). In effect, the mode
stirrer homogenizes the microwave field in the cavity so that energy peaks (modes) are
distributed more uniformly throughout the cavity. The desired effect of this redistribution
is to make any one position in the cavity like all others so that uniform heating is not
position-dependent.
Turntable
Even with the use of a mode stirrer the microwave energy field is not
homogeneous enough to achieve reproducible uniform energy exposure of identical
samples placed at different fixed positions in the cavity. Rotation of the containers
through the energy field by means of a turntable effectively maintains a uniform energy
exposure of all containers and minimizes the exposure differences between samples.
This provides the same digestion conditions for all samples and ensures that the vessels
are kept within temperature and pressure limits. Reproducible temperatures within 1-3%
have been achieved over the range from 23-180 °C with a 3 rpm 360° rotation about the
12 cm radius in a 600 W field.
Additional pieces of equipment distinguish this research unit from commercial
appliances and from most other analytical systems. They include a wavelength attenuator
cutoff for conducting tubing and wires in and out of the cavity and a torquing device for
sealing Teflon pressure vessels.
20
Wavelength Attenuator Cutoff
Microwave equipment can be modified by the attachment of a wavelength
attenuator cutoff to conduct tubing, thermocouple wires, and fiberoptics into the
microwave cavity. These devices have been made in many configurations. Two types of
attenuators, as shown in Figure 2, that have been tested are rigid stainless steel tubing (3)
and flexible, tin-plated copper braid (5). The common construction features of both
attenuators are metallic tubing of conducting metal with the smallest diameter possible to
allow the tube, wire, or device to pass through its inside diameter and the attenuator must
be grounded by making contact with the microwave cavity wall around the hole. This
cut-off device is effective only if the diameter is small and the length is long compared to
the wavelength of the 2450 MHz radiation which is approximately 0.12 m (12.25 cm)
long. Dimensions of the devices are approximately 0.7 cm inside diameter and between
35 and 70 cm. in length. Metallic devices passing through the attenuator must be
grounded to prevent the wire or metallic tube from acting as an antenna to transport the
electromagnetic energy out of the cavity and totally defeat the effectiveness of the
wavelength attenuator cutoff. The inside hole in the cavity wall should be kept to a
minimal diameter because a small hole less than a quarter wave in diameter is an effective
barrier to microwave energy, since the 2450 MHz wave cannot fit through.
Torquing Device
In order to adequately pressurize the Teflon PFA vessels and achieve the high
temperatures and pressures necessary for rapid and efficient microwave decomposition
the caps must be closed by applying a minimum torque of 16-20 N m. This force assures
good contact between the rim of the vessel and the inside sealing surface of the cap to
produce an effective pressure seal. The original torquing device (29) consists of a Sears
Figure 2.
iU id
Q__ I
.SSXS&
te w
\\t o i
05
O .Q .
tV lO T
CC
lD
U
JQ
C
V1
Equipment Setup for Microwave
Dissolution
Po
(Reprinted with permission from reference 6, copyright 1988 American Chemical Society)
21
22
automotive torque wrench, a restraining collar for both 60 mL and 120 mL vessels and a
socket that adapts the wrench to the vessels. A diagram of the torquing device appears in
Appendix A along with instructions for making such a device. Once sealed, the vessels
can be pressurized repeatedly to the same reproducible temperature and pressure levels.
Inadequate torquing permits vapor to escape and at high temperatures will force vapor
around the threads of the cap. This could result in substantial losses of solution and the
possibility of overheating a small volume. It is also likely that boiling will be observed in
these vessels since they are essentially still at atmospheric pressure. A schematic diagram
of the microwave system used throughout this research project is shown in Figure 2.
Theory of Operation
Generation of 2450 MHz Waves
Microwave generators are classified as crossed field because a magnetic field is
applied perpendicular to the electric field lines. In the linear beam group, the electric field
and focussing magnetic field are parallel, as in the electron tube of a mass spectrometer or
klystron. In the generation of microwave energy the magnetron may be thought of as an
electron tube or diode containing a central cathode surrounded by a cylindrical multicavity
anode. Mutually coupled resonant cavities are arrayed around the cathode and connected
to the cathode-anode space by slots. When a potential of several thousand volts is
applied across the diode, the electrons in transit from the cathode to the anode are
deflected by the field of a powerful permanent magnet superimposed on the diode. The
electron beam generated by the cathode may be thought of as a "wind" that passes across
a resonator much like air blown across the mouth of a bottle. The presence of a magnetic
field imparts an angular momentum to the electrons so that they rotate in a circle and pass
many resonators before being siphoned off at the anode, as visualized in Figure 3a.
Standing waves are created in the resonant cavities, much like the effect of
blowing air across the top of a glass or closed end tube, which produce oscillations that
cause the electrons to trace an angular path to the anode wall. Electrons are then
"bunched" in spoke-like fashion, as illustrated in Figure 3b, and create a swirling in
waves or pulses that appear to rotate in bursts across the cavity slots. Because the
cavities are joined together, the net output can be extracted by sampling any one of the
cavities with a probe and conducting the output, via coaxial cable or antenna, to a
waveguide or an applicator device (cavity). The size of the resonant cavities determines
the frequency generated much in the way the pitch of the "whistle" is determined by the
depth of the glass or tube. In the case of domestic microwave units the cavity is
dimensionally matched to the 2450 MHz wave; C= X f where 0, the speed of light
waves is 3.0 x 10 8 m s-1 and f, the frequency, is 2.45 x 10 9 cm-1, then h= 0.1224 m
(12.24 cm).
The field produced has both an electric and magnetic component that is
perpendicular to the travelling wavefront. If polar or charged, the molecule is influenced
by the energy and deflects (rotates) in response to the passing wave. Non-polar
molecules can experience an induced polarization in such a situation. As the half-wave
passes and diminishes, the dipole reverses direction to follow the field and consequently
undergoes a right /left oscillation or two swings for every passing wave; for the 2450
MHz microwave each particle will rotate 4.9 x 10 9 times s-1.
Magnet
(a)
Resonators
Anode
I I
Cathode }
I
/
Electron
“wind"
Output to
waveguide
Magnet
Magnet
Resonators
Anode
ff
Cathode
I
j
Output to
waveguide
Circulating
Electron
'
"wind"
Magnet
Figure 3. Generation of 2450 MHz Microwave in a Magnetron,
(a) Wind-like Path of Electrons in Tube (b) Vane Containing
Bunched Electrons
(Reprinted with permission from Trans.IMPI, vol 1,66-67, copyright 1973 IMPI)
25
Containers for Microwave Use
Sample preparations, organic synthesis, and other laboratory operations using
microwave energy will require many different types of sample vessels. The suitability of
the vessel material depends on the specific conditions, use, and solvent for which it is
intended. In choosing a container material for open-vessel work, the boiling point of the
acid or solvent being used is the maximal temperature to which the material will be
exposed. When closed systems are involved, temperature measurement is necessary to
prevent the liquid contents of the vessel from exceeding the upper limit of the vessel
material, which is normally the maximal continuous service temperature. At this
temperature a polymeric device will continue to function as intended without change for
an indefinite period of time. Above this point, such a material may deform, deteriorate,
or lose some essential property so that it no longer functions optimally. For example
high-boiling solvents like sulfuric and phosphoric acid will melt Teflon, but would not be
a problem when used with borosilicate glass or quartz vessels.
Criteria for Microwave Suitable Materials
When choosing containers for use in a microwave field, the composition of all
parts needs to be considered such as handles or screw-on caps, which tend to be made
from materials different from the vessel itself. Problems can be avoided by choosing
materials that do not significantly absorb microwave radiation or absorb very little.
Listed in Table 1 are dielectric dissipation factors expressed as tan 8, melting points, and
maximal continuous use temperatures of some common laboratory container materials.
Tan 8 = E"/E' where E" is the loss factor and E1is the dielectric constant. Tan 8, also
known as the absorptivity, may be thought of as a ratio of the ability to dissipate the
energy of the wave in a substance compared with its ability to block or store that energy.
Materials with small dissipation factors, where tan 8 <_5 at 25 °C, tend to be good electric
insulators, are transparent in the microwave environment, and exhibit little or no heating.
Dissipation factors and loss tangents for plastics and polymers at microwave frequencies
(1 x 1()9 Hz) are sparse, but when known (77) may be similar to values at the more
common frequency of 10^ Hz and may be used as a guidfe to heating behavior. For
example, fluorinated polymers are excellent materials for microwave containers because
of their small tan 8 values and are essentially transparent at a frequency of 2450 MHz.
Laboratory ware fabricated from or incorporating Bakelite, Lucite and other thermoplastic
resins may be acceptable for use, even though they absorb a small amount of energy. A
material that does not appear on the list and whose electromagnetic properties are
unknown can be tested by placing it in a microwave unit for 30-90 s at full power to
determine whether it absorbs energy as evidenced by a measurable increase in
temperature.
Other materials that can be used in a microwave environment include fiberglass
and polyurethane foams found frequently as vessel insulation or lining. If homemade
racks, carrousels, or sample holders will be used, some thought must be given to
construction materials to ensure compatibility. Ordinary polyethylene or Teflon
laboratory tubing can be used safely in the cavity for gas handling.
Metals and metallic products are generally unsuited for use in the microwave
environment because they reflect electromagnetic radiation and can accumulate an electric
charge powerful enough to arc weld metal. In general, metals are inappropriate for the
microwave environment, unless there is some other material in the cavity that has a very
high affinity for microwaves (a very large dissipation factor) such as a very large volume
of water or an absorbing dielectric like silicon carbide. In that case, great attention is paid
27
Table 1. Thermal and Microwave Characteristics of Laboratory
Container Materials
Material
Water
Sodium Chloride (0.1 molal)
Gasoline (100 octane)
Polysulfone
Phenol/Formaldehyde
Bakelite (asbestos filled)
Nylon 6/6
Nylon 6/10
Glass (Coming 0080) ’
Glass (Coming 7070) (Borosilicate)
Ceramic (depends on type)
Polypropylene
Polymethylmethacrylate
Polyacetal, copolymer
Porcelain (4462)
Polystyrene
Polyethylene
Kel-F, CTFH
Polymethylpentene
Tefzel,TFE+CE
Halon, (P)TFE
Teflon, FEP
Teflon, PFA
Polycarbonate
Quartz, Fused
Melting
Pt.,°C
<190d
dec
dec
253
>215
>1000
>1080
—
168-171
115c
165-175
242c
120-135
198-211
240b
271
>320
252-262
302
241
>1665
aTan d values at 25 °C and 3x109 Hz taken from Ref. 77.
CRef. 80
dRef. 81
eRef. 82
160e
120-190
200-218
102
80-102
—
—
—
100-105
76-88c
121
—
Note: Adapted from Ref. 78.
bRef. 79
Maximum
Service
Temp.°C
82-91
71-93
199
175d
200
260
204
260
121
-------
Tangent 3
(xlO4)
1570
2400
14
760
519
438
128
117
126
12-75
6-50
57
57
36
11
3.3
3.1
2.3e
—
2.0
1.5
—
1.5
0.7
0.6
28
to the shape and position of the metal vessel in the cavity. Shallow rectangular and
square metal foil containers (76) that are kept away from the walls of the cavity can be
used with care. Microwave furnaces (83,84) that use small platinum crucibles have been
developed for performing fusions and dry-ashing.
Open Vessels
Glass Kohlrausch vessels have been used with great success in microwave
sample preparations (51) and are of necessity open vessel digestions. The maximum
temperatures in solution in such cases are limited to the boiling points of the acid at
atmospheric pressure. Such conditions would not be feasible for the analysis of ultra­
trace level or volatile elements. As is the case for most open vessel digestions it is
advantageous to establish a refluxing condition in order to conserve the volume and
reduce the need for continual addition of acid. Glass volumetric flasks and Kohlraush
vessels are only minimally heated by microwave energy and are only secondarily heated
by conduction from the hot contents of the flask. With the cavity fan on this results in the
necks of the vessels remaining cooler than the rest of the vessels which permits refluxing
to occur. A slight but desirable concentration of the acid tends to occur during this
process. An additional advantage cited for open vessels are larger sample sizes, up to 2
g, which allows the CO2 decomposition of organics to evolve. Normal laboratory
glassware should not be sealed for closed-vessel use. Although pressurized glass vessels
have been used in conventional convection heated ovens they have not yet been
demonstrated safe for microwave digestions. In the determination of A1 and Si when HF
is needed in the digestion and glass cannot be used, semi-closed Teflon refluxing vessels
are a suitable alternative and provide similar advantages.
29
Closed Containers
Closed vessels are subject to stresses as the result of the partial pressures of the
heated solvents and decomposition products. Excess pressure can be dealt with by
appropriate venting. Pressure can be monitored and regulated by controlling the
temperature of the solvent or limiting the sample size to restrict the quantity of gaseous
decomposition products. For some vessel materials pressure can be contained at room
temperature without deformation of the vessel but not at high pressure and elevated
temperatures. This is especially true of polymeric materials. After chemical and
mechanical considerations, the microwave energy absorption characteristics of the vessel
material are important physical factors to consider. Many materials are compatible for use
in a microwave field because they do not absorb the energy. Among these are all types of
fluorocarbons, especially Teflon PFA which is ideal for use as a vessel because of its
exceptional chemical and thermal durability. Other common laboratory polymers, such as
polyethylene and polypropylene, are transparent to the microwave environment but do
not have the chemical or thermal durability of Teflon PFA.
The physical requirements of the sample containers used for closed-vessel acid
dissolutions are so demanding that only a few materials are suitable for vessel fabrication.
Although polycarbonate and other materials have been used for closed-vessel microwave
acid digestions (20), pressure limitations, acid resistance, and frequent vessel failures
have restricted the use of containers made from these materials. The safe operating
pressure of the vessel establishes the upper limits for pressure and temperature that can be
used in the digestion. For most microwave acid dissolutions the container material must
be inert to mineral acids at temperatures > 200 °C. Chemical resistance, tensile strength at
high temperatures, and microwave transparency of Teflon PFA make it the most
appropriate material. Innovative thread design and increased wall thickness in current
30
commercially available Teflon vessels have contributed to their safe use at elevated
pressure and temperature.
Moderate-Pressure Vessel
The development of special moderate pressure (60-120 psi) sample vessels for
use by the analytical chemist has greatly facilitated microwave acid decomposition.
Specifically designed 120 mL Teflon PFA digestion vessels for microwave use have been
manufactured by the Savillex Corporation. Because retrofitting a vessel to provide ports
for measurement probes can compromise its integrity, these vessels have also been
designed to monitor temperature and pressure during closed-container operation and are
manufactured with a variety of ports molded as integral parts of the cap. The
configuration with two 1/8 in ports is used for the monitored vessels, whereas caps
without ports are used on unmonitored containers. All of the experiments in this project
where temperature and pressure were monitored, have been conducted in 120 mL Teflon
PFA vessels with external relief valve and remote pressure sensing as pictured in Figure
4.
The fabrication of the chemically inert Teflon PFA polymer into closed vessels
has permitted advances in microwave digestion procedures not previously available.
Since most microwave acid decompositions are performed in closed containers, the
design and construction of these vessels is very important The specially engineered allTeflon PFA pressure vessel is also available equipped with a safety relief disk to prevent
overpressurization. Figure 5 shows an expanded view of the vessel and pressure relief
disk. This configuration will not allow the internal pressure of the vessel to exceed 120 ±
10 psi. Pressures greater than the rated use-pressure of the vessel will develop for nitric
acid at elevated temperatures >185 °C. The pressure relief disk seals against the inside of
31
Temperature Sensing Probe
Pressure Monitoring Tubing
Open Ferrule Nut
Deep V-shaped Groove
Double Port Cap
Teflon Sleeve
Vessel Body
Figure 4. M oderate-Pressure 120 mL Teflon Vessel with
T em perature and P re ssu re P robes
32
Vent Tubing
Venting Nut
Vessel C a p
Relief Valve
Vessel Body
Figure 5. Expanded View of 120 mL ModeratePressure Vessel with CEM Relief Disk.
(Reprinted with permission from reference 6, copyright 1988
American Chemical Society)
33
the cap around the pressure relief port When the gas pressure exceeds 120 psi, the top
of the cap acts as a diaphragm, flexing away from the disk seal and allowing gas to
escape through the venting port relieving the pressure >120 psi. A cross-sectional
diagram of this relief mechanism is shown in Figure 6. If the internal pressure remains
below 120 ± 10 psi, the ring valve remains sealed against the cap, the vessel remains
closed, and volatile components do not escape. A convenient way to verify venting is to
weigh the vessel containing the sample and acid before and after decomposition.
Depending on the degree of agitation occurring inside the vessel during
decomposition, small amounts of sample may also be expelled as an aerosol or
condensed liquid during venting. Analytes may be lost with the gas escaping from the
vessel. To prevent such losses, a venting tube may be fitted with a trap or condenser.
Other vessel sizes and configurations are available for a variety of applications.
For example, the two-port configuration has also been used as a trap in conjunction with
another decomposition vessel and for smaller volume work or inorganic dissolutions
where little gaseous decomposition products are formed, decomposition can be
performed in a 60 mL vessel of similar construction.
High-Pressure Microwave Vessels
Because the temperature in closed-vessel acid decomposition is frequently
controlled by the pressure limit of the container, new and stronger microwave-transparent
vessels are being designed and constructed. One type of high-pressure device (Parr
Instrument Co.) resembles the steel-jacketed bombs used in conventional convection
ovens. The basic construction is similar: a Teflon TFE liner for inertness and a strong
rigid retainer surrounding the core vessel that provides its structural support, as seen in
34
Vent Tube-
Venting Nut
Vent
Opening
G as Floi
Ring Valve
■Safety Disk-
Vessel
Digestion
Mixture
Sealed
Venting
Figure 6. C ross-Section of Release M echanism in Disk Valve
35
\
I
t
i
Figure 7.
Parr High-Temperature and H igh-Pressure Vessel
(Reprinted with permission from reference. 6, copyright 1988 American Chemical Society.)
36
Figure 7. A polymeric retainer material that is transparent to microwave radiation is used
to provide this rigidity.
The microwave bomb design incorporates replaceable Teflon O-rings in the liner
cap that seat against a narrow rim on the exterior of the liner and its cap when the
retaining jacket is screwed into place. Excess pressure in the vessel compresses a puck­
shaped disk above the liner cap and gas or vapor can escape through four outlets in the
retainer cap positioned at 90° around the circumference. When overpressurization does
occur, the O-ring is distorted and may rupture and when this happens, the sample is
compromised.
The vessel has been designed for a maximum working pressure of 80 atm,
approximately 10 times the pressure that can be sustained in the unsupported Teflon PFA
vessel. These high internal pressures permit increasing temperatures for the acids and
acid mixtures that would otherwise overpressurize the unsupported PFA vessels. These
temperatures cannot, however, be sustained for lengthy periods, because if the
temperature of the outer polymeric retaining vessel is allowed to exceed 50 °C (85), the
structural integrity necessary to maintain the seal and high pressure may be compromised.
Currently, because pressure and temperature conditions in the vessel cannot be
determined, calculated parameters cannot be confirmed.
As the result of the thickness of the inner Teflon core and light-weight but
massive outer jacket, the contents of the sample cup are well insulated. This significantly
reduces heat loss by conductive heat transfer from the sample to the vessel walls.
Conditions that would have taken > 1 h to reach in a conventional oven using the steeljacketed bomb can be now be achieved in < 1 min. To prevent blowout and subsequent
37
loss of sample, it may be best to approach conditions of extreme pressure slowly using
partial power and multiple-step heating.
Temperature and Pressure Measurement
It is more important to monitor conditions in closed vessels than in open vessels
at atmospheric pressure, because in open-beaker decompositions the temperature is
limited to the boiling point of the acid or the azeotropic mixture of acids. This
temperature maximum is maintained until all the acid has been evaporated. If multiple
acids are present, the most volatile acid will boil away first, followed by the next most
volatile, and so on, unless an azeotropic mixture is formed. In these cases the
decomposition reactions proceed as they would at the boiling point of the mixture. When
closed vessels are used, however, the solution temperature is not limited by the boiling
point of the acid at atmospheric pressure, and for many of the acids it can be raised
significantly higher than is possible in open-vessel procedures. The limiting parameters
now become the temperature and pressure that the vessel can safely contain. Once the
safety limit of a vessel is established, temperature and pressure can be monitored to
maintain the reaction within these limits.
Temperature measurement during operation is complicated by the high-energy
microwave fields in the cavity. While it is theoretically possible to monitor most physical
or chemical parameters of a material being heated in a microwave cavity, in practise, the
monitoring probes must not perturb the microwave energy in order to obtain accurate
readings. If the probes are metallic they may behave as antennas and reradiate the energy
or reflect the radiation and cause arcing in the microwave field. Conventional temperature
measurement devices are not recommended for use in this environment. Mercury and
alcohol thermometers in a microwave cavity with insufficient loads might explode since
38
the hot mercury or alcohol have no room for expansion. With proper precautions,
thermisters and thermocouples can be adapted for use in RF fields for temperature
measurement
Temperature Measurement in Microwave Fields
Three measurement systems have been developed for use in microwave
environments. One is a shielded thermistor normally found in conventional home
microwave appliances. This temperature probe is not suitable for use in microwave acid
digestion for several reasons. Temperatures during acid digestions range from ambient to
just over 400 °C in the case of sulfuric acid. Thermistors, which have an upper limit of
just over 100 °C, would not be adequate for the task. Shielding of the thermistor is
accomplished by encasing it in a stainless steel tube electrically connected to a copper
braid grounded at the oven wall. Successful operation in the microwave field requires
that the end of the probe be placed inside a microwave absorber where it is essentially
isolated from the field. Since the half-power depth of the 2450 MHz wave is ~2.5 cm, a
microwave absorber, such as water or hydrated tissue, totally consumes the 2450 MHz
field within a distance of approximately 5 cm. Thus, nearly all of the energy of the wave
will have been dissipated in the first 5 cm of dielectric. Because it is often necessary to
accomplish sample dissolutions in the smallest volume practical, generally under 10-15
mL of reagent volume, these massive probes would be exposed to the microwave field
without adequate protection and arcing would be observed.
Temperature measurement during acid decomposition in the microwave field has
been successfully accomplished with thermocouples (3,38). Temperature changes
resulting from microwave heating have been measured with thermocouples to determine
the kinetic parameters of both biochemical and chemical reactions (86). Dielectric
properties of minerals as a function of temperature have also been measured at microwave
frequencies with thermocouples (87). These devices have also been employed
successfully to determine thermodynamic functions of chemical reactions in aqueous
solutions (88), and thermal hemolytic thresholds of erythrocytes in saline solutions
heated by microwave energy (89). Specific details concerning thermocouple fabrication
in the narrow shielded configurations required for acid dissolution in microwave systems
have been thoroughly covered elsewhere (3, 6). The construction of a shielded
thermocouple sensor, with the proper dimensions for use in a microwave cavity can be
found in Appendix C. Type "T" (copper-constantan) thermocouples were used for early
temperature measurements with the Hewlett Packard 3497 Data Logger and an 020
twenty-channel option board. This board provides an electrical zero-degree reference
junction; emf values were derived from data tables based on thermocouple reference
tables compiled in NBS Monograph 125 (90).
The use of optical fiber devices for temperature measurement in microwave fields
is a relatively recent development (5,6,91). Unlike metal-shielded thermistors and
thermocouples, the fiberoptic device does not interact with microwave energy and is
transparent in the field. More recently, optical fiber thermometry has been used for
temperature measurement of microwave-induced hyperthermia (92), in the fabrication of
composite materials cured with microwave power (91), and in monitoring acid-initiated
sample decomposition in a microwave system (5). Localized heating (93) and electrical
interference problems common to metal probes (94, Olmi, R., IROE-CNR, Firenze,
Italy, personal communication, 1986) do not occur in optical fiber temperature sensors
(5). Although the current instrumentation is relatively expensive, it is reliable and
accurate (5,92,95,96). The use of this instrumentation in closed-vessel microwave acid
40
digestion has specific equipment configuration requirements (96,6). Figure 2 illustrated
the arrangement of both the temperature systems and pressure monitoring equipment used
in conjunction with the microwave system and was described in the section on
instrumentation (3, 6).
Since the Teflon PFA probe coating is permeable to acid vapor, the optical fiber
probe cannot be placed directly into the acid. The sensing phosphor at the tip of the
probe and the Kevlar protective covering are attacked by strong oxidizing acids and
should be protected by inserting the first 12 cm of the probe into a 1/8in-thick-walled
Teflon PFA tube sealed at one end. This sleeve can then be inserted into the digestion
vessel cap port and sealed to hold pressure by using ferruled Teflon nuts. The pressure
and torque of sealing the optical fiber probe directly could cause the glass fiber to break at
this point; direct pressure on the probe itself should be avoided to protect the fragile glass
filament. Figure 4 illustrates a configuration that protects the fragile probe from the
effects of high pressure and chemical attack by nitric, phosphoric, hydrochloric, and
dilute hydrofluoric acid (and these acid combinations) and yet provides reliable
temperature and pressure measurements. Problems encountered with the use of this
measurement system are related to the durability of the optical fiber probes; they are easily
broken and chemically attacked by long exposure to concentrated hydrofluoric acid
vapor. Degradation of the probe and subsequent failure have been experienced after 8-10
digestions of 10-15 min with mixtures containing 50% hydrofluoric acid. Electron
microscopy indicated that chemical attack of the glass fiber was the cause of probe failure
(Sturcken, E. E., Westinghouse Corp. Savannah River Laboratory, unpublished results).
Care and treatment of the probe have been discussed elsewhere (6).
41
A number of experimental probe configurations using optical links and remote
phosphor scanning are being investigated. A remote temperature measurement technique
using phosphors on the outside of glass digestion vessels has been employed with high
temperature sulfuric acid decompositions (98). This technique eliminates the necessity of
placing the probe in the digestion vessel but introduces a bias into the reading because of
the cooling effect of the air on the outer surface of the container. In this configuration the
temperature measured is that of the glass surface and not of the acid and sample inside.
This remote technique is not suitable for use with the Teflon vessels without modification
because of the tremendous insulating capability of this material.
Pressure Measurement
Since most pressure measurements are commonly performed either with metal
transducers or wheatstone bridges, they are necessarily remote from the cavity. Details
of such measurements made in closed PFA vessels have been documented elsewhere (3,
6). The functional range of both the transducer and the valve was matched to the
pressure range and vessel used. With current Teflon vessels the range of the relief valve
is from 5-10 x 10^ Pa (5-10 atm), thus the transducer should cover a pressure range from
ambient atmosphere to 17 atm. The pressure transducers used in connection with this
work were calibrated against NBS certified gauges and the calibration data appear in
Appendix D.
Whereas the vessel and pressure tubing are ordinarily Teflon®, the pressure
measurement device and safety valve are stainless steel and must be isolated from the
internal volume of the vessel and pressure line to prevent contamination of the samples
and degradation of the metallic components. A trap of distilled water can be used to
isolate the transducer. Water is noncompressible and transmits the pressure without
42
allowing the vapor to contact the metal portions of the remote pressure measurement
system. If this method is used, the water must be changed frequently, because it will
accumulate dissolved acid vapor from the vessel. A membrane can be placed at the end
of the water tube in the fitting to prevent the mixing of the acid vapor, thus effectively
isolating the measurement system from the sample.
Optical fiber measurement systems have recently been introduced that permit
direct measurement of the pressure inside a digestion container (99). Like the
temperature measurement probe it does not perturb the field and the system is capable of
instant readout compared to the remote system described above, where readings can only
be taken every 3 s because of a computer program limitation on updating.
Although the sample container does not absorb microwave energy directly, it does
allow heat to flow from the sample. This results in a thermal gradient inside the container
that reduces the temperature of the volatilized acid below that of the liquid phase. Since
the gas phase is cooled by heat loss from the vessel and absorbs little heat from the
microwave field, condensation occurs at the top of the container above the liquid phase.
Heat loss from the liquid phase is more than compensated for by the absorption of
microwave energy. When a sample is placed in 5 mL of nitric acid in the 120 mL Teflon
PFA vessel, > 95% of the vessel volume is filled with the gas phase. Less than 5% of
the volume is actually heated when the liquid phase absorbs microwave energy. Thus,
the gas phase is not in thermal equilibrium with the liquid phase. This phenomenon
prevents the use of partial pressure data accumulated under equilibrium conditions to
predict the pressure inside the vessel at a given temperature. The actual pressure in a
given container is dependent on the size and composition of the vessel, the type and
quantity of acid(s) used for the dissolution, the temperature of the acid, and the
43
temperature in the microwave cavity. These non-equilibrium conditions in the microwave
vessel make direct measurement of pressure the only practical way of relating pressure to
temperature.
Computer Controlled Acquisition of Data
Temperature and pressure data were obtained with a data acquisition system that
has also been described previously (3, 6). The ability to plot these parameters over time
or against one another has been useful for understanding the interactions of the acid with
the microwave energy and the acid decomposition of the sample. Temperature and
pressure data were used throughout the project in the development of decomposition
schemes and in quality control of routine dissolutions.
Numerous data acquisition equipment configurations can be assembled to record
and display the temperature and pressure data from the microwave digestion and depend
largely on the sensors used in monitoring and the needs of the analyst. If there is no need
to record data, direct reading devices that provide digital display for both temperature and
pressure are available.
Theory of Microwave Heating
Electromagnetic Energy and the Interaction with Matter
Microwave energy is electromagnetic (EM) radiation located in the frequency
spectrum between the infrared (IR) and radio waves as seen in Figure 8 and ranges from
~ 1 mm to around 30 cm. Like familiar visible light, microwaves can be focused by
lenses, reflected by mirrors or metals, refracted, polarized, absorbed and contains an
energy component. Unlike ionizing ultraviolet (UV) radiation whose wavelength is small
compared to the dimensions of an object, 2450 MHz waves, which are 0.1225 m (12.25
44
S
3
1
1
"S
DC
CM
o
CO
d>
c3
0) ;
O:
3)
« !
*o
O)
>- CO
CD O
5
T
s
CO
c
Ui
oT”"
o
X
CO
i J_
2
N
«- 0)
E
•* £
to W
- t
«* §g
IT)
o
<D
f
O
>
*-<0
"
1
I<3I4
2J
1£Eo
St?
c
E °
5
g>£
1CD ss
«
>
T
T T
;0|q|6!A
t
0)
(0Q
to©
o
5
CO §
O
X
CO
g§
«- w
u
m O
CM
o
I
©
©
E
n .s
<0
2 i
3a> g.
I
i
fa
O’S —
“1
T
®
If |I
S's
I
o
5
a 8
* ii
o
©
r~
®
E ^8
3
CO
CO
E
<D
-C
O
CO c
CO o(0
c
T
LU
£
* >1
rA s
a, "S
|
B
E
§a
O) giE B
■a
CO
T
3
“
45
cm) are comparable in size to the objects themselves and are non-ionizing at the small
amplitudes used in this study.
Materials may be classified into three distinctly different groups depending on
their behavior in a microwave field. The first class of materials are those that reflect the
waves and consist mostly of metals and mirrored surfaces. Most metals, where the
conductivity is high, have very large reflection coefficients. In the second group, energy
of the wave may be transmitted without any interaction between the molecules and the
wave; no energy is lost from the wave on passage through the material. Many plastics
( 100) and some space-age ceramics are transparent to microwave radiation, e.g. they
neither absorb nor reflect the microwave. Finally, materials may actually absorb a
portion of the EM energy on its passage through the material. These are dielectric
absorbers and tend to dissipate a portion of the wave's energy on each passage through
the material.
For recent reviews of the fundamental principles of material heating in the
microwave environment the reader is referred to the article by W. H. Sutton on
processing (ceramic) materials in a microwave environment (101) and Chapter 2 by Neas
and Collins in Introduction to Microwave Sample Preparation (102). Several of the
papers given at the 45th Annual Meeting of the Institute of Food Technologists, in
Atlanta, GA in June 1985 were reprinted in the June issue of Food Technology. Articles
by Mudgett (103) and Schiffman (104) that appeared in the Overview have covered the
subject well from a chemical perspective.
Mechanisms of Microwave Heating
Just how objects heat in a microwave energy field is an extremely complex
process. At least two principle mechanisms of heating, dipole rotation and ionic
conduction occur simultaneously in solutions irradiated at microwave frequencies. While
thorough for their purposes as introductions or for conference proceedings, the
discussions of microwave heating that have appeared in the literature (102,106-109)
have necessarily suffered from brevity and biased viewpoints. That is, the authors
discussed microwave heating as engineers allied with food science research, rather than
as chemists familiar with the interaction of microwave energy with molecules and
electrolytes in solution. Most discussions of EM heating focus naturally enough on the
mechanisms as they occur in water ( 110) since the subject of aqueous dielectrics has been
extensively studied at submillimeter and infrared wavelengths (111). The science of food
technology is now beginning to look at the chemistry of food interactions in the
microwave field (112). Perhaps then, this discussion will prove useful as a review as
well as provide a point of departure for further research into the chemistry of microwave
interaction with electrolytes in solution.
At all points in the EM spectrum, radiation is absorbed because resonance occurs
when the incoming energy matches the energy required for perturbation of the molecule
or the electronic structure of the atom. Molecules distributed in this energy field will be
induced to rotate while attempting to align themselves with the field. Such molecular
rotations would increase as the frequency increased and large amounts of heat would be
generated. Microwaves, however are considerably larger than aqueous ionic or
molecular species and energy transfer is accomplished by means of ionic conduction and
dipole rotation.
Dipole Rotation
Water is a good example of a molecular dipole, e.g. a non-symmetric, although
electrically neutral molecule in which partial (separated) electric charges exist on the
47
opposite ends of the atoms making up the molecule. Such molecules attempt to align
themselves in the presence of an electric field as the result of torsional forces, but they
cannot keep up with the oscillating field. Mechanisms of loss of energy from the field are
due to resistance by inertial drag and elastic and frictional forces that produce heat
throughout the volume of material. In the presence of an electric field, an entire molecule
may be polarized, or a polar molecule may twist or bend in response to the passing wave
as seen in Figure 9. With the decay of the field the molecules return to their disordered
state and with each collision some of the energy is transferred to the random motion.
Such rotational motion in a dipole converts energy from the field to potential energy of
the material and then to random kinetic (thermal) energy in the system. All of these loss
mechanisms are included in E", the loss parameter. The ratio of the loss parameter to the
dielectric constant E' is the loss tangent 5. As a general rule, the polarization of the
molecules lags behind the field even though the response times of the dipoles in aqueous
solutions are of similar magnitude to microwave frequencies, 2.45 x 109 cycles s-l.
Inertial forces exert a drag that opposes the rotation imposed by the torque that
seeks to orient the molecules with the field. Energy is thus abstracted from the field to
overcome this drag, adding to the random motion when the opposing torque is removed.
According to Walker (110), drag is important only at microwave frequencies; at lower
frequencies, the movements are too slow for drag to be important and at higher
frequencies, rotation is impeded and is not the perturbation observed.
A third contribution to the random motion in solution is offered by Hasted (111)
who suggests that microwave energy may be capable of breaking the bonds of aggregates
of water molecules that exist in solution. Groups of molecules are continuously making
and breaking bonds which lowers their potential energy but contributes to the kinetic
48
e
t- 0
8-
The water
dipole moment
is aligned
with the field.
54 -
W ater molecule
£
E
t = 0.1 ns
O.
H
/■Q
The dipole
moment rotates
in an attempt to
follow the field.
£
aE
a0
1a
t = 0.3 ns
54 -
The dipole
moment rotates
in the opposite
direction tying
to align with the field.
Figure 9. Dipole Rotation-Microwave Electric Field
Interaction with Water Molecule, (a) Aligned with
the Field (b) Rotation to Follow the Field
(c) Rotation in O pposite Direction to Keep up with
the Field
49
energy of the system. Since these occurrences are about equally balanced no net gain or
loss is observed. When the microwave field supplies energy to the system that can break
a hydrogen bond, the reformation potential energy of a hydrogen atom is lower and it
latches on to any suitably oriented nearby oxygen, adding the difference in energy
between bond breaking and bond making to the random motion of the group and thence
to the solution. Aggregates of water molecules consisting of as many as six or seven
molecules (111) are known, although 3-4 are the most common and are quite stable.
Many more bonds would have to break in order for the one hydrogen to twist off and
escape from a four- or seven-membered clique. Only one and two-membered aggregates
could reasonably be expected to undergo such events. When the solution is no longer
pure water, but is instead ionic or molecular and consists of charged particles, then
viscosity plays an important role in this transfer of energy as it contributes to heating by
offering resistance to ionic conductivity.
Ionic Conduction
In room temperature water only 1 molecule in 10? dissociates into constituent
hydrogen and hydroxyl ions. The OH" is a viable species; the H+ attaches to another
water molecule to form the hydronium ion. Both ions can be attracted by an electric field
that results in current flow. Because of its small dissociation constant, ionic conduction
plays only a small role in the contribution of microwave heating experienced in the pure
liquid. Since the dissociation of water increases with temperature this mechanism should
contribute increasingly more to the heating. This should be true of all pure liquids whose
dissociation constants are equal to or smaller than that of water. Never-the-less, for
water, dipolar rotation is the dominant heating mechanism.
50
By contrast, in dilute solution where the electrolyte is completely dissociated the
ions are presumed to be spherical and non-polarizable. The absorption of microwave
energy results in a rise in temperature because the conductivity of that solution gets larger
and larger so that the degree of dissociation is 100%. Pushing through the water in
opposite directions, the ions bump into each other and transfer some of the energy of
motion acquired from the field to the solution molecules raising the kinetic energy of the
solution which results in heating. Thus, substances with large dissociation constants
should be good absorbers of microwaves.
Such charged species can be influenced by the presence of an electric field which
causes the migration of the ions to their respectively (opposite) charged poles. In dilute
aqueous solutions the central ion is moving through the sea of solvent molecules which
by their sheer number exert a frictional drag on the ion. For H+ this "viscous effect" has
been thought to give rise to a form of proton hopping where the H+ is actually transferred
through solution by transport on the H30+ ion (7). This movement constitutes the flow
of current and serves to explain why the conductivities of the H+ and OH- ions are so
much larger than that of other ions. Electrical conductivity is just one mechanism which
contributes to heating of ionic solutions at microwave frequencies. Other phenomena
include the asymmetry effect, which is a renewal process and thus allied to the relaxation
frequency of the molecule and the electrophoretic effect (7,113). These effects are the
results of the properties of ion mobility in solution for the latter and equivalent
conductivity of the ions for the former. Both of these effects are tremendously influenced
by such solution properties as viscosity, concentration, temperature and the dielectric
constant, as well as ionic size and ionic charge.
Electrolyte solutions conduct electricity through the motion of the ions under the
influence of an electric field. At high concentrations, as is normally the case in
dissolutions, each ion is surrounded mostly by other ions and the field affecting any one
anion or cation exists partly because of its neighbor's influence. This is quite different
from the situation in infinitely dilute solution where the distance between ions is very
large and the individual ion feels only the effect of the applied electric field. Most of the
data in the literature is at infinite dilution and cannot easily be applied to acid dissolution
in a microwave field.
Ions in aqueous solutions that are infinitely dilute are normally solvated and carry
these solvent molecules around with them. As a result, the ion has greater bulk and thus
moves more slowly in solution than in its unencumbered state. Even though larger ions
have smaller electrical fields than smaller ions and would hold fewer solvent molecules,
the effective aggregate diameters would be similar, thus, comparably sized ions in
aqueous solution would have about the same effective diameters. When no external
electric or magnetic fields are present, ions in solutions that are not infinitely dilute can be
visualized as possessing an "ion cloud" seen in the schematic representation shown in
Figure 10a. The central ion is surrounded by a sphere of oppositely charged ions. In the
presence of an external electric field, the sphere becomes distorted as the individuals ions
move to their respective oppositely charged poles. The distorted cloud attempts to renew
itself, a process requiring some finite time, the relaxation time. In response to the
imposition of the field there is a relaxation of the atmosphere immediately surrounding the
core ion as it moves out of its central position with respect to its neighbors and
experiences a restoring force back to the center. But the atmosphere cannot keep up with
the ion because of its bulk and thus lags behind; the asymmetry of the particle exerts a
52
(a)
°
©
0
>£•
°
0
n
°
Symmetrical cation and ionic cloud
in the absence of electric field.
Distortion of ionic cloud in
presence of electric field.
(b)
r= ® y ~
Solvated central ion moving against the crowd
of solvated counter ions
Figure 10. Ionic Conduction (a). Asymmetric Effect
(b). Electrophoretic Effect
53
drag effect slowing the motion of the ion. Increasing the concentration increases the
number of neighboring particles which further slows the particle and lengthens the
relaxation time resulting in a more pronounced effect and additional heating.
Since the medium is constantly in motion the central ion is always moving against
the flow of other ions and their aggregate structures as seen in Figure 10b, a situation
that further reduces the mobility of the ion. This is the electrophoretic effect and like the
asymmetry effect, is enhanced as the concentration of the electrolyte increases because the
equivalent conductance decreases and the retarding forces increase. In very viscous
solutions, the electrophoretic effect may be diminished because both the solvent and ion
movement are slowed. Temperature increases are accompanied by increased conductivity
as the result of diminished viscosity effects and would show a reduced electrophoretic
heating component (7). Both field intensity and ionic charge contribute to the size
(intensity) of the electrophoretic effect. As the field increases, the diffusion potential
increases and the particles are all moving faster. Ions with more than a single plus charge
are holding their remaining electrons ever so much tighter in a more compact volume with
a larger charge to mass ratio and would be expected to contribute more to ionic
conductive heating processes.
In the absence of an electric field, ions in solution move according to their
diffusivity. In the presence of an electric field, ions move through solution according to
their conductivity and mobility which are affected by total charge, ionic size as well as
viscosity and concentration. At increased temperatures already hot viscous solutions will
get even hotter when irradiated with microwave energy. For liquids such as water and
small polar alcohols, tan 8 decreases with temperature ( 102).
In concentrated solutions, electrostatic charges become more important leading to
solvated ion pairs and the conductivity drops as the bare ion is now encumbered by
associated water or solvent molecules. These spheres tend to act like brakes on the
movement of ions in solution. The conductance of an electrolyte solution increases as the
valence of the ion increases and as its mobility and concentration increases. Small highly
charged ions have greater conductance. As the concentration of the ionic solution
increases so that retarding forces increase and equivalent conductance decreases, both the
electrophoretic and asymmetric effect are intensified. There is however, a limiting
improvement with concentration until viscosity effects reduce the mobility, e.g., there is a
maximum equivalent conductance with concentration beyond which conductivity
decreases and is different for different ions.
In the case of microwave acid dissolution, where the reagents are normally highly
concentrated solutions, microwave absorption may be diminished. Since most
concentrated acids dissociate completely in the presence of water as the result of
ionization, there may be an optimal concentration where microwave absorption is
enhanced. Heating due to both dipole rotation and ionic conduction would be increased
as result of dilution.
Behavior of Electrolytes
Solution Heating
Vessels frequently used in conventional hot plate and bomb methods are often
poor conductors of heat, thereby increasing the time necessary to accomplish the heat
transfer. Because the containers are heated from the outside, thermal gradients are
established across the liquid volume so that only a small portion of the vessel's contents
are actually at the temperature of the heating device. Temperatures in bomb devices
55
eventually reach equilibrium, but in open vessels a small amount of evaporation from the
liquid's surface exerts a counter-cooling effect that exacerbates the thermal gradient.
Thus at any one moment, only a small portion of the vessel's contents are at the boiling
point of the liquid. Compared with traditional conductive heating methods that require
the transfer of heat from the device to the vessel and then to the reagent, microwave
heating is almost instantaneous and heating is accomplished by means of direct coupling
of the radiation with the molecules in an energy transfer that produces heat as a by­
product. As the result of this coupling in solution, microwave heating occurs everywhere
in the liquid at the same time so that all of the vessel's contents are quickly raised to the
boiling point in open vessels, and well beyond that in closed vessels.
Fundamental Equation for the Determination of Absorbed Power
Equations for power measurements in a microwave system are derived from
thermal concepts based on the heat capacity of a mass at constant pressure. Heat
capacity, Cp, is that quantity of heat required to raise the temperature of one gram of a
substance by 1°C. The energy absorbed produces a rise in temperature, AT. If a quantity
of energy is delivered for a unit of time, then P, the power absorbed by a substance
(power density) in the microwave cavity may be expressed in the following relationship
Pabsorbed^
^
"
1^
(1)
where P is the apparent power absorbed by the sample in watts, (1W = 1 J s-1); K is the
conversion factor from thermochemical calories per second to watts (4.184 J caH); Cp is
the heat capacity, thermal capacity, or specific heat (cal g"*C-l); m is the mass of the
56
sample in grams; AT is Tf, the final temperature minus Tj, the initial temperature (°C); and
t is the time in seconds.
Equation 1 has been used, with minor modifications, to establish the significance
of local variations in tissue temperature as related to changes in the heat content of the
body as a whole (114) as well as to express the absorbed power density of tissue
exposed to electromagnetic radiation (115-117).
Heat Capacity
Equation 1 can be used to calculate the power uptake of any quantity of material
for which the heat capacity is known and for which initial and final temperatures can be
measured (3). Heat capacity values in the literature are frequently given as the apparent
molal heat capacities (cal °C-l mole-l). From values and equations given by Parker (118,
119), the heat capacities (cal g-1 °C‘l) have been calculated for commonly used acids at
several concentrations and are summarized in Table 2. The heat capacity of an acid varies
inversely with its concentration; for example, the more dilute the acid the greater its heat
capacity and the more nearly its value approaches that of water.
57
Table 2. Heat Capacity of Mineral Acids and Solutionsa
Acid Solutions
Acetic (100%)
Hydrochloric (37.2%)
tl
ll
Hydrofluoric(49%)
Nitric(70.4%)
ll
ll
Phosphoric(85.5 %)
Sulfuric(98%)
ll
II
Sodium Chloride
Water
Concentration
(mol L*1)
17.4
12
6
1
28.9
15.9
8
1
14.8
18
6.7
1.1
1
55
Heat Capacity
(cal g-l C-i)
0.4947
0.5863
0.7168
0.9378
0.6960
0.5728
0.7162
0.9497
0.4470b
0.3499c
0.6142c
0.9142c
0.9339
0.9997
Note: Data normalized to 25 °C (Reprinted with permission from
reference 6, copyright 1988 American Chemical Society)
aRef. 119
bRef. 120
CRef. 121
CHAPTER 4
EXPERIMENTAL DETAILS, RESULTS AND DISCUSSION
Experimental Methods
Determination of Absorbed Power
Temperature Measurements
For the determination of absorbed power, temperature measurements are required
on a known mass of material. Weighed to the nearest 0.1 g in Teflon containers, the
starting temperature of the water load was measured using a mercury thermometer
(calibrated against NBS standard thermometer) with continuous stirring. An average of 3
readings was taken as the initial temperature after approximately 1 min. The covered
container was placed in a conditioned (heated 1 kg of water for 5 min on full power)
microwave unit and irradiated for 2 min and 2 s. As soon as the program terminated the
container was removed to an insulated stirring platform. A small stir bar was inserted and
the thermometer was lowered into the container just off center. Slow stirring (no vortex)
was begun and temperature readings taken after 15-30 s. The final temperature is the
average of the three highest readings.
The same procedure was used for making all of the measurements on the acids
and electrolyte solutions to produce the data for the predictive equations for power
absorption. All of the samples are irradiated at full power, but the exposure time is
reduced as the mass is reduced so that the net temperature change is kept to between 5 and
50 °C where the changes in heat capacity are minimal.
58
59
Mixture Experiments
One-liter mixtures of 1:1,1:4,2:3 and 7:3 (v/v) hydrofluoric and nitric acids were
prepared daily and temperature measurements made as above. The volumes were then
irradiated for 2 min at full instrument power and the average temperature rise measured
was 14.4 °C. Variables in all these experiments included the mass, the temperature and
the line voltage. The first two were easily measured and form the core of the data
accumulated. Although line fluctuations of 1-2 v are normal, they cause less than 6W
change at full power. No power corrections were applied since they amounted to < 1%.
The time was fixed at 122 s.
Heating Profiles
Temperature and pressure data were obtained by monitoring the contents of one
vessel in any group of samples during irradiation. This container was designated the
monitor vessel and the monitoring equipment was attached by first inserting a closed end
1/8" Teflon PFA sleeve into one of the ports and securing it with a retaining nut in the cap
of the 120 mL vessel that had been torqued to a minimum pressure of 16 N m. The
optical fiber probe was then carefully inserted into the sleeve and the retaining nut was
secured using a wrench. A one meter pressure line was secured to the open port on the
cap for pressure measurements by a transducer. Pressure and temperature were read on
separate channels of a volt/ohmmeter (A Hewlett Packard 3497 Data Logger) and
converted to the appropriate temperature values by the Luxtron thermometer based on
look-up table of the phosphorescent decay, and to the corresponding pressure values
using the calibration tables for the transducers. They may be found in Appendix D.
Every 3 s a set of data points was taken that consisted of the observation or scan number
60
(multiplied by 3 to give the time in seconds), the temperature reading in °C, and the
pressure reading in psi. Data was stored on floppy disks for later retrieval.
Heating Rates
Temperature measurements for the heating rate data were made in the same
fashion as the heating profile measurements, however, the pressure line was connected to
a pressure monitor with analog output. The pressure monitor was set to limit the pressure
in the vessel to 100psi so that the heating program was terminated when that pressure
was attained.
Reflected Power Measurements
The heat rise in the sample container for the mass range from 70 to 1500 g of
nitric acid and water heated at 574 W of power for 120 s was measured as for the
absorption experiments. A copper-constantan thermocouple was mounted on the
waveguide end of the ferrite where the reflected power isolated by the circulator exits the
cavity. Temperature measurements during irradiation were made by connecting the
thermocouple to an empty acquisition channel on the Hewlett Packard Data Logger and
readings were taken every 3 s.
Heat Loss
Temperature measurements of the cooling were made with the Luxtron optical
fiber thermometer in the same way as the normal heating profiles. Ten grams of distilled,
deionized (DI) water were weighed into the 120 mL Teflon vessel and monitoring
equipment was assembled as described above for heating profile measurements. With the
exhaust fan on, the sample was irradiated at full power until the temperature of the water
in the vessel reached 180 °C (observed on the thermometer readout) at which point the
61
power to the unit was terminated. Continuous temperature readings were taken for the
balance of 45 min during the cooling of the vessel's contents.
Materials and Reagents
The source of the plastic containers used in this research effort are cited in the
section on equipment Glass Kimax beakers, used in the maximum power output study,
were obtained from Fischer Scientific. Reagent-grade concentrated mineral acids used for
the power absorption determinations were also obtained from Fischer Scientific.
Nitric, hydrochloric, hydrofluoric, sulfuric and perchloric acids used in the
dissolution of samples for certification analysis were specially purified reagents prepared
by subboiling distillation obtained from the NIST Purified Reagents Facility. Phosphoric
acid for the alumina digestion was prepared by hydrating Baker purified phosphorous
pentoxide with high-purity double-distilled DI water in a closed chamber constructed
from a cleaned polycarbonate covered tray.
Materials used as models for the biological and botanical target temperature study
were obtained from Sigma Chemical Company. Real world samples for the applications
studies consisted of Standard Reference Materials obtained from the Office of Standard
Reference Materials at NIST and were furnished as part of an ongoing certification
program. Alumina samples were proprietary materials submitted by Union Carbide and
the glass frit and simulated nuclear waste material were provided by the Savannah River
Laboratory of the Dupont Company.
62
Experimental Results and Discussion
Maximum Power Output
Power Absorption Determination
Based on the assumption that the majority of the power delivered to the cavity is
absorbed by a sufficiently large quantity of a dielectric absorber Equation 1 may be used
to evaluate the power output of a magnetron to the cavity. It is variously known as a
caloric power measurement and needs only a simple conversion factor (4.184 J cal-l) so
that it can to be used to evaluate the maximum output of commercial microwave
appliances (122). A simplified form of the equation is constructed by combining the heat
capacity, conversion factor, time and mass into a single constant for the evaluation of
microwave output power in commercial equipment (109,122,124). For a rapid ball­
park estimate of a unit's power a very simple microwave power instrument has been
devised (123) which can be calibrated in watts for a specific water load. The device
appears to function optimally with volumes of 1000 mL or larger.
Each microwave system is a unique device; the power delivered to the sample
depends not only on the power output of the magnetron but also on the tuning of the
waveguide and cavity dimensions. The apparent power absorbed by water irradiated at
100% power should be used to determine the maximum output of all 2450 MHz
microwave equipment that have power outputs between 500 and 850 W. Although
several different methods have been used (3,109,122,124), satisfactory measurements
are made on replicates of weighed, 1kg (1000g) samples of room-temperature distilled
water in thick-walled microwave transparent vessels.
Procedures for determining the power output of commercial microwave units
have been compared along with a procedure recommended in a report issued by Gerling
Laboratories (124). Several refinements that eliminate potential sources of error
introduced by variables retained in the recommended testing sequence are suggested.
One liter all-Teflon containers with lids were used so that heat lost through evaporation
from the surface was eliminated and the thermal mass of the container did not detract
from the measured output as it does with a glass beaker. Measurements made in 1000
mL Kimax vessels covered with parafilm are consistently higher at 587 W than the 574
W determined in Teflon vessels. Most commercial-sized microwave cavities are designed
for operating loads of between 500-1000 g. A single 1000 g load is easier to handle than
two such vessel volumes and reduces the measurement error involved in taking
temperature readings in two vessels simultaneously. Two readings in two vessels must
be averaged for the final result - a process that can smooth or mask cavity differences.
Multiple vessel measurements have larger uncertainties, i.e. larger standard deviations
(47) as seen in Table 3. In a homogeneous microwave field 1000 g of water absorbs
approximately the same amount of power in one container as it does when equally
divided between two or five containers, as the table shows.
Table 3. Maximum Microwave Power Output Measured in
Multiple Containers
mass, g
aF
Power, W
1
1000.3
16.34 ±0.15
570 ± 5
2
1000.2
16.54 ± .36
576 ± 15
5
1000.2
16.99 ± 0.56
592 ±20
# of Vessels
Multiple glass vessels (2 xlOOO mL) had differences as small as 1% and as large
as 8.4% between them. The mean of all measurements was 584 ± 24 W. As the AT gets
smaller the relative errors in measurement get larger. Proper pre-operating warmup is
essential to consistent measurements and we have found no significant difference in
measurements with the cavity fan on or off. Errors in the measurement of time by as little
as 1 s can introduce a 5 W error in measured output. The "time on" is fixed at 122 s and
incorporates a 2 s magnetron delay to reach full power. This produced a consistent
temperature rise of 16.6 ± .35 °C.
Table 4. Maximum Power Output Measured in Glass and
Plastic Vessels
# of Vessels
mass, g
Output Power, W
Glass
1
1000
587 ±8
2
2000
584 ±24
1
1000
574 ±6
2
2000
568 ± 9
Teflon
Note: ~4% average difference between Beakers 1 and 2. Glass
vessels are 1000 mL Kimax pyrex beakers. Teflon vessel is 1000 mL
Teflon PFA covered jar; n, number of measurements 3 > n < 6.
The mass of water was measured after the initial temperature was taken and
adjusted to 1000.0 ± 0.1 g. Differences of 2 °C in the starting temperature of the water
did not adversely influence the determination even though some of the samples were at
21 °C and some at 25 °C. Water for these experiments was left standing for a minimum
of 2 hr or overnight to equilibrate to the controlled room temperature of 23 °C; only ± 1
°C was tolerated.
In this test sequence a turntable was used to rotate the test sample through the
field in order to average the exposure of the vessel and contents to the field. The
recommended procedure assumes that the temperatures measured in 2 or 4 vessels, when
combined, give an average value for the heat rise. Microwave fields inside commercial
cavities have hot spots so that mode stirrers and circulating devices are frequently useful
for smoothing exposure to the field (109). Except for very large loads, the microwave
energy field is not homogeneous enough to allow the placement of vessels at fixed
positions in the cavity to achieve reproducible results. Even when mode stirrers are used
to homogenize the field, temperature differences of as much as 50% were observed in
identical 120 mL vessels placed at different locations within the cavity (3). Similar
variability was seen in a microwave unit that used a rotating antenna to homogenize the
energy field (47). Immediately upon completion of the test, the containers were removed
from the microwave to an insulated stirring platform where the lid was removed. A
Teflon coated stir bar was added and stirring, which is essential to distribute the heated
water uniformly, is commenced on slow speed so as not to create a vortex. The final
temperature is the maximum temperature read after equilibration, generally not less than
15 s but not more than 40 s after stirring begins.
In the same way that the addition of a salt such as K2SO4 to sulfuric acid
increases the boiling temperature in normal Kjeldahl digestion procedures, the addition of
10g sodium chloride to 1000 g water would be expected to increase the power
absorption of water samples as the result of the increase in conductivity of the solution.
Results obtained by Gerling (122) suggested that the absorbed power of 1% saline
solution as a function of time was lower than for pure water. This apparently anomalous
behavior was shown by Lentz (125) to be the result of evaporative losses from the
surface during the 60 s of heating in an open vessel that were unaccounted for in the
calculations. Substantial losses of heat through evaporation were accounted for by
measuring weight loss and the increased heating of the edges and surfaces to abet these
losses was confirmed by photographing the surface. Reduced penetration of the 2450
MHz wave with increased concentration is thought to account for a more rapid dissipation
of the energy as heat. In our laboratory, microwave absorption measurements for litersize loads are normally performed in covered Teflon containers irradiated for 2 min at
maximum power. Experiments conducted on DI water and 1 M (~6wt%) solutions of
sodium chloride under identical conditions indicate that 1liter of electrolyte solution is
only 3.6% more absorptive than 1L of water.
When the temperatures were corrected for the mass discrepancy (more material
would absorb more energy) the power absorptions were essentially identical within the
measured uncertainty. This again, was an unexpected result It is entirely possible that
the additional 60 s of irradiation in our experiment permitted more heat to be lost through
the container walls than gained from the heating rate. At the same time, the more
concentrated saline solution would have dissipated the heat faster. Thermally insulated
vessels should reduce the dissipative heat loss. However, preliminary experiments have
not shown discemable differences in power absorption between insulated and non­
insulated vessels with 100 and 200 g quantities of DI water and nitric acid as seen in
Table 5. Neither were significant differences found at larger masses of the two reagents,
as Table 5 indicates.
67
Table 5. Effect of Insulated Containers on the Measurement of Power
Absorption in Liquids at 2450 MHz
Volume, mL
Insulated
Power, W
Water
1000
1000
No
Yes
566 ± 5
565 ± 4
No
Yes
No
Yes
No
Yes
250 ± 7
255 ±6
313 ± 5
325 ± 3
433 ± 10
430 ± 12
Nitric Acid
100
100
200
200
1000
1000
Calibration and Proportional Power
Calibration
Measurement of heat transfer is critical to the calibration of microwave equipment
The many different commercial and laboratory microwave units being used for sample
preparation deliver from 500 to nearly 850 W of power to the cavity. Since these units
produce different maximum amounts of output power at full power the actual power
delivered by each magnetron must be determined. In addition, it is important to establish
the linear operating range of the microwave unit so that absolute power settings may be
interchanged from one microwave unit to another. When the power available to the
samples in both systems is known previously designed digestion schemes can then be
adapted to different systems and procedures may be transferred from one unit to another
(6,122). Single point calibrations based on maximum output power assume that the
power controller and associated electronics are both linear and precise over the entire 0100% range. Some units are very nearly linear over the entire range as seen in Figure 11
CM
o>
c
5
O
J=
0)
c
3
c
re
o
<D
E
<
O CO
o> 8
<D ( 0
CO
to
> DC o)
cc
5
Ul >
2 1 8
g
P
a
Q
~ ©
2
■s
□ «
Q.
a
O co
°i
«i
°
c 0)
o
_
■5
tc
0 III
—
“
»(
—
CO «
O >
2
o5
'5to
to
q. —>;
©
. c ,®
•3= o
°
CO
_> » - o
■2* 2 «
o
T-
i— c P
0)
i _ CD
3
O ) g
Z
500
LU
O
O
"«fr
m ‘uaMOd
o
o
CO
o
o
CM
aaaaosav
O
o
O
1 1
which represent a 5-point calibration including the zero point It is more common
however, to find deviations from linearity at both the highest and lowest proportional
settings, as seen in Figure 12 which shows that the region between 97 and 100% is
inaccessible, e.g these powers are indistinguishable. From Figure 12 it is also apparent
that a linear offset exists while operating at proportional power settings for a given unit.
Calibration produces a region, usually between 30 and 40% and 98/99% where the
percent power settings do have a linear correspondence with the watts of power delivered
to the cavity. Errors increase for power settings below 40% as the result of a) measuring
small differences in small temperature values, b) mechanical errors at low graduation of
power, and c) exaggeration of small timing changes at the low end of the duty cycle.
Thus, it is most practical to operate in the 40-95% range where the maximum error can be
kept to around 7%. A PC computer program (126) is available to aide in the calibration
of a microwave unit It is based on the microwave calibration procedure developed (127,
128) and used in this laboratory.
After the maximum power of the unit has been determined, calibration is
accomplished by measuring the temperature rise in 1000 g of water exposed to
electromagnetic radiation over the power output range from 40-99% for a fixed period of
time, usually 2 min. The step-wise procedure for the calibration of a microwave unit can
be found in Appendix C.
Teflon and polyethylene containers are among the most microwave transparent
and have been used successfully in these calibrations. Potential sources of error in this
determination include using microwave absorptive or reflective containers, not stirring the
water before measuring, and heat loss from the vessel. The container should be
circulated continuously through the field for at least 2 min at full power. Although 2 min
70
CM
0)
CO
in
o>
c
'§
0 tn
5 §
0)
c
CO
3 «ec
V
111
a> 5
5 > o
o co a.
a.
Q
CO
1 «
UJ o c
_i
a
hm
'<
co 2CL
*- O
0s
o
■M
"
.I s
s i
JQ _l
« -o
O c
co
009
o
u>
m
o
o
o
o
O
o
CO
‘uaMOd aaauosev
i
O
O
sis
£3 §3
O) 4^
•»
U- CO
71
represent a compromise between short and long exposures to produce a smaller change in
temperature, 1000 g samples can be exposed for 90-240 s at full power without
observing significant differences in apparent power absorption as seen in Table 6.
Table 6. Measurement of Maximum Microwave Power Output
as a Function of Time
Time, s
Power, W (± 1 SD)
90
570 ± 1
573 ± 7
569 ±19
573 ± 8
562 ± 9
565 ± 6
120
150
180
210
240
Note: For one kilogram of water, the number of measurements
n, 3>n<5. (Reprinted with permission from reference 6,
copyright 1988 American Chemical Society)
When the actual data are graphed, Figure 13, the maximum power output test .
shows a gradual erosion as the test time increases. This is probably not due to
evaporative losses since the vessels are covered, but it does suggest that, even at
relatively low temperatures and small AT, some heat may be lost through the vessel walls
to the surroundings. With increasing temperature, heating became less efficient because
the dielectric loss factor, E", decreases with temperature.
Proportional Power
To make power absorbencies relevant to another microwave unit they must be
corrected by using the calibration determined for that system (6). If the maximum power
610
72
M ‘UBMOd QBSUOSaV
73
of the calibrated microwave unit is 610 W for example, this value should be divided by
574 to give the correction factor between units (1.06 for this example). This ratio should
be multiplied by the absorbed power obtained from the model for a particular acid (for
this example, 250 W, therefore 1.06 x 250 W = 265 W) to obtain the amount of power
absorbed by that quantity of acid corrected for the 610 W microwave unit. Such
corrections are possible only on calibrated units that are linear throughout the power
range because proportional power delivered to the cavity results in proportional power
absorption by the quantity of reagent present The relationship between absorbed power
and proportional power applied, has been shown to apply to all analytical microwave
units tested (6). For microwave units that have biases in both the upper and lower
proportional power region, the correction factor may be derived by taking a ratio of %
applied power for both units in the linear portion of the power range. For example, if
300 W on the 574 W unit is equivalent to 52.2% ,and 300 W on the calibrated unit is
48%, then the ratio of 1.09 must be applied to the power (250 W in example above) to
arrive at 272 W of power for that mass of acid on the 2nd unit and then read off the
equivalent % applied power from the graph.
All microwave systems can deliver variable power to the sample cavity. This is
accomplished by time-chopping the power to the magnetron at full power. It is
frequently necessary to deliver a specific amount of reduced power per unit time to the
cavity to reach or hold specific temperatures. Full-power settings are not normally used
to digest 250- 500-mg biological and botanical samples in 5-10 mL of nitric acid.
Instead, fractional power provides a more controlled method of heating. When using a
partial power setting, one must know the number of watts absorbed at full power,
assuming that the linearity of proportional power has been maintained. The equipment
used to acquire the data presented here has been determined to be linear within 1% of
ideal over the entire power range. With all other parameters the same, a proportion of P
equivalent to the partial power desired is used. For example, at 20% power (115 W)
Equation 1 becomes
(.20)P = CK)(Cp)(m)(AT)
(2.
Since commercial home appliances have a 10- to 15-s time base which results in
full power for several seconds and no power during the balance of the cycle then the
partial power produces alternate heating and cooling that results in a saw-tooth profile of
the temperature curve. For multiple sample decompositions this could mean that some
containers are not equally exposed to the microwave field even if they are rotating
through the cavity. This is caused by the inhomogeneity of the microwave field. It is
more desirable for the magnetron to have a very short duty cycle, such as 1s, so that
when partial power is needed, the magnetron is on for only a fraction of a second and off
for the balance of the cycle. Rapid movement of the samples through the cavity
(360°/20 s) produces more uniform power absorptions among multiple samples. The
resultant heating curve is smooth and without wide variation in temperature as the result
of partial power settings. Multiple samples in such decompositions are more likely to
receive the same amount of microwave energy and therefore to experience more nearly
similar heating. Temperature profiles can be precisely and accurately reproduced in a
calibrated microwave system enabling the transfer of procedures from one microwave
unit to others.
75
Programming
Multiple programming steps are frequently required to maintain or gradually
increase or decrease acid temperature during digestion. Such programming can be
accomplished while the conditions inside the vessel are being observed. If the rate of
temperature increase is not as fast as desired or is not leveling off under some maximum
target temperature, the program can be incremented for 2 or 3 min and renewed at the end
-'O
of each timed cycle. Used in this way, the temperature and pressure measurements
provide feedback information which allows the operator a high degree of control over the
reaction conditions. This technique is especially practical during developmental
experiments.
Two different examples of the power programming technique are demonstrated.
In Figure 14 the temperature is increased rapidly for 3-5 min and then increased more
slowly over the next 13-14 min. An apparent inconsistency in the decomposition profile
of these biological and botanical matrices is the small reduction in pressure observed after
8min while the temperature continued to rise. This has been noted on many nitric acid
decomposition profiles and appears to coincide with a sudden change in the solubility of
NC>2, which is influenced by the total pressure in the vessel. At pressures > 8 atm the
solution's color changes suddenly from yellow-brown to dark green. This increase in the
solubility results in a decrease in pressure in the atmosphere above the sample in the
closed vessel. When the sample container is opened, and the contents allowed to stand,
NO2 slowly effervesces from solution like carbonated water, evolves the characteristic
NC>2fumes, and the solution returns to its more familiar yellow-brown color. The color
change from green to yellow-brown is the result of nitric acid digestion products and has
been observed by chemists after decompositions that were carried out in Carius tubes and
steel-jacketed bombs. These color changes were probably not observed during
uiie ‘gunSSBHd
o
(O
CO
CM
m
>.
CO
3
O
3
>»
a>
oO
i_
©
C .5
**
T
3 -*
CO
15
o
O
O ©
E
q
.E
ra > asz
o
o>C
Q cso
c
*k.
■— •*0
5 o
E
o
<
o
co
CM
£
O)
CD
c/> g
00
O)
"o
5= o
ow
>*
Q.
O
O
ui £ e
S
H 1■H &O
CO E
© Q.
CM
CD
0
O
c
£
£0
O)
It.
5
E
o
>
©
5
o
l>
o
e
2
o
"O c
*■
o
©
•— 1
^
O)
r
n
OL
o
<5
•z °4— x:
z I
i" c J ■o
0
■
200
CO
© w E gQ.
t; © «n £
o
in
o
Oo ‘3UfllVU3dW31
o
in
o
cn a
il £
77
decompositions before translucent Teflon PFA vessels were used for elevated pressures
in microwaves systems.
The approach to 140 °C, where carbohydrate decomposition takes place over a 5
min span is moderate enough to allow the evolution of C02 in a controlled fashion. With
sufficiently small sample sizes, generally between 100-250 mg, the evolution of carbon
dioxide decomposition products is completely contained inside the 120 mL Teflon
digestion vessel and the entire digestion can be accomplished in one step. These
temperatures are accompanied by pressure increases to ~7 atm. At the pressure peak it
can be seen that the temperature then continues to increase with increased power input
and is accompanied by a decrease in vessel pressure as the combination of temperature
and pressure increases the solubility of NO2 gases. The temperature of the digestate
easily approaches 175-180 °C where all but the aromatic ring compounds are completely
decomposed. Such a one step, multiple-stage digestion using power programming is
efficient and above all safe since the final pressures are well below the 140 psi maximum
limit at elevated temperatures. The most stressful conditions of temperature and pressure
are reached at the end of the heating program.
Power programming for digestions in the microwave unit creates a flexibility and
means of control over the preparation step that is unique and welcome from both the
safety and efficiency point of view. In a second example, shown in Figure 15 a digestion
sequence begun at high power setting can rapidly bring the temperature of the
decomposition reagent close to the target decomposition temperature of specific
components known to be in the matrix. Then the power setting is reduced so as not to
create conditions for a run-away heating situation that might cause premature release of
the vapor contents of the vessel. At reduced power settings the final target temperature
78
uiie ‘aunssaud
CO
<o
U)
CO
CVI
CO
0)
T3 >
Q. iZ ■o
CO
CC
CO 0>
$ a
T-
co
O) cl £'=
c P
io ■aD> zso
£ co
0) co
c
a2 E
Q)
d>
c
0 ir>
E O >. £
£ £
uT a.
S
„ s
H
1 o
CO o co
o> = <i>
CD
5 1 ? 'H
sa»
|
a> E
> CO >
co O *f$ ok. o
ok-
to
0 Q. co
0)
1
If
o CO
in D.
CO
fl> £ —•
i3 .S
4^ £
o
o
o
o
o
in
o
.E
*j» IO
U. 0 .
Do ‘aunivuadi/uai
»Vf
79
can be approached slowly so as to provide maximum stress on the vessel at elevated
temperatures only at the very end of the digestion where both the gaseous decomposition
products and the vapor pressure of the acid at that temperature combine to exert
substantial stresses on the caps, threads and bottoms of the vessels. In the example of
this approach, shown in Figure 15, five vessels each containing 45 mL of water and 5
mL of nitric and hydrochloric acids (1:1 v/v) were heated to 170 °C. This took 10 min at
574 W. Power was then reduced to 316 W for maintenance at 175 °C. At no time during
the program did the pressure exceed 100psi.
Microwave Absorption of Acids
Acid dissolution of a sample matrix is governed by many complex relationships
that must be evaluated. The acid or combination of acids is chosen for its efficiency in
decomposing the matrix. In addition, it is usually desirable for the acid to form a soluble
salt with the metal ion of interest. For these reasons nitric, hydrochloric, perchloric acids
are widely used in sample preparation for analytical chemical analysis. Knowledge of the
sample matrix, its major elements and compounds, is essential in choosing the
appropriate acid to assure complete sample dissolution.
Combinations of acids are frequently used in analytical chemistry to effect sample
matrix destruction. For example, hydrofluoric acid is inappropriate for the
decomposition of botanical material. However, if a siliceous component is present,
hydrofluoric acid is added to nitric acid to liberate trace elements that would otherwise
remain trapped with the silica. These combinations must be chosen on the basis of the
chemistry of the sample matrix (129,130).
Another important consideration is the interaction between the acid and the
digestion container. Hydrofluoric acid should be omitted in samples digested in glass
80
and quartz. Although most of the mineral acids traditionally used in decomposition are
good microwave absorbers, other properties, such as the stability of the acid in the
microwave field, its vapor pressure, and its interaction with other acids when used in
combination, must be evaluated before attempting a closed-vessel digestion.
Absorption characteristics
Microwave heating profiles of 5 mL of nitric, hydrochloric, phosphoric, sulfuric
and perchloric acids have been published previously (6). Two additional acids that are
frequently used with soil and geological decompositions to complex the silica are
hydrofluoric acid and tetrafluoroboric acid. Although hydrofluoric acid has a low vapor
pressure and boils at 106 °C, its heating profile is similar to that of nitric acid (6).
Tetrafluoroboric Acid
Tetrafluoroboric acid is used in some geologic decompositions (84,67) for
inorganic matrices that require the attack of silicates and high temperatures. At 227 °C the
partial pressure of tetrafluoroboric acid in the closed vessel was only 5.7 atm as shown in
Figure 16. No decomposition of the acid was observed, and temperatures much higher
than those practical with hydrofluoric acid can be achieved without high pressures.
Equation for Predicting Absorbed Power
Power absorption by small amounts of acids decreases proportionally as the mass
in the cavity decreases as shown Table 7 because some portion of the incoming radiation
never travels through the sample as the waves traverse the cavity (36). When microwave
power absorptions in Table 7 are compared, it is apparent that dilute acids absorb power
more strongly than concentrated acids. This is partly attributable to the larger fraction of
water present in dilute acids and partly the result of ionic heating losses that occur in
in
CM
o
CM
O o
of 5 mL
of
in
co
co
CM
CVI
o
in
o
‘3UniVU3dlAI31
o
in
o
Profile
co
Figure 16. Microwave Heating
T etrafluoroboric
A cid.
min
co
TIME,
w»e
‘3dnSS3Ud
CO
82
solution. Water is a better absorber of 2450 MHz radiation than any of the mineral acids
because the depth of penetration of the 2450 MHz wave is greater for water than for any
of the reagents in the table, since its relaxation frequency may be closer to 2450 MHz
than other acid solutions.
Table 7. Microwave Power Absorbed by Small Volumes of
Mineral Acids and Water
Reagent
Water
HNO3 (16M)
HN03(1M)
HF (29M)
H2S04(18M)
HC1(12M)
HC1(6M)
HCl(lM)
50 mL
Power Absorbed, W ± 1SD
200 mL
100 mL
344 ± 9
184 ± 2
212 ± 3
67 ± 3
231 ±6
148 + 3
138 ± 3
227 ± 4
408 ± 3
234 ± 5
269 ±6
238 ± 12
331 ± 4
173 ±6
190 + 4
287 ± 7
468 ± 5
313 ± 4
332 ± 3
315 ± 15
396 ±8
251 ± 2
253 ± 4
340 ± 5
Note: The number of measurements, n = 5. (Reprinted with
permission from reference 6, copyright 1988 American Chemical
Society)
In laboratories performing a number of routine sample dissolutions, it is not
efficient to digest one sample at a time. Multiple dissolutions can bring the total acid
quantity in the cavity to between 25 and 1,000 g, depending on the individual sample
sizes. The power absorbed for any mass of acid in the cavity can be calculated using a
set of equations derived from the experimental data for the acids and water.
The power absorption of 25-3,000-g samples of water, 1M sodium chloride and
mineral acids does not exhibit a simple linear relationship between mass and power as
demonstrated in Figures 17 and 18 for water and 16 M nitric acid. Power absorption
83
CD
T"
U .
<D
O
o
O
d) ■o C O
C
0)
“aj
o *• o
E
aQ
- <
D
sz
* o
O
c
o
CO
■*- ~o
aj
SZ
o
'u
a>
E §
o CQ
*= o>
<
E
oo
00
O)
T3 g
JC
WATER
O)
O
CO
.O )
'C
>.
Q.
*O I2
C
U)
w . § • £ _ !• O
o
<
d) CDa
to
o
s LU w CO 0)
«-
CD
^
fl) in u .
5 o> O
o
<*-
2
®
£
E
°- ~
—o
n co
a>
0.2
<
= 1
■o 5 ■d)O•*c=
O
CO
« w S O)
2
co ^m O Em
1^
T-
O
o
o
CD
o
o
U)
M ‘U 3M 0d
o
O
O
CO
o
o
CM
„
I-
Q)
05
•0 'S
d) c
TJc
3g!
3* Q.2* h- o £
LL. O
NITRIC ACID, 16M
84
M ‘U 3M 0d
85
continues to increase with sample mass until relatively large sample masses are reached.
For the materials studied to date, the absorbed power rises sharply with increasing mass
until it reaches approximately 500 g, where it levels out, and then increases very little
between 500 and 1,000 g. The measurements of absorbed power plotted against the mass
of the mineral acids and water are presented in Appendix E. These data are of practical
value because it is necessary to know the power absorption in order to predict
temperature conditions in the sealed vessels during digestion.
Because the amount of power absorbed is proportional to the quantity of acid or
water present in the cavity, that power can be predicted for a known mass of acid. Two
models have been used to relate the absorption of power to mass of either acid or water.
Equation 3 is a natural logarithmic based linear equation and Equation 4 is a natural
logarithmic based quartic model of the same data. The actual coefficients A' and B'
(linear model) and A through E (quartic model) used in these generalized equations for
the mineral acids and water are given in Appendix F. The generalized expressions are as
follows: (Susannah Schiller, Statistical Engineering Div., NIST, personal
communication, 1988).
First-order (linear) model:
In (absorbed power) = A’ + B 1ln(mass)
(3)
Fourth-order (quartic) model:
ln(absorbed power) = A + B ln(mass) + C ln(mass)2 +
D ln(mass)3 +Eln(mass )4
(4)
86
The difference between these two mathematical models is the accuracy with which
they predict the amount of power absorbed by a mass of acid or water. Table 8 shows
the average percent prediction error based on 95% confidence limits for both models.
The first-order model is biased and does not adequately represent the data but is provided
as a safety aid that can be used with hand calculators for quick estimates of power
absorption. The fourth-order equation represents the data with greater accuracy and
should be used for prediction of conditions. NOTE: This type of equation is especially
susceptible to rounding errors; it is very important not to round off any of the coefficients
provided for the quartic equation or any intermediate terms. Graphical results for both
models with the data and 95% confidence limits for each model are contained in
Appendix E. These graphs demonstrate the fit for the fourth-order equation and provide
a comparison for the predicted accuracy using the two different models. If greater
accuracy is needed for a particular mass of acid or reagent, then the temperature of the
actual mass of reagent needed should be measured under identical conditions and the
absorbed power calculated from Equation 1.
There is an upper power limit that is dependent on the output of the magnetron
and cavity tuning. For very large samples the power absorption is constrained by the
actual power delivered to the microwave cavity and by the unique conductance and
dielectric relaxation time of that particular acid (36). This upper limit and the mass at
which it is reached will be slightly different for each acid and for microwave equipment
delivering different amounts of power to the microwave cavity. Some power absorption
increase beyond 1,000 g can be attained for sulfuric and nitric acids, but for the others
the power absorption flattens at a mass of 1,000 g. One should not use Equations 3 and
87
Table 8. The Average Percent Error for the First-Order
and Fourth-Order Models
Reagent/concentration
Linear
Quartic
h 2o
15.5%
20.5
7.1
8. 1%
H2SO4.I 8M
HC1-1M
HC1-6M
HC1-12M
HF-18M
HNO3.IM
HNO3.I 6M
13.7
18.4
8.4
5.4
3.9
5.3
10.4
9.5
4.8
12.6
8.1
12.2
Note: At the 95% Confidence Limit (Reprinted with permission
from reference 6, copyright 1988 American Chemical Society)
4 to predict power absorption for masses outside the range used to estimate parameters
for these equations.
These ranges can be seen on the figures in Appendix E. Likewise, the lower limit
at which these models can be used to accurately predict is approximately 25 g. This is the
smallest quantity of acid that could be measured accurately.
Prediction of Temperature and Time
Once a value of P has been calculated for a specific volume of acid, both the target
temperature and the time that microwave power should be applied can be estimated by a
simple transformation of Equation 1 (3).
The final temperature can be estimated from
Tf - Ti + (K)(Cp)(m)
(5)
88
and the time it will take to reach some final temperature can be estimated from
t _ (K)(Cp)(m)(AT)
(6)
Trial and error evaluation can be minimized by using these equations to predict the
conditions that will result from specific power and time exposures of a sample. The
analyst can also decide on appropriate final conditions of temperature for a specific
sample and use these equations to establish the correct power and time settings.
These thermodynamic relationships can reliably predict temperature and time to
within a few percent, for the mineral acids tested, despite small changes in their dielectric
constants, when the parameters are AT < 140 °C or t < 2 min. In addition to acid
decomposition, organic synthesis and other low temperature applications can benefit
greatly from this predicting capability (31,1 32). For a hydrolysis in 200 mL of hot
water the fourth order equation gave an estimate o f425 W absorbed. When substituted in
Equation 4, the final temperature reached in 2 min is predicted to be 84.7 °C. This
compares with a final temperature of 87.0 °C that was actually observed. The equation
underpredicted the temperature by ~2.6% as the result of heat loss.
The absorption of power from the field by acid reagents like nitric acid is the
major influence on the rate of temperature increase. The presence of sample does not
change the absorption characteristics of the acid enough to cause a significant difference
in performance. Because the acids' absorption of microwave radiation appears to be
independent of the material being dissolved, realistic estimates for sample heating time
can be modeled by the acid's temperature profile. If the acid and sample in acid are
digested using the same program, power consumption and heat loss can be predicted for
a variety of samples digested under the same conditions. Table 9 shows a comparison of
89
the time necessary to reach various elevated temperatures for 5 mL of nitric acid and for
four 250 mg samples of biological and botanical materials in 5 mL of nitric acid.
Table 9. Actual Time Required for a Single 250 mg Sample of Biological
Material in 5 mL of Concentrated Nitric Acid to Reach a
Specific Temperature
sample
nitric acid
bovine liver
oyster tissue
wheat flour
rice flour
140°C
time, s
150°C
160°C
170°C
267
239
273
225
237
339
319
338
306
298
408
400
417
369
330
471
468
468
396
363
Note: Irradiated at 144 W for 8 min. (Reprinted with permission from
reference 6, copyright 1988 American Chemical Society).
When nitric acid is used as a model for the decomposition of bovine liver, the heating
profiles, as seen in Figure 19 and 20 are nearly identical. For samples that are smaller
than 500 mg, the time prediction for the oyster tissue is consistently within 2% of the
model, whereas the wheat and rice flours deviate significantly from the model at higher
temperatures.
It is at higher temperature differences (150-250 °C) and longer exposure times (520 min) however, that the actual conditions deviate substantially from their predicted
values. A major source of error is heat loss through the walls of the container. Table 10
demonstrates the deviation of the predicted values from the actual elapsed time for each of
eight samples containing 5 mL of concentrated nitric acid and evaporated Human Urine
SRM 2670, (originally 10 g each wet weight). Only the nitric acid significantly couples
with the microwave power, the mass of the sample is neglected in the calculations. This
90
wie ‘a a n ssa u d
CD
CO
o
<
©
*o
o
CO
CO
E
15
o
E
Q>
.C
o
c
coa
*C
(D
E
u>
<
Q
Oo>)
(D
jc
00
CO
c
S=
E
O
uT
144 WATTS
2
1mm
CL
CO
.?>
Q .
O
O
CD
®
C
(H
s
I
o>
CD
o
cm
■§
£
|
>
-i=
«
c
o
’w
CO
<N
5
1
s
s.
£
I
*D
O)
0)
3
O
CM
o
U)
O
Do (3U niV d3dlA I31
o
lO
O
i
O)
CD
Q.
(D
oc
91
wie ‘3bnSS3Ud
00
to
IO
^
C
O
C
M
•a
o
<
a
IHI
U.
++
00
z
**o
-1
E
to
c
1
uT
2
H
C
M
O
O
o
U)
o
Do
‘3dniVU3d!AI31
o
L
O
O
to
•*o
o
o
k_
Q.
>»
®
'5
o
CO
"ni
u
E
CD
-C
O
c
of
.o
*cz
(D
E
<
CO
a>
T~
00
s
z
#o>
*d
>,
a
o
o
<D
©
o
O)
c
C
£
a>
«■«
CO
a)
a) •
X <U E
> o
a> 3
c
>
o
CO
’(/)
5 a> CO
o1— c 1
®
o o> Q.
co
-E
2
4-4
«*— i
o o TJ
®
CM
c
o 'C
Q
a.
k.> E ®
cc
3
O) oo
LL CO
92
Table 10. Predicted and Actual Time for Urine Sample to Reach
a Target Temperature
Temperature, °C
Linear
Time, s
Quartic
Actual
110
130
150
160
170
180
183
64
80
95
102
110
117
120
69
86
102
110
118
126
129
54
72
81
102
113
144
150
Note: Irradiated at 574 W. (Reprinted with permission from
reference 6, copyright 1988 American Chemical Society)
table of predicted times at given temperatures was calculated by using the heat capacity
from Table 2, the power absorbed for eight samples (7.2 g each or 57.6 total g in 60 mL
Teflon PFA vessels) of acid using the equations 3 and 4 relating power consumption for
nitric acid, and the thermodynamic equation solved for time, Equation 6.
The values agree within experimental error for up to 1.8 min when a negative bias
in the predicted value becomes apparent. This bias is caused by heat loss through the
container walls. Such losses in Teflon PFA containers may reduce the effective power
provided to the acid by as much as 50% in 6 min. At a given microwave power setting,
the heat loss results in higher predicted temperatures than are actually measured.
Conversely, heat loss results in underpiediction of time to reach a given temperature.
The magnitude of both biases will increase with longer time or higher temperatures. For
this type of container the reproducibility of this heat loss under exactly the same
conditions is within several percent when measured between different lots.
These deviations vary with the heat loss of the vessel and thus become smaller
with greater thermal insulation of the vessel. Actual temperature conditions in the thick-
93
walled Parr bomb probably closely follow the theoretical predictions over a much wider
range of temperatures and for longer times because it is better insulated and has only a
fraction of the heat loss of the Teflon PFA vessels. In an experiment designed to test and
evaluate the mode of failure of the high pressure vessel, the time for 5 mL of nitric acid to
reach the 82 atm (1200 psi) pressure limit of the vessel was calculated from Equation 5.
At 1200 psi the temperature of nitric acid is between 219 and 256 °C (131). Using earlier
published estimates of the efficiency of nitric acid absorption (3), 5 mL of acid reaches
210 °C in 180 s and requires 150 s more to attain a temperature of 261 °C which is still
below the maximum use temperature of Teflon. Pressure release was accompanied by a
loud report and decompression of the microwave cavity at 5 min and 31s. When the
inner temperature of the Parr microwave bomb was still well in excess of 200 °C, the
outer jacket temperature was only 37 °C. By comparison, when the internal temperature
of the Teflon PFA vessel was 132 °C, the temperature of the outer wall at the bottom was
~ 92 °C.
The previous calculations should prove useful in designing specific conditions for
this microwave digestion vessel. Real-time temperature measurements in the Parr vessel
have not been obtained, because probes cannot be inserted without compromising the
integrity of the protective outer casing. Accurate use of these transformed equations is
limited to the initial increase in temperature, when uninsulated Teflon vessels are used.
Within these restrictions the calculations provide a method of determining the amount of
applied power necessary to reach a particular temperature or the approximate time this
temperature is obtained at some applied power. However, no predictive information can
be obtained about the amount of power required to sustain the temperature. Such power
requirements are dependent on the equilibrium that is established between the heat input
94
and the heat loss at a given temperature. These predictions also have application in the
safe use of the closed-vessel microwave technique. Unsafe conditions can be prevented
by using calculations to predict a maximum temperature or minimum time to reach given
conditions.
Microwave Absorption of Acid Mixtures
Acid combinations for microwave decomposition are practical for the same
reasons they are used in other wet-ashing procedures. Additional benefits accrue from
the closed-vessel microwave dissolution technique when the acids are heated to relatively
high temperatures in combination with one another. Acid combinations are chosen for
the ability of each acid to effectively decompose individual components of a particular
matrix. They are also frequently chosen for the effectiveness of one of the acids as a
digestion agent and for the resultant aqueous solubility of the complexed elemental salts
formed with a second acid during dissolution. This is true for all decompositions, both
open and closed.
Nitric, hydrochloric, and hydrofluoric are relatively low boiling and have large
accompanying partial pressures, whereas phosphoric, sulfuric, and fluoroboric acids
have low partial pressures at comparable temperatures and are relatively high boiling. By
combining one acid from each group these properties can be used advantageously to
produce a mixture with a partial pressure somewhat less than the pressure of the lower
boiling acid. Such acid mixtures are not ideal solutions and deviate from Raoult's law.
The result is a useful vapor pressure lowering of the solutions.
Effective combinations that have been investigated in the microwave system
include nitric and phosphoric acids for tissues (4,133, Patterson, K., Beltsville Human
Nutrition Research Center, USDA, personal communication, 1986) nitric and
hydrofluoric acids for biologicals and botanicals (14,15,48), (tetra)fluoroboric acid with
nitric for metallurgical matrices (67) and with a nitric and hydrofluoric acids for simulated
nuclear waste (68), and aqua regia for mine tailings and geologic samples (67). Alphaalumina has been successfully dissolved in 1:1 mixtures of phosphoric and sulfuric acids
(6). Mixed acid systems based on nitric acid with phosphoric, hydrofluoric and
hydrochloric have been evaluated for use in the microwave environment (6). Sample
decompositions in mixed acids will be discussed in more detail in the chapter on
applications.
Power Absorption of Acid Mixtures
Dissolutions of inorganic material that require elemental analysis frequently
necessitate complete decomposition of the matrix in order to liberate analytes of interest.
When that sample material is a soil, a single acid is often not adequate to produce
complete mineralization and mixtures of nitric acid with either hydrochloric or
hydrofluoric acids are often used to produce the desired result. In microwave assisted
wet-ashing methods it is essential to know the power absorption characteristics of such
acid reagents in order to determine the power needed to reach a specific decomposition
temperature. Such a problem would appear to be straightforward if heat capacities of
mixtures of concentrated acid solutions were routinely or easily available. Direct
estimates of the power uptake of mixed acids could be made by having values for the heat
capacity of mixed electrolytes in solutions. Such data are limited (134-136) and the heat
capacities of mixtures generally found in the literature are primarily for salts of common
ionic solutions and mixtures of infinitely dilute solutions, neither of which are suitable for
our purposes.
96
At present, the development of microwave applications using mixed acids is
restricted to actual measured results. High pressure acid dissolution with mixed acids has
been conducted in Carius tubes and steel-jacketed bombs where direct measurement of
actual temperature and pressure was difficult. Therefore, mixed acid systems have not
been well characterized for temperature and pressure.
A definitive experiment, the direct measurement of the power absorbed, heat
capacity data notwithstanding, is presently unavailable. It would be ideal if the energy
transferred to the reagent mixture could be measured directly as power uptake in watts. A
device to measure absorbed power in watts does not yet exist either. Those methods that
are presently available are indirect and require an energy balance between the output and
what is attributed to the sample uptake plus the unabsorbed power, usually reflected
power. It is not possible however, to measure the remaining power loss as there is no
suitable way, at present, to perform such measurements in fields of 600 W of average
power. It was assumed that the total output power from the magnetron source was
constant and that a mass of sufficient size would absorb all the forward power. The
small amount of power absorbed by the doors, walls and floor was assumed to be
negligible (Goetchius, R., CEM Corp., personal communication, 1986).
It would be helpful to have a means of estimating the power absorption of
mixtures and most desirable if they behaved in a simply additive fashion, eg. the power
of the mixture is simply equal to the linear sum of the individual components. It was
therefore, assumed that the power absorbed by the mixture was additive and that it was
equal to the sum of the individual power absoiptions using the heat capacity value of each
acid as if it were alone.
97
Estimates of the true power absorption of mixtures were calculated by taking the
linear sum of the fractions of each as an individual component of the mix using the Cp of
the single material and and the measured AT observed when the two materials were
mixed.
[K x M HF x Cp HF x ATmix] [K x M HNO3 x Cp HNO3 x ATmix]
Pabs = ---------------- 120----------------- + ------------------- m ---------------------
(7)
For a combination of acids to be a good model for the prediction of power
absorption expected in a mixed acid system several criteria should be met. First, the
individual acids should not react chemically, but rather simply mix in solution. In
addition, the individual acids should be well characterized, e.g. their microwave
absorption behaviors should be available for scrutiny at the combinations employed, and
the mixture should have demonstrated some utility to the practising chemist. Nitric acid
mixtures with hydrochloric acid form aqua regias and generate nitrosyl chloride (NOC1) a
reaction product that is one of the active agents in the mixture. It is no longer a simple
mixture and therefore unsuitable as a model. Nitric and hydrofluoric acids were selected
because they form a true mixture; they do not react chemically in solution, and no heat of
mixing is observed. The absorption of various masses of each acid individually has been
documented in the literature (6) and both acids have been extensively evaluated as acid
media for microwave dissolutions (3). The heating profile of a representative mixture of
nitric and hydrofluoric (5:3 v/v) is shown in Figure 21 and demonstrates the pressure
advantage gained in mixing the two acids. Since the heating and cooling curves retrace
themselves in the temperature and pressure profile shown in Figure 22, heating the
mixture to temperatures routinely used in digestions with these acids does not cause them
to decompose themselves or each other.
98
une ‘3dnSS3Ud
CO
o
W
>»
0
‘o
o
CO
75
o
E
x:o
a
E
aJ
o
<o
T“
•f-
0
E
<
o00o
o
o>
a>
c o
k. >
E a.
>
Ui O)
_
2
I-
CD
c co 0o
c
S
£
a)
X (0 *£
£
a)
in
.c
O)
'C
a!
o
o
T3
E
o
>
s i
CM
1
*
i
3
a o
. o
•c
o
*V)
«/>
0
Q.
T~ *-
CM ^
o) x
Is
U-
o
o
o
in
o
Oo ‘3HfUVH3dW31
in
o
o
(0
T3
0
Q.
0
a:
S
Xs
CO
O
o
UJ
EC
o
o
D
C
U
J
O
IO
CM
ui)e ‘3dnSS3Hd
<
a
2
UJ
H
of reference
o
10
with permission
6, copyright 1988
American
of
16
mL
Chemical
■
(Reprinted
S ociety)
»"
C
M
Pressure Profile
Acids (5:3v/v).
-1
Figure 22.Temperature and
of Nitric and Hydrofluoric
99
o
o
100
Quite by accident, the choice of these acids allows a built-in check on the validity
of the assumption of linearity since both nitric and hydrofluoric acids each absorb nearly
the same amount of power at all comparable volumes. Although there is a disparity in
density it is compensated by the appropriately-lower heat capacity of nitric acid. That is
smaller masses of hydrofluoric acid absorb similar amounts of microwave energy to
comparable volumes of nitric acid because the heat capacity of hydrofluoric acid is much
larger than that of nitric acid. Temperature change is the important variable for estimating
the power absoiption. The more dense material, nitric acid, has smaller heat capacity,
Cp = 0.5788 as compared to hydrofluoric acid, where Cp = 0.7034.
However the heat capacities are apportioned, or however the weights are ratioed,
the calculation of absorbed power approximates that of either one liter of nitric acid or one
liter of hydrofluoric acid. If Equation 1 is rearranged using SpG x V = mass for t = 120
s, then AT = constant x P/SpG x Cp. Thus, when the absorbed power is the same, the
heat capacity compensates for less dense material and ATxCpxSpG = constant. In fact,
they are within 2% of one another for all 1liter mixtures.
As seen in the table below the mean value of the absorbed power determined for 1
liter mixtures of HF/HNO3 are indistinguishable from one another and from either pure
acid.
The table shows clearly that there is a negative bias in the data, i.e. there are no
positive deviations from the expected value and that only the 20/80 HF/HNO3 mixture is
coincident with the range of the individual acids. Calculations of the power using
proportional contribution underestimates the absorbed power by 1-7 %, ranging from
405-426 W. Similar observations were made on a more useful mixture volume of 200
mL seen in Table 12. Again, only the 20/80 mixture overlaped the means of the pure
101
Table 11. Microwave Power Absorbed by of One-Liter Mixtures
of Hydrofluoric and Nitric Acids
Volume ratio
watts of each
HF
HNO3
P. abs
15.05
-
432± 8
432 ±8
1371
14.86
86± 2
340+6
426 ±8
40/60
1323
14.10
163± 6
242± 4
405 + 7
50/50
1304
14.40
209± 3
207± 3
416 + 5
70/30
1251
14.46
292± 8
124± 4
417 ± 10
100/0
1180
14.62
429+ 15
-
429 ±15
HF/Nitric
Wt of Mixture, g
AT, °C
0/100
1420
20/80
Note: Irradiated at 574 W for 2 min.
Table 12. Microwave Power Absorbed by 200 mL Solutions of
Hydrofluoric and Nitric Acids
Volume ratio
watts of each
HF/Nitric
Wt of Mixture, g
AT, °C
HF
HNO3
P. abs
0/100
284
29.55
-
315±14
315+14
20/80
275
29.46
62± 1
250+3
313±4
40/60
265
27.10
116± 6
173+8
288± 12
50/50
262
27.26
146± 2
145± 2
292±4
70/30
254
26.78
201± 8
89±4
290± 10
100/0
236
28.44
313± 5
-
313± 5
Note: Irradiated at 574 W for 60 s.
102
liquids; all of the others overlap each other, but not the pure liquids and, again, are
negatively biased
When actual data originally determined for equal 500 mL volumes of both acids
were combined, the individual power absorptions of a hypothetical, 1:1mixture, had a
value of 712 W. This suggests that the power determined for a mixture is not simply the
sum of each individual power absorption, but rather is a complex combination of their
relative individual contributions. Fractional combinations using the 4th order equation
for the single acids tend to give even lower estimates of the absorbed power than the
simple linear combinations, is seen in Table 13. Such estimates might be advantageous
in that no experimental work need be done, but as the table shows they do not overlap the
estimates based on measured AT of a mixture and would tend to far underestimate the
power absorption of the mixture.
Table 13. Microwave Power Absorbed by Mixtures of
Hydrofluoric and Nitric Acids Estimated From Quartic
Equations and by Linear Combination
HF/HN03
Mixture
Fractional Estimate
4° order
Linear Combination
Using AT
20/80
40/60
50/50
70/30
385± 16
363± 16
357± 16
356± 15
426 ±8
405 + 7
416 + 5
417 ±10
Underestimation of the power absorption in mixtures is probably the result of the
negative bias in the measured heat capaciity of dilutions when compared with a linear
extrapolation (121,137). If heat capacity were linear, and that is the underlying
assumption of the additivity concept, then all heat capacities should fall on a straight line
103
between 1 and 100%. In fact, because they lie below the line, underestimates of the
power absorbed by a mass for a given rise in temperature are the results normally
expected. Thus, if the true heat capacity data were available for use in these equations,
the estimate of the power absorption might approach the values actually determined. The
deviation from linearity is <8% which approaches the magnitude of error ~6.5%
observed in the measurements of AT for the nitric/hydrofluoric acid mixture.
Mixtures of nitric and hydrofluoric acids behave very much like dilutions of acid
and water and experience a lowering of the heat capacity, Cp. Therefore, mixtures of
acids which do not react in solution, but which retain their individual character,
experience a depression in their combined heat capacity value and absorb less microwave
energy than predicted by a linear extrapolation of the individual component absorptions.
Despite the slight negative bias, the power absorbed by a mixture of acids that meets the
criteria of a true solution can be usefully approximated by the method of linear additions.
Absorption Efficiency, Heat Loss and Power Rejection
Absorption Efficiency
During the course of any microwave heating program, many factors influence the
final temperature attained by the solution; the heat capacity of the reagent, the mass
present, the amount of power applied, and the irradiation time are among the most
important and Equation 1 correlates these parameters with AT. There are other factors
that account for dielectric heating, but do not appear in the fundamental power absorption
relationship; the dissipation factor and tan S, both of which are frequently dependent. In
order to have some measure of relative heating efficiency of the mineral acids, a
comparison was made between the power absorption of a fixed quantity of each solution
to that of water because that was the standard use to rate the power output of the 2450
104
MHz magnetron. Since the full power (574 W) is constant for this experiment and 1 L of
each acid was irradiated for 122 s, die amount of microwave energy absorbed by each
solution was calculated as the product of the change in temperature, the specific gravity,
and the heat capacity. The resulting value was compared with that of water and a table of
absorption efficiencies at 2450 MHz was created to express this relative ability of the
solution to heat at this frequency. Table 14 shows the absorption efficiencies of various
acid solutions along with their heat capacity constants and ionic conductivity. Absorption
efficiency is clearly dependent on a number of factors because acids with small heat
capacities are not necessarily among the most efficient, although it is clear that sulfuric
acid is the most efficient absorber at 2450 MHz. Other factors besides the
thermodynamic heat capacity and specific gravity expressed in Equation 1 are responsible
for the microwave heating observed.
As expected, materials that are largely dissociated in water, such as 1M HC1 and
1M nitric have large equivalent conductances and heat well. Concentrated sulfuric acid is
not highly dissociated, but has a large doubly-charged ion that carries current and so heats
well. It is well known that polar liquids heat much more rapidly than non-polar liquids
(137) in the microwave field; perhaps the minimally dissociated sulfuric acid molecule has
a strong dipolar heating component.
A series of heating curves are shown in Figure 23 that demonstrate the
relationship between cavity loading and the rate at which the load heats. As the number
of vessels containing 10 mL increases, the larger masses take longer to heat. Such data
are useful because the abosrption efficiency can be determined from calculations made
during the initial heating portion of the curve which is the first few minutes where the
heating rate is rapid and before the vessels have had a chance to lose heat appreciably.
105
s o >—i o o o o m
Table 14. Microwave Absorption Efficiency of One-Liter Acid Solutions at Full Input Power
1**^• •
g 8>
■8 S
S &
W
h-
^
■ i i i i i i i i
CO «-i t ' « l O ' D h O O
0 \ O i O N O \ ' t H g j o o
o \ a \ o \ o q r ^ t - ^ t ^ v o v q
o o o o o o o o
c o o t s v o - « t c s r ~ ' n r >
on
cr\/a J*
O
r-
O N «
o o «o o •r - •r ~ •r - •' - r«' -
oo oo h
m
*—
vo
t-~ f " v©
in h o
I O O t—< > — l < —I
+i
+io c+io m
+i +i ±| +i
+j
oo oo+1
T
r -to
^H
f ' V DtscooNc C
s gSoro^
< P
i lh
53
I
£
%
trH
c3
c/a
gO
O
•js
8
"a3
'Q
• #»
U
0in
(S
a
o
■!
O
o
w OO
■
m
in
r-~
0 (U
1t-i i&
MM
11u
CO (j
&°-a
o
O
(i
co
o
c
o
U
2 *>
•^r
I
cd
oo
a
1
-H O o
© © ,-s
mm ©
© £ «0 ©
in c o c o m om mo
w
co co
3
*
z oC
VO ON OO s o
t" 7 1O r j
-u o
$ co u
+
VO
oo Ov
C O Ov
O n On
O O
CO O
©
rH
-
©
t—
H
CO
in o
VO
C O Ov
d
o
ts
oo ©
i-H
rH
§ § k
00
r-~ 00 C O
o
ON in r "
<s
o
o
,&&
k*
r—
ic o t—
i m
co i
o
NO NO'
E5 £
U U
I
&
u
cD5
JS
1
in
oo 00
oo H
in
o
o
C O CM 00 oo o
©
»—I 1—1
i-H
.-H
00 - - !S g$ £| *>
00
g
gp
<«
05
co co
PuOO
'z B B x R S e k e
co
g5
106
o
'Z
00
o
sz
o
o
V.
•o
_l
>.
X
o>
<o
(0
(A
0)
a
E
<a
u>
05
c
I
llT
s
p
CO
CM
(A
a>
(A
(A
0)
£ >
1-
a>
(A is
*to 2-I
x
■a
o> c
c «
o5 a>
S f
CO w
01 c
0) ■“
3 -O
o
o
CM
o
o
Go ‘3UniVd3dlAI31
o
to
O
O) o
il <
107
The empirical equations used to determine the temperature of a mass of reagent exposed
to microwave energy function effectively only in this region. The influence of heat loss
on the final temperature is not accounted for in the calculations. From the graph it is
possible to see the bending of the curve where cooling exerts a major influence on the
temperature observed. As was seen in the Tables 9 and 10, there is a negative deviation
from the true temperature at higher temperatures and longer times. That is, the observed
temperature is always less than predicted from the equation as the result of heat loss.
Besides the influence of microwave frequency and the nature of dielectric heating
in a material, other factors such as size and shape of the individual mass, total cavity
mass, and % applied power influence the apparent absorption of energy experience by
materials in a microwave field. Reduced absorption of energy occurs as the resonance
frequency moves away from 2450 MHz and as the polar and ionic character of solutions
diminish. This section is concerned with absorption efficiencies of comparable volumes
of acids at full and reduced power as well as for small and large volumes of solutions.
Such information is necessary for efficient decompositions.
From heating rate data (41), a comparison was made of the absorption efficiency
of acids frequently used as dissolution reagents. Absorption efficiencies were determined
by comparing the calculated power absorbed during the first 90 s of heating to the value
of the input power to derive at a fractional value for individual acids and mixtures
normally used to digest a specific matrix. As the number of vessels containing 10 mL of
acid is increased from one to 12, the absorption efficiency of both single acids and their
mixtures increases with the mass. It can be seen clearly in Figures 24 and 25 that a
plateau is reached at 10-12 vessels. Alt is also obvious from the curves that the greatest
reduction in efficiency occurs when there are fewer than 6 vessels; it is not efficient to run
CO
CO
0) E
■o o •
o
■o
CVI
<
O)
c
CO E
3
TJ '5
'> c
CO ( O
L
L
■o oo £ •
CO rf
o
co co
>* w M
O ® .2
c
.2
'5
2
to
® *5:
ui jq
o
E
_
c
o
o
co
n
<
CM
o
in
o
o
CO
o
o
% ‘A0N3l0lddB NOIidUOSQV
o
O
3
3
0)
Z
5
o
© GL
: i
° i
= x
CM .2
co
S s
a>
3 ©
O)
u_ CO Z
109
CO J
0) E
wo
CM
CO T“
u.
CM
2 o>
O'
3< -j=
c £
■o CO - »
S e t
■
.5 o
CO
_l
HI
CO
co CO
IU
>
0E
<0
s o «,
o w
° © «
O c
>* C
O3
a C
0>
c
0) > a>
■ 5 -5
l
CO U
CD £
UJ °u
53
O
<D
c -Q*
o E a>
5
s-z£
| 2 E
CM
in ° *CO
CM
e l
)
+0
■>
.1 1 o
UL UL
o
io
o
■O'
O
CO
o
CM
% ‘AON3IOIdd3 N O Ild U Q S eV
z
110
just one vessel in the cavity. For all acids and mixtures, there is only a modest increase in
efficiency realized between 9 and 12 vessels. Ten vessels may be the optimum loading
configuration for these acid combinations. Except for nitric acid and its mixture with
sulfuric acid, the greatest efficiency is realized with the largest mass.
Large temperature increases are accompanied by heat losses and changes in heat
capacity; neither condition is considered in Equation 1. Empirically one observes that
very small samples get hotter than large samples for the same power setting and exposure
time which seems logical, since there is less mass to heat. However, small acid quantities
actually absorb proportionally less power than do large quantities because they absorb
less energy from the wave. For example, at an exposure to 574 W for 2 min, 200 mL of
6 M HC1 increases 50 °C, whereas 500 mL of 6 M HC1 increases only 35 °C.
Energy absorbed is a product of the heat capacity of the material, the mass and the
change in temperature that results from energy absorbed.
For 200 mL of HC1
= 0.718 cal g-1 °C-l x 200 mL x 1.18 g mL-l x 50 °C
the Energy absorbed = 8472 cal
and for 500 mL
= 0.718 cal g-l °C-! x 500 mL x 1.18 g mL-l x 35 °C
the Energy absorbed = 14827 cal
14827 -*■8472 = 1.75 times more energy absorbed by the larger sample. Thus, larger
samples are more efficient absorbers of microwave energy.
For more nearly typical volumes used in actual sample digestion, the reduced
efficiency of small volumes is even more marked. One vessel with 10 mL and nine
vessels with 10 mL each of nitric and sulfuric acids (1:1 v/v) absorb 88.5 W and 291 W
respectively, at 646 W. This translates to an efficiency of 14% for the smaller volume and
45% for the larger volume. Here, the 10 mL in each of 9 vessels is nearly four times as
111
efficient as 10 mL in one vessel. Except for decomposition research or material
characterization, single sample decompositions do not make efficient use of the
microwave energy availableo in the cavity.
The absorption efficiency of a given mass does not change with different vessel
combinations. One hundred and twenty mL of nitric acid distributed in 6 vessels of 20
mL each absorb the same as 12 vessels each containing only 10 mL of the same acid.
Digestions that include less than six vessels may not absorb efficiently unless the total
mass is increased to compensate for the fewer number of vessels. If, however, the vessel
volume is reduced from 20 mL to 5 mL, and the number of vessels is kept constant, small
but measurable reductions in efficiency are observed in some of the acids, as seen in
Table 15.
Table 15. Microwave Absorption Efficiency of Individual and Mixed
Acids as a Function of Vessel Volume
Absorption Efficiency, Percent
Vessels
Volume
6
6
6
12
20 mL
10 mL
5 mL
10 mL
HNO3 HC1
41.0
40.0
42.0
39.0
47.0
37.0
34.0
43.0
Aqua Regia
H2SO4:
HNO3 (1:1)
Aqua Regia:
HF (7:3)
39.0
37.0
25.0
41.0
42.0
41.0
42.0
46.0
35.0
26.0
31.0
Nitric acid, by itself and in combination with sulfuric acid, does not change its
absorption efficiency, whereas aqua regia and its combinations experience a marked
decrease when the volume is reduced. These mixtures contain both nitric acid and
hydrochloric acid. Something is clearly different about the microwave absorption
112
efficiency of nitric acid since by itself and in mixtures it does not follow the expected
declined inefficiency seen in hydrochloric acid. Perhaps nitric acid is a weak dielectric at
all masses. Since most digestions are not run at maximum power it is important to have
some idea of the reduction in performance expected as the result of reduced or
proportional power output. Inspection of the data in Table 16 suggests that for volumes
from 65-100 mL of all acids and combinations decreases in absorption efficiency of 1025% can be expected when fractional power settings below 40% are selected for nitric
acid and many of its mixtures.
Reduced absorption efficiency is apparent when small volumes are compared with
larger volumes such, as 1-L of nitric and hydrochloric acids. Additional attrition occurs
with proportional power and becomes especially significant below 40% applied power.
As suggested in the section on proportional power, differences in small numbers result in
large errors and the mechanical offset of small % applied power is unreliable in this
region.
Microwave absorption efficiency of similar masses of water and nitric acid, as
seen in Table 17, shows little dependence on the power input but rises dramatically when
the mass is increased to 90 g of HNO3 and 200 g of water. Comparison of the
absorption efficiencies of large masses of aqua regia and water with trace nitric acid to
those of small masses of HF, HCIO4, HNO3, and H3PO4 makes it clear that the best
efficiencies are realized at the largest masses and that efficiency drops sharply when the
mass falls below 10 g.
113
Table 16. Microwave Absorption Efficiency of Mineral
Acids and Their Mixtures as a Function of
Proportional Power
Acid
Mass, g
Power In, W
% Efficiency
HC1
66
635
420
270
109
37
39
34
34
Aqua Regia
71
648
417
266
112
37
32
35
30
Aqua RegiarHF
(7:3 v/v)
71
651
422
266
26
27
24
HNO3: HF
(1:5 v/v)
73
648
424
277
28
24
29
HNO3
85
651
490
357
235
175
40
36
32
35
32
H2S04:HN03
(l:lv/v)
98
648
424
261
41
40
32
Note: All experiments consist of six vessels each with
10 mL of acid
Although better overall efficiency is achieved with larger masses good results are
obtained for sample dissolutions when the cavity mass is between 100 and 200 grams.
Microwave units that are nominally 600 W may be power limited, as will be seen later for
90 g of acid, whereas units of 800 W will be able to manage 200 g with ease. A balance
114
Table 17. Microwave Absorption Efficiency of Mineral Acids and
Their Mixtures as a Function of Mass and Power Input
Reagent
Water
H20:HN03 (9:1 v/v)
HN03
HN03:HF (20:1 v/v)
HN03:H3P04 (1:1 v/v)
H 3PO4 (90% w/w)
H 3PO4 (100% w/w)
H3P04:H2S04 (1:1 v/v)
HF
HCIO4
Aqua Regia
Mass, g
Power In, W
%Efficiency
5
20
20
25
250
7.2
7.2
90
90
10.5
13
8.6
6.3
11
6
8.5
179
212
144
431
144
574
144
574
373
574
574
212
212
144
92
212
212
350
26
28
27
23
79
7
7
31
28
33
22
15
22
33
13
11
63
Note: Average of > 4 measurements
is necessary between the additional time needed to attain adequate digestion temperatures
and the power requirements that a larger mass necessitates when processing more
samples.
As Mudgett showed for oil and water (103), the energy transfer efficiency in a
2450 MHz cavity is clearly mass dependent, and independent of the unit and maximum
power because, as seen in Table 17 and in Mudgett's data, 200 g of water have
efficiencies of 79 and 80%, respectively. Although the efficiency increases for larger
volumes, only small gains are realized on going from 500 to 1000 mL (3).
An equilibrium can be reached where the power supplied to the system is exactly
compensated by the heat lost from the vessels. When this happens, a target temperature
115
can be maintained. In such a situation, heat loss is constant and balanced by the power
input at the given target temperature. Multiplying the efficiency of each acid or
combination by the number of watts required to sustain that temperature gives a rough
mass-dependent estimate of the heat loss. The power output requirements for the
maintenance of different target temperatures for hydrochloric and a sulfuric and nitric acid
(1:1 v/v) are shown in Table 18. As expected, the power requirement drops as the
maintenance temperatures decrease. More power is required at higher temperatures as the
result of greater heat losses and perhaps because the dielectric heating measured as
absorption efficiency may actually be decreasing at highest temperatures. Knowledge of
the heat loss can be used to construct a digestion protocol that is specifically tailored to the
number of vessels and the volume of acid normally needed for a particular dissolution.
Table 18. Microwave Power Output Requirements for the Maintenance
of Target Temperatures in Mineral Acids
Acid
Hydrochloric Acid
Sulfuric:Nitric
(1:1 v/v)
Temp., °C
Power In, W
Heat Loss, W Vessel"!
170 ± 2
149 ± 2
106 ± 1
249 ± 3
190 + 3
166 + 2
221
92
37
278
300
114
37
17
7
45
25
19
Note: All data based on 6 or 12 vessels each containing 10 mL of acid.
Heat Loss
When Equation 1 is used to calculate the absorption of energy by measuring the
heat production, the conservation of energy conditions (125) for thermal heating in the
microwave requires a balance as shown in the following equation:
116
the Microwave Power Absorbed by a Mass in the Cavity = [increase in thermal
energy of the load] + [increase in thermal energy of the container] + [energy lost to
surroundings from the container] +[energy lost directly to the cavity walls, door and
ceiling].
(8)
In our experimental setup there is no energy loss by conduction to any cavity shelf
or tray and none lost by evaporation or radiation from the surface since the vessels are
closed. To use Equation 1, it is clear that all of the terms on the right hand side, except
for the first, must be equal to zero. It is also assumed that no microwave energy is
dissipated into the walls of the Teflon container. This, too, is a reasonable assumption
based on the thermal characteristics of the Teflon. Errors in the absorption calculation are
therefore caused by the extent to which any of these factors is non-zero. The energy lost
to an empty cavity has been estimated to be about 50 W (Goetchius, R., CEM Corp.,
personal communication, 1986). Since that is a constant condition of the cavity it may be
neglected in calculations made in the same unit. Heat loss is a function of the specific
vessel geometry its wall thickness and heat capacity of the material. Since all of the tests
and measurements of heat loss were performed in the same vessels with the same mass
loading the energy lost from the containers should be similar. It, too, is considered zero
when estimating the power absorption of a specific mass. For accurate power absorption
measurements close control of the temperature in solution and minimal variation in
temperature are necessary. If an estimate of the energy lost from the container as heat
could be determined, then this information could be added to the right hand side to
balance the equation.
Teflon is an excellent insulator and a poor conductor of heat Its thermal
conductivity, k= 6.0 X 10'4 cal cm-l cm^ s-1 O 1 (79), is comparable to other plastics
117
and similar to wood whereas glass and ceramics have heat transfer quotients one order of
magnitude larger. Metals are 4 orders of magnitude larger still (141). Inside the Teflon
vessel, reagent molecules couple directly with the radiation and heating occurs
everywhere at once in solution so that the container is heated by conduction from the
inside out. Although Teflon’s insulating properties reduce the loss of heat somewhat, the
thermal mass of a 120 mL vessel still requires an energy input to maintain the solution
temperature because the vessel is constantly cooled by air at 24 °C brought into the cavity
by a variable speed exhaust fan. Since air is a poor capacitor and does not absorb the
2450 MHz wave, the air in the cavity is not heated by microwave energy when the
magnetron is on. Digestion vessels loose heat by forced air convection and this
continuous loss of heat at the outside walls of the vessel is not obvious until the vessel
has attained temperatures of >120 °C, or when the vessel is heated for periods longer than
3-4 min. This temporal factor was seen earlier in the Table 10, because heat loss was not
considered. A two-stage power program could be run to keep the temperature constant
by taking heat loss into account, or such losses could be estimated for an average
digestion volume in a single vessel as in Table 18 and applied as a correction factor to
Equation 1. The problem then, is to determine the rate at which heat is lost from the hot
vessel to the cooler cavity. Since the rate of heat loss at target temperatures has been
shown to be constant but increases with increasing temperature, it is necessarily of
greatest consequence at higher temperatures because most digestions are usually run
between 140 and 180 °C.
The heat loss was evaluated by modeling the cooling of 10 mL of distilled DI
water in a 120 mL Teflon PFA vessel normally used for digestions. Because the depth of
penetration of the 2450 wave is ~2.5 cm and the diameter of the vessel is 4.7 cm rapid
118
heating occurs as the water reaches 180 °C in ~ 7 min. For cylindrical shapes with small
diameters (103-105), as we have here, heating occurs from all directions at once, thus the
entire vessel in contact with liquid is essentially at a uniform temperature at any given
instant. Unlike conductive or convective heating on hot plates and in ovens, a transport
phenomenon is not involved because Teflon does not heat in the microwave. However,
microwave energy does couple with all reagent molecules with the same probability and
efficiency so that the vessel is heated by conduction of the hot liquid inside to the plastic.
A transfer of heat occurs at the interface of the vessel with the cooler cavity air and vessel
cooling occurs as the hot surface comes into contact with the cooler moving fluid (air) in
the cavity. Since a temperature gradient exists, energy is transported from the hot vessel
to the cooler air. At the same time, the air in the cavity is being heated as the result of
intimate contact with the hot Teflon vessel. Temperature increases in the cavity are
approximately 8 °C for a 15 minute heating program that attains 180 °C.
It was assumed that the rate at which the temperature in the vessel decreased was
proportional to the temperature difference between the vessel and its surroundings, the
Newton Cooling Law (142). A vessel at Tt (~180°C) cools to the surroundings Tf
(~30°C) and that dT/dt = -k(Tt -Tf), where T =f(t) = Tt. That derivation, in Appendix G
gives the expression
ln(T-Tf) = ln(T0-Tf)-kt
(9)
which has been graphed in Figure 26. Simple regression yields \he equation y = ln5.0392 0.12937x where the slope, -k = 0.12937 is the cooling constant. The correlation coefficient,
= 0.995 suggests that the model adequately predicts the
11 9
J= o
■ - 2:
s i
«) 2
o
CO
Si
>«
o
| c
jS Q
“ o
-J o
E«
r-
in
o>
£
°
CM
E r_ ■o
CM h i
CM
o _0)
CO
o> o
c X
O *-
S3
CO
CM
o>
2 ■o
? 2
0) =
z to
co■ Q
O
)
CO
N
-
.
fl>
to
in
CM
CO
Oo ‘( u - l )
U|
O
|« >
- oT- O
«
U.
120
cooling behavior of the 120 mL Teflon vessel with 10 mL of water that has been heated to
180°C.
At a sustained temperature of 180 °C, as occurs during decomposition, 10 mL of water
in a 120 mL Teflon vessel is a hot cylinder 3 cm tall and 4.7 cm in diameter with a radiating
surface of approximately 62 cm^. This area contains approximately 100 g of Teflon whose heat
capacity is 0.28 cal °C"lg"l. Heat is lost from the vessel at the rate of ~0.13 °C m in'l, and
using the appropriate conversion factors, 0.25 W for the 62 cm2 of vessel surface translates to ~
16 W. The area in contact with the hot liquid conducts heat, albeit slowly, to the rest of the
vessel so that perhaps 5-6 W may be lost from the remaining surfaces. Thus, approximately 22
W of heat can be accounted for by cooling from the surface of the vessel. This theoretically
derived value correlates well with the average heat loss of 20 and 22 W vessel-1 respectively
exhibited by hydrochloric acid and the sulfuric:nitric acid mixture that had been heated to within
10°C of the model.
At the equilibrium condition that exists for temperature maintenance in a 225 g water
sample containing 35 g of nitric acid distributed in 5 Teflon vessels, 362 W was lost from the
system. That translates to 52 W vessel" 1. From the table on efficiencies in the previous section
we know that really represents ~42 W vessel" 1. Therefore, 22 W of heat lost by radiation
means that at least half of the heat lost from the vessel can be accounted for by conductive
cooling losses from the surface of the container.
Power Rejection
It has been shown in Table 7 that small volumes of each of the mineral acids
investigated absorb less microwave energy than comparable volumes of water. Even
when a mass >500 g, which is sufficiently large to absorb all of the power output of the
magnetron is irradiated, some power is not absorbed but rather is rejected or reflected
121
back toward the waveguide. This reflected power can be monitored by measuring the
heat rise in the dummy load that results from power isolated by the circulator. A small
amount of forward power leaks to the dummy load (Goetchius, R., CEM Corp, personal
communication, 1987) however, detailed observations of the behavior of this shunt with
various acids and water of different volumes has not before been documented. One liter
of water experiences an average rise of 13 °C at the heat sink compared with an average
rise in temperature of 26 °C for 1L of nitric acid irradiated for 2 min. When 1500 g of
water are heated for 2 min the heat sink shows a rise of ~7 °C. A plot of the rate of
temperature increase with the mass in the cavity can be seen in Figure 27, and is inversely
proportional to the irradiated mass; e.g., the larger the mass in the cavity, the slower the
dummy load heats. For both nitric acid and water, the heating rate approaches a limiting
value, i.e., there is no mass that absorbs all the output power of the magnetron and the
small energy leakage from the magnetron to the dummy load is confirmed.
Figure 27 also shows that more heat is generated in the dummy load for nitric acid
than for water. These findings support the absorption efficiency data which suggests that
nitric acid is less efficient at absorbing the 2450 MHz energy than water. This may be
due to the fact that the wavelength of maximum energy absoiption for nitric acid is much
farther away from 2450 MHz than that of water (Bruce R., Vanderbuilt University,
personal communication, 1989).
A single set of measurements made on 100 mL of concentrated (86% w/w)
phosphoric acid indicate that this volume of acid absorbs 305 ± 12 W compared with 408
± 3 W and 252 ± 7 W for water and nitric acid respectively and heats the dummy load at a
rate between that of 100 mL of nitric acid and 100 mL of water. As suggested in Table
122
o
o
10
O
-o
C
o
*-*
o
c
3
IL
<0
co
co
T3
CO
o
_4
o
o
>*
o
E
E
O) 3
Q
CO
(/} o>
<
£
0)
o
o
U)
CO
cc
o>
.E co
s «
X s
S i
£3 00i
i
O
CO
i
in
o
o
S/Oo
co
d
04
d
‘31VU 9NI1V3H
1
o
o
.2>*LL O
123
15, phosphoric acid may be a more efficient absorber of the 2450 MHz wave than nitric
acid, and so reflects less power to the dummy load than nitric acid, but more than water.
Decomposition Reactions of Specific Matrix Components
in Biological and Botanical Materials
When biological and botanical materials are decomposed in nitric acid a unique
pattern of temperature and pressure dependence is consistently observed. For example,
in Figure 28 the heating profile of one of six one-gram samples of Wheat Flour SRM
1577 in 10 mL nitric acid irradiated at 374 W shows that when the temperature reached
140 °C, 8 atm from organic decomposition to C02 had accumulated in 1 min. Figure 29
shows that this rapid decomposition is correlated with specific temperatures above the
normal boiling point of nitric acid. Such behavior is consistently identified with samples
having high carbohydrate content. Carbohydrates are one of three basic constituents that
comprise biological and botanical tissue samples; protein and lipid molecules are the other
principal classes of compounds present. These biological building blocks were also
studied and similar results were observed for each pure material.
Model Compounds
Carbohydrate
With this correlation identified, isolated materials representative of the
carbohydrate in botanical tissue, two polysaccharides and one sugar monomer, were
decomposed in the same manner to determine whether the oxidation of carbohydrate
components of the botanical matrix was responsible for the sudden production of CO2 in
nitric acid at 140 °C. The polysaccharides were 3:1 amylopectin-amylose and soluble
o
in
o
Do ‘aunivuadvuai
o
in
CM
o
CL
in
W
LU
UJ
F
2
6, copyright 1988
American
Chemical Society)
Flour
O
reference
o
from
CM
with permission
CO
1 g of Wheat
00
(Reprinted
of
o
Profile
Acid.
CM
Figure 28. Microwave Heating
SRM1577a in 10 mL of Nitric
ime ‘aunssaud
o
CO
0>
XJ
o <
O) o
in
h- z
o
oo
LU
o
■
0c
3
h
<
DC
U
a
2
UJ
0) O
^ -I
2 E
CL
O
ak_> T"
3 C
CO —
</> _
a> c
£.2
?'S
a |
a> o
!= O
Ui
US
I
”
E in
fO
h- r_
o> f l
" Si
a>
Li L
o
OO
CD
C\J
o
3 3
O) O
iZ u.
u ijb
‘3 d n S S 3 U d
126
starch; glucose was the monomer. Figures 30-32 show the temperature and pressure
decomposition profiles of soluble starch in nitric acid, the mixed polymer and the glucose
monomer, respectively. A plot of pressure and temperature for this data emphasizes the
dramatic trend in pressure production. All three graphs clearly show that the pressure
increases dramatically at 140 °C, producing up to 10 atm of CO2 with virtually no
temperature increase. Finally, the temperature continues to rise, but with only a minor
additional increase in pressure as a result of the partial pressure of nitric acid. These
results demonstrate that carbohydrates decompose rapidly in nitric acid at 140 °C and that
this component of a sample will decompose completely in 1-2 min under these
conditions.
Protein
Bovine Serum Albumin SRM 926, a 99% pure protein material, was selected as
the protein model. As seen in Figure 33, the protein decomposes rapidly in nitric acid at
145-150 °C. When the protein model compound was digested in nitric:phosphoric acid
(3:1 v/v), a similar temperature and pressure profile showing decomposition at 140-145
°C was observed, as seen in Figure 34. This behavior indicates that the decomposition
temperature is characteristic of the material and not dependent on the acid medium.
The temperature and pressure profile of the first decomposition of Albumin
(Bovine Seram) at 140 °C for 15-20 min is compared with the profile of the second
decomposition carried out at 170-180 °C in Figure 35. The evolution of copious
quantities of CO2 is evident in the nearly vertical pressure rise at 145 °C. At this
temperature, the carbon-to-nitrogen and carbon-to-oxygen bonds are easily and rapidly
broken. At the end of the decomposition approximately 1.2 atm of residual pressure was
measured in the vessel which corresponds to the 1.03 atm estimated for 250 mg of
127
0 )2
o _
^ noJ
E
S a l
N* 5
ca
o
<D o I
^
o .<
oo
h.
- E oo>o
a.
m JC
„
o>
O t° 3 c f
„ tn •“ al
LU (A
cc d) c 8
p fi
K
T3
CL
CO
uj
o
UJ C
s
.2
co
■—
<D
0
CD
CO O
B 5
1 «
H 5 8 2
ca ° c
b.
o
a> J Z
a
E
a>
«
O .52
CO I
55 a
JZ
©
a> I
CO JQ -a
0)
3
oo
<0
in
C
O
uije ‘3UnSS3Hd
C
M
o
o
3
®'
O
'= 2
</) © o
o>
u- o
128
o
o
XJ
CM
O)
E <
o o
in
0
z
CM
•*—
§
1
in
o
- f
a.
o
o
■
UJ
s
UJ
H
o
o
ca o
o>
a
a>
o
u>
E
D
JO<Z
a>
LU 3
oc tn
tn CO
3
a> o
Q.
cc
E
Q.
15
o
E
in <E
co
c. oo
0)
*_
■
CO
c
<
oo
L—
a>
o
o
®
o ‘oo
.D)
>*
Q.
O
a
CD
oc
£
©
£
E
o
co
hm E
>. c
0) o
0
a.
E OL
a>
a
■_
‘co
CD
E
8.
co 'S
*-•
1
CO x>
a>
a>
>_ T3
CO
I___I
o>
3 a>
O) X
CO
<o
in
co
UJ}B (3tinSS3Hd
CM
iZ
Q.
a>
CC
o
w
(O
tn
CO
Uiie ‘3anSS3Ud
CM
T—
o
o
with permission
from
reference
6, copyright 1988
American
■ ■
(Reprinted
Chemical Society)
of 240 mg
of Nitric Acid
CM
Figure 32. Temperature and Pressure Profile
of Glucose Monomer Decomposition in 5 mL
o
130
o
o
c
E
3
CM
£ O
<
oz
>\
o
isJC
%
<
0 Z
H—
o
U)
o o
CL -J
■■
E
m
o
o
T-
%
111 co
O
oc C
0 c
ID
I—
c
<
o
DC
UJ •a
c 'co
a
S
\
UJ
CO
o
0 a.
E
o
0k. o0
0
o
u>
a
a
E
0 E
H k3_
0
eo
co
CO
0
0 •c
MM
I I 1
00
h
CO
w
co
Uiie ‘HUnSSHUd
CM
>
o> o
il m
o
o
CM
3 CO
T3
O
■f—<
o O
k.
0)Bo
■
•(M
*> xz
ok. a
Q. oCO
£
a) a.
k.
3On
111 C
OcCO
cc C>
3 a
k.
H CL
o° <
o
•»- DC
Ui T
O •k^B
a CZ
O
S C
Ui
H o -i
k_ E
+3C-•O
k. c
CD 1
o
Q.
u>
E
CDE
1“ k3>
■ a>
COCO
CD c0)
k.
3 ■o>
•O
M)
u_ ffi
ja
<
a
z
-¥
X
§I
‘
\
\
I
00
L
<0
J
U)
co
une ‘Bunssaud
1
I
CM
-
132
2 c
M E
□□
CD 3
0)
X3
Im —
O)
*o
a. <
■o t :
c
CO
S
0)
c
0)
h. o
o
u>
43-» +w■*
(0 o> 5
oo 0)
*“ a) a
Q. 5 <
uT E
g *
o>
■ ° ■§
c £
o 3;
o z
S € OT 'S
sHI O
I-
C
o
CO c
.
2 % 10
'Z
m
CO £
C
O
h-
a
E
o £ E
+= 3
•*_ CD
if)
CO
o
<n
CO CD
fl> C
a>
i» e:
’>
3
o>
u>
co
Uiie ‘aunSSBUd
o
Cl o o
il & “
133
organic material. Additional pressure observed in the vessel at the end of the digestion
may be due to gaseous decomposition products of nitric acid such as brown/red nitrogen
oxides. Inspection of the temperature and pressure profile of the second digestion at
175 °C shows a trace that matches the general heating profile of pure nitric acid run at the
same conditions. Almost no additional carbon dioxide can be detected in the sample trace
above the acid background.
Lipids
Tristearin, a simple lipid material which is a C-18 triacylglycerol was selected as
the lipid model. Its temperature and pressure decomposition profile in HNO3is shown
in Figure 36 and follows the pattern for the carbohydrate and protein models except that
the greatest increase in pressure is at ~ 160 °C. This higher decomposition temperature
reflects the increased energy requirements for the oxidation of a higher percentage of
C-H, O-H, and C=0 bonds than needed for C-N or C-0 so abundant in proteins and
sugars (143).
Analysis
In order to better estimate where the real decomposition temperatures occur,
pressure and temperature data for the decomposition of all three model biological
constituents were modeled mathematically. With the help of the Statistical Engineering
Group (Susannah Schiller, Statistical Engineering Group, NIST, personal
communication, 1988) an analytical curve was fitted to the data in order to reduce each
data set to a few workable parameters. The best model was a logistic curve which looks
like an "S" with a vertical center, an example of which is shown in Figure 37. Between
the "lower asymptote", which is influenced by the boiling point and partial pressure of
134
O)
CM
E
©
CM
CM
■O
O
<
•* -
O
o
’JI
*->
a>
H- z
o
k- •fQ. o
U>
a>
3
E
CO
CO CO
CD
LU
oc
3
H
<
cc
k.
Q. c
a.
c
UJ
2
UJ
h-
■a
CO
c
.2
V '{/>
k_
o
3 CL
*■»
(0
k_ o
a> o
a>
E
a.
EQ
a>
1- c
'k_
<o (S
0)
CO «->
CO
Q
k-> b
3 H
h»
CO
in
CO
uije ‘3dnSS3dd
CM
O
•o>
■■
Li. «*o
135
©
a>
>
3
o
U)
oo
LU
CO
‘5>
cc
i-
a>
3
“
g f
K i
rv
co
a>
O)
il
i
■
(O
CO
ui)e ‘3UnSS3Hd
CM
136
the acid reagent, and the "upper asymptote", which is dependent on the length of time the
decomposition program was run, lies the center of the logistic curve called the "transition
region". This is where decomposition takes place. The equation for the logistic function
is:
y _ A + As(X-Xo) +(X-Xo)2 t B + BS(X-X0) + Bq(X-Xp)2
1 + ez
1 + e-z
( 10)
(X-Xo)(l + eQ(x‘xo))
z = --------------------------2D0
(11)
where X is the temperature and Y is the pressure. The quadratic portions in the
numerator describe the lower and upper asymptotes and the exponential in the
denominator describes the transition region. Xq is an estimate of the center of the
transition region and corresponds to a point half-way through the decomposition. This
program also provides a parameter, Xio, which is the temperature by which 10% of the
transition has taken place for each of the data sets. Such a parameter is influenced by the
partial pressure of the acid and its boiling point and thus, would not be a good candidate
for estimating the decomposition temperature of the respective biological component.
X q,
which occurs after the decomposition is well under way is a more meaningful
estimate of the decomposition temperature. When an analysis of variance is performed,
the Xq values are influenced only by the sample type. By the same analysis technique,
neither the duration of the heating program nor the power level significantly influence the
Xq value. In Figure 38 the different transition temperatures predicted by the fitted data
for the three components are observed. They are indeed distinguishable, even though the
protein and carbohydrate both fall between 140 and 145 °C. Yet, the lipid material is
137
D>
C
00
£ _
</> (0
o
O)
2 O
CO
3
.2
O ffl
00
oo
UJ
cc
3
I<
cc
UJ
Q.
s
UJ
1 1
.tr
o
<0
o *
E o
O •*“
o
<D
TJ 2
(1)
0
0 -3
00
CO
Q ©
CO
0)
■o Q.
0) E
©
H
CO
oo
co .2
o ®
c
.tr o
CO Q-
2 S E
- 2 o
u. I- O
!O
O
CO
CM
wie ‘B d n s s a a d
o«
distinctly different. In Table 19 are found predicted values for the transition temperature,
Xo, for the model compounds and two real materials. When these real materials were
analyzed by the same method, the decomposition temperatures predicted for wheat flour,
largely a carbohydrate, were supported by the actual graphed data in Figure 39. Bovine
liver was easily distinguished from the carbohydrate or the protein and clearly suggests
that higher temperatures are needed for lipid-like materials.
Table 19. Predicted Values for the Transition Temperature (Xo) of
Model Biological Components and Real Samples
Carbohydrate
Protein
Lipid
Flour
Liver
Predicted
Lower
Bound
140.3
142.5
161.8
142.8
175.2
128.3
132.0
147.0
132.4
163.2
x0
Upper
Bound
152.4
152.9
176.5
153.2
187.2
The decomposition temperature of a complete tissue sample, Bovine Liver SRM 1577a
was much higher than any of the model components reflecting its more complicated
composition. Since liver tissue contains carbohydrates, proteins, and lipids, thorough
destruction requires a temperature that will decompose all of these molecular species (54,
144). A simple triacylglycerol lipid model was adequate to demonstrate the trend in
decomposition temperatures of tissure although a complex lipid-like sterol might reflect
the liver composition more accurately. When the decomposition profile of the bovine
13 9
O)
c
5
o
f
C/3 O)
ca
om •r>
R
o
in
oo
uf
CC
3
o
cc
HI
a
S
UJ
H
a>
+
-•
m
s
c _
O
■-
CO
0)
OT =
o
a. *E £
if
.? cg.
o
"D
0)
2:
Q.
ac>
H
a> c
CO o
£ w
3>£
a
o
00
CO
uiie ‘3UnSS3Ud
o
o
140
liver was compared with that of the Albumin seen in Figure 40, the profile clearly does
not resemble the protein model. When compared with the profile of the Tristearin, as
seen in Figure 41, the liver tissue decomposition more nearly resembles the lipid than the
protein. The presence of multiple-ringed structures, such as cholesterol and aromatic
amino acids, certainly influence the decomposition temperature since they require more
energy to break the rings and sustain decomposition than can be supplied by nitric acid at
161 °C.
Problems
Chromatographic separation of the liver digestate was undertaken to determine
what organic molecules remain after a 10 minute nitric acid decomposition that sustained
temperatures of > 175 °C for several (3-4) minutes. Only peaks for ortho, meta, and para
nitro benzoic acid were found; the one organic bond not decomposed under these
conditions is the 7t bond of the benzene ring (54,144). It is assumed that the aromatic
amino acids are the origin of these ring structures (54,144). These results indicate that
the benzene ring will not be decomposed by pure nitric acid under these conditions.
Many inorganic analytical instrument determinations, such as Inductively Coupled
Plasma (ICP) or Atomic Absoiption (AA), are not seriously affected by trace quantities of
these molecules; other methods, such as polarography, will be severely affected by the
electroactive species. No significant traces of carbohydrates, proteins, or fatty acids are
detected by high performance liquid chromatography (HPLC), suggesting that
decomposition may be sufficiently complete for many techniques with a 10 min
digestion.
©
> ©
■o
o
o
0) 2
e
> c
o ©
CD *■»
oL_
a
O .
±+
+ J.«
O
++ +
+<*■♦,
+\ ••
+.
LU
a>
o
o
UJ
oe
=3
I<
cc
UI
o
a . •o
k.
a>
© k.
ca
3 CL
CO E
CO
o
© O
a.
■o
UJ
H
o
« <
c«
5CO ■z
d) c
Q. .=
» §.2
h*
4*«
d g
* £
© E
o
00
CO
uiie ‘au n ssau d
o
3
°
.5* ©
ll Q
142
a>
>
fl>
a> ■o
c o
> 2
CVI
o
ffi
■p
a
o -I
a> £
W
■X
oo
U
J
DC
3
<C
DC
UJ
a
S
UJ
H-
5o ?>
k_
0- -O
0)
a> <
^0
3 a.
co E
co
a> o
O
■g
■a a
c <
CO
2
-
3 -«
CO z
in
a> c
a.
E
Q> c
h* o
'co
o
a.
O
k.
CO
CD
lujb
‘3UnSS3Ud
cvj
o
E
3 o
O) o0)
u. a
CHAPTER 5
PRACTICAL APPLICATIONS
Throughout the course of this research I have been involved with the real business
of the National Institute of Standards and Technology (NIST), which is the certification
of Standard Reference Materials. Such materials serve as well-characterized
representatives of a particular matrix type, like soil or sediment, an ore, metal solids,
foodstuffs, clinical materials, organics, and biological and botanical tissues. As such,
these materials may be regarded as the ultimate example of a "real world sample" and it is
altogether appropriate that microwave dissolution techniques be applied during the
preparation phase to release the analyte into solutions suitable for elemental determination.
Based on the fundamentals described in the last chapter, several examples of the digestion
i
technique are presented in this chapter as they have been applied to a broad variety of
material types.
Biological and Botanical Matrices
With information on the specific decomposition temperatures gained from the
analysis of the major biological components, two examples are presented that
demonstrate the results obtained from a decomposition procedure designed specifically to
take into account the actual constituents of the matrix. In addition to the protein and fats
typical in an American diet, Total Diet SRM 1548 is also high in carbohydrates. Since
these are the most easily digested of the 3 major classes of materials (6) and
decomposition to carbon dioxide occurs at the lowest temperature in nitric acid, a two-
143
144
step digestion procedure was conceived. In the first step, the experimental conditions
were designed to attain a temperature of 140-145°C in approximately 5 min. That
temperature was then maintained for fifteen minutes - enough time for the decomposition
to proceed to near completion as evidenced by the large increase in pressure measured in
the vessel. At these temperatures, the Teflon vessels can easily withstand the added
pressure of nearly 7 atm of combined C02 and the contribution from the vapor pressure
of nitric acid.
Certification analyses of Reference Materials at NIST frequently have very rigid
protocols regarding the total number of replicates, blanks and controls that must be
incorporated into the experimental procedure. It is not uncommon to have 25-30
individual samples depending on the number of bottles sampled. Eight different samples
in duplicate, 6 process blanks, two controls and 3 "dummy" samples were divided into
three batches of nine samples each; blanks are "samples" for the purposes of total mass
and power absorption estimates since the acid is the principal absorber of microwave
energy. From fourth-order equation calculations, 90 g of concentrated nitric acid absorb
197 W of power. This is the power absorbed during the first 2 min and if the system
were 100% efficient, a temperature of 141 °C could be attained in -128 s. However,
only ~80 W of power are actually absorbed and thus a factor of 2-3 must be incorporated
to bring the solution to the target temperature. That can be accomplished by doubling the
time as well as doubling the power output to the cavity. At a power setting of 373 W,
140 °C can be attained in about 5 min and held for 15 min after which time the bulk of the
CO2 would have been evolved. Figure 42 shows that 6 atm are measured in the vessel.
With ~3-4 atm due to nitric acid at 140 °C, the balance of 11/2 atm represent the
decomposition of the diet material and 6 atm is well below the safe upper limit of the
145
Uiie ‘au n ssau d
CD
U)
CO
o
co
o
*o ^°
LO
CVI
.5? c
•H O
2 «
ui
Li.
O)
O
£ cP
"
■
*o
•»- in
o cm
C O o
in
1.2
; § »«
UJ Q. o .
373 WATTS
s O) E
CO
.i= Ui
4-*
CO
m
2>
CO
■
2
o ^
o s
m
= s 2
CM
*- o
.2 <
Q
2 - 0
3
o
o
in
Oo ‘3UniVU3dW31
o
o
CO I
.2>o ■
■
=
LL I— Z
146
vessel. Inspection of the temperature and pressure profile in Figure 43 shows the
dramatic evolution of C02 at 140 °C. Because the matrix contains lipids and proteins, a
second decomposition at temperatures between 170-180 °C is needed in order to attack
the lipids and the rest of the proteins. Figure 44 shows that six samples weighing a total
of 60 g required 574 W to reach the targeted digestion temperature of 175 °C. 'Hie reason
for reducing the number of vessels may be seen by inspection of Figure 45 which shows
the second-stage digestion attempt with nine vessels. The maximum temperature
attainable with a cavity load of 90 g of concentrated nitric acid is 153 °C. This
temperature was reached at about 17 min and could not be increased to 175 °C because
the amount of heat lost from the vessels was greater than the amount of energy heat being
pumped in at 574 W with the same percentage efficiency. Full power can only raise and
sustain a temp of 153 °C with 90 g of HNO3; the microwave unit was power-limited and
higher temperatures could not be attained with this configuration. Reducing the total
mass by reducing the number of vessels permits 545 W to raise and sustain 60 g of nitric
acid at the target temperature of 175 °C. Upon completion of the digestion, the samples
are ready for ICP analysis or any other technique where ultra traces of organic materials
do not interfere in the analysis.
The temperature and pressure profile of the second digestion seen in Figure 46,
shows that little evolution of gas took place during that step and that the pressures at these
elevated temperatures are due almost entirely to the vapor pressure of nitric acid. All of
the easily oxidized material is decomposed during the first digestion.
Results of the analysis for Mn, Cu, and Ni by Laser Enhanced Ionization (LEI)
for the SRM diet material prepared by microwave digestion followed by robot separation
are shown in Table 20. Values for each of the elements in the controls are in good
147
■»**>
C
O
E -o
* 3
to
O
u
W 4o z
*5 °>
£
>=
2
3 ”
O L
U
O«
O D
C C
2
>S
Q. T_
■a 2
e cc
C
O(/)
(I) *<
>- o
to °
in
£O- *?
®
e £
CO _
* o
<o
in
co
me ‘g u n ssau d
C
M
o
o
®i iC
sO
W
3O) .O’
&
il Q
148
m
UJIB ‘3dnSS 3dd
U)
CD
co
CM
o . .
oo
T3
C
o
.2 <
CO o
p
Q Z
0) £
o»
T3 O)
C
o o
O
T-
0)
C
©
CO
©
0>
£
fc
- g° S
•*“ 00
- o <?
in
H ~c S
DC
CO U>
© *X ©
Q) Q
>
2
co
O ©
2
0
*
o
© E
o
o
CM
O
to
o
o
0o ‘3HniVU3dW3±
o
W
^ ©
o> w
i l co
14 9
uije ‘3UnSS3Hd
<o
c
o ■
4- u
(0 o
0) <
o>
Q £
CO
?
!
8
MAXIMUM TEMP (153°C)
O)
CO o
0)
to
CM
® r-
■c .£
4 . CO
o in
^
0 T"
CM
1
2
S DC
CL CO
U>
o> c5
.E
4m 5
■»
CO
£ a
1s
s :
s °
I
in
£
i. |A
u> CO
a>
a> c
3 Z
o
o
ru. -o
o
Do ‘3UniVU3dW31
150
-a
CM
* ■
■
O
o
0> c
3
CO CO
CO
0k-> in
T“
Q.
CC
LU
O. ■o s
s c oc
U J <0 0)
H
a> a>
k_
■3H Q
C
k-O
O
o> ■CH
a. o
E (a>
I- *♦—
o
cd c
o
*■»
a>
k. CO
>
3 Q
O) O)
iZ Q
ui
cc
=3
H
<
co
in
CO
uije ‘3 d n S S 3 d d
CM
o
o
c
o
o
<D
(0 T3
■M
a
a> <
£
o
4k
o
d> z
**■» O)
o
ko. oT~
151
Table 20. Elemental Analysis of Standard Reference Materials by Laser
Enhanced Ionization (LEI)
Concentration, jig g-l
Material
Element
Bovine Liver SRM 1577A (Control)
Mn
Cu
Ni
Pb
10.33 ± 0.16
151 ± 1.4
0.292 ± 0.085
0.136 ± .017
Total Diet SRM 1548
Mn
Cu
Ni
Pb
5.04 ±0.13
2.43 ±0.10
0.408 ± 0.080
0.050 ± 0.017
Citrus Leaves SRM 1572
Mn
Ni
22.13 ±0.02
.562 ± .059
Apple SRM 1515
Mn
Ni
54.8 ± 0.66
.966 ± .037
...
---
Peach SRM 1547
Mn
Ni
98.4 ±1.1
.736 ± .034
---
Reported
Certified
9.9 ± 0.8
158 ± 7
---
0.135 ±0.015
23 ±2
.600 ±.300
Note: Microwave decomposition followed by robot separation
agreement with their previously certified values. This gives a justifiable measure of
confidence in the certificate values for the new diet SRM and the uncertainty suggests that
blank control is much improved using closed vessel microwave decomposition.
Perhaps one of the most significant advances in blank reduction has been the
introduction of laboratory apparatus and containers fabricated from fluorocarbons. From
a trace element standpoint they are the purest container material available (8). Closed
vessel decompositions in Teflon PFA have significantly reduced the limits o f detection
for many elements in a variety of analytical techniques. Containers made from
unadulturated Teflon PFA have the lowest level of impurities of any plastic (8,145).
152
Because of their low coefficient o f friction they do not require plasticizer or mold-release
agents and can be injection molded at high temperatures. Vessels fabricated from Teflon
PFA can be cleaned and leached in hot acids (3,147).
Chemical blanks in a vanadium study conducted in open beaker labvvare of Teflon
p p A that been thoroughly cleaned with hot hydrochloric and nitric acids (148) ranged
from 0.15-3.16 ng. In a study of clinically important but "difficult" elements such as Al,
As* Cd, Cr, Mo, Mn, Sb, and V (149) acid leached specially fabricated 60 mL Teflon
pp A containers were used in the sample preparation step. This time, the samples were
digested by a closed vessel microwave decomposition procedure. Blank concentrations
ranged from <3.5 pg g-l As to 1.4 ng g-l of Al and Cu.
A dramatic improvement in digestion blanks was shown recently (65) when
conventional digestion blanks using open vessels in a normal laboratory setting were
cornpared with those obtained during closed vessel microwave digestions in normal and
clean laboratory facilities (65). Not only had the mean value of the blank been reduced
by one order of magnitude when microwave digestion had been combined with clean
laboratory fcailities, but the range hadbeen narrowed with a similar ten-fold reduction of
the maximum. A concomitant benefit has been the reduction in the limit of detection for
aluminum by GFAAS.
Elements that are of clinical interest because of their importance from a
toxicological or nutritional viewpoint, such as mercury (47), selenium (62), and
vanadium (149) can now be analyzed in many matrices because exceptionally clean,
closed containers are available for decomposition.
In an analogous fashion, apple and peach leaves were prepared for analysis by
microwave decomposition. Since there is a small soil component in the matrix, 0.5 g HF
153
is added to the nitric acid for decomposition of the silicon dioxide in order to release any
trace metals that might be tied up with the silicate. Inspection of the first digestion curve
in Figure 47 shows that the temperature program could be maintained at 373 W, the same
as for the diet. After cooling and degassing, six samples were digested in the second
step. Leaf material has very little protein, but does contain lipid-like phytic acid
hydrocarbons that need the higher temperatures to be completely destroyed. Results for
the determination of the same elements are shown in Table 20.
Since both leaf and diet material behave predominantly like carbohydrate matrices
they are amenable to similar heating programs and digestion sequences. A first-step
digestion at 373 W to attain 145 °C for the decomposition of the easily digested
components is satisfactory for both tissue types. For the leaf matrix, a second step at
545 W decomposes any remaining lipid component and for the diet material, 574 W was
needed to attain the somewhat higher temperatures needed for decomposition of the more
complex proteins and lipid structures. Not only are lst-step heating profiles of diet
replicates superimposable, as shown in Figure 48, and 2nd-step heating profiles of leaf
replicates, shown in Figure 49, but the lst-step profiles of diet and leaf material are
coincident as well, as may be seen on inspection of Figures 42 and 47.
Soils and Sediments
Soil
Soil samples constitute an increasingly important reserve for the study of disposal
wastes. Leach tests and multiple site sampling of these repositories can afford clues to
the behavior of environmentally important (toxic) elements. Peruvian soil was selected as
a model for the development of a microwave extraction procedure that was designed to
mineralize those elements that are thought to be present in the soil and that migrate
154
CO
u iie
‘aunssaud
CO
CO
♦
c
co
co
a>
O)
ICO f^
O
m
.
a>
<0
<
0) u
o
CO C X
iZ
5
1
a> **• 2
S
-o
T" >
X
o
IO
C ®
Li
UJ CL
S
T3
C O)
CO
in
2
T - O
10
- Tn3
t
C
CO CO
373 W
^ ■c■M S
s
CO
(/) 1=
X
CO z
o
a>
>
> co
a>
O)
CO
£ " S
in
o
o 8 “
a) 2
L.
•■■■
0- o
r»
ct
Q.
T J
q
CO o )
£ 0) E
o
in
o
Oo ‘aum vuadiA rai
o
in
O
o>
iZ
l CS
M
<
155
LO
CM
co —
a> 2
■J5 °
Sa. <o
a> k.
CC .t:
Z
CM
in
c
E
373 W
uT
£
P
o
5 .E
H
(0
o ®
a
co
E
a> co
W
2 ~
a. .2
Q
O)
C
CD
co
•*?
Q) O
X
CO
CO
£
>
CO
> .2
(0
+*
O 2,
.2 o
2
_
co
o o ■—
u.
(0 0)
»- £
3
■*-
.5>ho
in
o
o
in
Oo (3UniVU3diAi3JL
o
o
u. o
U>
U)
o
CM
o
in
o
Oo ‘3UniVU3dW31
in
o
49.
Microwave Heating Profiles of Two R eplicates
Second Digestion of 6 Leaf Samples in Nitric Acid
CM
Figure
of the
545 W
W
1 5 7 <r
through the strata as a result of their solubility in water and in the organic wastes that
might be present in the deposit.
Since soil components rarely form gaseous decomposition products when treated
with concentrated nitric acid, larger sample sizes can be tolerated in the 120 mL Teflon
digestion vessels. If hydrofluoric acid is needed to break down silicates, then volatile
SiF4 would be expected to add to the vessel pressure. Two one-gram samples of soil in
10 mL of nitric acid w ere irradiated for 10 min in the microwave at 344 W so that a
minimum temperature o f 175-180 °C could be attained. At this temperature, all of the
easily oxidizable organics that might be present in the soil will decompose adding no
more than ~1 atm to the 8 atm of nitric acid vapor pressure normally measured at this
temperature. This pressure is still below the limit of safe operating pressures for these
containers.
Because a consistent practise o f running two samples at a time is not efficient use
of the microwave system , the procedure was scaled up for six samples simultaneously so
that the same temperature conditions would be achieved since that is the key monitoring
parameter that governs th e decomposition of specific molecular structures in nitric acid.
Six samples weighing 2 5 0 mg each w ere treated with 10 mL of nitric acid and irradiated
at 574 W. As can be seen in Figure 50, the temperature profiles are essentially
superimposable over th e 10 min dissolution. This shows that the same decomposition
conditions can be devised for different sample multiples and very clearly illustrates the
relationship between sam ple size and pow er requirements. From the plateau at 180 °C it
is also evident that each o f these programs is capable of maintaining the target temperature
for essentially an indefinite period of tim e. It is also clear that sample size for non­
interactive materials plays little if any ro le in the absorption of microwave energy and that
158
.2
■*- w
in
1
”
| s
is
0)
q :5
o> m
£ c
os
o >
tn <u
«
a.
•- *£ H“
E 2 o
Ui
“ ■ (0
0)
=w JS
I
« (/)
o>
X
W
0)
>
co
%
-O
to o
_ <
?c
<o .2
2 CM £
«
2
°*5
M
in co
2 ? o
o
o
CM
o
W
o
Do ‘3dfllVd3dlAI31
o
in
o
o
i>I
u. O .E
159
the acid medium is the major absorber of power and governs the temperature profile
produced. A sediment leach test was proposed to EPA based on the results of these
experiments.
River Sediment
Environmental samples have become increasingly important as gauges of the
quality of life or conversely as measures of how badly we have been polluting the
surroundings we live, work and play in. Monitoring the concentration levels of the
biologically important trace elements is a preoccupation in the fields of medicine and
health, food production, manufacturing and the regulatory agencies that must enforce the
standards and respond to public as well as commercial pressures to "do something about
the environment". A long-standing problem in the analysis of such materials has been the
difficulties associated with the analysis of volatile elements. When present in ionic form
associated with the matrix, mercury, arsenic, and selenium are normally determined in
solution by hydride formation provided that they have not been converted to a halogen
form during decomposition. For that reason, hydrochloric acid is not used as a digestion
reagent when these elements are sought since their chlorides are highly volatile (146).
But chlorides are frequent contaminants in soils and sediments, therefore the recoveries
of the toxic metals is usually quite low (21,37). Closed vessel sample decomposition
procedures like Carius tubes and Parr bombs are capable of retaining these elements, but
they are time consuming and hazardous.
By combining several of the techniques discussed in the experimental section, the
conditions for decomposition of river sediment were designed specifically for that matrix.
Firstly, actual dissolution was based on the behavior of the acid mixture model
commonly used to digest sediments. Then, the irradiation conditions were estimated
160
based on the power absorption of the total reagent quantity. For 0.5 g samples a 9:4
(v:v) ratio of nitric acid and hydrofluoric acid was selected. From the fourth-order
equation it was determined that 85.5 g of the mixture would absorb ~250 W. Since the
absorption efficiency of the acid mixture is only about 20%, a cavity mass approaching
90 g needs the total output power of the microwave unit Organic wastes are frequently
dumped into rivers and fresh water life forms inhabit the river beds so that these
sediments frequendy contain one, two or all three of the basic biological components.
For that reason, the temperature of the digestion must reach 170-180 °C and be
maintained for ~ 5 min in order to decompose the protein, lipid and carbohydrates
present, and to mineralize the elements. Trace quantities of organic material did not
contribute measurably to the pressure in the vessel, but the formation of SiF4 produced
an increase in the pressure observed above that of the pressure of the solution without
sample.
As can be seen in Figure 51, the temperature curves of the sample digestions
follow that of the acid model. A small amount of oxidation contributes additional heating
in solution during the first 7-8 min of decomposition, but as the target temperatures are
reached there is no more than 6-7 °C difference in temperature between the samples.
Two sets of five samples each were decomposed in this fashion and their thermal
histories are compared in Table 21. All 10 samples were digested at the same average
temperature. Sampling any one vessel was the same as sampling any one of 5 or any one
of 10 as was shown for wheat and rice flour (100). Therefore, the temperature
differences between samples did not vary more than the differences measured between
samples in the two batches.
161
« S
£ £
Z o>
o w
^ °
§
E £•■M c
co ^ x :
r-
V
*- C
O
CO
O
0 co
c
« o ”
® -O CO
.E 4= o 0>
E 2 s
Q. ~
>
sHI O )^
h .E
^ “.
(D 0>
IT
d)
g|
®
w
w
..
Q)
x ~ 3
o > -o z
>Co
in
lo w
1 o s
S J2 DC
^ O W
>» m
£3 1 E
200
5 , TJ -O
o
in
o
in
3o
‘3UniVU3dlAI31
o
o
.2* c 0)
U_ CO </)
162
Table 21. Reproducibility of Temperature and Pressure in Digestion of
Buffalo River Sediment SRM 2704
Batch 1
Time, min
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
4.00
4.50
5.00
5.50
6.00
6.50
7.00
7.50
8.00
8.50
9.00
9.50
10.00
Temp., °C
26.4
50.0
85.0
116.6
133.3
138.7
143.8
150.0
155.3
159.0
164.4
169.9
172.9
176.5
179.1
180.2
178.3
177.2
178.0
178.0
179.6
Batch 2
Abs.
Pressure
atm
0.00
0.12
0.20
0.48
0.90
1.30
1.68
2.08
2.54
3.17
3.58
4.34
4.90
5.44
6.01
5.65
5.44
5.16
5.11
5.10
5.29
Temp., °C
27.9
45.6
81.7
114.7
132.5
142.2
148.2
155.2
160.8
166.5
170.9
174.1
176.7
176.8
176.9
178.6
178.8
177.8
180.3
179.5
180.0
Abs.
Pressure
atm
0.00
0.08
0.19
0.41
0.99
1.45
2.12
2.69
3.22
3.85
4.50
5.05
5.61
5.94
6.29
6.58
6.80
6.76
6.17
5.90
5.76
Because the vessels were completely sealed during decomposition and opened
only when thoroughly cooled, any volatile elements present were retained in solution. As
seen in Table 22 (150), not only could arsenic, mercury and thallium be done well by
ICP-MS, but selenium was certified for the first time in sediment, and the thallium results
are corroborated by a second technique. In dilute solutions phosphorous can be volatile
(97) leading to low results. This would not be a problem with the closed vessel method.
In the past, when multiple elemental determinations were required on sample
matrices, mercury, selenium, arsenic, tellurium, tin, germanium, and antimony were
1 63
Table 22. Elemental Analysis of Buffalo River Sediment SRM 2704 by ICP-MS and
GFAAS
ICP-MS
GFAAS
Element
Determined
jig g-1 analyte
Certified Value
pg gr1 analyte
As
Hg
P
S
H
U
23.4 ± 2.6
1.49 ±0.14
1.016 ± 0.016 mg
3.56 ±0.16
1.07 ± 0.07
2.97 ± 0.04
23.4 ±0.8
1.44 ± 0.07
0.998 ± 0.028 mg
Se
11
1.13 ±0.9
1.15 ± 0.22
----
1.2 ±0.2
3.13 ±0.13
...
1.2 ±0.2
Note: Data from Ref 150
often determined separately because of their volatility under incomplete oxidative
conditions or because chlorine was present to create volatile chloride salts (151). When
the possibility of thermal losses exists, closed-vessel decomposition is essential to
complete and accurate determination of trace element concentrations. The same
conditions that have been instrumental in lowering the chemical blank now permit the
traditionally volatile elements to be determined routinely in the same digestion sequence
with other traditional elements thus eliminating the need for separate procedures such as
hydride generation or cold vapor AAS. Provided there is no release of the pressure
valve, microwave decompositions carried out entirely in closed vessels are capable of
preventing such losses because they simply cannot get out. Elemental species that are
traditionally considered volatile (151,152) are increasingly being documented as retained
in solution when closed vessel decompositions are performed and the vessels opened at
room temperature (30,47,62).
164
Small quantities of water and mineral acids such as nitric, hydrochloric, and
hydrofluoric will diffuse into the vessel wall and escape during decomposition. Metal
cations and most anions remain in solution and therefore do not penetrate the Teflon
PFA. Exceptions to this rule of thumb are substances that can pass through the polymer
such as metallic mercury, osmium tetroxide (153), and the hydrogen halides (Greenberg,
R.; Kingston, H.M., NIST, unpublished results). In a 30-day leach/clean test conducted
on 60 mL molded Teflon PFA vessels using high purity water at 90 °C, the pH of the
water dropped to 3.4 from nearly neutral. This is the result of HF trapped in the
polymer lattice during molding that was leached out of the wall (147). At the elevated
pressures of 6-7 atm generated in the vessels during digestions conducted in nitric acid,
molecules of HBr, HF, HC1 and HI can conceivably be forced into the lattice. These
specific compounds have been detected by measuring the residual radioactivity of a
Teflon vessel after an irradiated sample matrix had been decomposed in a microwave unit
(Greenberg, R.; Kingston, H.M., NIST, unpublished results).
In addition to retention of the vessel contents, retained vapor pressures lead to the
higher temperatures necessary for the efficient oxidation of numerous elements. This
advantage may eliminate the need for time-consuming, dangerous perchloric acid
digestions. Patterson et al (62) found that decompositions which previously required the
high temperatures and oxidation power of perchloric acid were just as efficiently digested
by closed vessel microwave decomposition. Selenium is a nutritionally important
element (154) but is often difficult to determine because bound organic forms are difficult
to oxidize fully and reduced forms of the element are volatile. All of the biological
matrices in this selenium study (62,40) were decomposed directly without a predigestion
step and were oxidized fully to the Se+6 form necessary for complexation. This was
165
confirmed by the results of chelation and determination by hydride generation AAS
(Patterson, K., Beltsville Human Nutrition Research Center, USDA, personal
communication, 1986). Recoveries of selenium were essentially complete, which
suggests that none of the metal was lost by adsorption on the walls of the vessels.
Organic matrices are frequently subject to mild heating predigestion steps to
reduce the chances of vigorous oxidation splattering sample material. To minimize the
stress exerted on small 60 mL Teflon PFA vessels organic matrices were frequently
predigested at low temperatures on hot plates so that the evolution of copious quantities
of carbon dioxide would not cause a problem during the microwave decomposition.
Such predigestion steps can lead to elemental losses and were identified as the reason for
low phosphorous results in Wheat and Rice Flour SRM (5,29). With the introduction of
new, stronger 120 mL Teflon PFA vessels, predigestions are no longer necessary. The
entire digestion sequence can be accomplished in one or two steps in the microwave
system. Recoveries of 98.1 ± 2.4% of mercury in biological samples have recently been
reported (47) using 100 mg samples in 60 mL Teflon vessels. Clearly, no mercury was
lost due to volatilization or adsorption.
Simulated Nuclear Waste
A sequential approach was developed combining classical dissolution concepts in
the application of a microwave digestion procedure for process control analysis in the
vitrification of high-level radioactive waste at the Savannah River Plant (68). It is
essential to be able to monitor the composition of the waste at each stage of the process
from arrival of the slurried nuclear waste to the final vitrified material that will be cast for
storage. The ideal dissolution would dissolve material sampled from the waste at every
166
stage in the process and produce a transparent digestate, free from particulates, that could
be simply diluted for determination of the elemental composition by ICP.
Samples of the simulated sludge at each stage in the processing included the waste
slurry as received, after calcining, before and after formic acid treatment to recover
cesium, after a mercury distillation step, and after addition of the glass frit. It was
expected that each sample could be prepared and analyzed in 30 min. At several steps,
notably after the addition of the organic acid, the precipitated material as well as the waste
in the processing smelter were sampled. The glass frit is analyzed before its addition to
the simulated material and after it has been used to "sandblast" the outsides of the waste<*>
containing cannisters. Because homogeneity of the material sampled is essential for a
meaningful analysis, dried samples are easier to analyze than wet or slurried materials,
hence the drying steps. The digestion procedure was developed using dry, homogeneous
powders of the simulated material in each of the process steps.
The dissolution procedure devised consisted of digesting two 250 mg samples,
each in 10 mL of 1:1 (v/v) nitric and hydrofluoric acids in 120 mL Teflon PFA vessels
that have been torqued to 14 N m to ensure reproducible temperature and pressure
conditions during the microwave dissolution. Sample sizes of 250 mg generate gaseous
decomposition products which when combined with the vapor pressure of the acid
mixture at that temperature can be completely contained in the 120 mL volume of the
vessel. Although volatile elements are not specifically assayed, they could be determined
since the containers remain closed throughout the dissolution process. The digestion
sequence was designed not to exceed a maximum of 100 psi near the end of the digestion
and to attain temperatures > 160 °C to be sure that the matrix containing the organic acid
was completely digested. A single-step power program of 287 W brought the 20 mL of
167
acid to 160 °C in 2.5 min. During the balance of the 10 min digestion, the temperature
rose gradually to 170-180 °C. The heating program for the 10 min digestion of the
calcined material is shown in Figure 52. After cooling for 5 min the vessels are untorqued
and placed on a clean (class 100 air filter) bench where the caps are removed. Forty mL
of hot DI water containing 1.5 g of boric acid are added with constant mixing. The
vessels can be returned to the microwave for an additional 5 min of heating at 230 W to
complex and solubulize the metal fluorides. The vessels are cooled and removed as
before. Finally, 5 mL of concentrated hydrochloric acid are added and the solutions
returned to the microwave for an additional 15-30 min at 230-287 W. In this step, strong
chloride complexes are formed that dissolve any insoluble transition metal precipitates
and the result is a clear, pale-colored solution suitable for direct analysis by ICP. For
samples without the glass frit, a ratio of 7:3 (v/v) nitric:hydrofluoric may be more suitable
because the matrix has less silicon and more iron.
Results of the determination of several elements by ICP are shown in Table 22 for
a glass standard and the frit material used in vitrification.
Inorganic Matrices
Alpha-alumina
Refractory aluminas used in ceramics, as catalyst supports and for other
specialized purposes that are produced by sintering have a high alpha-alumina content
which makes them very difficult to degrade. When compared with metallurgical grade
alumina the trace element content is normally ten times lower (Ferland, P., Alcoa of
Australia, Ltd., personal communication, 1987), often between 5-50 mg g-l Fe and Si,
for example. Methods for the dissolution of this material include fusions, either acid or
caustic attack, and dissolution in acid mixtures in a PTFE bomb.
168
uite ‘3dnSS3dd
m
CO
co
"O
0)
«
=
CO
E2
</>
«
•» -
o o
O) o
E
o
3
o
in TJ
CM
>•
X
■o
0) c
CO
.E H=
E 2
_ CL
UJ
2 O)
F= .E
to
(0
0) -1
X
00
w
£
0)
>
CO
5
CM
o
2 Q)
1 *■*
CO
0
CO
CM £
200
in
o
UJ
o
Go ‘3 U n ± V H 3 d W 3 1
o
to
o
o
o>
k_
CO >
O)
O
3
® >
il
IE
T~
169
Table 23. Elemental Analysis by ICP of Glass Frit Used for the Vitrification of
Simulated Nuclear Waste
Weight, %
Oxide
Glass Standard
Glass Frit
% RSD (n=6)
A1
B
Ca
Fe
K
Li
Mg
Mn
Na
Ni
Si
Ti
U
4.1
9.27
1.3
13.23
3
3.06
1.21
3.54
11.78
1.12
45.42
1.05
2.32
3.56
nda
1.32
13.96
3.09
3.05
1.23
3.44
12.02
1.25
ndb
1.02
2.27
3.99
nda
2.27
0.71
1.03
2.34
1.56
0.8
1.13
5.81
nd*5
2.35
2.55
aBoron not determined because of boric acid
^Silicon not determined because of contamination with torch
All of these procedures are lengthy and none is free of problems and interferences.
Silicon cannot be determined if glass or even quartz vessels are used and most fluxes are
contaminated with traces of iron.
Several digestion reagents and specific conditions were investigated as potential
candidates for the microwave-assisted dissolution of alpha-alumina. They include HQ
and HF (4:1, w/w) in both ordinary moderate-pressure as well as high-pressure vessels;
neither of which succeeded. Nitric acid with added HF (7:1, w/w) left the alpha-alumina
unchanged, as did concentrated HQ in the high-temperature Parr vessel. Even "super"
phosphoric acid was incapable of effecting dissolution of this refractory matrix. A
170
phosphoric-sulfuric acid combination was suggested by several sources (155, Ferland,
P., Alcoa of Australia, Ltd., personal communication, 1987).
Microwave sample preparation conducted in Teflon vessels using sulfuric and
phosphoric acids (1:1, w/w) accomplishes dissolution at the same time that it eliminates
nearly all of the analysis problems previously associated with alumina. Vety low analyte
levels are possible because the dissolution reagents can be obtained or prepared in very
high purity; sulfuric acid by sub-boiling distillation (145) and a "super" phosphoric acid
with very low water content by hydrating high purity phosphorous pentoxide
(Slobodkin, J., Pennsylvania State University, personal communication, 1986) with high
purity water. Closed Teflon vessels that have been leached in hdt acid provide a clean,
non-contaminating environment in which to conduct the digestion. The
sulfuric/phosphoric acid combination proved to be a strong absorber of microwave
energy as can be seen in Figure 53 which shows that temperatures over 200 °C are
attained in less than 3 min at a mere 92 W of input power. The targeted temperature of
260-265 °C is reached in 7 min and is accompanied by a negligible vapor pressure;
conditions that can be maintained with just 60 W. This mixture has an absorption
efficiency of about 30%, calculated during the first 60 s. At that ratio, equilibration at 60
W of power represents a heat loss of ~18 W. After 40 min the transparent, colorless
solution is ready for analysis by ICP or other AA or AES techniques.
Water
Whenever newspapers bring the subject of pollution and toxic substance finds to
public attention we are reminded of the basic fragility of the environment and its
homeostatic mechanism. In no other sphere is this more critical and vital than water;
ground water, rivers and streams as well as the oceans, for water is the source of all life.
171
uiie ‘3UnSS3Ud
co
CM
«• 245
-C o
“ ■<
O S
sz
57 W
o) a
c co
c o
o a
a> o
c = t
1 o i
Uj
w
2 O) 4'
i= .E o
CO _ J
63 W
£
e
ra .E
'
O
I -■M +a>
a C-■O
0)
O)
i
92 W
CO
in
5
CO >
c
a> .E
•3 E
O
O
CO
O
O
O
O
Go ‘3UniVU3dW31
o
O) 3
il <
-»
>
172
It's needed to germinate, to grow, and to sustain all life forms. As the lifeblood of man,
it is the principal means of transport of all the bad as well as the good. Anything that
finds its way into water can eventually find its way into the human body. It is for that
reason that the establishment of guidelines and the analysis of water quality as well as the
enforcement of regulations to protect water are so vital to mankind.
The U. S. Environmental Protection Agency has been charged with the
responsibility of enforcing regulations specifying the impurity levels mandated by
Congress. Determination of trace elements in high particulate water sources has been
greatly simplified and improved by the development of a microwave assisted method of
decomposition. Water sampled in the field is normally stabilized by the addition of 5 g of
concentrated nitric acid per liter (nominally 1000 g) of liquid. For heavy particulate water
samples the procedure recommended to EPA for adoption was modeled using 50 mL
samples made up of 5 mL of concentrated acid and 45 mL of distilled DI water. The
higher concentration of acid is needed for the digestion of organic residues normally
found in such matrices. Because high particulate water might possibly contain oily
wastes, the target temperature for such digestions is 170-175 °C. The decomposition of
organic materials to carbon dioxide will add to the vapor pressure of 170 °C water
measured inside the vessel. Therefore, monitoring the pressure is also important so that
the total pressure in the vessel does not exceed the safe pressure limit and does not cause
the relief valve to open thereby compromising the determination of any volatile elements
that might be present. The goal is to arrive at the target pressure of 100 psi at the very
end of the digestion. In this way, the vessel would not be under stress during the bulk of
the digestion sequence. This was accomplished by incorporating a 10 min inductive
heating phase at 545 W designed to arrive at a target temperature of 160 °C, then reduce
173
the power to 344 W for the remainder of the two-step digestion sequence. Inspection of
Figure 54 shows that this power level produces a gradual increase of about 1°C min-l
over the 10 min maintenance portion of the program. The graph also gives some idea of
the excellent reproducibility of digestion conditions that can be expected when repeating a
given dissolution program on succeeding batches of samples. With monitoring, all
twenty samples have the same thermal history so that temperature variations in the
preparation step areeliminated. It is possible, however, that multiple-ringed aromatic
compounds may have survived these decomposition conditions since they were found
even at 240 °C in Teflon bomb decompositions. Unfortunately, nitrated derivatives are
electroactive and interfere with the polarographic determination of lead, cadmium, and
zinc (54).
Temperature and pressure parameters are sensitive to small variations in power
programming which allow the digestion conditions to be fine-tuned to meet specific
f
temperature requirements. As shown in Figure 55, when higher power is applied for a
longer period of time during the initial inductive heating program it results in a higher
temperature at the beginning of the maintenance portion even though the water and acid
were heated at the same rate, 13.5 °C min-l. At this higher initial temperature, 362 W is
capable of maintaining the temperature at 175 °C for the balance of the 20 min program.
In contrast, an initial small drop in temperature occurs with 316 W which can maintain a
slightly lower temperature. As was shown in the heat loss calculation, 52 W vessel-1 are
needed to maintain equilibrium at 362 W. In contrast, 40 W vessel-l at an input of
316 W does not maintain the steady state and the system sustains a heat loss of
~1°C min-l.
174
LO
CM
CO
<D
Q co
Q. a.
O) E
(0
v. ©
(0
©
X
0)
>
(0
tn
V.
a>
co
£
7= T3
£
E O
o
k.
in
lf> ^
2 * o
© *Z
© > .t;
T -
•Oan
LU
s
I-
•S o ©
c
o
'35
O CQ £
a
k.
a>
Q.
3 2 "D
o)
•*©
— jS
</>
w
CD
k.
A
to
a
cd
O)
3 5
£
•*=
3
O) ih o
CD
o
o
CM
o
in
o
Do ‘3dniVd3dW31
o
o Ui
175
in
E
CB
hm
I s
•“ k.
(0
~
0> Z
X H—
o °
CM
m
co
T“
c
E
CO
win
V) -
fi
uT CO
S £ tf>
H O «>
©
$
I
CB
o w
■S
O *(B
in
E
. in
in *r
in
0)
fl) >
o
o
CM
O
in
o
o
Do ‘3UniVU3dVU31
o
in
o
3 il
.2>h-
u. o
A more vigorous digestion scheme that combines 45 mL of water with 5 mL of a
mixture of nitric and hydrochloric acids (1:1, v/v) was developed for high particulate
water. For estimating the power absorption of 200 mL of a 10% aqua regia solution the
solution may be thought of as 100 mL of a 10% nitric acid solution and 100 mL of a 10%
HC1 solution which are approximately equivalent to a 1 M solution of nitric acid and a 1
M solution of HC1. The fourth-order equation was solved using coefficients for 1M
solutions of nitric and hydrochloric acids to arrive at the power absorptions. Even
though 200 mL of water absorbs 450 W, it requires nearly all of the output power of the
magnetron to reach 140 °C because its absorption efficiency is only ~70%. A 5 min
inductive heating phase was targeted to 140 °C. Figure 56 shows that 5 more min at 531
W was needed to reach the 160 °C for the beginning of the decomposition phase. A
reduction in power to 327 W for the next 10 min of the decomposition step produces a
gradual increase to the final targeted temperature of 165-170 °C. Only at the end of the
program did the pressure in the vessel approached the limiting pressure of 100 psi.
Another approach using this acid combination can be seen in Figure 57. The
water is heated for 1 min longer during the initial phase, but at higher power, to achieve
the same heating rate. At this heating rate the longer inductive phase brings the initial
target temperature to 172 °C, which can be maintained by 362 W for an additional 9 min.
A second batch of samples could barely be maintained in the 170-180 °C range with a
second-stage program of 316 W and is not quite sufficient to maintain the desired
temperature so that cooling is observed.
200
min
CM
CM
O
in
O
O
Go ‘3dniV d3dW 31
O
Figure 56. Superposition of
the Microwave Heating Profiles of
Four Different Batches of Five 45 mL Water Samples D igested
with 5 mL of Nitric and Hydrochloric Acids (1:1 v/v).
TIME,
W
178
E
CO
k.
o>
o
_
o>
CO
Io>*=
CO
I
z
H—
<D O
JZ
CO CM CM
CO <0
CO CO
U)
c t:
o
IO
■= d> .e
c
CO
E c n .tr
c
5
uT *
s S <0 ■
F ° i >
o
<5 E £
5 w -
o to r ,
o. k.
•£ JH «
° « 2
* * <
£ E o
w
.c
200
0) o
0) > o
u>
in
3o
‘3dniVd3dW31
o
o
3 £ -o
- oO IZ '
Li.
CHAPTER 6
CONCLUSIONS
Importance of Power Measurements. Calibration, Prediction and
Absorption Efficiency
Power Measurements
Microwave equipment manufacturers produce many units of varying size and
power output so that it is frequently difficult to gauge just how long it will take for a mass
to heat. When the actual magnetron power output is known then it is possible to
determine empirically the relationship between the power and heating program needed for
a task so that determination of the maximum power output gives some idea about the
optimal cavity loading. Once the maximum power has been determined it is also possible
to rank or scale several units, determine a proportionality constant, and thereby relate
them to one another. As with most laboratory equipment, knowledge of the unit's
specifications, in this case, maximum power, is an acceptable way of checking its
performance. As part of a complete calibration, the maximum power helps establish the
non-linear range of the unit's power control regulator.
As with any piece of laboratory equipment the calibration is an essential ingredient
of the confidence with which the apparatus gives reproducible results. Each unit has a
specific or unique calibration, i.e. the response, the actual power absorbed by a dielectric,
as a function of the input stimulus, the amount of power applied, which is perhaps the
most important criterion of the unit's performance. So long as the calibration of the unit is
reproducible then the results are comparable and have meaning on a relative scale. Matters
179
are simplified if the calibration is truly linear, i.e. no heat is observed in a dielectric
absorber when the unit is off and exactly half the heat or half the full power is absorbed
when the scaled input is 50%. Unfortunately, very few microwave units are absolutely
linear. The majority have a linear operating range over the bulk of the applied power
input scale with an offset at low power where no heating occurs and an inaccessible
region at large applied powers where the graduations are indestinguishable. These biases
are the result of capacitor changes in the high-voltage circuity made by manufacturers to
preserve the magnetron and associated electronics.
Once the linear operating region has been established, then heating programs can
be implemented with the confidence that a partial power setting will deliver the
appropriate number of watts to reproduce specific temperatures and digestion conditions.
Another benefit of calibration is that successful digestion procedures can be transferred
from one unit to another with the assurance that a specified amount of power can be
delivered. It is also essential to know the correspondence of proprotional power settings
in order to effectively use the empirically derived curves for the power absorptions of the
acids.
When working with materials that are potentially dangerous such as hot,
concentrated mineral acids, it is helpful to know what to expect from a particular action.
Using a ruler and the empirically derived data graphed on regular scale, it is possible to
estimate the power absorbed by a quantity of acid during the first 2-4 min of heating
simply by reading off the power corresponding to the mass of acid. If this information is
used with the transformed Equations (2) or (3), then either a final temperature can be
estimated for a given mass in a desired time or the time required for the mass to arrive at a
predetermined temperture can be predicted. Both pieces of information give some idea of
1 81
the heating rate of the dielectric material. Since most acids, bases and ionic solutions are
good absorbers of microwave radiation, it should not be surprising that small volumes
can heat very quickly and that exothermic reactions are possible when the energy is
furnished to an oxidative reaction. For a better estimate (within ~ 10%), the fourth-order
equation and appropriate coefficients should be used. Together with information on the
absorption efficiency of a reagent or combination of reagents the predictive equations
should provide safe, sound operating conditions for a trial dissolution.
Although the lack of heat capacity data did not allow a similar set of predictive
curves to be derived for acid mixtures, their absorptions can be estimated by taking the
simple linear sum of the individual acid absorptions based on the temperature rise in
solution. Within a first approximation this technique can be used to estimate the power
absorption of a mixture even though the actual mixtures underestimate the real heat rise
expected in such a mixture.
Microwave absoiption efficiency data of the acids and solutions show that for
volumes between 5-10 mL, the greatest efficiencies are realized with 9-12 vessels. As a
result, successful digestions of biological and boltanical materials could be achieved with
9 vessels each containing 10 g of nitric acid. Reduced or partial power levels between
650 and 200W did not exhibit reduced absorption efficiency, thus, partial power
programming of dissolutions provided safe and controlled digestion conditions. Although
many polar species and ionic solutions have absoiption (resonant) frequencies in the RF
and microwave frequency region, the absorption efficiency at 2450 MHZ was more
dependent on the size of the individual samples and to a lesser degree on the nature of the
absorber. Small volumes of all materials tested were substantially less efficient absorbers
than larger volumes, especially when there was only one vessel. Since the absorption
182
efficiency is a function of both vessel geometry and depth of penetration, successful
heating requires at least one dimension to be much larger than the other.
Not only is it necessary to consider the absorption efficiency of the digestion
medium, it is important to make some allowance for heat loss in the system when
digestions are much longer then 5-10 min. In fact, the average heat loss of
~25 W vessel-1 must be considered when selecting a targeted temperature that needs to be
maintained for a specific number of vessels in the cavity.
Many reactions could benefit from this heat loss knowledge. For instance,
hydrolysis of proteins in 6 N HC1 (156) is accomplished in 2 h of irradiation in a
microwave cavity. Successful cleavage of the proteins without subsequent degradation of
the fragile amino acids requires carefully controlled temperatures that do not exceed
122 —2 °C. Uncompensated heat loss would prolong the hydrolysis or it might be
incomplete in the programmed time. A more vigorous heating program than necessary
could lead to excessive final temps that would further degrade the fragile amino acids.
Comparison of the Quality of the Chemistry Between Microwave
and Traditional Digestions
Chemical Blanks
Contamination from sources external to the sample are reflected in the magnitude
of the blank correction required in accurate trace element analyses and has been elegantly
discussed by Murphy (8). Good control of the four principal sources, namely the
environment, the reagents, the containers (apparatus) and the analyst, will reduce their
variability and improve the lower limit of elemental concentrations that can be determined.
Microwave decomposition in closed Teflon PFA containers can make very real
183
contributions to the improvement of blank levels because all four sources of
contamination are addressed during the process.
Digestions in closed containers restrict the environmental contribution since the
containers need never be exposed to ordinary laboratory air if the samples are prepared in
a clean facility (8). Unlike open beaker work on hot plates where not only airborne
particulates but shedding from analyst's manipulations may gain access to the sample, the
closed vessel essentially removes the analyst and the environment as possible
contributors to blank contamination. Ultra pure reagents that have been prepared by sub­
boiling distillation (145,147) have very low impurity levels (145,157-159) and can also
help improve the blank determination for numerous analyses. Closed containers
conserve the original reagent volume and eliminate the need for constant additions to
maintain volume. Thus, reagent impurities are not concentrated.
Microwave digestion vessels fabricated from a "super clean" Teflon PFA polymer
have recently become available (Saville, R., Savillex Corp., personal communication,
1987). Microwave assisted wet ashing techniques now make it possible to control and
reduce the four principal sources of contamination by preparing samples in closed Teflon
PFA vessels under class 100 clean air conditions using specially purified reagents.
Elements Retained and Lost
Elemental losses can occur at any stage during the analytical procedure but
considerable losses are often experienced during the destruction of the sample matrix
especially biological and botanical matrices (160). Both wet and dry ashing in open
vessels risk the physical loss of analytes of interest; by volatilization and spattering when
acid digestions are taken to dryness, and by physical loss to air currents and volatilization
during furnace work. Additional opportunities for loss may be present if HC1 is needed
184
for digestion since many elemental chlorides have high vapor pressures. Indeed, the
form of the element is the key to its behavior and important metals can be lost in their
bound organic form as well as their elemental or ionic forms.
Separate sample preparation solely for the purpose of analyzing mercury, arsenic,
selenium and other "volatile" elements may be a thing of the past With the exception of
elemental mercury, osmium tetraoxide (153), and the hydrogen halides (Greenberg, R.;
Kingston, H. M. NIST, unpublished results), all cations and anions appear to be retained
in solution as the result of the closed vessel decomposition. Because Teflon PFA is
permeable to certain gas molecules like hydrogen fluoride and oxygen, it is not entirely
unexpected that at high temperatures and pressures elemental compounds of the halogen
gases as well as the acid-halides pass through the walls of the Teflon PFA . Finding
osmium tetraoxide both trapped in the vessel walls as well as outside the vessel (153)
was totally unexpected. Such a large molecule would not normally be expected to
penetrate the walls of the polymer, but perhaps its geometry is such that it can be forced
through the polymer lattice under pressure. Despite such specific problems, these vessels
are being used routinely in service labs and process control situations. Chemists can now
capitalize on the advantages of the multi-element techniques such as ICP.
Significance of Monitoring Parameters
At any temperature, the pressure developed in the vessel depends on the partial
pressure of the solvent(s) which are under nonequilibrium conditions and the gaseous
digestion products formed during decomposition. Because these are complex, it is
difficult to exactly predict reaction mechanisms. Measurement of pressure produced
during these reactions aids in the investigation of microwave acid decomposition and
other microwave and molecular interactions. Mechanisms of decomposition are reflected
185
in the temperature and pressure curves and thus provide valuable information about the
reactions taking place. Through real-time measurements during acid decomposition we
have maintained a safe laboratory environment, have achieved some ability to control the
rapid reactions, and have begun to understand some of these complex interactions.
Because the acid decomposition of organic materials produces higher pressures in
closed vessel digestions than occur with highly oxidized materials such as geologies,
ceramics and metals it is important to observe the added pressure produced from the
digestion products. Monitoring the temperature and pressure is especially important to
observe the added pressure from sample digestion products. Once the behavior of a
material has been established and the amount of the material that can be digested safely is
determined, that material may routinely be decomposed with confidence that repeatable
pressures are being produced in the vessel.
Because the fundamental relationships and predictive (transformed) equations
could not be used to establish conditions where the long term maintenance of temperature
or its slow gradual rise was desired, empirical data from the measurement of the
temperature and pressure in the vessel were used for method development. Effective
power consumption was reduced by the heat loss of the container and changed with each
maintenance temperature. By measuring the temperature rise with power over time for
the acid or acid combination that was used it was possible to compensate for vessel
variation and temperature by adjusting the input power to match each particular heat loss.
After evaluating the slopes of the temperature and pressure curves the power can
be adjusted manually. This interactive mode provides a means of achieving variablestaged conditions. The program of a decomposition can be extended or renewed if the
rate of temperature rise is not as fast as desired. Conversely, the power can be attenuated
186
in a reaction that is not levelling off at some maximum targeted temperature. This would
be especially useful in methods development
Another benefit of monitoring the digestion parameters is the ability to investigate
the behavior of previously untested reagents or solvents with unknown heat capacities or
heating characteristics. Monitoring parameters also helps to insure the reproducibility of
the the thermal profile that has been previously determined to be effective for a specific
matrix dissolution. If the same sample size and power program are used then the thermal
history of multiple samples can be controlled and repeatedly reproduced.
Real-time measurements of the temperature and pressure during decomposition
can provide insight into the mechanisms of decomposition and thus provide valuable
information about the reactions.
'**
Multiple programming steps have frequently been required to maintain or
gradually increase or decrease acid temperature during digestion and was accomplished
while the conditions inside the vessel were observed. Used in this way, the temperature
and pressure measurements provided feedback information that allowed a high degree of
control over the reaction conditions. This technique was especially practical during
research and methods development
Establishment and Control of Decomposition Conditions
When closed vessels are used for microwave decomposition, the solution
temperature is no longer limited by the boiling point of the acid at atmospheric pressure
and for many acids the boiling point can be raised significantly higher than in open vessel
procedures. The limiting factors in closed-vessel decomposition are the temperature and
pressure that the vessels can safely contain. Once these limits have been established,
temperature and pressure can be monitored to maintain the reaction within these limits.
Experimental work with numerous acids and acid combinations has shown that
the solvent, usually the mineral acid in digestions, but equally applicable to base
hydrolysis and solvents used in organic synthesis, is the principal absorber of the
microwave energy. Each of the individual acids has its own characteristic temperature
and pressure profile. These are important parameters that must be taken into account
when considering a dissolution medium. For instance, nitric acid is limited to ~ 180 °C in
the Teflon vessel because its vapor pressure at temperatures above 180-190 °C exceeds
the safe pressure limit of the vessel at that temperature. HF is limited to somewhat lower
temperatures since it has an even higher partial pressure than nitric acid. Maximum safe
temperatures for HC1 in the microwave system are not as well documented, but may be
similar to nitric acid. However, HC1 decomposes to chlorine gas at very high
temperatures and pressures. Although the presence of corrosive chlorine gas may
facilitate certain dissolutions it is also toxic. If proper precautions are not taken, then the
analyst could be exposed to dangerous vapors when uncapping the vessels. Perchloric
acid similarly decomposes to chlorine gas in the microwave at temperatures between 205
and 245 °C (6). Because hot, concentrated perchloric acid is extremely
dangerous it is not recommended for use in the microwave (161).
A group of relatively high boiling point acids such as phosphoric and
tetrafluoroboric acids that have low partial pressure acids were evaluated specifically for
use in microwave dissolution to take advantage of these useful properties.
Tetrafluoroboric acid can attain temperatures much higher than can be achieved with
hydrofluoric acid and can be used in decompositions that require attacking silicates at
high temperatures. In some dissolutions, the need for boric acid might be eliminated with
this ashing reagent. "Super" phosphoric acid(>99%, w/w) is the strongest coupler of
188
microwave energy that has been investigated in this laboratory. It easily attains
temperatures of 250 °C in barely 3 min. It is also an excellent dehydrating reagent.
Because the boiling point of sulfuric acid is >400 °C it must be used in glass or quartz
vessels, but it is still the reagent of choice for Kjeldahl digestions. It is therefore,
essential to monitor the temperature inside the Teflon PFA vessels when using these acids
so that the maximum use temperature (262°C) is not exceeded during the digestion.
Because many of these acid combinations are not ideal solutions and thus do not
obey Raoult's Law, the result of mixing these acids is a useful lowering of the vapor
pressure of the solution. These properties were used advantageously to produce mixtures
with partial pressures somewhat lower than that of the lower boiling acid. Nitric and
hydrofluoric acids were one such combination that was especially effective because some
reactions that don't take place at room temperature do occur at the elevated temperatures
attainable in closed-vessel microwave dissolutions. In addition, it is believed that
retained HF helps suppress passivation (53A). Combined with phosphoric acid, nitric
acid solutions can attain temperatures of 180-200 °C and have been more effective in
oxidizing certain biological matrices (Patterson, K., Beltsville Human Nutrition Research
Center, USDA, pesonal communication, 1986). When the Pair digestion vessel was
used with this acid combination temperatures in excess of 200 °C were attained. Perhaps
the most strongly absorbing mixture is phosphoric and sulfuric, two high boiling acids
that have a barely measurable partial pressure in solution at 250 °C.
Different acid combinations have rather different microwave absorption
characteristics. For example, nitric and hydrochloric acids (1:1, v/v) require only 531 W
as compared with 545 W for nitric acid alone to achieve the same heating rate and take the
same mass of water to the same first-stage temperature and ~3% less power to go from
189
160 to 170 °C in the second stage. These systems can be manipulated sufficiently well so
that precise conditions can be designed to maximize the efficiency of a microwave
digestion and accomplish very specific goals. If the components of the matrix of the
samples are known, then the digestion conditions can be individualized to take advantage
of each acid's best characteristics and digest a matrix at a specific temperature rather than
run some general procedure that is not optimized but represents compromise conditions to
achieve a general level of efficiency.
Modeling
Because the exact conditions may not be predictable from theory, digestion
parameters of an acid combination (e.g. aqua regia), can only be approximated before
use. Digestion can be modeled by measuring the temperature and pressure curves for the
acid mixture and evaluating the specific quantity of reagent (total number of sample
vessels) needed in the reaction as was shown for digestion of river sediment. Monitoring
equipment is necessary for this step, but once the conditions are worked out, the
temperature and pressure no longer needed to be monitored. Systems can be "run blind",
in vessels equipped with pressure relief disks (102) or valves that will relieve the
pressure in the vessel before a critical pressure and temperature are reached. This
technique was also used to devise the microwave digestion test recommended to EPA for
the sample preparation step in the trace element analysis of high particulate waters. Acid
combinations are ideal for many materials and can be monitored during procedure
development to confirm the exact conditions under which the sample was decomposed
before analysis. Because the major absorber of power, in most cases, is the acid and not
the sample, many of the procedures that use the same quantity of the same acid should
exhibit similar temperature profiles. Temperature and pressure measurement of
decompositions with new reagents can provide empirical descriptions of expected
conditions for which predictions cannot yet be made.
Because microwave energy is a directly coupled power source and is available at
relatively high wattages, it is possible to create unsafe conditions in a very short period of
time. Real-time monitoring provides the opportunity to observe the conditions in the
vessels and reduce or discontinue the power before the vessel specifications are
exceeded. The vessel temperature and pressure limits depend on the design of the
container and the type and grade of material. Thus, these limits will be unique for each
container model. Furthermore, if conditions are not monitored, there will not be enough
warning to prevent unwanted or uncontrolled vessel venting. In addition to providing
information for research or quality control, real-time monitoring of conditions is good
laboratory practice.
Even if decompositions are monitored, it is a good idea to use a pressure relief
valve, as described, to minimize the possible occurrence of unsafe situations. The form
of pressure relief depends on the particular type of vessel and equipment configuration.
If the release pressure of the valve is set to just below the safe operating limit of the
vessel, the device will compromise the closed system only if the pressure becomes
unsafe.
Reactions by Specific Matrix Components
Superimposed on the temperature and pressure profile of the acids is the specific
decomposition of biological and botanical matrix components. In nitric acid, these occur
at characteristic reproducible temperatures that have been identified through temperature
monitoring and are specific to the component, but characteristic enough to appear in
composites that contain sometimes, one, two or all three components. With knowledge
191
of the matrix, specific decompositions can be tailored to the achieve the most efficient
digestion. For instance, because of the large partial pressure developed in the Teflon
vessel, nitric acid alone cannot attain temperatures much above 180 °C. This temperature
is usually not adequate for the complete destruction of complex aromatic rings that are
nitrated during the reaction. Addition of small amounts of phosphoric acid, which lowers
the partial pressure, can raise the temperature to 200-210 °C, where additional
decomposition occurs. Such combinations may prove useful in eliminating the perchloric
acid finishing step required in so many digestions of biological materials that must be
analyzed by methods that are sensitive to the presence of trace organic species.
When the specific composition of the sample is known, then the digestion
procedure is basically an application of the specific decomposition temperature needed for
those major components applied within the framework of the modeling technique. For
instance, when bovine liver samples must be digested, the target temperature selected
should reflect the composition of the tissue (wet) which is nominally 16.5% protein,
nearly the same amount of lipid and about 6% carbohydrate; the balance is water and
electrolytes. The temperature and pressure profiles shown, Figures 40 and 41, suggest
that the matrix behaves more like a lipid in nitric acid, than like a protein. By collecting
temperature and pressure data during the decomposition of liver samples the transition
temperature of 175 °C, at which half the decomposition is complete, is observed to be
well above that of the lipid model. Thus, materials containing complex lipids such as
cholesterol require target temperatures of at least 170-180 °C for microwave
decomposition.
When analyzed by the same predictive technique Wheat Flour SRM 1577a
behaves exactly the way a carbohydrate is expected to behave, that is, it is 50%
192
decomposed at a temperature of 142 °C. Target temperatures of 140-145 °C cause the
decomposition of the carbohydrate to proceed essentially to completion.
This semi-quantitative assessment of the decomposition of organic materials
provides a historical record of a sample preparation process that targets decomposition
temperature for the specific components of a matrix. By specifying the decomposition
temperature and the number of steps in the process, similar materials may be prepared for
analysis by similar digestion programs. It is completely reproducible. With specific
temperatures of decomposition identified for carbohydrates, proteins and lipids it is
possible to optimize the sequence by decomposing the easily-oxidized carbohydrate and
protein components first and then conduct a more vigorous attack of the sample in a
second step to digest the remaining lipids and complex ring structures.
Completeness of Decomposition
Targeting a specific decomposition temperature for the individual components of a
biological matrix allows the material to be degraded efficiently and safely. That is,
degradation of the carbohydrate components of a 250 mg organic sample that is digested
in 5-10 mL of nitric acid at 140 °C generates 100 ± 10 psi which is easily contained in the
120 mL Teflon PFA vessel with little risk of valve release. Materials that are all
carbohydrate may be satisfactorily digested in a one-step procedure that is conducted at
140-145 °C for a minimum of 6-10 min. A hypothetical carbon compound
-(C2H4)n-
+ 3 n 0 2 ---------> 2nCC>2+ 2nH20
(12)
with a molecular weight of 100 generates 0.005 moles of C02 and occupies a volume of
112 mL and exert a pressure of 1.09 atm at 25 °C. Confined to a volume of 120 mL, the
0.005 moles have a pressure of 1.04 atm; at 145 °C, they exert a pressure of 1.42 atm in
193
the same volume. The pressure in the vessel at 145 °C then, is a combination of the
vapor pressure of the acid, the matrix decomposition products, and nitric acid
decomposition products as well. From the heating profile shown in Figure 40, it can be
seen that 5 mL of nitric acid contribute about 6 atm to the vessel at 145 °C. When the
vessel is cooled to 65 °C, 1.03 atm remain.
Tissue samples that are high in protein or that have high lipid fractions require a
second digestion to be carried out at temperatures >160 °C, usually between 170-180 °C
so that the long-chain hydrocarbons and the easily oxidized aromatic rings are destroyed.
If no lipid-like components are present, then high carbohydrate biologicals are digested in
a single-stage prolonged decomposition at 140 °C with slow rise to 150 °C near the end of
the digestion. The second digestion at 170-180 °C reduces the remainder of the matrix
to carbon, hydrogen, and nitrogen components. Similar optimum dissolution
temperatures are expected for many of the inorganic matrices such as ores, metals and
other inorganic compounds. At the higher temperatures, any remaining undigested
organic material is usually decomposed and its trace metals successfully mineralized.
However, not all of the trace organic constituents will have been completely decomposed
during this process (54,162,163). Nitrated aromatic benzene rings have been found in
the nitric acid digestates of Bovine Liver SRM and pure Bovine Serum Albumin (Pratt,
K., NIST, unpublished results). Methods of analysis such as polarography are sensitive
to nitrated organic species so that a further decomposition using perchloric acid is needed
to get rid of the residual organic moieties. Many inorganic analytical instrument
determinations, such as ICP or AA, will not be seriously affected by trace quantities of
these molecules. In such cases a final treatment with concentrated perchloric acid may be
necessary to completely decompose the resistant ring species.
194
Complete decomposition of organic matrices is only rarely achieved using special
reagents such as perchloric acid. Since it is desirable to avoid the use of this strong
oxidizer, higher pressure and temperature decompositions using nitric acid are an
acceptable alternative. Oxidation aids such as hydrogen peroxide are occasionally added
to digestions of biological materials and may be safely used in microwave systems,
(Littau, S., CEM Corp. personal communication, 1988). The extent of nitric acid
dissolutions of organic samples in steel-jacketed PTFE bombs at high temperatures and
pressures has shown that the decomposition of these samples was incomplete even after
treatment at 180-200 °C for 3 hours (2,162,163). When the residues from those
dissolutions of biological tissue were analyzed, many of the same molecules were found
in the PTFE bomb decompositions that had been found in the 10-15 min microwave
decompositions at 180 °C. Similar ratios of ortho-, meta-, and para-nitro benzoic acids
were found, as well as small amounts of dinitrobenzoic acid and several other aromatic
compounds (163).
Although nitric acid decompositions of organic samples may not be complete even
when high temperatures and pressures are applied for extended times, these high
pressures and temperatures do appear to achieve reproducible decompositions. Under
controlled conditions, nitric acid at high temperatures and pressures decomposes all but a
few organic moleculesand they have been identified in reproducible amounts in solutions
resulting from dissolutions of both biological tissue and individual biological
components. Since reproducible decompositions are possible from both microwave and
steel-jacketed bomb dissolutions, elemental analysis using techniques which do not
require complete decomposition of organic materials should be evaluated for their
tolerance to these molecules. Although the closed-vessel microwave nitric acid
195
decomposition of organic tissue gives comparable results to traditional steel-jacketed
PTFE bomb decompositions, these same conditions are achieved in a much shorter time
due to direct transfer of the microwave energy. The usefulness of these high temperature
and pressure decompositions in conjunction with a variety of instrumental trace element
analyses has been demonstrated in the literature for several decades (2).
Reproducibility and Thermal History
Accurate and reproducible temperatures are nearly impossible on a hotplate
because of uneven heating of the surface. Therefore, unless only one sample at a time is
prepared, multiple sample preparations on a hot plate are, of necessity, digested under
different temperature conditions. Differences in contamination exposure from different
parts of a hood are also possible. All techniques that rely on conductive heating for acid
decomposition suffer from extreme thermal gradients in the vessels.
One of the most important factors in successful trace element determination is
homogeneity and uniformity of the analyte medium. As far as is practical, sample and
controls are the same for similar matrices and the goal of sample preparation is to produce
a
solutions for analysis that are as alike as possible to eliminate matrix effects i.e., to
eliminate "problems with the chemistry". To that end, it is desirable to have sample
preparations that are as free as possible from uncontrolled variables. Since the analyst
controls the sample and reagent weights, and controls the length of the digestion program
it is advantageous as well to control the thermal history of the sample during
decomposition.
Solutions in closed vessels heated with microwave energy tend to get much hotter
because of the direct coupling of the energy with the molecule. Solution temperatures
196
also tend to be more uniform because there is no thermal gradient and all parts of the
solution are heated at once. Uniform heating is facilitated by circulating the vessels
through the microwave field using a carousel. Better distribution of the wave pattern in
the cavity and a more homogeneous field is achieved using a mode stirrer to deflect the
*
incoming waves. The direct coupling of energy with the solvent molecule in a relatively
homogeneous field condition means that sample containers in the microwave field
experience very similar heating conditions and have very similar thermal histories. If the
acid and sample weights were similar and the microwave power program was the same,
then the temperature and pressure profiles of the different batches were similar.
Reproducible digestion conditions can be achieved in single-stage, one-step heating
programs as demonstrated for the river sediment, see Figure 51, as well as in single-stage
two-step dissolutions for the biological and botanical matrices that were shown in Figures
48 and 49. One-step power programming of the dissolution is also capable of replicating
the temperature and pressure conditions within 2% for successive batches as seen in
Figures 54 and 56 for the water samples with different acids. Reproducibility allowed
the implementation quality control so that the conditions of digestion became the criteria
for good sample preparation, rather than the results of the analysis.
Benefits of Microwave Dissolution
Applicability
Decompositions using microwave energy have been demonstrated for a broad
range of materials. Its suitability for the preparation of dissolved samples from nearly all
matrices requiring analysis for trace elements and minerals continues to be documented in
the scientific literature on a routine basis. Although wet ashing by acid hydrolysis has
been the predominant appplication attempted to date, base hydrolysis is certainly a logical
197
extension of the technique. The feasibility of microwave assisted acid hydrolysis of
proteins for the analysis of the amino acid content has recently been demonstrated (156).
For many years, Gedye and Smith have conducted classic organic hydrolyses and the
syntheses of esters in microwave units (31-33,132). A vety elaborate microwave
extraction technique for soils and sediments has been developed and refined by Mahan
(37) and a one-step extraction technique using nitric acid has been proposed to EPA (127)
for the analysis of solid waste samples.
Speed
The increased oxidizing power of nitric acid at elevated temperatures results in the
efficient decomposition of carbohydrates. Because nitric acid boils near 120 °C, this
higher temperature cannot be reached in open vessels; thus, a closed-vessel microwave
dissolution technique is much more efficient than traditional open-vessel dissolution.
Among the most notable advantages of the microwave dissolution technique is the
reduction in sample preparation time. This is a direct consequence of the increased rates
of reaction that result from the higher temperature and pressure conditions that are
obtained in closed vessels. Some comparisons between conventional digestion
procedures and microwave dissolutions in ours or similar laboratories are shown in Table
24. No longer is sample dissolution tedious and labor-intensive; sample preparation is
now on the same time scale as the analytical technique which quantitates the elements.
The need for constant addition of reagent to maintain volume has been eliminated and
throughput of 25-30 samples h-l is quite common.
198
Table 24. Comparison of Traditional Digestions with Microwave Sample
Preparation Methods
ji-wave
method
jx-wave
time
times
faster
Source
Biol,SRMshotplate,ICP
~8h
HN03,HC1
HF,PFA
30min
16x
USDA.ND
Biol-hotplate,ICP
1.5min
lOOx
Stripp&Bogen
Geol-hotplate,ICP
6h
HN03,HC1, PFA
4min
90x
Kidd Creek
Biol, Kjeldahl
IC,Spectra (N2)
2h
H 2SO4, glass
4-6min
20x
CEM
40min
36x
Union CarbideNBS
1
HN03.HC1
Parr
CC-AI2O3, ICP
JS
Traditional
lime
00
Sample-Prep.
Time
24h - H2SO4, H 3PO4,
270 °C, PFA
NBS 179,BCS452
Fe alloys-Parr.ICP
lh
PFA
80s
45x
Allegheny
Ludlum
Water, Waste W. etc
ICP
6h
HNO3, PFA
lOmin
60x
EPA-NBS
Bio-hotplate,AAS,
GF
6-8h
HNO3, PFA
38min
9.5x
USDA,MD
Organic synthesis
l- 2h
H20-Et0H, PFA
2min
12-60x
Laurentian
Univ.
Safety
Microwave acid decomposition in closed vessels introduces unique factors that
many chemists have not encountered. Acid combinations traditionally used to digest a
particular matrix may be inappropriate when microwave energy is used as the heat source.
Although numerous papers on microwave dissolutions have appeared in the literature
since the inception of this research, the combinations of acids, reagents, samples and
199
methods that are safe and reliable are just beginning to be well documented. A distillation
of the knowledge acquired through many carefully conducted experiments, as well as
from discussions with experts in combustion chemistry and microwave engineering has
been presented in the Chapter 11 in Introduction to Microwave Sample Preparation
(100). Situations such as equipment failure, arcing of metallic samples, flammable
solvents and exothermic reactions that could lead to problems in the laboratory are
discussed.
Knowledge about the specific makeup of a sample matrix is essential to determine
the most efficient temperatures necessary for decomposition and allows the analyst to
more readily control the sample digestion process. Efficiently designed microwave
digestions require reaching and maintaining minimum temperatures that rapidly
decompose the major organic and inorganic components in the matrix. Whereas,
previously, quality control in sample dissolution has been assessed by comparing the
results of elemental analysis, perhaps the specific characteristics of the digestion more
properly form the basis for judgment prior to analysis.
APPENDIX A
TORQUING DEVICE
WW h
Materials
4.
5.
One adjustable torque wrench covering 0-36 ± 1% N m (0-75 ft lbs)
One sheet of stress-relieved polypropylene, 2'x 2' x 1/4 or 3/8
One set of Savillex torque wrenches (for 60 and 120 mL Savillex Teflon PFA
vessels)
316 stainless steel sheet, l'x l'x 1/4 "
Assorted stainless steel bolts
Construction
Refer to Diagram:
1.
Vessel holder fabricated from two sheets of 1/4 (3/8) inch stress-relieved
polypropylene sheet.
2.
One Savillex torque wrench bolted to polypropylene cut out to match the wrench
3.
Barrel of wrench socket fabricated from three pieces of stress-relieved
polypropylene.
4.
A stainless steel adaptor for the drive is attached to the top of the socket
5.
One-half of the second Savillex wrench with 1/2" cut from perimeter is bolted to
the bottom of the socket.
200
Figure 58.
Torquing
Device
201
APPENDIX B
THERMOCOUPLE CONSTRUCTION FOR TEMPERATURE
SENSING IN A MICROWAVE CAVITY
Materials
1.
One-piece thermocouple assembles such as Type "T" (suitable to 450 °C)
ungrounded, in a 1/16 inch (.062") diam. 304 or 316 stainless steel tube, with
ceramic insulator, 3.2 and 6.0 inches long, closed at one end, with 36 inch
(minimum) leads of glass/glass or Teflon/Teflon insulator.
2.
Thermocouple extension wire (AWG 30) to connect meter or recorder.
3.
Copper braid, 1/8 inch i.d.
18 inches for small thermocouple
15 inches for tall thermocouple
Assembly
1.
Solder small gold collar to stainless tube by means of a 11/2 inch solid gold coil
under collar.
2. Weld extension wires as needed if thermocouples are short. Otherwise next step.
3. Trim braid and clean for plating.
4. RTV at junction of wires and tube, if not already sealed.
5. Attach connector to small collar on tube. This step may be done after plating, if
carefully crafted.
6.
a) Electroplate stainless tube to entire length with 10-14 pM OF 24 K gold,
b) Electroplate braid with 10-14 pM 24 carat gold.
7.
Attach connector to braid for tube attachment and slide over thermocouple extension
wire. Crimp tight.
8.
Thread knurled subminax connector over wires with braid and attach to back end of
braid with collar (sleeve). Crimp tight.
9.
Cover entire braid with flexible, microwave transparent heat shrink material.
202
203
10. Attach thermocouple extension wires to suitable miniature connectors for Data
J.x>gger or temperature readout
11. Ground subminax connector at back end of braid to unit wall in a panel mount.
12. Test thermocouple in RF environment
These last two steps are best done in the laboratory just before use.
12. Make a Teflon sheath for thermocouple tip to fit in port of 60 or 120 mL PFA vessel.
13. Check electrical integrity.
14. Calibrate thermocouple to ±1.0 °C.
204
(On
d>
JD
kO.
Q.
a>
k.
3
CO
a>
III
■o -o 2
O g -9
uE«
CL
E
£
2.
H
G>
Q.
3
O
O
o
E
»_
0 )
£
H
<0
o
V
‘5.
a>
c
O
•
a
■
in
G l a s s or T e f l o n ®
in s u la tio n
2
3
O)
il
APPENDIX C
MICROWAVE CALIBRATION PROCEDURE
1.
Fill two 2500 mL glass beakers with DI water and let them come to room
temperature. Fill a small "squirt" bottle with DI water.
2.
Have four 1-L Teflon PFA containers and their covers (closed ports) clean and at
room temperature.
3.
Condition the microwave unit by heating a glass beaker with 500-1000 mL of tap
water at full power for 5 minwith the fan on full.
4.
Assemble the thermometer (0.1 °C divisions and estimable to 0.05) and stirring
platform. Have two magnetic stir bars and a pair of Teflon tongs available.
5.
Weigh 1000 g of room temperature DI water into a 1-L Teflon container.
6.
Measure the temperature of the water to .05 °C while stirring gently. It is best to
have water at 24 ±1°C. Remove the stir bar.
7.
Reweigh the water. Add or remove a drop or two of water, if necessary, to bring
the weight to 1000.0 ±0.2 g. Cover the container.
8.
Place the filled, covered container on a shallow flat turntable near the center post.
9.
Program the microwave unit for 2 min and 2 s at full power (100%). Press run/on
and start the turntable rotation (reversing mode). Keep fan on full all the time.
10. Remove the container to the stirring platform as soon as the program has
terminated.
11. Carefully remove the cover and insert a stir bar. Lower the thermometer into the
container near the center.
12. Measure the temperature to 0.01 °C (estimate the second place) while stirring gently.
Record the highest constant reading taken in the first 20-30 s.
13. Remove the thermometer and stir bar. Discard the heated water and wipe the
thermometer dry. Let the container and cover cool.
205
206
14. Repeat the process twice more. Refill the empty 2500 mL glass beaker with fresh
DI water after the 2nd trial.
15. Repeat the measurement process at 99,98,97,95,90, 80,70, 60, 50 and 40%
power.
16. The calibration measurements can be run through the MICROWAVE
CALIBRATION ASSISTANT written by Peter Walter (126). This program will
determine the linear operating region of the microwave unit. It will also produce a
plot of the percent power and watts calibration for the unit
' APPENDIX D
CALIBRATION OF MONITORING EQUIPMENT
Transducer Calibrations
Table 25. Transducer Calibrations
#2
#1
EMF, v
Pressure, psi
EMF, v
-0.008627
-0.013961
0.21424
0.69548
1.1909
1.6968
2.2428
2.6837
3.1474
3.7122
4.1670
4.6952
0.0
14.5
26
50
75
100
128
150
173
201
224
250
-0.014320
-0.012560
0.21819
0.74275
1.2071
1.6914
2.1896
2.6800
3.2034
3.6810
4.1886
4.6204
Pressure, psi
0.0
14.5
26
52
76
100
125
150
176
199
225
246
NOTE: NBS standardized air pressure gages in the gas and
particulate division used in calibration.
Coefficients of linear regression
#1 Transducers
A = 12.043
B = 51.057
R2 = 0.998
#2 Transducers
A = 12.036
B = 51.037
R2 = 0.998
Optical Fiber Calibration
Calibration of the optical fiber thermometer consists of look-up tables based on
the fluorescent decay of a proprietary (Luxtron Corp) phosphor sensor. An analog
207
208
output board produces an analog voltage that is proportional to temperature according to
the following equation,
V out= Scale Factor ( T meas ■T offset)
where the Scale Factor is 10 mv °C-l and T offset Is where the analog output is zero
volts.
(13)
APPENDIX E
GRAPHS OF QUARTIC FIT OF REAGENT DATA
209
210
o
CM
>,
(0
o
c
0)
c
o
'o
o
w
o
z:
o
75
o
.5
ea>
sz
ui a. O
c
nJ
o
o
'C
©
o
CO
E
(0 ■=
2
<
E ■o
CO
o c
00
(0 . O)
OQ
05 T 3
<r
(0- o a»
o ^ >*
Ol
co
co
__
CO
< E a* o 8
■ COI s& x ^
NITRIC ACID, 1M
CL 5
X k.
Ui
o
a>
5
C
O
o
vP
o to
a . o>
■o
CD
n
o
CM
CO CO
CO
CO
n
<
o
CO
o
o
o
o
LO
M ‘U3M0d
o
o
O
o
3
.5>
u.
2
Z ©o
c
>. £
O £
*_
a?
0
o
“O O
c
*
<0
CO
r s.
Sf
01 -Os
XJ
3
0) .E
■D Q
.
■2£
o
*5
o c
a.
211
o
&
m
<0
o
~
c0)
c
’o
o
5
.2
CO
o
15
o
If
X Ui s ' I
LU Q.
8 “
m
•> o
TO a>
O E
< <
oo
o
a oa>
ffl u
HYDROFLUORIC ACID, 29 M
0) t
0 0
a "O
„ a> a>
o /»
o <
as “
to <o
< E 5>
o>
> v
Q
.
O
O
„
2 ~(A 2 ■a
> 1 to
Ui C 1 s
O
a>
3
O
vP
o'o in
Q. o> a> E
o
■o
o
"S £ 2 co
'55
\ «*
co
o
0)
CM
<
(A
CO
CO
2
<D o
o
o
to
o
o
o
CO
M ‘UBMOd
o
o
O
a>
3
.2> o
u. a.
wO E
k.
r s.
5 ■=
o 5
XS ffl
a> c
■o ■§.
3 ffl
o £
*•
SULFURIC ACID, 18M
212
m
‘aaMOd
213
T3
o
fl>
XJ
3
UL
o
O
c
is*
CD
•75
M c
+-»
'o
o
c «o
d> o-•
c
O
a.
X
CO
75
o
T3
E
a>
O
c
ro
o
'k.
CD
E
<
©
UJ Q.
a>
HYDROCHLORIC ACID, 12 M
oo
CO
E XI
o c(0
S
CO
co
CM °>
CO m
•> D)
„TJ
o m a> 0)
o 2
c a a.
CO
a> <
<
g
sE ■o
a
w
•—
“
CD
k. o
LLl
o2 £
£ ffi
a>
vO
5 o ' aO ao
±= E
a. in
a> T 3 o
o
>» -
X
■o
0)
S3
o
(0 to
.Q to
a
to
o
o
■'t
o
o
00
M ‘H3M0d
o
o
o
“l_
O E
**- E
-
8.
< S
2■o ■«=
05 0
_
*S ^ffl
(O
o
o
c
o
w
</)
O 5
**
c
O =
0)
Io
t
3 1 <5 £
O)
3
iZ
a
V
214
CM
O g)
■o
u= o
_ c
>»
*-*
CO “
0
*o
o
.2 c
c .2
co
15
o
2S
E
O T3
0
o. a>
x Qi
qj
.c
O
c
0
O
*w
0
E
HYDROCHLORIC ACID, 6 M
fl>
00
_
<
CO
E -o
o>
o c
w
Q
Q
00
<D £
CD
c/>"§ g
‘k.
Si.g
< CO8
c 2 £
to
CO ^
o
00
CD
O 0
w
i ■= g
*- O o ®
<0
E
o'*
>2
o £
O UO
E
"O o
>»
■o -c X co
'in
2 ■.a
> i_ in
ft
o— E
w
O
H
o
Q . CO
o
co co
JQ
CO
<|
s
o
o
in
o
o
o
CO
M ‘H3M0d
o
o
CM
o
o
a.
a> _
S = |
O ■=
£ 0>r t
.®0 3
ILQ.O
O
O
I
T3
G)
CM
ii”
O
.2 c
o
o
O TJ
.2
’5
o
E .2
i.2
CO
15
o
O T3
Q. 0)
o
o
HYDROCHLORIC ACID, 1 M
CO
E
0)
X '
-C
111
0>
O
c
nJ
.2
*aZ
_
co
<
E "2
CO
CD
°
o °>o
CO®
o
<o « C 0
E
CO
o
§
CD
c
1 1 5
CO ^
x:
* o>
s
< °
CD
.2
o
o
•st
o
fl>
sP
2o% in
o'
Q. o>
*o JC
2 —
o
o
CM
€
5
S
CO
CO
CO
.D
<
■C £
O
J— i— i— i— i— JL-i—
CO
O
O
lO
O
o
i— i— i.
O
O
CO
M ‘UBMOd
» l-i.
o
o
CM
O
O
£
| >•l
—
*
,|
n -p
P
©
Q
.
o>
2 ■o
o
in o 2
co 2
o
o
®
w o o §
13
<8
0) i- •“O i§•
S: «u
® C DC
CO
3 O 3
U. 0 . O
216
00
O *0
a>
co -o
c
3
.2 o
*
<D
O
S £
o
co
u- o
ai
o
oB
E
r<a =a>5
a>
sz
O
cai
o
*u
(D
■a
c
E
<
as
to
\
■o t ;
o> (0 C CO
a> CO 3
co .;=
ffl 0
< J
00
00
O)
.c
05
*u
a> w Si
a> oa
z w c0) □c
co
-J T3 2
CD
in
a> IP
U
c
C
T
J
(0
£
E oo ra> Q
C
Oa
WATER
2
0)
ui
k_
-
o
£a
in
a>
■ •<
»-
E
o
0) •
a) «a>
5 Si
co
t:
£
'25
f
a>
CL
•
c
o
</>
E
»_ C
O JC
-* o a>
. «- c
co
■o
CD
CD
M;
CM
CD
CD
o
CD
M ‘U3M0d N1
00
ui
CD
ID
00
M1
® 8
CL
s i s
®
3
7; cc
ID
•
2*
co t
L i. - J
^
217
n oo
CD
c
o
«->
o
c
CD 5
3 <D
a g
3
»>
£"
W
□ s1i
U- «*O a> 75
C .2
o
~
2* c •O Oc
a> os
5
w «—
O ffl
IS '=
•a o> O ®
c o° I
c
CO
NITRIC ACID, 16M
co
a>
■o
ct o £
«
LT “
CD
V
.E
.E
o
CO
O —I•
0)
<
CD
s CD
0)
w & C
73 ®
U5
J
2
~
-i
>.
a.
o
»
«o
S
c
£ s
(0 (O 73 ®
E *"”o -2£
co *o <2. E
o
UI o
^
<
a>
1 =
a g
73 o
Q) To
73 |
2 -
o 8.
»- — £
O
5
C -a
I i
CVJ
cb
cb
o
cb
oo
ui
co
ib
rjib
M ‘d3M 0d N1
C
M
ib
o
ib
coCO
’t
CO CO O ®
CO
a) cd ~O aI .
■—
<D
■o cc
1 =
9c
U. -J
CD —
218
-t fs.
O
C
O
c
.o
o
S
io
o
c
Crt «
3
o
CD
O
-
c !
C « J 1
® c ■o
=> § a> c(0
0 S +*
o
o ©
?m °® a <E
C
Oc
.» « 0
k. ® k - CO
C
O®
o> 2 S a> _
co .E § ■—c £a i
W _j Q
LQ<
w
I >.
0) O
C
O Q.
O
NITRIC ACID, 1M
2— m
<0 -5
o''
co
- 1 T3 O .E
mJ ©
o
® .
-a
—■ §©
■I
MM?
O £
IS
C
OTl- W E
uj
m-
--
o
k <
■o §
a> -m
1o k52 TO S2
a . .-s 3 E
a>
O g.
£
i \ \
CO
sz
s
TJ
C
OC
Oo ©
l i
Ti­
ed
E
C
O*5 .1
0) C
B£ a-
C\l
CD
O
OO
CD
ui
CD
to
M ‘d3MOd NT
CVI
ui
ui
o
ui
CO
5 2
.5? c
U_ _l
T3 CC
a> *=■
id
Mr
CO
in
id
CVI
o
id
co
in
co
in
min
M ‘U 3M 0d N1
in
CM
o
in
co
CO
co
Mr
(Reprinted
with permission
from
reference
6, copyright 1988
American
Chemical Society)
Figure 69. Ln Power Estimated as Linear and Quartic Functions of
Ln Mass for Hydrofluoric Acid, 29M- 95% Confidence Bands for
Prediction
Included. (Solid Lines-Linear; Dotted
L ines-Q uartic)
LN MASS, g
HYDROFLUORIC ACID, 29M
co
220
00
O
co
c
o
CO
4-*
3
—■ > ;
Ih
o
®
CO
°
3| o H- Q m
<0 c m
O 13 = 1
5 £
!=
01 x j 5
a m
U.
®
0
o) S - CO
u Q
O
c
©
X) c
E
c
0)
(0 T3 . .. <
c
o> <5 o
CO
-
CO
SULFURIC ACID, 18M
co
g O
w - 1 vO
m 00
00
CO
Q)
=i
<
O'
s i
s_ C
«O J2
C U
®
_ J co'
5 -0 i
®
® £ TJ g
CO
E -o
=
O
£
3
Q CO £
CD *2
E
III H
S
o
ID
>-
O
5
3
13
o
0)
T3
3
—
CL 3
CO
N
co
CM
CD
CD
O
CD
CD
ID
M ‘H3MOd N1
CD
ID
•MID
CM
ID
■*
«
c
o
w
u>
®
Q.
C
CD
0>
c
O
S
®
(0 ? c
23 S Sj? CS’
C
.S> c
U
£ ~
IQ.
221
-i Is*
O
C
O
<0
o '•£
*■
Im
p
co >:
c
o
V-«
o
c
3
u.
C
O <om
ffi a>75
o
Oj
E
(0
3
O
•o
c
§
C
O3 ®
TJ 0 0
C
i °
C O
3) T3
XI
■MM
s §
"
C o ®
c o
C
O
o
q
<E
sP M
L '«O)
O
O) C
d>
S2:
HYDROCHLORIC ACID, 12M
w> .£ , £ %
LO <
5
2
MC
Oo
C
OC
rW
„
_J TO ■o
Ci o
0> °
.E
CO
- I ffl
§
C
O
E <2 g
o o £
• MM
H -
‘5
C
OI
( 0 E
UI o ^ 5
. JC . ■*"
o>
o
3
o
-O
g
0) -55
o -o -o 1
t
1 O <Q
“.
a
d
c
-1
c ^
“ £
5
-r - C
s
O o ffl
CM
cd
o
CO
00
id
CO
to
Tjid
CM
id
M ‘U3M0d N1
©
id
co
■M
"
CO
CO
2
C
O^ .E
0) C
OO
w"
a.
5
_
5 ir
.2> c
U. J Q.
HYDROCHLORIC ACID, 6M
CM
cd
10
o
cd
co
id
CO
id
J
■M
"
id
i
CM
id
L
M ‘H 3 M 0 d N 1
o
id
co
■M
"
(Reprinted
with permission
from
reference
6, copyright 1988
American
Chemical Society)
Figure 72. Ln Power Estimated as Linear and Quartic Functions of
Ln Mass for Hydrochloric Acid, 6M-95% Confidence Bands for
Prediction
Included. (Soild Lines-Linear; Dotted
L ines-Q uartic)
COCO
Tf
LN MASS, g
co
223
h-
•*—
C
O
c s o S ts:
O©
o (0 C
3 'u
c
3
«
u. ?CO? CO
«
m a) o
o
c e
co
r ® □ 5
(0
3
g
o
»
■o tc: oS e o
c o q <
C
O
O
CD
co
O) (0
-
HYDROCHLORIC ACID, 1M
C
O
d>
vO IT 2
c So> 2z: -o)
I ■“_J >Q.,
C
OS I O
in < □
S
Z C
O
0) °
co"
■o •a —i ®
o
o .3
O
C
C
O< T J S
E
■3 o o £
C
OI d ) c
Hi
cu
o
w
p
o -a
§
S P ® w
sO O
i tZj in
r
0 . >* . 2 E
lo g .
c
c _
-J £O — ■
§
3
■
*"
e
CO
r«- co —
o xjeg
co 3
I
Ttco
i
CM
cd
o
cd
00
co
■M
-
CM
id
id
id
id
co
» ( £ -
Li. _J 0 .
M ‘d3M 0d N1
.E
0) C
O O Q.
APPENDIX F
TABLES OF COEFFICIENTS FOR THE FIRST-ORDER
AND FOURTH-ORDER EQUATIONS
Table 26. First-Order Model Coefficients
Acid
A’
Water
H2SO4 (18 M)
HC1 (1 M)
HC1 (6M)
HC1 (12 M)
HF (29 M)
HNO3 (1 M)
HNO3 (16 M)
B’
5.1593378
4.36467137
4.51149281
3.70091195
3.7240524
3.961217
4.39165309
4.11758785
0.17242256
0.25646561
0.2421269
0.32487487
0.31907256
0.30172725
0.25572899
0.27124904
NOTE: Reprinted with permission from reference
6, copyright 1988 American Chemical Society.
Table 27. Coefficients for the Fourth-Order Model
Acid or Water
H2O
H2SO4I 8M
HC1-1M
HC1-6M
HC1-12M
HF-29M
HNO3-IM
HNO3-I 6M
A
3.2003974
-8.29138
14.0885888
28.5892761
19.5198223
36.4337146
16.8621705
27.1480917
B
C
1.18320369 -0.160514
5.79513997 -0.666696
2.49106325
-7.86067
6.18708653
-20.2258
3.42081805
-11.8319
7.91099431
-26.2649
3.16719797
-10.1777
4.9371303
17.3782
E
D
0.0079261
0
-0.330417
-0.806255
-0.419341
-1.0178
-0.414305
-0.598649
0
0.00271
0.01599077
0.03845682
0.01894251
0.04785757
0.01976774
0.02661368
NOTE: Reprinted with permission from reference 6, copyright 1988 American Chemical
Society.
224
APPENDIX G
DERIVATION OF THE NEWTON COOLING EQUATION
The rate at which T decreases is proportional to the temperature difference between the vessel and
its surroundings.
when t = o
when t =«»
T(o) = To
T(oo) = Tf
T = f(t) = T(t)
statement
^
= -k (T(t) - Tf) = -k (T-Tf)
Let T '
(A)
and since T=T(t) then T ' = -k(T-Tf)
T ' + kT = kTfs which is like y' + py = q
the integration factor =
if p = k eft*11= e^kdt
and the integration factor = ekt
multiply (A) by e^
T'ekt + kektT = kektTf
CTekt)' = kTfekt
Integrate
Tekt = JkTfekt = kTf ( -|e k9 + C
Tekt = Tfekt + C
(B)
T = Tf+Ce-kt
at the boundary where T(0) = T0 and T(oo) = Tf
in eqn (B)
for t = oo, T = Tf,
and when t = o, T70) = T0 = Tf + C
and C = T0 - Tf
225
226
(C)
T = Tf+(T0-Tf)e-to
at any time "t"
(T-Tf) = (T0 -Tf)e-to
ln (T-Tf) = ln (T0-Tf) + log e*kt
ln (T-Tf) = ln (T0-Tf) - kt
A graph of this function is a straight line whose slope is -k.
BIBLIOGRAPHY
1.
Abu-Samra, A.; Morris, J. S.; Koirtyohann, S. R. Anal. Chem. 1975, 47.
1475-1477.
2.
Jackwerth, E.; Gomiscek, S. Pure Appl. Chem. 1984, 56(41. 480-489.
3. Kingston, H. M.; Jassie, L. B. Anal. Chem. 1986, 58. 2534-2541.
4. Kingston, H. M.; Jassie, L. B. Presented at the National Bureau of Standards
Symposium on Accuracy in Trace Analysis, Gaithersburg, MD, September,
1987.
5.
Kingston, H. M.; Jassie, L. B. Presented at the 25th Eastern Analytical
Symposium, New York, NY, October, 1986, #76.
6. Kingston. H. M.: Jassie. L. B. Introduction to Microwave Sample Preparation:
Kingston, H.M.; Jassie, L.B.. Eds.; American Chemical Society: Washington,
DC, 1988; Ch 6.
7.
Castellan, G. W. Physical Chemistry. Addison Wesley: Reading, MA, 1964; Ch
27.
8. Murphy, T. Accuracy in Trace Analysis: Sampling. Sample Handling. Analysis
vol I; LaFleur, P. D. Ed.; NBS Special Pub. 422; U.S. Gov't Printing Office:
Washington, DC, 1976; p 509.
9. Brown, A. B.; Keyzer, H. Contrib. Geol. 1978, 16, 85-87.
10. Barrett, P.; Davidowski, Jr, L. J.; Penaro, K. W.; Copeland, T. R. Anal. Chem.
1978,2, 1021-1023.
11.
Cooley, T. N.; Martin, D. F.; Quincel, R. H. J. Environ. Sci. Health Part A
1977,12 (1&2), 15-19.
12. Andoh, K.; Saitoh, T.; Takatani, A.; Takahashi, F.; Tazuya, Y.; Tsunajima, K.;
Motoki, C.; Yasuoka, K.; Yamaji, Y.; Natsuoka, C. Kenkvu Kivo - Tokushima
Bunri Daigaku 1982,2$, 113-125; Chem. Abstr. 1982,22, 125965c.
13. Keyzer, H. Chemistry in Australia 1978, 45(21. 44.
14. White, R. T.; Douthit, G. E. J. Assoc. Off. Anal. Chem. 1985, 68(41.766-769.
15. Nadkami, R. A. Anal. Chem. 1984, 56, 2233-2237.
227
228
16.
De Boer, J. L. M.; Maessen, F. J. M. J. Spectrochim. Acta Part B, 1983, 38.
379-746.
17.
Matsumura, S.; Karai, I.; Takise, S.; Kiyota, I.; Shinagawa, K.; Horiguchi, S.
Osaka Citv Med. J. 1982.28. 145-148.
18.
Blust, R.; Van der Linden, A.; Decleir, W. Atomic Spectroscopy 1985,6 (6),
163-165.
19.
DeMenna, G. J. Presented at the 23rd Eastern Analytical Symposium, New
York, NY, October, 1984, #166.
20.
Matthes, S. A.; Farrell, R. F.; Mackie, A. J. Tech. Prog. Rep-US, Bur. Mines
1983, No. 120.
21.
Lamothe, P. J.; Fires, T. L.; Consul, J. J. Anal. Chem. 1986. 58. 1881-1886.
22.
Fernando, L. A.; Heavner, W. D.; Gavrielli, C. C. Anal. Chem. 1986, 58. 511512.
23. Fischer, L. B. Anal. Chem. 1986, 58, 261-263.
24. Westbrook, W. T.; Jefferson, R. H. J. Microwave Power 1986, 21(1), 25-32.
25. Smith, F.; Cousins, B. Anal. Chim. Acta 1985,177.243- 245.
26.
Foner, H. A. "Rapid Pressurised Acid Dissolution of Rock and Mineral Samples:
An Assessment", Geological Survey of Israel, Report No. G/83/2, Nov., 1983.
27. Papp, C. S. E.; Fischer, L. B. Analyst March 1987, 112. 37-338.
28. Bettinelli, M.; Baroni, U.; Pastorelli, N. J. Anal. At. Spectr. 1987, 2(5), 485489.
29.
Jassie, L. B.; Kingston, H. M. Presented at the 36th Pittsburgh Conference and
Exposition, New Orleans, LA, March, 1985, #108A.
30.
Nakashima, S.; Sturgeon, R.; Willie, S.; Berman, S. Analyst 1988, 113. 159163.
31. Gedye, R.; Smith, F.; Westaway, K.; Ali, H.; Baldisera, L.; Laberge, L.;
Rousell, J. Tetrahedron Lett. 1986, 27 (3), 279-282.
32.
Gedye, R. N.; Smith, F. E.; .Westaway, K. C. Can. J. Chem. 1988, 66, 17-26.
33.
Gedye, R. N.; Smith, F.; Westaway, K. Ed. Chem.1988, Sept.. 55-56.
34.
Giguere, R. J.; Bray, T. I.; Duncan, S. M.; Majetich, G. Tetrahedron Lett.
1986, 27(41), 4945-4948.
229
35.
Hwang, D. R.; Moerlein, S. M.; Lang, L.; ’Velch, M. J. J. Chem. Soc.. Chem.
Commun. 1987, 1799-1801, Com.#959.
36.
Neas, E.; Zakaria-Meehan, Z. Introduction to Microwave Sample Preparation:
Kingston, H. M.; Jassie, L. B., Eds.; American Chemical Society: Washington,
DC, 1988; Ch 8.
37.
Mahan, K. I.; Foderaro, T. A.; Garza, T. L.; Martinez, R. M.; Maroney, G. A.;
Trivisonno, M. R.; Willging, E.M. Anal. Chem. 1987, 59, 938-945.
38.
Kingston, H. M.; Jassie, L. B.; Fassett, J. D. Presented at the 190th National
Meeting of the American Chemical Society, Chicago, IL, September, 1985,
#ANAL 10.
39.
Veillon, C.; Patterson, K. Y.; Kingston, H. M. Presented at the 28th Rocky
Mountain Conference, Denver, CO, August, 1986, #8.
40.
Veillon, C.; Patterson, K. Y.; Kingston, H. M. Presented at the 13th FACSS,
St. Louis, MO, September, 1986, #689.
41. Revesz, R.; Hasty, E. Presented at the 38th Pittsburgh Conference and
Exposition, Atlantic City, NJ, March, 1987, #252.
42.
Copeland, T. Presented at the 2nd Annual U.S. EPA Symposium on Solid Waste
Testing and Quality Assurance, Washington, DC, July, 1986.
43. Binstock, D. A.; Grohse, P. M.; Swift, P. L.; Gaskill, A.; Copeland, T. R.;
Friedman, P. H. Presented at the 3rd Annual Symposium on Solid Waste Testing
and Quality Assurance, Washington, DC, July, 1987, #5-1.
44.
Bettinelli, M.; Baroni, U.; Pastorelli, N. J. Anal Atomic Spectrometry 1987, 2,
485-489.
45. Que, H.; Shane, S.; Boyle, J. R. Anal. Chem. 1988, 60, 1033-1042.
46. Van Delft, W.; Vos, G. Anal. Chim. Acta 1988, 209, 147-156.
47. Vermeir, G.; Vandecasteele, C.; Dams, R. Anal. Chim. Acta 1989, 220, 257261.
48.
Schelkoph, G. M.; Milne, D. B. Anal. Chem. 1988, 60, 2060-2062.
49.
Beauchemin, D.; McLaren, J. W.; Berman, S. S. J. Anal. At. Spectr. 1988.3,
775-780.
50.
Van Wyck, D. B.; Schifman, R. B.; Stivelman, J. C.; Ruiz, J . ; Martin, D. Clin.
Chem. 1988. 34. 1128.
51.
230
White, Jr. R. T. Introduction to Microwave Sample Preparation: Kingston, H.
M.; Jassie, L. B., Eds.; American Chemical Society: Washington, DC, 1988; Ch
4.
52.
Black, S. S.; Babo, J. M.; Stear, P. A. Introduction to Microwave Sample
Preparation: Kingston, H. M.; Jassie, L. B., Eds.; American Chemical Society:
Washington, DC, 1988; Ch 5.
53.
Kingston, H. M; Jassie, L. B. Presented at the 3rd Annual U. S. EPA
Symposium on Solid Waste Testing and Quality Assurance, Washington, DC,
July, 1987.
54.
Pratt, K. W.; Kingston, H. M.; MacCrehan, W. A.; Koch, W. E. Anal Chem.
1988, 6Q, 2024-2027.
55.
Burguera, M.; Burguera, J. L. Anal.Chim. Acta 1986, 179. 351-357.
56.
Kammin, W. R.; Brandt, M. J. Spectroscopy 1989, 4 (3), 49-55.
57.
Kruse, D. Report to Sanitary District of Rockford, IL, May 15,1986.
58.
Fernando, A. R. PhD. Dissertation, University of Alberta, Alberta, Canada,
1988.
59.
Rantala, R. T. T.; Loring, D. H. Anal. Chim. Acta 1989, 220. 263-267.
60.
Zhong-quan, Y.; Bao-hou, L.; Zhong-hou, T.; Kai, H. Presented at the 40th
Pittsburgh Conference, Atlanta, GA, March, 1989, #1372.
61.
Bao-hou, L.; Zhong-quan, Y.; Kai, H. Presented at the 40th Pittsburgh
Conference, Atlanta, GA, March, 1989, #1375.
62.
Patterson, K. Y.; Veillon, C .; Kingston, H. M. Introduction to Microwave
Sample Preparation: Kingston, H. M.; Jassie, L. B., Eds.; American Chemical
Society: Washington, DC, 1988; Ch 7.
63.
Wandt, M. A. E. PhD. Dissertation, University of Cape Town, Cape Town,
South Africa, 1986.
64.
Introduction to Microwave Sample Preparation. Kingston, H. M.; Jassie, L. B.,
Eds.; American Chemical Society: Washington, DC, 1988.
65. Skelly, E. M.; Distefano, F. T. Appl. Spectroscopy 1988, 42(7), 1302-1306.
66.
67.
Labrecque, J. M. Introduction to Microwave Sample Preparation: Kingston, H.
M.; Jassie, L. B., Eds.; American Chemical Society: Washington, DC, 1988; Ch
10.
Matthes, S. A. Introduction to Microwave Sample Preparation: Kingston, H. M.;
Jassie, L. B., Eds.; American Chemical Society: Washington, DC, 1988; Ch 3.
231
68.
Sturcken, E.; Floyd, T.; Manchester, D. Introduction to Microwave Sample
Preparation: Kingston, H.M.; Jassie, L.B., Eds.; American Chemical Society:
Washington, DC, 1988, Ch 9.
69.
McGill, S.; Walkiewicz, J. Presented at the 23rd Microwave Power Symposium,
Ottawa, Canada, August, 1988.
70.
Walkiewicz, J. W.; McGill, S. L.; Moyer, L. A. Presented at the Microwave
Processing of Materials Symp, Materials Research Society Meeting, Reno, NV,
April, 1988, #4.7.
71.
Walkiewicz, J. W.; Kazonich, G.; McGill, S. L. Minerals and Metallurgical
Processing 1988,1(1), 39-42.
72. Fanslow, G. E.; Hou, C.; Richardson, C. K.; Bluhm, D. D .; Markuszewski, R.
Presented at the 22nd International Microwave Power Symposium, Cincinnati,
OH, August, 1987.
73. Baghurst, D. R.; Chippindale, A. M.; Mingos, D. M. P. Nature 1988,332. 311.
74. Day, J. D. A.; Leidigh, W. J. Trans. IM PI1973, JL 185-198.
75. Carr, K. L. Presented at the 19th Annual Meeting of the International Microwave
Power Institute, Minneapolis, MN, September, 1984.
76. Hertzig, R.; Keefer, R.; Lorenson, C. Presented at the 23rd Microwave Power
Symposium, Ottawa, Canada, August, 1988.
77. Von Hippel, A. R. Ed. Dielectic Materials and Applications: Technology Press of
M.I.T. and Wiley: New York, 1954; p.301.
78. Plastics 1980: Gasness, R.; Kusy, P. F.; Keimel, F. A.; Miller, H. L.; Pebly, H.
E. Eds.; International Plastics Selector: San Diego, CA, 1979.
79. Handbook of Plastics and Elastomers: Harper, C. A. Ed; McGraw-Hill: New
York, 1975.
80. User's Practical Selection Handbook for Optimum Plastics. Rubbers and
Adhesives: I'll: Tokyo, Japan, 1976.
81.
Polymethylpentene "TPX", Mitsui Petrochemical Industries, Ltd., Kasumig
Aseki 3-Chome; Chiyoda-Ku: Toktyo 100, Japan.
82.
Modem Plastics Encyclopedia 1988: Juran, R., Ed.; McGraw-Hill: New York,
1987.
83.
Collins, M. J.; Hargett, W. P. U. S. Patent 4 565 669, 1986.
232
84.
Matthes, S. Presented at the 25th Eastern Analytical Symposium, New York,
NY, October, 1986, # 72.
85.
Microwave Acid Digestion Bombs, Parr Instrument Company, Bulletin 4780
2/87.
86.
Bacci, M.; Bini, M.; Checcucci, A.; Ignesti, A.; Millanta, L.; Rubino, N.; Vanni,
R. Proc. 14th Microwave Power Symp., Monaco, 1979, p. 42-44.
87.
Salsman, J. B. Presented at the 23id International Microwave Power
Symposium, Ottawa, Canada, August, 1988.
88.
Bacci, M.; Bini, M.; Checcucci, A.; Ignesti, A.; Millanta, L.; Rubino, N.; Vanni,
R. J. Chem. Soc. Faraday Trans. 1 1981. 77. 1503-1509.
89. Checcucci, A.; Olmi, R.; Vanni, R. J. Microwave Power 1985, 20(3). 161-163.
90. Powell, R. L.; Hall, W. J.; Hyink, C. H. Jr.; Sparks, L. L.; Bums, G. W.;
Scroger, M. G.; Plumb, H. H. Thermocouple Reference Tables Based on the
IPTS-68. NBS Monograph 125, March, 1974.
91. Wickersheim, K. A.; Sun, M. H. J. Microwave Power 1987, 22(2), 85-93.
92. Papoutis, D. Photonics Spectra 1984, March. 5360.
93. Bini, M.; Ignesti, A.; Olmi, Rubino, N.; Vanni, R.; Millanta, L. Abstracts of
Papers, XXIst General Assembly URSI, #8.8.
94.
Baker, R. J.; Smith, V.; Phillips, T. L.; Kane, L. J.; Kobe, L. H. IEEE
Transaction. Microwave Theory and Techniques 1978. MTT-26(#8). 541-545.
95. Sandberg, C.; Gerling, J. Am. Soc. of Mech. Eng. 1984, 84-HT-50.1-6.
96.
Wichesheim, K; Alves, R. B. Ind. Research/Development 1979, December
97.
Kingston, H. M.; Jassie, L. B. Presented at the 25th Eastern Analytical
Symposium, New York, NY, October, 1986, #7.
98.
Zakaria-Mehan, Z.; Neas, E. Presented at the 25th Eastern Analytical
Symposium, New York, NY, October, 1986, #75.
99.
Saaski, E. W.; Hard, J. C.; Mitchell, G. L. Advances in Instrumentation 1986
41X3), 1177-1184.
100.
Kingston, H. M ., Jassie, L. B. Introduction to Microwave Sample Preparation:
Kingston, H. M.; Jassie, L. B., Eds.; American Chemical Society: Washington,
DC, 1988; Ch 11.
Sutton, W. H. Ceramic Bulletin 1989, 6§X2), 376-386.
101.
233
102.
Neas, E. D., Collins, M. J. Introduction to Microwave Sample Preparation:
Kingston, H. M.; Jassie, L. B., Eds.; American Chemical Society: Washington,
DC, 1988; Ch2.
103. Mudgett, R. E. Food Technology 1986, June, 84-93.
104. Schiffman, R. F. Food Technology 1986, June, 94-98.
105.
Kashyap, S. C.; Wyslouzil, W. International Microwave Power Symposium
Proceedings, Philadelphia, PA, July, 1983,19-23.
106. Van Koughnett, A. L. Trans. IM PI1973, 1, 17-39.
107. White, J. R. Trans. IMPI 1973, 1, 40-64.
108.
Smith, R. D. Appendix A "Fundamentals of Microwave Heating" in Elec. Power
Res. Inst. Report # EM-3645, August 1984.
109.
Copson, D. A. Microwave Heating: AVI: Westport, CT, 1975; Ch 1.
110. Walker, J. Scientific American 1987, 256. 134-138.
111. Hasted, J. B. Aqueous Dielectrics: Chapman and Hall: London, 1973; Ch 3.
112. Katt, J. L. Presented at the meeting of the Intemation Microwave Power Institute
Symposium, Ottawa, Canada, August, 1988.
113.
Weiss, J. Handbook of Ion Chromatography: Johnson, E. L., Ed.; Dionex
Corporation: Sunnyvale, CA, 1986; Ch 6.
114.
Minard, D. Physiological and Behavioral Temperature Regulation: Hardy, J. D.;
Gage, A. P.; Stolwijk, J. A., Eds.; Charles C. Thomas: Springfield, IL, 1970;
Ch 25.
115.
Guy, A. W.; Lehmann, J. F.; Stonebridge, J. B. Proc. IEEE 1974, 62(1), 5575.
116.
Lehmann, J. F.; Guy, A. W.; Stonebridge, J. B; Delateur, B. J. IEEE Trans.
Microwave Theory Tech. 1978, MTT-26(8). 556-563.
117.
Johnson, C. C.; Guy, A. W. Proc. IEEE 1972, 60(6), 692-718.
118.
Parker, V. B. Nat'l Stand. Ref. Data Ser. (U. S. Nat'l Bur. Stand.) 1965,
NSRDS-NBS2.
119.
Parker, V. B. CRC Handbook of Chemistry and Physics: 66th ed.; Weast, R. C.,
Ed.; CRC: Cleveland, OH, 1985; p D-122.
234
120.
Lange's Handbook of Chemistry: 12th ed.; Dean, J. A. Ed.; McGraw-Hill: New
York, 1979.
121. Kunzler, J. E.; Giauque, W. F. J.A.C.S. 1952,74, 3472-3476.
122. Gerling, E. E. The Microwave Energy Applications Newsletter 1978, volXI(3).
20-27.
123.
Voss, W. A. G.; Greenwood-Madsen, T. J. Microwave Power E^ 1987,22(4).
209-211.
124. Gerling, J. E. Gerling Laboratories Report No. 80-013, July, 1980.
125. Lentz, R. R. J. Microwave Power 1980, 15(2)T 107-111.
126.
Walter, P. J. Microwave Calibration Assistant version 1.00, NIST, 2/28/89.
127.
Kingston, H. M. EPA Report No. DWI-393254-01-0, Quarterly Report,
September 30, 1988.
128.
Kingston, H. M.; Jassie, L. B.; Neas, E. a Short Course "Microwave Sample
Preparation", Presented at the 40th Pittsburgh Conference and Exposition,
Atlanta, GA, March, 1989.
129.
Johnson, W. M.; Maxwell, J. A. Rock and Mineral Analysis: John Wiley and
Sons: New York, 1981; Chap. 4.
130.
Dolezal, J.; Povondra, P.; Sulcek, Z. Decomposition Techniques in Inorganic
Analysis: American Elsevier Publishing Company Inc.: New York, 1968; Ch 1.
131.
Gordon, C. L. J. Res. Nat'l Bur. Stds. 1943, 30,. 107-111.
132.
Gedye, R.; Smith, F.; Westaway, K. Presented at the 26th Eastern Analytical
Symposium, New York, NY, September, 1987, #058.
133.
Jassie, L. B.; Kingston, H. M. Presented at the Society of Analytical
Chemists of Pittsburgh Meeting, Pittsburgh, PA, November, 1988.
134.
Smith-Magowan, D.; Goldberg, R. N. A Bibliography of Experimental Data
Leading to Thermal Properties of Binary Aqueous Electrolyte Solutions: NBS
Spec. Publ. 537; U.S. Government Printing Office: Washington, DC, 1979.
235
135.
Staples, B. R.; Garvin, D.; Smith-Magowan, D.; Jobe, Jr.,T. L.; Crenca, J.;
Jackson, C. R.; Wobbeking, T. F.; Joseph, R.; Brier, A.; Schumm, R. H.;
Goldberg, R. N. Bibliographies of Industrial Interest: Thermodynamic
Measurements on the Systems CQo-EbO. CuCb-HoO. HoSO/i-HoQ. NH^H?Q. H?S-H9Q. ZnCb-HoO. and HctPO^-HoO: NBS Spec. Publ. 718; U.S.
Government Printing Office: Washington, DC, 1986.
136.
Cobble, J. W.; Murray, Jr., R. C.; Turner, P. J.; Chen, K. High-Temperature
Thermodynamic Data for Species in Aqueous Solution: EPRINP-2400; San
Diego State University Foundation: San Diego, CA, 1982.
137.
Steel, B. J.; Stokes, J. M.; Stokes, R. H. J. Phvs. Chem. 1958, £2, 1514.
138.
Watkins, K. W. Chem. Ed. 1983, £Q(12), 1043-1044.
139.
Moore, W. J. Pvsical Chemistry. Prentice-Hall, Inc.: Englewood Cliffs, NJ,
1972; p. 362.
140.
Handbook of Chemistry and Physics: 66th ed.; Weast, R. C. Ed. CRC:
Cleveland, OH, 1985; p. E-55.
141.
Wu, Y. C. J. of Materials 1972, 2(4), 573-579.
142.
Rainville, E. D.; Bedient, P. E. Elementary Differential Equations: 5th ed.
Macmillan Publishing Co.: New York, 1958; p.47.
143. Huheey, J. E. Inorganic Chemistry: 3rd. ed., Harper and Row: New York,
1983; Appendix E, p A -28.
144.
Kingston, H. M.; Jassie, L. B. J. Res. Natl. Bur. Stds. 1988, 93(3), 269-274.
145. Moody, J. R.; Beary, E. S. Talanta 1982, 29, 1003-1010.
146. Gorsuch, T. T. Analyst 1962, §7, 112-115.
147.
Kingston, H. M.; Cronin, D. J.; Epstein, M. S. Nuclear and Chemical Waste
Management 1984. 5. 3-15.
148.
Fassett, J. D.; Kingston, H. M. Anal. Chem. 1985, 52(13), 2474-2478.
149.
Greenberg, R. R.; Zeisler, R.; Kingston, H. M.; Sullivan, T. M. Fresenius Z.
Anal. Chem. 1988,222, 652-656.
150.
Epstein, M.; Prustowski, E.; Kingston, H. M. 1989, in press.
151.
Hillebrand, W. F.; Lundell, G. E. F.; Bright, M. S.; Hoffman, J. I. Applied
Inorganic Analysis: 2nd ed., J. Wiley & Sons: New York, 1953.
236
152.
Bock, R. A. Handbook of Decomposition Methods in Analytical Chemistry:
translated and revised by Marr, I. L.; Wiley & Sons: New York, 1979; Ch 4.
153. Walker, R. J. Anal. Chem. 1988,
1), 1231-1234.
154. Awashthi, Y. C.; Beutler, E.; Srivastava, S. K. J. Biol. Chem. 1975,240.
5144-5149.
155.
van de Walt, T. N.; Strelow, F. W. E. Anal. Chem. 1985,5 L 2889-2891.
156.
Margolis, S.; Jassie, L. B.; Kingston, H. M. in press.
157. Moody, J. R.; Wissink, C. E.; Beary, E.S. Anal. Chem. 1989,61(8), 823-830.
158. Kuehner, E. C.; Alvarez, R.; Paulsen, P. J.; Murphy, T. J. Anal Chem. 1972,
44, 2050-2056.
159.
Paulsen, P. J.; Beary, E. S.; Bushee, D. S.; Moody, J. R. Anal Chem. 1988,
60, 971-975.
160.
Gorsuch, T. T. The Destruction of Organic Matter. Pergamon Press: New York,
1970; Ch 4 &5.
161.
Schilt, A. A. Perchloric Acid and Perchlorates. The G. Frederick Smith Chemical
Company: Columbus, OH, 1979.
162.
Stoeppler, M.; Muller, K. P.; Backhaus, F. Fresenius Z. Anal. Chem. 1979.
297. 107-112.
163.
Wurfels, M.; Jackwerth, E.; Stoeppler, M. Fresenius Z. Anal. Chem. 1988.
330. 160-161.
Документ
Категория
Без категории
Просмотров
0
Размер файла
8 653 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа