&EPA
United States
Environmental Protection
Agency
Municipal Environmental Research EPA-600/2-80-098
Laboratory' '•• August 1980
Cincinnati OH 45268
Research and Development
Design
Considerations for a
Thermal
Decomposition .
Analytical System
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution-sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-80-098
August 1980
DESIGN CONSIDERATIONS FOR A THERMAL DECOMPOSITION
ANALYTICAL SYSTEM
by
Wayne A. Rubey
University of Dayton Research Institute
Dayton, Ohio 45469
Grant No. R805117-01-0
Project Officer
Richard A. Carnes
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental
Research Laboratory, U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the U.S.
Environmental Protection Agency, nor does mention of trade names
or commercial products constitute endorsement or recommendation
for use.
-------
FOREWORD
The U.S. Environmental Protection Agency was created because
of increasing public and government concern about the dangers of
pollution to the health and welfare of the American people. Nox-
ious air, foul water, and spoiled land are tragic testimonies to
the deterioration of our natural environment. The complexity of
that environment and the interplay of its components require a
concentrated and integrated attack on the problem.
Research and development is that necessary first step in
problem solution; it involves defining the problem, measuring its
impact, and searching for solutions. The Municipal Environmental
Research Laboratory develops new and improved technology and sys-
tems to prevent, treat, and manage wastewater and solid and
hazardous waste pollutant discharges from municipal and community
sources, to preserve and treat public drinking water supplies, and
to minimize the adverse economic, social, health, and aesthetic
effects of pollution. .This publication is one of the products
of that research and provides a most vital communications link
between the researcher and the user community.
The safe, permanent disposal of highly toxic organic wastes
is vitally important. The study reported here presents a
sophisticated laboratory system that has been designed and
assembled to provide fundamental thermal decomposition data on a
wide assortment of organic waste materials.
Francis T. Mayo, Director
Municipal Environmental Research
Laboratory
111
-------
ABSTRACT
The safe, permanent disposal of highly toxic organic wastes
is vitally important, and controlled high-temperature incin-
eration is one of the most promising methods for achieving it.
For safety reasons, the thermal decomposition properties of a
toxic organic substance must be well established before large
quantities are subjected to high-temperature incineration. How-
ever, there is a serious lack of basic experimental data concern-
ing the thermal decomposition behavior of these organic
materials.
In response to this need, a sophisticated laboratory system
has been designed and assembled to provide fundamental thermal
decomposition data on a wide assortment of organic materials.
This thermal decomposition analytical system (TDAS) is a
specially designed, closed, continuous system consisting of a
versatile and fully instrumented thermal decomposition unit which
is connected to dedicated in-line gas chromatography, mass
spectrometry, and data reduction equipment. .With the TDAS, a
wide variety of organic samples can be subjected to analysis,
including gases, liquids, solids, and even polymeric materials.
In the TDAS, precisely controlled thermal decompositions
are conducted in a narrow-bore quartz tube reactor. The various
thermal decomposition products are subsequently collected using
a cryogenic adsorptive trap. After thermal desorption, the
products are subjected to high-resolution gas chromatographic
analysis, and identification of the various decomposition pro-, ::
ducts is accomplished using in-line coupled mass spectrometry,
A dedicated minicomputer is used for reducing the analytical
data. From the product analyses obtained through a sequence of
thermal exposures, the thermal decomposition properties for a
given organic sample can be readily obtained.
The TDAS has been designed to generate fundamental thermal
decomposition data rapidly, economically, and safely. These
laboratory data should be instrumental in establishing acceptable
criteria for controlled high-temperature incineration of toxic
organic wastes , ,
This report was submitted in partial fulfillment of Grant
No. R805117-01-0 by the University of Dayton Research Institute
under the sponsorship of the U.S. Environmental Protection
Agency. The report covers the period September 1, 1977 to
April 15, 1979, and work was completed as of May 31, 1979.
-------
CONTENTS
Foreword
Abstract v
Figures ] * vj_
Tables '. '. '. '. '. ix
Abbreviations and Symbols '.'.', x
Acknowledgments
1. Introduction ^
2. Conclusions 4
3. Recommendations 5
4. System Precursors 6
5. Concept and Basic Design of TDAS 14
6 . Thermal Decomposition Unit . . , . 20
Sample handling and insertion 24
Transport of sample and effluent products ... 29
Quartz tube reactor 30
Collection of effluent products 48
TDU instrumentation . . . 50
7. Chromatographic Separation and Analysis .......... 59
8. Mass Spectrometer with Dedicated Minicomputer
and Associated Data Reduction Equipment 68
9. Utilization of the TDAS in Thermal
. Decomposition Studies 73
References 75
Bibliography 89
Appendices : .
A. Gas flow analysis: a detailed analysis of
gas flow and dispersion in a" narrow-bore
circular cross section tube of folded
configuration g^
Straight tube of circular cross section ..... 93
Folded tube of circular cross section 102
B. Error analysis of mean residence time
determination 109
C. Quartz tube reactor heat transfer analysis ....... 112
D. TDAS transducer calibrations .,..„,*...«« 123
E. Preparation of gas chromatographic open
tubular columns -j.31
F. Application of thermal focusing and programmed
temperature gas chromatography in the TDAS ....... 139
G. Approximate costs of TDU components 142
v
-------
FIGURES
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Transport between adjacent analytical components
Schematics of the first generation discontinuous
system
First generation thermal equipment
Variety of residence time distributions
Block diagram of TDAS .
Simplified schematic of TDAS ......
Artist's rendering of assembled TDAS
Diagrams of the TDU
TDAS residence time distribution
Sample insertion assembly
Glass rod probe
Packed tube probe
Sample handling devices
Transport tubing tee connector
Heated transport lines
First generation reactor configuration
Proposed second generation reactor
Secondary flow and circular cross sectional
flow path
Sectional view of furnace interior
Quartz tube reactor components
Page
8
• 9
10
12
15
16
18
21
23
25
27
27
28
31
32
34
34
. . . . 36
38
39
VI
-------
Number Page
21 Assembled quartzware 40
22 Quartzware assembly in three-zone furnace 42
23 Detail of quartz tube reactor 43
24 Additional details of reactor assembly 45
25 Square wave profile of.gas phase temperature
versus time 46
26 Scanning electron microscope photographs of
quartz surfaces ........ 47
27 Schematic of collection trap assembly 51
28 Adsorption trap insert . 52
29 TDU console and instrumentation modules 53
30 Interior of gas flow control module 55
31 Fundamental schematic of mass flow transducer 55
32 Pressure transducer schematic '. 58
33 Schematic of GC column installation ,«,.„,..„... 60
34 Glass OTC installation in modified Pye 104 61
35 Glass OTC separation of normal alkanes 63
36 Chromatographic separation of a complex mixture .... 64
37 Separation of complex chlorinated hydrocarbon
mixture 65
38 LKB 2091 GCMS and TDU 69
39 Minicomputer for data reduction , 71
40 Mass spectra searching terminal 72
A-l Effect of residence time bandwidth 92
A-2 Quartz tube reactor data . . 105
A-3. Relative bandwidth as a function of mean
residence time ..„,.... 108
Vll
-------
Number
C-l
C-2
C-3
C-4
D-l
D-2
D-3
D-4
E-l
E-2
E-3
Page
Details of quartz tube reactor assembly 113
Simplified version of fine bore connecting
tube
114
Graph of bulk temperature versus distance 120
Sketch of heat transfer condition 121
Mass flow transducer calibration 125
Calibration of transducer readout
instrument 126
Calibration of T2 readout instrument
129
Calibration of temperature selector
readout instrument . . 130
Capillary cleaning apparatus 133
Whisker-textured tubing bore 135
Apparatus for drying glass tube interior 136
vi a, i
-------
TABLES
Number Page
1 Sample Handling Modes 26
2 TDU Temperatures 33
A-l Calculated Fluid Dynamic Values ......,., ..'. , . 106
D-l Differential Pressure Transducer Calibration 124
D-2 Mass Plow Transducer Calibration 125
D-3 Calibration of EP-3465A Digital Multimeter ....... 126
D-4 Reactor Temperature Measurements 127
D-5 Omega K-l Thermocouple Chromel-Alumel 127
D-6 Calibration of Omega Digital Temperature 129
D-7 Calibration of Omega Digital Temperature
Indicator Model 175, Type K 130
IX
-------
ABBREVIATIONS AND SYMBOLS
ABBREVIATIONS
AMP
CHAZ
DDT
GC
GLC
GCMS
GCMS-COMP
ID
MS
OD
OTC
PBB
PCB
PGC
PTGC
RTD
TDAS
TDU
UV
amu
eV
kV
ms
mv
tc
w
operational amplifier
N-cyclohexyl-3-azetidinol
dichloro-diphenyl-trichloroethane
gas chromatography
gas-liquid chromatography
gas chromatography mass spectrometry
gas chromatography mass spectrometry with
dedicated computer
inside diameter
mass spectrometer
outside diameter
open tubular column
polybrominated biphenyl
polychlorinated biphenyl
pyrolysis gas chromatography
programmed temperature gas chromatography
resistance-temperature detector
thermal decomposition analytical system ..
thermal decomposition unit
ultraviolet
atomic mass unit
electron volt
kilovolt
milliseconds
millivolts
thermocouple
watts
SYMBOLS
A — cross-sectional area
m
C — concentration
C — average radial concentration
D — effective diffusion coefficient
e
D — gas diffusion coefficient
F — flow
x
-------
m
F
K
L
L
N.
N.
N.
N
Dn
Nu
Re
Sc
R
R,
R
n
-------
k.
i
m
P
Pa
P2
q
r
ro
rs
rw
v
v
ve
vm
x
y
z
21
n
e
x
V
IT
P
a
cr.
— initial partition coefficient
— mass rate of flow
— pressure
— differential pressure '.
— outlet pressure
— pressure at condition 1
— pressure at condition 2
— rate of heat transfer
— rate of heat transfer by convection
— rate of heat flow by conduction
— radius
— inner tube radius
— radius of inner surface
— radius of outer wall surface
— time
— residence time
— average residence time
— time coordinate
— velocity
— average velocity
— exit velocity
— mobile phase velocity
— coordinate
— coordinate
— distance and coordinate
— revised coordinate
— fluid viscosity
— cylindrical coordinate
— coil to tube diameter ratio
— kinematic viscosity
— geometric constant
— density
— statistical standard deviation
— time-based standard deviation
-------
0)
distance-based standard deviation
identity in Appendix A
probability
flow profile parameter
xi li
-------
ACKNOWLEDGMENTS
The author wishes to acknowledge the support and encourage-
ment of the EPA Project Officers, R. A. Carnes of the
Municipal Environmental Research Laboratory and L. Weitzman of
the Industrial Environmental Research Laboratory. Thanks are
also due to D. S. Duvall, the Principal Investigator for this
grant and Head of the Environmental Sciences Division at the
Research Institute. During the course of this research effort,
many of my colleagues graciously offered suggestions and advice
pertaining to the development of this system. Particular
appreciation is extended to R. A. Grant for his exceptional
scientific glass blowing and his many ideas concerning the
fabrication of the system quartzware. We also wish to express
our gratitude to J. A. Mescher for her consultation with
respect to the mass spectrometry requirements, and to R. N._Ely,
who assisted with the transducer installations and calibrations.
xiv
-------
SECTION 1
INTRODUCTION
Thousands of organic compounds are used daily in our
modern industrial society. Unfortunately, it seems that many of-
the_more useful industrial chemicals also have associated
toxicities of one form or another. Even though there are • :
continual searches to find chemical substitutes that are less
toxic,for the foreseeable future, there will be large quantities
of toxic substances that must eventually be disposed of once they
have fulfilled their intended industrial use. . , , :
Many procedures have been proposed for the disposal of toxic
organic compounds (1,2). Some of these procedures are temporary•
that is, they consist essentially of containment methods. Other'
methods are more permanent in nature, as they produce chemical
changes in the substance with the intention of rendering the
material safe. Of the various methods involved with changing the
structure of the toxic organic molecule, chemical and thermal
techniques have received the most attention (3). in addition to
the straightforward chemical and thermal reaction procedures,
special techniques such as microwave plasma detoxification (4,5)
and molten salt decomposition methods (6) have been employed.
Controlled high-temperature incineration is one of the most
promising methods for disposal of hazardous organic wastes. Both
land-based incineration and incineration on ocean-going vessels
have been utilized for the permanent disposal of organochlorine
wastes (7) and large quantities of other types of organic
industrial wastes (8). Incineration-has also been employed for
the disposal of various pesticide formulations (9). Although
waste disposal by incineration at sea is expensive, it is
receiving increased consideration (10).
Before methods can be developed for the immense variety of
organic materials, precise information is needed concerning their
thermal decomposition properties. Much remains to be learned
about the thermal decomposition of these organic materials and
the nature of the decomposition products formed. It is antici-
pated that this basic information can be obtained in the
laboratory using very small quantities of sample—micrograms or
J-6SS •
-------
There are many advantages to generating fundamental thermal
decomposition data in the laboratory. First, it is anticipated
that these thermal decomposition experiments can be conducted
safely in a properly equipped laboratory. There would be a
much higher element of risk if the initial studies were conducted
in larger throughput units. Second, data generated in the
laboratory can be much more precise and comprehensive than
thermal decomposition data generated in field-scale studies.
Third, laboratory thermal.decomposition data- can be obtained
economically and in a short period of time.
Once the thermal decomposition properties of a particular
material have been characterized in the laboratory, the pre-
liminary decision can be made as to whether high-temperature
incineration is a viable disposal route for that particular
material. If no adverse characteristics are detected during the
laboratory experiments then,with due caution, the material can
be subjected to larger-scale thermal decomposition studies.
However, if the laboratory data indicate difficulties or problem
areas, then incineration is probably not a viable disposal method
for that particular substance.
As an example, chlorinated hydrocarbons are indeed valuable
industrial chemicals; however, they are also known for their
chronic and sometimes acute toxicity. Many of the compounds on
the recent priority pollutant list are members of the chlorinated
hydrocarbon family. High-temperature incineration is being con-
sidered for the disposal of many of these chlorinated hydro-
carbons. Prom previous thermal decomposition experiments with
chlorinated hydrocarbons and chlorocarbons, it was determined
that detailed thermal decomposition data were necessary before
conducting large-scale thermal disposal operations. Specifi-
cally, it was concluded that although the parent molecule may be
dealt with at certain thermal exposures, other secondary or
intermediate reaction products may be produced that are more
toxic than the parent substance. Consequently, there is an
urgent need to determine the full thermal destruction data for
toxic organic materials before .progressing to the larger-scale
disposal units.
Laboratory-scale thermal decomposition studies of various
organic materials are being performed at the University of Dayton
Research Institute. These experiments are being conducted using
recently developed second generation thermal decomposition
instrumentation. The present instrumentation is referred to as a
thermal decomposition analytical system (TDAS). This system
incorporates a versatile in-line thermal decomposition unit with
sophisticated analytical instrumentation capable of analyzing the
various decomposition products.
The objective of this report is to describe, in detail, the
various aspects associated with the design of the TDAS. This
-------
second generation system has been designed specifically to pro-
vide precise and accurate thermal decomposition data for a wide
variety of organic samples. The TDAS has-, also been designed to
provide this information safely and in a short period of time.
-------
SECTION 2
CONCLUSIONS
A versatile thermal decomposition analytical system (TDAS)
has been designed, assembled, and tested. With .this laboratory
system, the thermal decomposition behavior of toxic organic ..,
substances and a wide assortment of other organic compounds.can
be experimentally determined. The effects of the major thermal
decomposition variables (i.e., exposure temperature,.composition
of atmosphere, mean residence time and residence time distribu-_
tion, and reactor internal pressure) can be individually investi-
gated with the TDAS. .
A number of specific design features have been incorporated
into this comprehensive system, with the following results:
o Because of the closed continuous design concept of this
system, toxic samples can be safely tested.
o A sample can be subjected to a very precise thermal ex-
posure in the TDAS. At selected reactor temperatures
ranging from 200° to 1150°C, the maximum temperature
variation is less than ±2°C.
o This system is capable of subjecting a sample to a
precise mean residence time ranging from 0.25 to 5.0
seconds. In addition, this system provides a narrow
Gaussian residence time distribution.
o With the flexibility designed into the TDAS, thermal de-
composition studies can be conducted using pure organic
substances or organic mixtures, and only small quantities
of sample are needed (micrograms). Samples can be
either gases, liquids, or solids. In addition, these
samples can be subjected to thermal decomposition studies
in any of a wide variety of atmospheres.
o Comprehensive analyses of the thermal decomposition
products are performed through the use of gas chromato-
graphy/mass spectrometry (GCMS) analytical instrumen-
tation .
o The TDAS can provide fundamental thermal decomposition
data economically and on a quick response basis.
4
-------
SECTION 3
RECOMMENDATIONS
The thermal decomposition properties of organic substances
can be determined with the use of a specially designed thermal
decomposition analytical system (TDAS) . It is therefore •••-',-
recommended that this laboratory system be utilized for estab-
lishing the thermal decomposition behavior of individual toxic
organic substances and multicomponent industrial organic wastes.
Such fundamental thermal decomposition data would be instrumental
in determining the thermal requirements necessary for environ-
mentally acceptable disposal of hazardous materials.' '-•• •
-------
SECTION 4
SYSTEM PRECURSORS
The incineration of toxic organic wastes is profoundly
different from the typical incineration of, municipal refuse. The
temperatures required for adequate thermal decomposition of
hazardous organic materials are.much higher than those employed
in other types of incineration. These incineration temperatures
must be sufficient to decompose not only the parent organic
molecule, but also the various intermediate thermal reaction pro-
ducts (11,12.).
The residence time of molecules at these elevated tempera-
tures must be sufficient to produce complete thermal decomposi-
tions. Therefore, to insure that organic compounds experience
sufficient temperature and residence time, high-temperature
afterburners are usually incorporated in the process stream. The
design and selection of materials for these high-temperature
afterburners has received a considerable amount of study, and
work is ongoing in this area (13,14).
Recommended guidelines have been prepared for selecting the
exposure temperature, residence time, and excess oxygen needed
for the incineration of hazardous organic waste materials (15).
Most of the data compiled thus far have been associated with^the
decomposition of the parent molecule. Detailed data pertaining
to the various products formed during the thermal decomposition
are practically nonexistent.
Instrumental chemical analysis techniques have been employed
for many years in the monitoring of effluents from thermal re-
actors and various thermal analysis instruments. Some of these
effluent monitoring techniques (15-22) rely upon periodic GC
separations and quantitations of the emerging products. Other
methods of monitoring effluents employ different forms of
analytical instrumentation (23-26). Gas chromatography (GC) has
been used extensively for monitoring combustion devices (27), and
also for monitoring the effluents from microreactor devices that
have been incorporated into the GC flow path (28).
Thermal reactors with in-line effluent sensing instrumen-
tation have been used in basic reaction kinetic experiments (29,
30). The coupling of thermal analysis instrumentation with^GC
has been used extensively in working with nonvolatile organic
-------
materials such as plastics and polymers. The term pyrolysis gas
chromatography (PGC) has been applied to this coupled instrumen-
tation technique, and it is used routinely for the analysis and
characterization of organic 'polymeric substances (31-35).
A gas chromatograph is an extremely versatile and sensitive
analytical instrument. If due consideration (36) is given to
the transport of sample between adjacent analytical components
(see Figure 1), it is then possible to assemble analytical
systems that by their nature can provide a vast amount of
information. There are many benefits associated with the
analytical system approach. Briefly, precise information can be
obtained rapidly from analytical systems that cannot be gathered
from isolated instrumental analysis techniques.
Prom previous experiences with systems (37,38) that used GC
in conjunction with thermal analysis instrumentation, a system
was developed that involved a high-temperature quartz tube re-
actor and a GC. Schematics of this discontinuous system are
shown in Figure 2, and a photograph of the assembled thermal de-
composition equipment is presented in Figure 3. With this
system, a sample was. first gradually vaporized in a continuously
flowing carrier gas and then passed through a narrow-bore, high
temperature quartz tube. Upon emergence from the controlled
high-temperature exposure, the various condensable effluent pro-
ducts were captured by an adsorptive trap. This trap was then
inserted into the modified chromatograph, where the products '
were thermally desorbed and subsequently analyzed by programmed
temperature GC (39-42).
^A number of environmentally important compounds have been
examined for their thermal decomposition properties with the
above relatively simple discontinuous system: DDT, Mirex,
Kepone, hexachlorobenzene, PCBs, hexachlorocyclopentadiene, and
PBBs. Other workers (43-44) have employed similar instrumen-
tation to determine the thermal decomposition properties of some
volatile compounds of high environmental interest such as
benzene, vinyl chloride, and acrolein.
The determination of thermal decomposition properties in the
laboratory can be accomplished with very small amounts of
sample; in addition, the various parameters' affecting the thermal
decomposition can be closely controlled. How,.well these
laboratory data extrapolate to field scale units remains to be
seen. However, in one particular case (the Kepone Incineration
Test Program at Toledo, Ohio), close agreement was found between
laboratory thermal decomposition data and that obtained from a
special pilot-scale test facility (45).
Although the relatively simple discontinuous system as
shown in Figure 2 provided some much needed information in a
relatively short period of time, it was already apparent that
-------
EXTENDED CHROMATOGRAPHIC SYSTEM
UPSTREAM
ANALYTICAL
INSTRUMENT
DOWNSTREAM
ANALYTICAL
INSTRUMENT
CHROMATOGRAPH
Figure 1. Transport between adjacent analytical components
-------
A. . COMPRESSED AIR,, BREATHING .QUALITY GRADE
B. TWO-STAGE PRESSURE REGULATOR
C, "HYOROPURGE" FILTER " "• •- '• •". 5-
D FLOW CONTROL VALVE
'E PRESSURE TRANSDUCER " ''.',.-,-
F SAMPLE ' HOLDER", PY.REX ' .. ' '
Gi • HEATED INLET CHAMBER ' '.-'••
H QUARTZ TUBE . - ' '
I HEATED OUTLET CHAMBER
J EFFLUENT TRAP, TENAX-GC OR CHARCOAL
K FLOW METER -. . . „,.... . r ... ,.
INSULATION
EXTERNAL POWER
• SUPPLY AND
TEMPERATURE
CONTROL
A
INJECTOR
CAP
'
\\\\\\\\
ir^1
WUi
vm,
OVEN DOOR
TRAPPING
MEDIUM
h^^%$M
^\ ^^
ALUMINUM BLOCK
WITH EMBEDDED
CARTRIDGE HEATER
t
HELIUM
CARRIER
v\\\\\V<
; HFID
^s^
x
^INSULATION
1» VENT
GLC COLUMN
NPYREX)
I.8M OV-I
Figure 2.
Schematics of the first generation discontinuous
system.
-------
TRANSDUCER
CONTROL UNIT
DIGITAL READ-OUT
OF TEMPERATURE
AND PRESSURE
LINDBERG
FURNACE
FLOW METER
TRANSDUCER
POWER SUPPLY
PRESSURE
TRANSDUCER
HYDRO-PURGE
FILTER
COMPRESSED
AIR
. HEATED
OUTLET
SATURATOR
Figure 3. First generation thermal equipment.
10
-------
this thermal concept could be vastly improved to provide a highly
versatile in-line closed system capable of dealing with a wide
range of thermal decomposition problems. An expanded system with
increased versatility is definitely needed to handle a wider
range of toxic organic samples and to accommodate the very com-
plex mixtures of thermal decomposition products. The large
number of products formed at intermediate temperatures and the
inability to sufficiently separate and identify these products
necessitated the use of more sophisticated analytical instrumen-
tation. Also, the advantages presented by an in-line closed
system would be such that gases, liquids, solids, mixtures
thereof, and even nonvolatile organics could be tested. At the
same time, it was realized that a more sophisticated reactor
could be designed and fabricated. This device could be fully
instrumented and thereby provide very accurate thermal decom-
position data along with the corresponding detailed chemical
analysis of the effluent products.
GC is an extremely valuable technique for separating and
quantitating organic compounds that can be volatilized. However,
it has a weakness when it comes to qualitative analysis or
identification of the various constituents of a complicated
organic mixture (46). This shortcoming of GC has been all but
overcome by the combined instrumental analysis technique of GCMS.
This powerful instrumental combination is ideally suited for the
analysis of organic pollutants (47-49). Also, coupled GCMS seems
to be compatible with an upstream in-line thermal reactor.
With respect to the thermal reactor, it is important that
the mean residence time and the residence time distribution be
accurately known. This is one of the distinct advantages of
determining a material's thermal decomposition properties while
in the laboratory. For example, in the much larger pilot- and
field-scale units, a wide variety of residence time distributions
might be encountered, as typified by the residence time profiles
shown in Figure 4. For experiments conducted in the laboratory,
a narrow bore flow path can be used that produces residence
distributions as shown in Figure 4(f). Thus if the residence
distribution is known, one can avoid the situation in which some
molecules rapidly pass through the high-temperature zone, and
other similar molecules experience a sustained exposure.
After considering these concepts and desired improvements,
we decided that a suitable thermal decomposition system could
be designed and assembled that would have a wide range of
versatility. Such a system would be capable of handling and
testing various toxic organic waste materials—both relatively
pure substances and very complicated organic mixtures. Also,
broader ranges of exposure temperature and mean residence time
were anticipated for this system. Likewise, with appropriate
instrumentation, this system should be capable of conducting
tests at different reactor pressures. In addition, this system
11
-------
(a)
MIXING CHAMBER
(b)
MIXING CHAMBER
WITH CHANNELING
id
EXPONENTIAL DECAY
<
CO
o
tdi
BI MODAL OR
MULTIMODAL
lei
GAUSSIAN WITH
LARGE VARIANCE
(0
GAUSSIAN WITH
SMALL VARIANCE:
TIME-
Figure 4. Variety of residence time distributions
12
-------
should be capable of conducting thermal decomposition, tests in a
variety of flowing atmospheres — for example, air, nitre ?,n,
helium, and mixtures thereof.
13
-------
SECTION 5
CONCEPT AND BASIC DESIGN OF TDAS
The rationale behind the design concept of the TDAS has not
changed basically from that of the earlier discontinuous system
(39,40). The sample is still inserted into the system and then
gradually vaporized in a flowing carrier gas. The vaporized
compounds are subsequently subjected to a controlled, high-
temperature exposure. The components that emerge from the high-
temperature environment are then collected and subjected to
instrumental chemical analysis. This same thermal analysis
format has been employed with respect to the TDAS, but each
operation within the system is much more sophisticated, thereby
producing greatly increased experimental versatility. Each of_
these procedural operations will be discussed at length later in
the report.
The major design changes over the earlier system are
centered around the design of the reactor, the closed continuous
system concept, and also the vastly increased analytical cap-
ability that is now provided by an in-line gas chromatograph/
mass spectrometer/dedicated computer (GCMS-COMP). Numerous re-
finements have also been designed into the TDAS, and these are
detailed in Sections 6 and 7.
Many design objectives were associated with the develop-
ment of the TDAS. This system should be capable of conducting
precise thermal decomposition tests. More precisely, it should
be capable of experimentally determining the effects of the five
prominent thermal decomposition variables—exposure temperature,
gaseous atmosphere, pressure, mean residence time, and residence
time distribution. In addition, the TDAS should be able_to ,
accommodate almost any type of organic material. Also, it
should be capable of analyzing all of the thermal decomposition
effluent products. This closed continuous system should be
capable of dealing with toxic materials. Also, the TDAS should
be capable of generating data on a quick response basis.
Figure 5 shows a block diagram of the TDAS, and Figure 6
is a simplified schematic of the assembled TDAS components.
Studies can be conducted with the TDAS using almost any com-
pressed gas as the carrier and thermal decomposition atmosphere.
Indeed, pyrolytic studies can be performed using inert gases,
and oxidative studies can be conducted using air or any other
14
-------
Q
O
o
VI
C/3
-------
IN-LINE
GC/MS/DS
A. HELIUM GAS
B. COMPRESSED AIR
C. INERT CARRIER
D. PRESSURE REGULATOR
E. .OXYGEN SCRUBBER ;
F. DIRECTIONAL VALVE ,
G. FILTER
H. FLOW CONTROL VALVE
.). FLOW TRANSDUCER
J. PRESSURE TRANSDUCER
K. INSERTION CHAMBER"
L TEMPERATURE PROGRAMMER
M. REACTOR IN FURNACE
N. PRODUCT COLLECTION TRAP
Figure 6. Simplified schematic of TDAS.
16
-------
oxygen-containing carrier. Accurate measurements of pressure and
flow can be readily obtained with the TDAS. The internal
pressure in the reactor and mass flow rate (thus mean residence
time) can be continuously monitored by in-stream instrumentation.
In^the earlier work with the discontinuous system, only low-
volatility organic samples could be tested. The TDAS has been
designed for measuring the thermal decomposition properties of a
wide range of organic samples—gases, liquids, solids, and even
polymers. Complicated organic mixtures can also be tested with
the TDAS by using the process of slow vaporization of the sample.
After vaporization, the gas phase molecules are subjected to pre-
cise high-temperature conditions (these range between 200° and
1150°C and are held with ±2°C) in a fused quartz tube reactor.
The emerging products are then rpaidly, swept into a cryogenic,
adsorbent in-line trap where the condensable products are
captured at temperatures down to minus 110°C. The collected
effluent products are subsequently thermally desorbed from the
trap and then subjected to GC analysis using high-resolution
glass, open tubular columns. The separated compounds are then
subjected to detailed analysis by mass spectrometry (MS). The
thermal decomposition test parameters and the chemical analysis
data are_then retained, and a complete analysis of the thermal
decomposition process is carried out.
Although the TDAS is a complicated system with many inter-
acting components, the operation of this system and the conduct-
ing of thermal decomposition tests can be accomplished with
relative ease (that is, after suitable familiarization and
experience). Most of the crucial instrumentation for conducting
experiments with the TDAS have been mounted in an instrumentation
console, and most of the test functions can be continuously
monitored on this console.
An artist's rendering of the assembled TDAS is presented in
Figure 7. This equipment is located in a temperature and
humidity controlled laboratory that was specifically designed for
conducting sensitive chemical analyses.
The basic procedure for testing a particular sample* begins
with selection of the thermal decomposition atmosphere or carrier
gas. Next, the mode of sample insertion must be selected. Then
the actual exposure temperature is chosen, and each of the
furnace zones are set to this temperature. The internal pressure
under which the test is to be conducted can now be established
by adjusting the gas flow restrictors situated downstream of the
reactor. The next parameter to be established is the mean
*A11 samples subjected to analysis by the TDAS would be screened
beforehand using other chromatographic instruments to verify that
the sample would indeed be conducive to a TDAS examination.
17
-------
0)
H
tn
CQ
to
4-1
O
tn
G
•H
in
(U
cu
tn
•H
-P
M
t~-
Q)
18
-------
residence time. Once the other variables have been stabilized
and measured, it is a simple matter of calculating the desired
carrier flow and dialing off that value with the adjustable flow
control.
After establishing the thermal decomposition test parame-
ters, the next step in the procedure is to cool down the
effluent collection trap so that the condensable products can be
captured in the adsorptive cryogenic trap. The remaining steps
are relatively routine. The sample must then be admitted, the
products collected, and after switching to helium carrier gas,
the captured products must be subjected to in-line analysis bv
GCMS-COMP. - ; • ' •• Y
The above test procedures would yield one set of data points
in the thermal decomposition procedure. Any one of the above
variables could now be changed for the next thermal decomposition
test, and so on, until the thermal decomposition properties of
the sample 'are adequately characterized.
Extensive design details covering the various TDAS com-
ponent parts and assemblies are presented in the following
sections.
19
-------
SECTION 6
THERMAL DECOMPOSITION UNIT
The TDAS is a continuous system that can be viewed as three
in-line instrumental stations or units. The first unit in the
system is designated as the thermal decomposition unit (TDU).
Block diagrams of the TDU are shown in Figure 8. Specifically,
this unit consists of the various gas supplies, an instrumen-
tation control console, the sample insertion chamber along with
the high-temperature reactor-, the effluent collection trap, and
the various associated transfer lines. All of these components
will be discussed later in this section.
The primary function of the TDU is to subject gas phase
molecules to well-defined thermal exposures and to collect the
various emerging chemical compounds. The remaining stations of
the TDAS will subsequently separate and analyze these collected
compounds.
The schematic shown in Figure 8(a) can be redrawn as in
Figure 8(b) to depict the physical relationships within the gas
flow path of the TDU. From this sketch, it is seen that the
pressures, temperatures, and volumes can be related according
to the ideal gas law as follows:
V
2p2
V p
0^0
T
(1)
where V is volume, p is pressure, T is temperature, and T J-S
temperature averaged over the length of the reactor. The
subscript 2 denotes the quartz tube interior, and the subscript
o represents the ambient outlet conditions.
If the impedance or resistance to gas flow presented by the
downstream transfer lines %, the packing within the trap Rt, and
the restrictor Rr, is such that
R
R
(2)
where RCT represents the resistance to gas flow of the quartz tube
reactor assembly, then the differential pressure Pd can be
20
-------
SAMPLE
HEATED
TRANSFER
FILTRATION
AND
CONTROL
CRYOGENIC'
TRAP
i t
a) BASIC SCHEMATIC
IMI FT
F
~m
PO
REA(
INSERTION
p
'd
\ / CHAMBER
TRANSDUCERS
\J
V
;TOR
P
2
T2
EFFLUENT
TRAP
VARIABLE
RESTRICTOR
^
pl I
; VACUUM
^>
V F P
V0'r0'r0
OUTLET
b) SCHEMATIC REDRAWN TO SHOW PHYSICAL RELATIONSHIPS WITHIN
THE GAS FLOW PATH
Figure 8. Diagrams of the TDU.
21
-------
expressed as
= Pl " Po
(3)
Now, volume flow F can be written as
where t is time, and equation 1 can be rewritten as
V
2p2
Fotpo
(4)
(5)
or
irr
F tp
o ^
T
(6)
where r and L are the tube radius and length, respectively.
Therefore, it is seen that time, or, in this case, residence
time, can be written simply as
(7)
where tr represents the calculated mean residence time of a
substance in the quartz tube interior.
The equipment used to measure these parameters will be dis-
cussed at length later in this section. Also, an error analysis
associated with these measurement variables is presented as part
of the appendices. ,
One of the distinct advantages in conducting thermal decom-
position experiments using the laboratory scale TDAS is that each
molecule is subjected to essentially the same thermal exposure.
To illustrate, consider the transport of like molecules through
the high-temperature reactor. First, the individual molecules
experience random entry into the gas flow over a relatively long
period of time. However, as the pressure, temperature, and
linear gas velocity within the reactor are constant, the
individual molecules generate a residence time distribution as
shown in Figure 9.
There are many advantages in having a Gaussian or normal
residence time distribution. Linear combinations of Gaussian
random variables lead to new random variables that are also
Gaussian. Such simplicity is not the case for most other
22
-------
-------
distribution functions. This is but one of the desirable
features of working with the narrow Gaussian function as opposed
to some of the skewed and distorted distribution profiles depict-
ed earlier in Figure 4.
Before the sample handling in the TDU is described, it is
interesting to note from Figure 8 that both the mass flow trans-
ducer and the pressure transducer are upstream, yet in-line with
the reactor and product collection trap. Therefore, with the
use of these in-line sensing devices, thermal decomposition
experiments can be conducted at subatmospheric, atmospheric, or
elevated pressure.
SAMPLE HANDLING AND INSERTION
The sample insertion section of the TDAS was designed to be
as versatile as possible. This fabricated sample insertion
assembly is rigidly mounted on a machined aluminum support locat-
ed between the TDU instrumentation console and the high-tempera-
ture reactor furnace. A photograph of this assembly is presented
in Figure 10. The actual sample reception chamber consists of
fused quartz tubing surrounded by an isolated aluminum block that
can be readily heated and temperature programmed.
The exit end of the sample reception chamber empties into
1.0 mm ID quartz tubing that is connected to the inlet of the
quartz tube reactor assembly. An outer jacket of quartz tubing
surrounds this small bore transfer tube, so that by using a
counter flow of heat transfer air, a uniform temperature
distribution is obtained for the transport of the gas sample to
the reactor assembly.
There are numerous modes for inserting samples into the
TDAS, and because of the built-in flexibility, almost any type of
organic sample can be examined. Some of these sample handling
modes are listed in Table 1. A sketch of the glass or quartz
rod probe is shown in Figure 11, and the special packed tube
probe is depicted in Figure 12. A photograph of the different
sample handling devices is shown in Figure 13.
The one major requirement with respect to sample handling
with the TDAS is that substances leaving the sample insertion
chamber must be in the gas phase. Precautions must be taken to
prevent particulate material from migrating into the heated
transfer lines and the high-temperature quartz tube assembly.
TDAS samples are either volatilized or thermally degraded
in the sample insertion chamber. Volatilization is accomplished
by controlled heating of the insertion chamber with a special
temperature programmer. If sample thermal degradation is
required, it can be accomplished using the probe sample holders
24
-------
Figure 10. Sample insertion assembly,
25
-------
TABLE 1. SAMPLE HANDLING MODES
Types of
organic samples
Mode of insertion
into TDAS
Gases
Liquids
(low and high
volatility substances)
Solids
(molecular weights
less than 800)
Non-volatiles or
polymeric organics
• Gas sampling valve
• Gas-tight syringe
• Syringe
• Ampoule
• Quartz rod probe
• Platinum ribbon probe
o Quartz rod probe
• Packed tube probe
• Platinum ribbon probe
• Packed tube probe
26
-------
SAMPLE DEPOSIT REGION
2.0 cm BY,0.4 cm DIAMETER
(SAND BLASTED)
GAS-TIGHT SEAL
(3-SILICONE 0-RINGS)
PYREX ROD
25 cm BY 0.6 cm
\
ADJUSTABLE
SET-SCREW
RETAINER
Figure 11. Glass rod probe
COILED PLATINUM HEATER — CONTROLLED BY
FLASH PYROLYSIS POWER SUPPLY
GRANULATED POLYMER SAMPLE
MIXED WITH QUARTZ WOOL
OUTGOING GAS
PLUS 0 1
VOLATILES
ft c\
b /o \3 V)
<] 1 INCOMING GAS
QUARTZ
WOOL
THIN WALL QUARTZ TUBING
Figure 12. Packed tube probe,
27
-------
Figure 13. Sample handling devices
28
-------
and a versatile pyrolysis unit. These two programmable thermal
controllers will be described later when discussing the TDU
instrumentation.
The sample insertion section of the TDAS has been designed
so that a Plexiglass glove box can be installed that will
completely encompass the sample entry region. If needed, samples
could actually be prepared in the isolated environment of the"
glove box and then inserted into the TDAS.
TRANSPORT OF SAMPLE AND EFFLUENT PRODUCTS
Three basic criteria must be met for proper transport of gas
phase samples. First, the transport tubing material must not
interact with any of the chemical substances that constitute the
flowing media. Second, it is important that a constant flow of
sweeping gas be maintained throughout every portion of the trans-
port line. Third, the gas transfer must be maintained at
elevated temperatures that are compatible with the flowing media.
Each of these criteria require some elaboration relative to the
transport of sample and products in the TDAS.
Wherever possiblef fused quartz was used as the tubing
material for sample transport in the TDAS. However, certain
portions of this continuous system required metal to metal
connections, and the 'decision was made to use glass-lined
stainless steel tubing (from Scientific Glass Engineering) at
these locations. Other possible tubing materials were nickel 200
and stainless steel. Indeed, these latter materials are physi-
cally more rugged, but the glass-lined SS tubing was selected
primarily because of its increased inertness. Other types of
materials are also encountered in the transport path downstream
of the reactor. Stainless steel, Teflon, and polyimide surfaces
are found in such components as Swagelok unions, the switching
valve interior, and tubing ferrules.
The second basic requirement for good transport of sample
involves the continual gas sweeping of the transfer tubing. It
is desirable that the transport gas flow be maintained at
relatively high linear velocities (that is, greater than 20 cm
sec -1-) . Thus the inside diameter (ID) of the transport tube
should be relatively small, e.g., less than 1.0 mm ID. Unswept •
recesses in a gas flow path can present major transport problems.
Consequently, the use of various tube fittings should be
minimized. Several unions were necessary in this system;
however, to minimize difficulties at these junctions, the ends
of the tubing material were faced off with a grinder before
assembly.
A common problem in transporting gases arises when a gas
stream must be split into two or more branches. In the TDAS,
29
-------
this splitting was accomplished with the use of tee connectors
that had been filled with quartz wool (Figure 14). With the
placement of quartz wool at the point of branching, considerable
gas phase mixing is induced. This mixing essentially avoids such
difficulties as molecular weight fractionation and non-lxnear
splitting.
The selection of transport temperature and the maintenance
of uniform temperatures along the transport path, constitutes the
third criteria for proper transfer of sample. The selected
transport temperature must be high enough to avoid any serious
condensation and possible adsorption effects. However, it must
be low enough that additional gas phase reactions do not occur
during transport. Localized cold or hot regions must be avoided.
The transfer tubing used in the TDAS was not directly heated
but was actually jacketed with much larger diameter glass or
quartz tubing. The outer tube was first instrumented with
thermocouples located at key positions along the tube. The
thermocouples were attached to the outer tube with the use of
glass adhesive tape (No. 27 from 3M Co.). After electrically
insulating the various conductors along the transport line with
woven glass tape (Ware Apparatus Co.), the tubular assembly was
wrapped with high-temperature heating tape (Electrothermal
flexible tape). The heating tape was then wrapped with another
layer of woven glass tape. A photograph of an assembled trans-
port line is shown in Figure 15. Table 2 also shows the
various thermocouple locations and their corresponding maximum
temperatures for gas transport in the TDAS.
During the course of conducting,experiments with the TDAS,
certain vital components in the transport path are intentionally
cooled to very low temperatures. This localized cooling is
necessary (50,51) for capturing the products emerging from the
TDAS. Details concerning this low temperature procedure will
be discussed later in this section.
QUARTZ TUBE REACTOR
The design and performance of the high-temperature reactor
is vitally important to the entire TDAS. Consequently, 'con-
siderable attention has been centered upon this component. As
stated earlier, the important thermal decomposition variables
(i.e., gas phase environment, mean residence time and distri-
bution profile, exposure temperature, and pressure) apply direct-
ly to the interior of the reactor.
In the earlier thermal decomposition studies conducted with
the first generation unit, a narrow-bore folded quartz tube was
used (Figure 16). This particular quartz tube reactor was of
heavy wall construction and occupied most of the internal volume
30
-------
SPLITTER
AND
VENT TRAP
TRANSPORT
LINE
GCMS
QUARTZ
WOOL
Figure 14. Transport tubing tee connector.
31
-------
Figure 15. Heated transport lines.
32
-------
TABLE 2. TDU TEMPERATURES
Temperature ' . ,,
selector . Ma™m
_ • , • temperature
positxon Thermocouple location (<>c)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Room temperature (tc in connector)
Insertion block
Interior of pre-reactor tube
Mid point of inlet heat transfer cavity
Reactor end region of inlet cavity
Reactor end region of outlet cavity
Mid point of outlet heat transfer cavity
Junction of glass-lined and quart tubing
Transfer tube between valve and GC
Glass-lined tubing upstream of collection
trap
Effluent collection trap
Glass-lined tubing downstream of
collection trap
Junction of glass-lined tubing and
valve inlet
Switching valve
GC oven interior
30
300
300
400
350
350
400
350
300
300
300
300
250
240
300
33
-------
GAS FLOW
HEAVY 'WALL FOLDED QUARTZ
TUBE WITH CONSTANT BORE SIZE
(FLOW PATH • 0.84 m by 0.8 mm ID)
\
INSULATION
HEATING ELEMENT ,
OF UNDBERG FURNACE
(R.T. to 1000°C)
Figure 16. First generation reactor configuration,
GAS FLOW -»
FINE BORE
ENTRANCE
QUARTZ TUBE
(0.5 mm ID)
\
/////A/////
THIN WALL QUARTZ
(FLOW PATH =1.0
/
///////////,
HELICAL
TUBING
m by 1.0 mm ID)
1
/
/////// X /
1 1 \ rK-rrrr?rtf7??v?v?r>
t
'///////////
//////////Z^.
\
/X///.
\
.1
//A
FINE BORE
INSULATION
HIGHER TEMPERATURE
FURNACE (R.T. to 1150°C)
Figure 17. Proposed second generation reactor
34
-------
of_its high-temperature furnace. Many tests were conducted with
this reactor, and it experienced considerable use at temperatures
up to 1000°C. Even so, a number of design refinements were con-
sidered for the second generation quartz tube reactor. Better
control of exposure temperature throughout the gas flow path was
desired, and the minimization of end effects would be beneficial.
Also, the ability to conduct thermal decomposition tests at some-
what higher temperatures was considered most advantageous.
Therefore, the initial conceptual design for the second genera-
tion quartz tube reactor was as.shown in Figure 17. Again, the
major portion of the proposed new reactor was a narrow-bore flow '
path of considerable length. (It is this design concept that is
of paramount importance in obtaining narrow residence time
distributions of Gaussian profile.)
A variety of studies have been conducted on laminar fluid
flow through narrow-bore flow paths (52-57). Methods for deter-
mining gas diffusion coefficients have been developed using the
precise control of dispersion afforded by the' harr.ow flow path
(58,59). Thermal decomposition testing using this narrow-bore
flow path concept requires a high-temperature tubing material,
and_fused quartz (also referred to as fused silica and vitreous
silica) seems to be the best material presently available for
high temperature.use (60,61) . The proposed quartz tube reactor'.
shown in Figure 17 uses very fine bore entrance and exit tubes
to transport the gas rapidly into and out of the central portion'
of the reactor. Later in this section even further refinement of
this transport concept is discussed.
Unfortunately, fabrication difficulties are encountered in
preparing tightly coiled quartz tubing. Specifically, when
requests were sent to various quartz fabricators, the common
reply was that they could not provide helical quartz tubing with
a precise circular cross section flow path. It was our under-
standing that when quartz tubing is tightly coiled, the circular
cross section bore distorts into an ovalized cross section and
even takes the shape of a "D" with the flat side being on the
inner diameter of the helix. As a result of the apparent
fabrication difficulties, and before contemplating other geo-
metric configurations for the reactor, it was decided to assess
the possibility of bending narrow bore quartz tubing without
changing the circular cross section of the tubing bore.
Maintenance of the circular cross section flow path is
essential because it affects the ability to describe the effects
of secondary flow on the residence time distribution. Secondary
flow occurs whenever a fluid is passed through a tube that
exhibits a significant amount of axial curvature. Figure 18(a)
shows^the typical streamlines associated with secondary flow,
and Figure 18(b) depicts the corresponding radial mixing pattern.
Secondary,flow has been observed and studied for decades (62-67).
Dean (62) studied the effects of secondary flow as early as 1927."
35
-------
(a) STREAMLINES
(b) RADIAL MIXING
(c) CURVED QUARTZ TUBE OF
CIRCULAR CROSS SECTION
Figure 18. Secondary flow and circular cross section flow path
36
-------
Also, pressure losses (68) were associated with secondary flow
in bent tubes as early as 1937. To characterize accurately the
residence time contribution in a tube that exhibits some degree
of secondary flow, it is imperative that the flow, path be of
circular cross section. The description of secondary flow in an
elliptical, ovalized, or noncircular flow path is practically
an impossible task.
Therefore, upon giving this fabrication problem to the
University scientific glass blower, thin-wall , 1. 0. mm' ID quartz
tubing was eventually obtained that retained its circular
cross section when bent through a 180° arc at, a radius of 2.1
cm. Figure 18 (c) shows a photograph of a typical cross section
of curved tubing. Indeed, there is no visible ovalization
of the tubing bore. Circular cross section was maintained by
locally heating the tubing on a mandrel and allowing the quartz
to bend by its own weight.
The ability to maintain the circular cross section flow
path in a curved tube permitted the eventual design of a compact
folded tube reactor. The actual configuration and dimensions
of this quartz tube reactor will be presented later, along
with a detailed description of the gas flow behavior.
the ovalization difficulty overcome, it was then
possible to design a complete quartz tube reactor assembly that
could be fitted into a high -temperature tubular furnace. The
furnace selected for heating this quartz assembly was a Lindberg
Model 54357, a three-zone furnace of hinged construction that
was designed for continuous operation at temperatures up to
1200°C. This furnace has its own control console (Model 58744-A)
which separately regulates the temperatures of each of the three
zones . Only the. middle portion of the furnace ' s central zone is
used to heat the high-temperature reactor, although during
operation the entire length of the tubular furnace is held under
isothermal conditions. Figure 19 shows a sectional view of the
furnace . interior along with the locations of the thermocouples
used for controlling temperatures within the three separate
zones.
The assembled quartzware, which was eventually positioned in
the high-temperature furnace, is detailed in Figure 20. This
quartzware, and all of the other glassware employed in the TDAS ,
was fabricated in the University Glass Shop by a scientific
glassblower. Even though the assembled quartzware is delicate,
it has .experienced considerable use without problems at
extreme temperatures.
A better understanding of the actual behavior of the entire
quartz tube assembly can be obtained by close inspection of the
schematic shown in Figure 21. In this figure, three separate
gas flow paths are depicted. The main flow is the carrier
37
-------
28.0
STEEL HOUSING
[DIMENSIONS ARE IN CENTIMETERS!
Figure 19. Sectional view of furnace interior.
38
-------
rf-
S E!
_o
E
©
©
C£
g
S
o;
©
g
o
ce:
©
©
cr
S
oo
g E
O 5
Q- E
CL. ^,
3= O
O Di
r=c ar
"' Q
>- o
oo
<
D;
P
o
a
i—i
oo
CO
-P
0)
O
i1
O
o
S-l
o
-P
o
fd
Q)
Q)
-P
N-
-P
d
O!
O
CN
(L)
S
tn
•H
PL,
39
-------
n -> o
°- UJ O
V "- ffl
E CM
o o —
IIR
III
55
v /
0)
Csl
-P
-------
stream through the transfer tubing, into the reactor, and out
the exit transfer tubing. This gas flow path carries the
vaporized sample molecules into the reactor, where they are
subjected to the controlled high-temperature conditions and
then passed downstream where the various products are collected
and eventually subjected to analysis. The other two gas flow
paths consist of adjustable flows of heat transfer air that
cool the entrance and exit chambers of the quartz tube assembly
Thus the main gas flow into the reactor is maintained at lower
nondestructive temperatures right up to the point of entry into
the high temperature reactor. As soon as the sample or product
molecules leave the reactor, the gas temperature is abruptly
decreased to the same nondestruction transport level (approxi-
mately 275°C). The gas then flows on to the collection trap at
approximately the same temperature. Figure 22 shows the
quartzware assembly centrally located in the three-zone furnace.
The reactor was constructed from quartz tubing obtained
from Heraeus-Amersil, Inc., and their smooth-wall tubing was
found to possess uniform wall thickness and bore diameter over
the entire tubing length. The tubing diameter measurements, as
shown in Figure 23, were obtained by carefully filling the
particular tubing section with mercury and then removing the
mercury and obtaining its respective weight with the use of an
analytical balance. This mercury displacement technique was
very similar to that described by Barr and Anhorn (69).
A narrow-bore, multiple-fold quartz tube reactor of race
track configuration (Figure 23) was selected for several reasons
First, with the relatively short overall length of the main
portion of the reactor and with utilization of the multiple-fold
tube design, any existing longitudinal temperature gradient
within the furnace would be effectively averaged out over seven
cycles, thereby assuring that sample molecules encounter
essentially isothermal conditions during their controlled
traverse through the reactor. Second, these flow path di-
mensions would permit mean residence time studies of between
0.25 and 5.0 seconds, which was a TDAS design objective. In
addition, as there are no protrusions into the flow path, and
as the tubing diameter transitions are very gradual, eddy
currents would be negligible. With this size reactor, gas
phase thermal decomposition studies would be conducted in the
laminar flow region at Reynolds numbers that are orders of
magnitude below the onset of turbulence. Calculated Reynolds
number extremes for this quartz tube reactor range from 0.7 to
140. Furthermore, there would be a very low pressure drop
across_this reactor even at the very high linear gas velocities.
Also, in view of the narrowness of the gas flow path, back
diffusion in the reactor would be nonexistent.
In the TDAS design, gas phase molecules are subjected to
a narrow residence distribution while maintaining, within the
41
-------
CARRIER
SAMPLE
QUARTZ TUBE ASSEMBLY
IN 3-ZONE FURNACE
(CAPABLE OF I20O C) .
CARRIER
fe +
PRODUCTS
Figure 22. Quartzware assembly in three-zone furnace,
42
-------
RACE-TRACK
CONFIGURATION
(3.5 CYCLES)
CONNECTING TUBING
70 mm by 0.43 mm ID
2.1 cm RADIUS-
•11.8'cm
70 mm by 0.43 mm ID
980 mm by 0.97 mm ID
Figure 23. Detail of quartz tube reactor.
43
-------
reactor, an acceptable ratio of gas volume to surface area. The
design of this quartz tube reactor permits a detailed descrip-
tion of gas flow behavior within the reactor. (Such a detailed
analysis is presented in Appendix A.)
As derived earlier in Section 6, the mean residence time
for gas phase molecules in the above reactor is
1 +
(7)
and the standard deviation of the residence time distribution is
1
^ _ fc> T.
a =
2D (I, + L
(v)3
24 v D
24 v' D
(8)
as obtained from the derivation presented in Appendix A (see
list of symbols for the designations of the various terms).
An error analysis pertaining to the mean residence time
determination for this reactor has also been conducted (see
Appendix B).
Additional details concerning the high-temperature gas
flow into and out of the quartz tube reactor are illustrated in
Figure 24. Basically, this design permits negligible flow path
end effects and entrance length distortions. The steady state
heat transfer effects for this reactor were studied in Appendix
C, and a very rapid temperature transition time was found for
molecules passing through the quartz tube reactor. Specifically,
Figure 25 shows the incoming gas reaching the' elevated tempera-
ture in less than 5, milliseconds. This essentially square-wave
profile of temperature versus time is accomplished through the
use of the low-temperature heat transfer principle, whereby the
incoming gas is maintained at transport temperatures right up to
the point of entry into the high-temperature region. Then by
employing fine bore connecting tubes, the linear gas velocity is
increased by a factor of five at the entrance and exit portions
of the high-temperature reactor.
This high-temperature quartz tube reactor is capable of
withstanding intense temperatures for long periods of time.
However, it can be easily damaged by some corrosive gases. For
example, the scanning electron microscope photographs shown^in
Figure 26 were obtained from various sections of quartz tubing
that were intentionally exposed to corrosive and destructive
gaseous environments. From this figure, it is observed that the
smooth quartz surface can be drastically changed by corrosive
agents. Consequently, care must be exercised with respect to
determining what samples are to be subjected to analysis by the
44
-------
HIGH -
VENT TEMPERATURE
i' INSULATION
CARRIER
4-
SAMPLE
CARRIER
+
PRODUCTS
Figure 24. Additional details of reactor assembly,
45
-------
TEMPERATURE VARIATIONS
CYCLIC OVER SEVEN
CYCLES
±2.0°(cOl2p00K
• tr = 2.00 SEC -
< 5ms
Figure 25. Square wave profile of gas phase -temperature versus
time. i '
46
-------
CLEAN QUARTZ SURFACE
(10,OOOX)
QUARTZ SURFACE AFTER ATTACK
BY HF AT HIGH TEMPERATURE
(10,OOOX)
QUARTZ SURFACE AFTER EXPOSED QUARTZ SURFACE AFTER EXPOSED
TO NF4HF~ AT 400°C FOR 4 HOURS TO NF.HF.-, AT 400°C FOR 4 HOURS
(3,OOOX) (10,OOOX)
Figure 26, Scanning electron microscope photographs of quartz
surfaces.
47
-------
TDAS. If indeed the surface of the quartz tube is attacked, and
increased surface area and adsorptive behavior result, then the
precise residence time control no longer applies. A certain
amount of sample screening is necessary before an unknown sample
can be subjected to thermal decomposition analysis by the TDAS.
COLLECTION OF EFFLUENT PRODUCTS
As the gas phase sample and its various decomposition pro-
ducts emerge from the high-temperature reactor, they are swept
downstream through the heated transfer line and into the effluent
product collection trap. This product collection trap is an
important component in the TDAS and requires considerable dis-
cussion.
As stated in Section 4, the product collection trap must
possess a slight resistance to gas flow, as should the smaller
bore transfer tubes. However, the primary function of the pro-
duct collection trap is to quantitatively capture the_unreacted
sample and the various products of thermal decomposition that
emerge from the high-temperature reactor.
The utilization of collection traps at cryogenic
temperatures has been in practice for many years in the field of
gas chromatography (70-74). Indeed, the benefits of low-
temperature trapping of samples are well known. However, there
are also numerous problems associated with low-temperature
condensing traps. One of the major difficulties is that of the
condensing out of very large quantities of water along with the
various organic compounds of interest.
Many procedures have been developed for thermally desorbing
collection traps, including cryogenic sample collections (75-79).
Some of these techniques can liberate the captured sample in a
very short period of time. It can then be passed downstream into
either a chromatograph or some other analytical instrument.
One of the most useful and intriguing sample trapping
methods was devised by Kaiser (80-83) . He employed the
properties of an adsorbent trapping media along with cryogenic
trapping for the collection of sample. He then used thermal
desorption for liberating the trapped contents. Basically,
Kaiser employed the concept of a reversible countercurrent heat
exchanger for capturing and liberating substances while in an
in-line system. This trapping technique has seen a considerable
amount of use in both in-line and discontinuous sampling systems.
The selection of trapping media is very important, and there
are many different trapping materials commercially available._
Some of the commercially available trapping media used primarily
for collecting organic compounds are as follows: carbon
48
-------
molecular sieves, graphitized carbon black, Ambersorb adsorbents,
XAD resins, chromosorb polymer series, Porapak polymer series,
Tenax GC, activated charcoal, and polyurethane foam. Studies
have been conducted concerning the physico-chemical behavior and
trapping properties for many of the adsorbent trapping media (84-
89). There are numerous applications to which these various
trapping materials have been applied (90-100). Also, to increase
trapping versatility, some of these trapping media have been
placed in series arrangements and then used in the same trappincr
device, - . . .. . c.c ^
The adsorbent trapping material that was eventually selected
for use in the TDAS was Tenax GC. There were many reasons for
selecting this particular porous polymer. Tenax GC can withstand
high temperatures before appreciable thermal degradation occurs.
High molecular weight materials can be trapped and readily
liberated by thermal desorption from this particular adsorbent.
Also, water in the gas flow path does not adversely affect the
behavior of this adsorbent to any great extent (101) . Tenax GC
does have some disadvantages relative to some of the other
trapping materials, but for the intended uses in the TDAS, it
seemed to be the best material presently available for in-line
cryogenic trapping of organic products.
Tenax GC was introduced in 1969 and was initially intended
for use as a gas chromatographic column packing. -However, it
was quickly recognized that it could also be used most
beneficially as an adsorbent trapping material. Tenax GC is a
porous polymer based on 2,6-diphenylparaphenylene oxide and is
reported to have an upper temperature limit in the vicinity of
450°C (102). Even so, for sensitive analytical work where a '
minimum of degradation or bleed can be tolerated, its upper
temperature limit is significantly below this value. Many
workers routinely condition Tenax GC traps at 325°C in a flowing
inert gas.
In the past five years, the use of Tenax GC as a trapping
medium has seen wide use. It has been used to capture and
analyze organic compounds in environmental samples (103,104),
volatile polar organics found in water (105), organic com-
ponents present in human breath (106) , pesticides, poly-
chlorinated biphenyls and polynuclear aromatics found in surface
and drinking waters (107,108), the gas phase of cigarette smoke
(109), organics in marine biota (110), and the exhaust of net
engines (111) .
If a Tenax GC trap is swept with a high velocity gas
stream while at high temperatures, high molecular weight
substances can be quantitatively liberated from the adsorbent
particles (12,112). Accordingly, this high velocity sweeping
procedure has been incorporated into the subsequent design of
the in-line trap used in the TDAS.
49
-------
A schematic of the effluent collection trap used in the
TDAS is shown in Figure 27. This design is modeled after the
trapping concept as introduced by Kaiser. However, rather than
using a bidirectional flow of nitrogen, the gas flow through the
jacket is discontinued during the thermal desorption mode. To
liberate the sample from the in-line trap, the temperature of
the aluminum heating block that surrounds the trap tubing is
linearly increased to an eventual desorption temperature of
280°C. Note in Figure 27 that the trap contents are .designated
simply as in-line trapping media. The reason for this is that
for most of the work done with the TDAS, an adsorbent porous
polymer trapping medium is used (specifically, an 80/100 mesh
Tenax GC trap as shown in Figure 28). The remainder•of the
sample collections are, obtained using a trap containing
phosphoric-acid-treated glass wool or chemically bonded packings
at cryogenic temperatures.
Very recently a problem concerning irreversible adsorption
of certain types of organic molecules by various porous polymers
has received considerable attention by analysts and environmental
scientists. Therefore, to avoid any difficulties with the very
low levels of residual adsorptivity, the Tenax GC trap is
employed for the thermal decomposition studies in which the
concentrations that go onto the trap are relatively large and
the interest concerning low-level trace components is relatively
minor. When the objective is to determine the trace level
remnants of the unreacted parent sample after exposure to very
high temperatures, a glass wool trap or a chemically bonded
packing trap is used along with cryogenic trapping temperatures.
TDU INSTRUMENTATION
Most of the measurements pertaining to the thermal
decomposition unit have been instrumented, and the various con-
trol and measurement devices* have been incorporated into an
instrumentation console. A photograph'of this instrumentation
console is presented in Figure 29, along with the descriptive;
designations of the modules that measure or control the various
operations performed by the TDU. These modules are mounted in
a Cabtron Series C modular electronic cabinet. ,,-. .
The temperature programming function for both the sample: •'
insertion process and the desorption of the collection trap is
accomplished with the use of a Theall Engineering TP-2000 _
temperature programmer. This device is capable of generating'
linear temperature programs from minus 100°C to plus 700°C. •
A switch has been installed in this module so that power can be•
*calibration data are presented in Appendix D,
50
-------
GLASS-LINED'..as. TUBING
/INSULATED S WRAPPED\
V WITH HEATING TAPE ' /
IN-LINE TRAPPING MEDIA
3i_ [—1 KNN\\y^ , — | i L.UW
T -^==3-lJ-T~
FLOW FROM: [
THERMAL i
DESTRUCTION
UNIT
N« V F M T -<- f3a
p V L. 1 M 1 ^ txjr -•
jJ ^^^'t
1 S\\ \\\ X\
' t
ALUMINUM
BLOCK ,
a :
CARTRIDGE]
HEATER 1
1
1
1
1
1
• INSULATED;
LINES !
1
LIQUID — -U
,J - - ^CHROM/
k N2GAS
INLET
(N2-!60°C)
v,
*i
r^ ^ N? GAS
SUPPLY
N2 ^^
Trapping Mode, valve V1 ls open and trap temperature approaches.- 160°C.
.Desorb.Mode: Valve V1 is closed and trap is subjected to a temperature
1 ' rate"™ °f "16°° tO +3°°°C at a Pre-selected programming
Figure 27. Schematic of collection trap assembly.
51
-------
GLASS LINED S.S. TUBING
0.16
TENAX-GC
(8tng, 8O/IOOMESH)
. GLASS WOOL
(S1LANE TREATED)
0,31
U-r- 1.5 —J 1-0 I*—'1.5 —*
18.9 '-. -
7.4
TCHMENSIONS ARE IN CENTIMETERS|
Figure 28. Adsorption trap -irisert,
52
-------
TEMPERATURE PROGRAMMING
•SAMPLE INSERTION
•.PRODUCT DESORPTION
PRESSURE AND FLOW READ OUT
PROBE TEMPERATURE/TIME CONTROL
TEMPERATURE MEASUREMENT
GAS FLOW CONTROL
f **
*•'•«« • • • »-<
m m w_4 * -m
TRANSPORT TEMPERATURE CONTROL
Figure 29. TDU Console and instrumention modules.
53
-------
applied to either the sample insertion chamber or the product
collection trap.
The pressure and flow readout module consists essentially
of a Hewlett-Packard 3465A digital multimeter. The signal
output from the mass flow transducer is in the range of volts
DC, whereas the output from the differential pressure trans-
ducer is read out in millivolts DC.
Certain low volatility samples can be inserted into the
TDU with the use of an electrically heated probe. A modified
Chemical Data Systems Pyroprobe 100 solids pyrolyzer_is used to
perform this sample insertion and subsequent volatilization. A
number of sample insertion techniques have been developed using
this type of programmed heated probe. This pyrolyzer can also
be used to study nonvolatile materials such as polymers.
The temperature measurement module consists of 18
electrically shielded thermocouple inputs, a thermocouple
selector switch (Omega MTG-OSW3-18), and two digital temperature
indicators (Omega 175-KCl and Omega 250-KC1-05). The average
temperature of the quartz tube reactor is continually monitored
by the Omega 250-KC1-05 instrument.
The gas flow control module contains the gas input , •'.•:.
connections, a carrier gas filter (Applied Science Laboratory's
Hydro-Purge), a gas switching valve (Carle Instruments'. 5511).,
needle valves that control the flow of heat transfer air in the
high-temperature reactor assembly, the variable flow controller
(Porter Instruments' VCD 1000) that controls the actual gas
flow to the reactor, a switch for the solenoid valve that
controls the nitrogen gas flow in the cryogenic trapping unit,
and the flow and pressure transducers along with their
associated electronics.
Because of its direct interaction with the high-temperature
reactor, the gas flow control module is a very crucial component,
A photograph of the interior of this isolated module is
presented in Figure 30. (Notice the positioning of the mass
flow and differential pressure sensing elements relative to the
thermometer and the mounted circular level.)
The principle behind the operation of the mass flow meter
is well established (113,114). Although there is'a number_of
different methods for measuring in-line gas flow (115), this_
particular flow measurement device has performed quite well in
laboratory studies where accurate measurements of in-line gas
flow were required (116) . A schematic showing the principle of
operation of the transducer and the simplified associated
electronic circuitry is shown in Figure 31. The mass flow
transducer is position sensitive. Therefore, a circular level
was mounted on the same flat base that supports the mass flow
54
-------
MASS FLOW
TRANSDUCER
DIFFERENTIAL
PRESSURE
TRANSDUCER
Figure 30. Interior of gas flow control module.
55
-------
f TO
CONSTANT
1Y
HEATER
FLOW-
TO
EXCITATION"
SUPPLY *l
RTD.,: RESISTANCE-TEMP
'1,2
DETECTOR
R3>4 : STABLE RESISTORS
TO
EXCITATION
SUPPLY # 2
,.(•*-) .' ,
0-5V OUT
* \ FLOW
v ' INDICATION
Figure 31. Fundamenetai schematic of mass flow transducer;
56
-------
sensing element, thus providing the means of establishing the
required perpendicularity of the device.
The differential pressure transducer used in this module is
somewhat similar to the device used in our earlier work with
the_ first generation system. However., a number of circuitry
revisions have been made that resulted in improved signal
stability. A schematic drawing of the transducer used in this
module is shown in Figure 32.
Both of these transducers have a fixed working range. The
mass flow transducer is calibrated to handle flows of air up to
50 cm-5 min L, and the differential pressure transducer is
designed for use up to 2.0 absolute atmospheres. Each of these
sensitive components has a built-in safety factor that allows
them to handle somewhat higher gas flows and pressures. However,
the recommended upper values are,as stated above. If TDAS work
is anticipated at flows and pressure beyond these limits,
different transducers would have to be substituted.
Two additional measurements are routinely made in the
course of conducting a test with the TDAS. ;One of these
measurements is ambient temperature. This temperature
measurement is made with a mercury-in-glass -thermometer, which
is located as shown previously in Figure 30.: The other important
measurement is-that of barometric, pressure, and this measurement
is taken using a Sargent-Welch S-4565 vernier rack barometer,
which reads the height of the mercury, column to the nearest 0.1
mm. ' ; ....-.,.
The transport temperature control module consists of a
number of different variacs, adjustable power supplies,
temperature controllers, and assorted fuses and switches, which
are used for heating the transport lines, gas switching valve
chamber, and the GC inlet block. These power supplies have
adjustable outputs, and the temperatures of the various heated
components are routinely measured using the temperature
measurement module. The power lines for the various heaters are
sufficiently separated from the shielded thermocouples and the
temperature measurement module.
There are
tation modules
advantageous in
to have at one
test parameters
modular concept
to the TDAS can
several advantages to housing these instrumen-
in one compact console. First, it is particularly
conducting detailed experiments with the TDAS
location the controls and readouts of the various
In addition, by employing the versatile
for the TDU instrumentation, future modifications
be readily accommodated.
57
-------
SOLA 81-05-230-1
CONTROLLED
POWER .
SUPPLY
(5.0VDC)
500ft
STATHAM STRAIN GAGE
DIFFERENTIAL PRESSURE
TRANSDUCER
PL-295TC-I5-350
DIGITAL
* VOLTMETER
• (mv)
Figure 32. Pressure transducer schematic.
58
-------
SECTION 7
CHROMATOGRAPHIC SEPARATION AND ANALYSIS
The TDAS concept relies strongly on the analysis of the
chemical compounds that emerge from the high-temperature
reactor. This analysis establishes the extent of thermal
decomposition of the parent substance and at the same time
provides information on the various thermal reaction products.
The seven stages of a typical instrumental chemical
analysis are as follows: . .
1. ; Collection . ,.. •.••>•
2. Storage :
3. Extraction , ,;. <
4. Concentration , =.
,5.......Isolation
6. Identification
i 7. Quantification
Now the primary function of the GC located in the TDAS is to
separate the constituents of the various complicated chemical
mixtures. In addition, this GC unit is also responsible for
stages 4 and 7 .of the above analysis scheme.
The GC installation within the TDAS is shown schematically
in Figure 33. The glass open tubular column (see Appendix E) and
the bypass tubing are located in the interior of a modified Pye
104 gas chromatograph (Figure 34). This particular instrument
is part of the LKB-2091 GCMS-COMP system. A high-temperature
gas switching valve has been attached to the side of this GC,
and the interior of the heated valve chamber is maintained at a
constant elevated temperature.
Water and volatile acids can present difficulties to most
glass open tubular columns. Therefore, a sample bypass feature
has been incorporated into this installation so that harmful low
molecular weight species bypass the open tubular column (OTC)
and pass on to the mass spectrometer (MS), where they can be
continually monitored. This bypass mode can also be used for
MS analysis of the low molecular weight products. The columns
that have been prepared and used in the TDAS can accept small
quantities of water without adversely affecting the performance
59
-------
GC COLUMN OVEN
HEATED
'JUNCTION
AUXILIARY
HELIUM
TRAP.
MASS
SPECTROMETER
§£ n
RESTRICJOR
s,.;.
DETECTOR
(TIC)
RESTRICTOR /
. = /
— • ^nnnrinri ~ •f-*-ii
HR-OTC L-H
CARBON -^$1
^-^
^f
!
'
MOLECULAR ; ^
SIEVE TRAP ' I
' HEATED
J VALVE
CHAMBER
CHARCOAL
TRAPS
CAPILLARY
RESTRICTORS
(VARIABLE)
VENTS
Figure 33. Schematic of GC column installation.
60
-------
SOT^,
Figure 34. Glass OTC installation in modified Pye 104
61
-------
of the column. Even so, if possible, it is desirable to keep
oxygen, water, and various harmful substances from entering the
glass capillary.
As mentioned earlier, another vital function of the TDAS
chromatographic column is to trap the condensable products that
are liberated from the effluent collection trap during the _
thermal desorption process. A chromatographic column can indeed
be an excellent trap, and if held at room temperature, it can
quantitatively capture most substances having a molecular weight
of 100 or greater. If the column oven is maintained at sub- ;
ambient temperatures with the use of auxiliary coolant (e.g.,
dry ice, etc.), much lower molecular weight species can be
successfully trapped (117-119). This thermal focusing process
(see Appendix P) is a crucial step in the sample handling with
the TDAS. Also, for the analysis of products from the TDAS,
programmed temperature GC is essential.
, ' i
The TDAS depends on high-resolution separation of the
various chemical compounds. Therefore, this necessitates the
use of glass capillary GC columns.
The gas-liquid chromatographic separation of volatile
organic compounds was first accomplished in 1951 (120) , and ,
OTC's (121-124) were introduced in 1958. The latter continue
to receive the attention of many researchers. The various
mechanisms that influence chromatographic separation have been
placed on a firm technical foundation (125), and today it is not
uncommon to find a sample containing more than a hundred
different compounds being adequately separated for qualitative
and quantitative analysis. Research work with glass OTC's is
relatively recent, although glass capillaries were introduced
as early as 1960. Much of the recent interest in glass OTC's
has come about through increased attention being devoted to .
surface chemistry and physical compatibility with ,the various
stationary liquid phases. A bibliography has been prepared_
(126) concerning some of the more recent developments in this
field of open tubular gas chromatographic columns. Also, a text
on glass OTC's has recently been published (127.). .
The various columns used in the TDAS, and those used in ,the
other supporting GC's, are fabricated in our laboratory.
Appendix E describes the procedures used, for preparing these
OTC's. Our OTC preparation procedures have been influenced
collectively by the works of many researchers (128-137) in the
field of capillary columns. The columns that are used_in the
TDAS are first subjected to a performance examination in a
separate GC instrument. Figures 35, 36, and 37 show chromato-
grams obtained by such columns. Figure 35 indicates the
separating efficiency for a series of normal alkanes; Figure 36
indicates the tremendous complexity.of some samples (e.g., a
distillation fraction of crude oil), and Figure 37 shows the
62
-------
GAS CHROMATOGRAPHlb CONDITIONS
Instrument
Analyst
Date
Varian 1860-1
W. Rubey
5-23-79
'Column:
tubing material ". ' Pyrex-'
tubing length 15 .
tubing bore ' 0. 25
-meters
Stationary Phase SE-30 Silicons
weight . - ,. .,'•','' '
Suppojt
mesh
Carrier Gas_
inlet pres
Helium
, 6
linear velocity_
outlet flow ^
Chart Advance
Read Out
Sample
size
30
•\-l.l
abs atmo *
_cm/sec •.
0.25
cm/min
1.. 0
my full scale
normal alkane mixture
0.5 ul
Detector_
range
HFID
AFS.
attenuation
64
Gas flows, ml/min
hydrogen 30
air
split ratio_^
Temperatures, °C
detector 300
injector • -200
column:
initial 35
final '.. 220-
300
40 to 1
program rate_
Sample Solvent
concentration
n-Hexane
°C/min
98
solvent
T
T
30 ... . ,.",, 60
TIME,, minutes ....
,90
Figure 35. Glass GTC separation of normal alkanes;
63
-------
. s
CU
X
•H
g
(1)
•H
o
o
(0
IH
o
o
•r-\
-P
(0
-------
uJuntojLj[jUt_Jvj
GAS CHROMATOGRAPHIC CONDITIONS
Instrument Varian 1860-1
Analyst W. Rubey
Date 5-25-79
Column:
tubing material Pyfex
tubing length 15
tubing bore Q . 25
meters
mm
Stationary Phase SP-210Q Silicone
weight ~ ~
Suppo rt • - .
mesh
Carrier Gas_
inlet pres
Helium
1.6
linear velocity_
outlet flow
25
0.25
1.0
abs atmo
cm/sec
ml/min
cm/min
Chart Advance
Read Out '
Sample Industrial Waste Mixture
size
mv full scale
0.5 ul
HFID
Detectoi
range ; IP"1
attenuation 32
AFS
Gas Flows, ml/min
hyd rogen 30
air
300
split ratio
40 to 1
Temperatures, °C
d et e cto r 3QQ
inj e cto r 260
column:
initial 35
final 2~TO~~
program rate_
Sample Solvent
concentration
"C/min
Benzene
95 i
solvent
20
40
60
—i r
100
TIME, minutes
Figure 37. Separation of complex chlorinated hydrocarbon
mixture.
65
-------
separating capability for a very complex mixture of chlorinated
hydrocarbons and other organic compounds.
The columns that are presently used in the TDAS are
relatively thin film OTC's in that the stationary phase film
thickness is on the order of 0.1 microns. This produces a
relatively low stationary phase bleed level and also permits
low temperature elution of relatively heavy compounds (138).
The one disadvantage is that these thin film OTCs have low
sample capacity. The columns that are usually used in -the TDAS
are from 10 to 20 meters long. However, the columns that are
used in the other GC's which support the TDAS operation, are
much longer; consequently, higher efficiency separations can be
obtained with these instruments. Presently, the separation
capability within the TDAS is adequate for the samples currently
being tested. In the future, other types of OTC's will also be
incorporated in the TDAS. These will contain chemically bonded
stationary phases (139,140) and other types of high-temperature
substrates.
The gas flow to and from a gas chromatograph must be
especially free of extraneous organic contaminants. _Otherwise
the chromatographic separation and subsequent analysis are
adversely influenced by these contaminants (141). Also, the
geometry of the flow path should be free of mixing chambers and
unswept regions (142). In a system as complicated as the TDAS,
there are a number of potential sources for organic contami-
nation. However, by using thermally stable materials throughout
the,gas flow path, the level of bleed contaminants can be
minimized.
With the TDAS, there is an,additional trapping procedure
that can be invoked when it is desirable to conduct an_analysis
with a very low background level. Specifically, this is
accomplished by cooling the cryogenic adsorptive trap, immediately
after the collected products have been thermally desorbed from
the trap. In short, by cooling the in-line trap to subambient
temperatures, any residual compounds that would normally be
migrating during the time period of the GC analysis would now be
captured by this cryogenic trap. These contaminants are
therefore not continuously migrating through the GC column
during the programmed temperature GC analysis. Any condensable
contaminants that are admitted to the column before or during
the thermal desorption of the product collection trap are sub-
sequently captured by the thermal focusing feature of the GC':
column before the start of the programmed temperature GC
analysis. And of course, these condensable contaminants, would
eventually be chromatographed'as typical solutes.
In the TDAS, the GC effluent is passed into an electron-^
impact source mass spectrometer, which is described briefly in
Section 8. The influence of the GCMS coupling on the
66
-------
performance of the OTC can be negligible, providing the gas
transport between the two instruments is fully understood (143-
145). In reality, the MS probably has far less influence on the
behavior of the chromatographic column than the column has on
the MS (e.g., column bleeding, etc.)
The ability to separate complex volatile mixtures is at a
relatively advanced technological state. Even so, recent
developments indicate that this already high/resolving cap-
ability will probably be advanced even further (146-149).
Therefore, if situations arise in the, future in which greater
resolution is needed in the TDAS, it would appear that such
could be readily accomplished.
67
-------
is an
SECTION 8
MASS SPECTROMETER WITH DEDICATED MINICOMPUTER
AND ASSOCIATED DATA REDUCTION EQUIPMENT
In the TDAS, the identification of the various thermal
decomposition products is accomplished through the use of_
in-line MS. As shown earlier in Figures 5 and 7, the MS i
integral and vitally important component of the TDAS. The
particular MS instrument incorporated into the TDAS is an
LKB 2091 GCMS. A photograph of this instrument along the up-
stream TDU is shown in Figure 38.
A combined GCMS is a very powerful analytical tool (46,
150, 151) and is especially applicable for analyzing complex
mixtures of organic compounds. The LKB 2091 is a single focusing
MS equipped with a 90° sector magnetic analyzer. The MS
assembly consists of one unit, housing both analyzer and con-
trols . There are individual high-capacity vacuum, systems _ for
the analyzer tube, the source, and the inlet systems. This
instrument employs the Ryhage two-stage molecular separator
(152), which is capable of operation at temperatures up to 350°C.
The MS incorporated into the TDAS uses an electron impact
source, and this instrument has a mass range of 0 to 2000 at
1.17 kV acceleration voltage. In the GC mode, the resolution
of the mass spectra is better than 700 at 3.5 kV acceleration
voltage, and at an ionization potential of 70 eV. This MS can
be automatically scanned at different rates over the full
spectrum range or parts thereof. Parabolic scanning rates range
from 1 to 240 seconds for a full-scale scan.
There are two detectors in the LKB 2091 GCMS. A total ion
current detector is used for generating a chromatographic
signal that represents the ionized effluent at the exit of the
ion source. The output from this detector is displayed on a
potentiometric recorder. The other detector is an electron
multiplier that is positioned at the outlet of the magnetic
analyzer tube. The output of the electron multiplier-
preamplifier can be displayed on a UV-recording oscillograph,
or data can be fed to the data reduction system. This instru-
ment has its own mass marker (LKB 2091-110) , which has an
accuracy of ±0.3 amu at 3.5 kV acceleration voltage. This MS
is also equipped with a direct inlet system (LKB 2091-210)
68
-------
Figure 38. LKB 2091 GCMS and TDU.
69
-------
whereby a sample can be admitted to the spectrometer through the
use of a special probe.
The data reduction equipment that supports the GCMS is an
LKB 2130 data system. A photograph of this equipment is shown
in Figure 39. The data system consists of a Digital Electronics
Corporation PDP 11/04 Minicomputer. The graphics terminal is a
Tektronix 4012, and the line printer/plotter is a Versatec 800.
Included with the data system is the LKB applications software
package and the LKB 2130-220 library.
In addition to the above data handling components, another
terminal has been incorporated into the TDAS. This additional
terminal is a Texas Instruments 733 ASR, which uses dual
cassettes. This terminal functions as an external data base
interface through which searches can be conducted of the EPA-
NIH Chemical Information System (153, 154) mass spectral library,
A photograph of this mass spectra searching terminal is shown
in Figure 40.
70
-------
Figure 39. Minicomputer for data reduction.
71
-------
Figure 40. Mass spectra searching terminal,
72
-------
SECTION 9
UTILIZATION OP THE TDAS IN THERMAL
DECOMPOSITION STUDIES
_The TDAS has been designed with the objective of conducting
precise thermal decomposition experiments in the laboratory.
This system is capable of investigating a wide variety of organic
materials and experimentally examining the effects of the major
thermal decomposition variables. The TDAS can handle relatively
pure substances and also very complex samples (e.g., multi-
component industrial organic wastes).
The most important immediate use of the TDAS is in deter-
mining the thermal decomposition behavior of toxic organic
substances. This is a very sensitive problem area in which
information is urgently needed with respect to the tremendous
variety of toxic organic materials.
With the flexibility built into the TDAS, many types of
thermal decomposition experiments can be conducted. Essentially
the various types of information that are obtained from TDAS
experiments can be grouped into four categories:
1. Establishment of the thermal decomposition profile
of a substance.
2. Determination of residence time effects and associated
kinetic behavior.
3. Investigation of different gaseous atmospheres,
pressure effects, and other related thermal :
decomposition variables.
4. Collection of data for prediction of thermal
decomposition behavior.
_The thermal decomposition profile of a substance is
obtained by subjecting a known quantity of the material to a
controlled thermal exposure. Analyses are then conducted to
determine the amount of the unreacted parent substance and also
to identify and quantitate the various thermal decomposition
products. By subjecting a sample to a series of thermal
exposures covering a broad experimental range, the thermal
73
-------
decomposition behavior of that sxi±>stance can be established.
Although the thermal decomposition behavior of the parent sub-
stance is of major importance, the detection and identification
of various thermally stable decomposition products is also of
vital concern (12). There are many instances where an organic
substance decomposes and forms other chemical species of equal
or greater toxicity (155). The major objective in obtaining a
thermal decomposition profile is to determine the thermal
exposure conditions that are necessary to produce essentially
complete destruction of the parent substance and its various
products. This profile also shows the thermal regions where
toxic products are formed (12,156). such thermal decomposition
data are of considerable value in providing guidance for large-
scale thermal disposal operations. The thermal decomposition
profile of a substance can form the basis for establishing its
safe incineration conditions (15,45). This same type of TDAS
experiment can be conducted with very complex industrial wastes,
only here the objective is simply to determine at what thermal
conditions no toxic substances are emerging in the effluent.
The effect of residence time on the extent of thermal
decomposition can be readily measured with the TDAS. This
system can also be used for a variety of reaction kinetic
studies, as the reactor incorporated into the TDU is ideally
suited for determining residence time and kinetic effects (12,
41). Some excellent data have been obtained for lower molecular
weight organic substances using a similar laboratory reactor
(43,44). -
The third group of experiments encompasses studies in
which the thermal decomposition behavior of a substance is
evaluated while it is in different 'gaseous environments. Also,
studies can be conducted where the macroscopic reaction pressure
is varied to determine the effect of pressure on the thermal
reaction.
The fourth experimental area in which the TDAS can have
broad application is in the prediction of the thermal_decom-
position behavior of different organic molecules. _This is a
promising experimental area of considerable.potential. From
laboratory data generated thus far (12,41,43), there seems to be
a good possibility that eventually thermal decomposition be-
havior can be predicted (at least in part) by simply knowing
the molecular structure of a substance (157)-. It is anticipated
that by conducting special TDAS experiments aimed at collecting
thermal decomposition data on different chemical families or
bond structures, some level of predictability, will emerge for
organic compounds of similar structure. The ability to predict
the thermal decomposition behavior of a waste substance strictly
by knowing its molecular structure would be of considerable
practical value.
74
-------
1.
REFERENCES
Paulson, E. G. How to Get Rid of Toxic Organics.
Eng./Deskbook Issue, Oct. 1977, p. 21.
Chem.
Bell, B. A., Welday, J., and Gilley, W. Alternate Methods
for the Disposal of the Pesticide Kepone. Paper presented
at the Third Annual Conference on Treatment and Disposal of
Industrial Wastewaters and Residues. April 1978, Houston,
TX.
Kennedy, M. V., Stojanovic, B. J., and Shuman, F. L., Jr.
Chemical and Thermal Aspects of Pesticide Disposal. J.
Environ. Qual., 1:63, 1972. ~
Bailin, L. J., Hertzler, B. L., and Oberacker, D. A.
Development of Microwave Plasma Detoxification Process for
Hazardous Wastes-Part 1. Environ. Sci. Tech., 12:673,
1978. ~
Oberacker, D. A., and Lees, S. Microwave Plasma Detoxifi-
cation of Hazardous and Toxic Materials. NERC Topic,
Oct. 1977.
Molten Salt Decomposes Pesticide Wastes.
News, Sept. 1977, p. 44.
Chem. and Eng.
Wastler, T. A., Offutt, C. K., Fitzsimmons, C. K., and
Des Rosiers, P. E. Disposal of Organochlorine Wastes by
.Incineration at Sea. EPA-430/9-75-014, U.S. Environmental
Protection Agency, Washington, DC, 1975.
.Paige, S. F., Baboolal, L. B., Fisher, H. J., Scheyer,
K. H. , Shaug, A. M. , Tan., R. L. , and Thorne, C. F .
Environmental Assessment: At-Sea and Land-Based Inciner-
ation of Organochlorine Wastes. EPA-600/2-78-087, U.S.
Environmental Protection Agency, Research Triangle Park,
NC, 1978.
Shih, C. C., Tobias, R. P., Clausen, J. F., and Johnson,
R. J. Thermal Degradation of Military Standard Pesticide
Formulations, Contract No. DADA 17-73-C-3132, U.S. Depart-
ment of the Army, Washington, DC, 1975.
75
-------
10,
11,
12,
Waste Disposal by Burning at Sea.
1979, p. 44.
Chem. Week, April 11,
13,
14,
15.
16.
17.
18.
19.
20,
21,
22,
Benbow, A. W., and Cullis, C. F. The Formation of Toxic
Products During the Combustion of Halogen-Containing Poly-
mers. European Polym. J. 11:723, 1975.
Duvall, D. S., and Rubey, W. A. Laboratory Evaluation of
High Temperature Destruction of Kepone and Related Pesti-
cides. EPA-600/2-76-299, U.S. Environmental Protection
Agency, Cincinnati, OH, 1976. •
Rolke, R. W., Hawthorne, R. D., Garbett, C. R., Slater,
E. R.f Phillips, T. T., and Towell, G. D. Afterburner
System Study. EPA-R2-72-062, U.S. Environmental Protection
Agency, Research Triangle Park, NC, 1972. ,
Barrett, R. E. Environmental Assessment of Afterburner
Combustion Systems. In: Proceedings of the Second
Stationary Source Combustion Symposium, p. 107. Vol.. 111.
EPA-600/7-77-073c, U.S. Environmental Protection Agency,
Research Triangle Park, NC, 1977.
Hazardous Waste: Proposed Guidelines and Regulations and
Proposal on Identification and Listing. Federal Register,
43, Dec. 1978.
Girard, L., and Harkness, A. C. Method for Analyzing the
Effluent from a Microreactor. J. of G.C. 5:516, 1967.
Keulemans, A. I. M., and Perry, S. G. Identification of
Hydrocarbons by Thermal Cracking. In: Gas Chromatography
1962. Van Swaay, M., ed. Butterworth, London, 1962, p. 356.
Kokes, R. J., Tobin, H. , Jr., and Emmett, P, H, New
Microcatalytic-Chromatographic Technique for Studying
Catalytic Reactions. J. Am^ Chem. Soc. 77: 5860, 1955.
Munday, C. W., and Primavesi, G. R. A Modular Gas-
Chromatography System for Analyzing Exit Streams from
Reactors. Analyst 88:551, 1963.
Chiu, J. Polymer Characterization by Coupled Thermogravi-
metry-Gas Chromatography. Anal. Chem. 40:1516, 1968.
Chang, T., and Mead/ T. E, Tandem Thermogravimetric
Analyzer-Gas Chromatograph-High Resolution Mass Spectro-
meter System. Anal. Chem. 43:534, 1971.
Shen, J. In-Situ Sampling System for Thermal Volatilization
Analysis of Nonvolatile Materials by Gas Chromatography-
Mass Spectrometry, Anal. Chem. 49:886, 1977.
76
-------
23.
28.
29.
30.
31.
32.
33.
34.
Johnson, D. R., and Smith, J. W. Bureau of Mines Apparatus
for Thermal Analysis: Simultaneous DTA-TG-EGA for Oil
Shale and Solid Fuels. Report of Investigations 7429, U.S.
Department of Interior, Bureau of Mines, Washington, DC,
1970.
24.
25.
26.
27.
Lodding, W., ed. Gas Effluent Analysis.
Inc., New York, NY, 1967.
Marcel Dekker,
Ayres, W. M., and Bens, E. W. Differential Thermal Studies
with Simultaneous Gas Evolution Profiles. Anal. Chem.
33:568, 1967.
Wendlandt, W. W. A New Apparatus for Simultaneous Differen-
tial Thermal Analysis and Gas Evolution Analysis. Anal.
Chem. Acta 27:309, 1962. —~—
Norem, S. D. A Combustion Device for Use in Conjunction
with Chromatographic Columns. In: Proceedings of First
ISA Symposium on Gas Chromatography. Academic Press
New York, NY, 1958, p. 191.
Steingaszner, P. Microreaction Gas Chromatographic
Techniques. In: Ancillary Techniques of Gas Chromato-
graphy, Ettre, L. S., and McFadden, W. it., eds. Wiley-
Interscience, New York, NY, 1969, p. 11.
Kwart, H., Sarner, S. F., and Olson, J. H. Homogeneous
Gas-Phase Thermolysis Kinetics. An Improved Flow Technique
for Direct Study of Rate Processes in the Gas Phase. J.
Phys. Chem. 73:4056, 1969. ~
McCarter, R. J. Apparatus for Rate Studies of Vapor Pro-
ducing Reactions. In: Proceedings of a Symposium on the
Current Status of a Thermal Analysis. National Bureau of
Standards, Washington, DC, April, 1970. ' . .
Iglauer, N., and Bentley, F. F. Pyrolysis GLC for the
Rapid Identification of Organic Polymers. J. Chrbm. Sci.
12:23, 1974. : ;
Derby, J. V., and Freedman, R. W. Vapor-Phase Pyrolysis
and GC Analysis of Fire-Retardant Materials Containing PVC.
Amer. Lab. May 1974.
Cramers, C. Pyrolysis Gas Chromatography of Volatile
Hydrocarbons. In: Gas Chromatography 1968. Elsevier Pub.,
Amsterdam, Netherlands, 1969, p. 395.
Uden, P. C., Henderson, D. E., and Lloyd, R. J. Compre-
hensive Interfaced Pyrolysis Gas Chromatographic Peak
Identification System. J. Chromatog. 126:225, 1976.
77
-------
35.
36.
37
. 38
39
40
41,
42,
43
Uden, P. C., Henderson, D. E., and Lloyd, R. J. An Inter-
faced Vapor Phase Instrumental System for Thermal Analysis
and Pyrolysis. In: Proceedings of the Third International
Symposium on Analytical Pyrolysis, Amsterdam, Elsevier
Scientific, Amsterdam, Netherlands, Sept. 1976.
Rubey, W. A. The Measurement of Concentration Zone Profile
in Elution Chromatography. Technical Report, UDRI-TR-70-45
University of Dayton Research Institute, Dayton, OH,
Dec. 1970.
Rubey, W. A. Design of a System for Simultaneous Thermo-
gravimetric-Gas Chromatographic Analysis of Stabilized
Polyacrylonitrile Fiber. Technical Report, UDRI-TR-71-28
University of Dayton Research Institute, Dayton, OH,
Sept. 1971.
Rubey, W. A., and Duvall, D. S. Description of Apparatus
for Conducting Isothermal Aging-Evolved Gas Analysis
Tests with High Temperature Polymers. Technical Report,
UDRI-TR-75-01, University of Dayton Research Institute,
Dayton, OH, Jan. 1975.
Duvall, D. S., Rubey, W. A., and Carnes, R. A. High_
Temperature Destruction of Kepone and Related Pesticides.
Report presented at 173rd American Chemical Society
National Meeting, New Orleans, LA, March 1977.
Carnes, R. A., Duvall, D. S., and Rubey, W. A. A Laboratory
Approach to Thermal Degradation of Organic Compounds.
In: Proceedings of 70th Annual Meeting of Air Pollution
Control Association, Toronto, Ontario, June 1977.
Duvall, D. S., and Rubey, W. A. Laboratory Evaluation of
High-Temperature Destruction of Polychlorinated Biphenyls
and Related Compounds. EPA-600/2-77-228, U.S. Environ-
mental Protection Agency, Cincinnati, OH, Dec. 1977.
Duvall, D. S., Rubey, W, A., and Mescher, J. A. Laboratory
Characterization of the Thermal Decomposition of Hazardous
Wastes. In: Proceedings of SWRI-EPA Conference on Dis-
posal of Hazardous Wastes, EPA-600/9-78-016, U.S. Environ-
mental Protection Agency, Cincinnati, OH, 1978.
Lee, K., Jahnes, H. J., and Macauley, D. C. Thermal
Oxidation Kinetics of Selected Organic Compounds. In: Pro-
ceedings of 71st Annual Meeting of the Air Pollution Con-
trol Association, Houston, TX, June 1978.
78
-------
44,
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
Lee, K., Hansen, J. L., and Macauley, D..S. Predictive
•-,Model of the Time-Temperature Requirements for Thermal
Destruction of Dilute Organic Vapors. In:' Proceedings of
72nd Annual Meeting of Air Pollution Control Association,
Cincinnati, OH, June 1979.
Bell, B. A., and Whitmore, F. C. Kepone Incineration
Test Program. EPA-600/2-78-108, U.S. Environmental Pro- .
tection Agency, Cincinnati, OH, 1978.
Leathard, D. A., and Shurlock, B. C. Identification Tech-
niques in Gas Chromatography. Wiley-Interscience, New
York, NY, 1970. , ..
McFadden, W.H. Introduction of Gas-Chromatographic Samples
to a Mass Spectrometer.. Sep. Sci. 1:736, 1966.
Giger, W. , Reinhard, M., Schaffner, C., and Zurcher, F.
Analyses of Organic Constituents in Water by High-Resolution
Gas Chromatography in Combination with Specific Detection
and_Computer-Assisted Mass Spectrometry. .In: Identifi-.
cation and Analyses of Organic Pollutants,in Water. Keith,
L. H., ed. Ann Arbor Press, Ann Arbor, MI, 1977, p. 433.
Gudzinowicz, B. J., Gudzinowicz, M. J.,'and Martin, /H. F.
Fundamentals of Integrated GC-MS. Marcel Dekker, New York,
NY, 1976. ......
Kirkland, J. J. Residue Analysis by Gas Chromatography.
Paper presented at ACS Analytical Summer Symposium,
Houston, TX, June, 1960.
Harris, W. E., and Habgood, H., W., Programmed Temperature
Gas Chromatography. Wiley, New York, NY, 1966.
Westhaver, J. W. Theory of Open-Tube Distillation Columns.
Ind. Eng. Chem. , Jan. 1942, p. 126. , ,.-.,
Westhaver, J. W. Concentration of Potassium 39 by Counter-
current Electro-migration: Some Theoretical Aspects of the
Operation. J. Res. NBS 38:169, 1947.
Taylor, G. Dispersion of Soluble Matter in Solvent Flowing
Slowly Through a Tube. Proc. Roy. Soc. A 219:186, 1953.
Aris, R. On the Dispersion of a Solute in a Fluid Flowing.
Through a Tube. Proc. Roy. Soc. A 235:67, 1956. .
Golay, M.J.E. Theory of Chromatography in.Open and Coated
Tubular Columns with Round and Rectangular Cross-Sections.
In: Gas Chromatography 1958. Desty, D. H., ed. Butter-
worths, London, 1958, p. 36.
79
-------
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
Bode, H. Transfer Behavior of Narrow Bore Tubes under
Laminar Flow. Chromatographia 3:113, 1970.
Giddings, J. C., and Seager, S. L. Rapid Determination of
Gaseous Diffusion Coefficients by Means of Gas Chromato-
graphic Apparatus. J. Chem. Phys. 33:1579, 1960.
Fuller, E. N., Ensley, K., and Giddings, J. C. Diffusion
of Halogenated Hydrocarbons in Helium: The Effect of
Structure on Collision Cross Sections. J. Phys. Chem.
73:3679, 1969.
Sutton, R., and Harris, W. E. Pyrolysis Gas Chromatography
of Some Hydrocarbons. Can. J. Chem. 45:2913, 1967.
Carter, M. K., Moore, M., and Nadel, F. High Efficiency
Oxidation Pyrolysis Tube for Analytical Determination of
Trace Elements. Paper presented at 1974 Pittsburg
Conference on Analytical Chemistry and Applied Spectro-
scopy.
Dean, W. R. Note on the Motion of Fluid in a Curved Pipe.
Phil. Mag. S7 4:208, 1927.
Dean, W. R. The Stream-line Motion of Fluid in a Curved
Pipe. Phil. Mag. S7 5:673, 1928.
Koutsky, J. A., and Adler, R. J. Minimization of Axial
Dispersion by Use of Secondary Flow in Helical Tubes.
Can. J. Chem. Eng. 42:239, 1964.
Tijssen, R. Effect of Column-Coiling on the Dispersion
of Solutes in Gas Chromatography. Part I: Theory.
Chromatographia 3:525, 1970.
Tijssen, R., and Wittebrood, R. T. Effect of Column-
Coiling on the Dispersion of Solutes in Gas Chromatography.
Part II: General Theory. Chromatographia 5:286, 1972.
%
Janssen, L.A.M. Axial Dispersion in Laminar Flow Through
Coiled Tubes. Chem. Eng. Sex. 31:215, 1976.
Keulegan, G. H., and Beij, K. H. Pressure Losses for
Fluid Flow in Curved Pipes. J. Res'. NBS 18:89, 1937.
Barr, W. E., and Anhorn, V. J. Scientific-and Industrial
Glass Blowing and Laboratory Techniques. Instruments
Publishing Co., Pittsburgh, PA, 1949, p. 198.
Rasmussen, R. A. A Quantitative Cryogenic Sampler.
Lab. July 1972, p. 19.
Amer.
80
-------
71. Brenner, N., and Ettre, L. S. Condensing System for Deter-
mination of Trace Impurities in Gases by Gas Chromatography,
Anal. Chem. 31:1815, 1959.
72. Drozd, J., Novak, J., and Rijks, J. A. Quantitative and
Qualitative Head-space Gas Analysis of PPB Amounts of
Hydrocarbons in Water. J.. Chroma tog. 158:471, 1978.
73. Tyson, B. J., and Carle, G. C. Gas Chromatographic
Separation of Water from Cryogenically Collected Air
Samples-Enhanced Concentration of Trace Organic Volatiles
Prior to GC-MS Analysis. Anal. Chem. 46:610, 1974.
74. Des Marais, D. J. Variable-Temperature Cryogenic Trap
for the Separation of Gas Mixtures. Anal. Chem. 50:1405,
1978.
75.
76.
77.
78.
79.
80.
81.
82.
83.
Aue, W. A.,,Vogt, C. R. H., and Younker, D. R. A Gas
Chromatographic Cartridge Desorption Port. J. Chromatog.
114:184, 1975. • "
Nawar, W. W., Sawyer, F. M., Beltran, E. G., and Pagerson,
I. S. An Injection System for Gas Chromatography. Anal.
Chem. 32:1535, 1960. ' . —
Russell, J. W. Analysis of Air Pollutants Using Sampling
Tubes and Gas Chromatography. Environ. Sci. Tech. 9:1175,
1975. ;—: .
Russell, J. W., and Shadoff, L. A. The Sampling and Deter-
mination of Halocarbons in Ambient Air Using Concentration
on Porous Polymer. J. Chromatog. 134:375, 1977.
Hopkins, B. J., and Pretorius, V. Rapid Evaporation of
Condensed Gas Chromatographic Fractions. J. Chromatog.
158:465, 1978. T
' • » ' ' •
.Kaiser, R. E. Temperature Gradient Chromatography: Part
1. Reversion Gas Chromatography. Chromatographia 1:199,
1968.
Kaiser, R. E. Ultra Trace Analysis of Volatile Substances
with GC Methods in the Sub ppm and ppb Range Down to 10~10%.
J. Chromatog. Sci. 9:227, 1971.
.Kaiser, R. E. Enriching Volatile Compounds by a Temperature
Gradient Tube. Anal. Chem. 45:965, 1973.
Kaiser, R. E. Gas Chromatography in Environmental Analysis.
.;J. Chromatog. Sci. 12:36, 1974.
81
-------
84.
85
86
87
88.
89
90
91.
92.
93.
94.
Feairheller, W., Mara, P. J., Harris, D. H., and Harris,
D. L. Technical Manual for Process Sampling Strategies
for Organic Materials. EPA-600/2-76-122, U.S. Environ-
mental Protection Agency, Research Triangle Park, NC, 1976.
*
Snyder, A. D., Hodgson, F. N., Kemmer, M. A., and
McKendree, J. R. Utility of Solid Sorbents for Sampling
Organic Emissions from Stationary Sources. EPA-600/2-76-
201, U.S. Environmental Protection Agency, Research
Triangle Park, NC, 1976.
Adams, J., Menzies, K., and Levins, P. Selection and
Evaluation of Sorbent Resins for the Collection of Organic
Compounds. EPA-600/7-77-044, U.S. Environmental Protection
Agency, Research Triangle Park, NC, 1977.
Jones, P. W., Wilkinson, J. E., and Strup, P. E. Measure-
ment of Polycyclic Organic Materials and Other Hazardous
Organic Compounds in Stack Gases: State of the Art.
EPA-600/2-77-202, U.S. Environmental Protection Agency,
Research Triangle Park, NC, 1977.
Gallant, R. F., King, J. W., Levins, P. L., and Piecewicz,
J. F. Characterization of Sorbent Resins for Use in :
Environmental Sampling. EPA-600/7-78-054, U.S. Environ- •
mental Protection Agency, Research Triangle Park, NC,'1978.
Vidal-Madjar, C., Gonnord, M. F., Benchah, F., and Guiochon,
G. Performances of Various Adsorbents for the Trapping and
Analysis of Organohalogenated Air Pollutants by Gas
Chromatography. J. Chromatog. Sci. 16:190, 1978.
Bruner, F., Ciccioli, P., and Di'Nardo, F, Use of .
Graphitized Carbon Black in Environmental Analysis., J.
Chromatog. 99:661, 1974.
Mueller, F. X., and Miller, J. A.
Vapors in Industrial Atmospheres.
Determination of Organic
Amer. Lab. May 1974.
Zlatkis, A., Kaufman, H. R., and Durbin, D. E. Carbon
Molecular' Sieve Columns for Trace Analysis in Gas
Chromatography. J. Chromatog. Sci. 8:416, 1970.
Grob, K., Grob, K., Jr., and Grob, G. Organic Substances
in Potable Water and its Precursor. J. Chromatog. 106:299,
1975.
Thomas, T. C., and Jackson, J. W. A Technique for Sampling
of 27 4-D; 2, 4, 5-T Herbicides in Air. Paper presented at
1975 Pittsburg Conference on Analytical Chemistry and
Applied Spectroscopy.
82
-------
96.
97.
98.
99.
95. Coburn, J. A., Valmanis, I. A., and Chau, A. S. Y.
Evaluation of XAD-2 for Multiresidue Extraction of Organo-
chlorine Pesticides and Polychlorinated Biphenyls from
Natural Waters. J. AOAC 60:224, 1977.
Ambersorb Carbonaceous Adsorbents: Technical Notes. Un-
published data. Rohm and Haas Co., Philadelphia, PA, 1977.
Ciccioli, P., Bertoni, G., Brancaleoni, E., Fratarcangeli,
R., and Bruner, F. Evaluation of Organic Pollutants in the
Open Air and Atmospheres in Industrial Sites Using
Graphitized Carbon Black Traps and Gas Chromatographic-Mass
Spectrometric Analysis with Specific Detectors. J. Chroma-
tog. 126:757, 1976.
Lee, K. Y., Nurok, D., and Zlatkis, A. Combined Headspace
and Extraction Technique for Profile Analysis by Capillary
•Gas Chromatography. J. Chromatog. 158:377, 1978.
Lewis, R. G., Brown, A. R., and Jackson, M. D. Evaluation
of Polyurethane Foam for Sampling of Pesticides, Poly-
chlorinated Biphenyls, and .Polychlorinated Naphthalenes in
Ambient Air. Anal. Chem. 49:1668, 1977.
100. Averill, W., and Purcell, J»E. Concentration and Analysis
of Organic Compounds from Air and Water, Paper presented
at 1978 Pittsburgh Conference on Analytical Chemistry and
Applied Spectroscopy.
101. Janak, J., Ruzickova, J., and Novak, J. Effect of Water
Vapour in the Quantitation of Trace Components Concentrated
by Frontal Gas Chromatography on Tenax-GC. J. Chromatog.
99:689, 1974. ~
102. Van Wijk, R. The Use of Poly-Para-2, 6-Diphenyl-Phenylene
Oxide as a Porous Polymer in Gas Chromatography. J.
Chromatog. Sci. 8:418, 1970. ~
103. Bursey, J. T., Smith, D,, Bunch, J. E., Williams, R. N.,
Berkley, R. E., and Pellizzari, E. D. Application of
Capillary GC/MS/Computer Techniques to Identification and
Quantitation of Organic"Components in Environmental
Samples. Amer. Lab. Dec. 1977.
104. Bellar, T. A., Lichtenberg, J. J. The Determination of
Volatile Organic Compounds at the yg/1 Level in Water by
Gas Chromatography. EPA-670/4-74-009, U.S. Environmental
Protection Agency, Cincinnati, OH, 1974,
83
-------
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
Kuo, P. P. K., Chian, E. S. K., DeWalle, F. B., and Kim,
J. H. Gas Stripping, Sorption, and Thermal Desorption
Procedures for Preconcentrating Volatile Polar Water-
Soluble Organics from Water Samples for Analysis by Gas
Chromatography. Anal. Chem. 49:1023, 1977.
Gearhart, H. L., Pierce, S. K., and Payne-Bose, D. A
Sampling Technique for Organic Components in Human Breath.
J. Chromatog. Sci. 15:480, 1977.
Leoni, V., Puccetti, G., and Grella, A. Preliminary
Results on the Use of Tenax for the Extraction of
Pesticides and Polynuclear Aromatic Hydrocarbons from
Surface and Drinking Waters for Analytical Purposes,
J. Chromatog. 106:119, 1975.
Leoni, V., Puccetti, G., Colombo, R. J., and D'Ovidio, A.
M. The Use of Tenax for the Extraction of Pesticides and
Polychlorinated Biphenyls from Water: II. Tests with
Artificially Polluted and Natural Waters. J. Chromatog.
125:399, 1976.
Zeldes, S. G., and Horton, A. D. Trapping and Determi-
nation of Labile Compounds in the Gas Phase of Cigarette
Smoke. Anal. Chem. 50:779, 1978.
Chesler, S. N., Gump, B. H., Hertz, H. S., May, W. E., and
Wise, S. A. Determination of Trace Level Hydrocarbons in
Marine Biota. Anal. Chem. 50:805, 1978.
Black, M. S., Rehg, W. R., Sievers, R. E., and Brooks,
J. J. Gas Chromatographic Technique for Compound Class
Analysis of Jet Engine Exhaust. J. Chromatog. 142:809,
1977.
Peterson, H., Eiceman, G. A., Field, L. R., and Sievers,
R. E. Gas Chromatographic Injector Attachment for the
Direct Insertion and Removal of a Porous Polymer Sorption
Trap. Anal. Chem. 50:2152, 1978.
Katys, G. P. Continuous Measurement of Unsteady Flow.
Pergamon Press, New York, NY, 1964,
Laub, J. H. An Electric Flow Meter. Electrical Engineer
Dec. 1947, p. 1216.
Grey, J., and Liu, F. F. Methods of Flow Measurement, J.
Amer. Rocket Soc. May-June 1953, p. 133.
84
-------
116. Lochmuller, C. H., and Gordon, B. M. The Determination of
Binary Gas Diffusion Coefficients Through Measurement of
the Time-Based, Chromatographic Peak Variance. J.
Chroma tog. Sc i . 16:74, 1978. ~~
117. Baumann, F., Klaver, R. F., and Johnson, J.'P. Subambient
Programmed Temperature Gas Chromatography. - In: Gas
Chromatography, 1962, Butterworths, London, 1962, p. 152.
118. Rasmussen, R. A., Harsch, D. E., Sweany, P. H., Krasnec,
J. P., and Cronn, D. R. Determination of Atmospheric
Halocarbons by a Temperature-Programmed Gas Chromatographic
Freezeout Concentration Method. J. Air Poll. Control
Assoc. 27:579, 1977. :
119. DeRuwe, H. J. J., Schulting, F. L., and Van Grondelle, J.
A Simple, Inexpensive Cryogenic Unit for Subambient Gas
Chromatographic Analysis. HRC and CC 1:211, 1978.
120. James, A. T., and Martin, A. J. P, Gas-Liquid Partition
Chromatography: The Separation and Micro-estimation of
Volatile Fatty Acids from Formic Acid to Dodecanoic Acid.
• Biochem. J. 50:679, 1952.
121. Golay, M. J. E. Theory and Practice of Gas Liquid
Partition Chromatography with Coated Capillaries. In:
Gas Chromatography. Coates, V. J., ed. Academic Press, New
York, NY, 1958, p. 1.
122. Golay, M. J. E. Theory of Chromatography in Open and Coat-
ed Tubular Columns with Round and Rectangular Cross-
Sections. In: Gas Chromatography 1958. Desty, D. H., ed.
Butterworths, London, 1958, p. 36, p. 62. '••'.'
123. Golay, M. Brief Report on Gas Chromatographic Theory.
In: Proceedings of Third International Symposium on Gas
Chromatography 1960. Edinburgh, Scotland, Butterworths,
Washington, 1960, p. 139.
124. Golay, M. Height Equivalent to a Theoretical Plate of an
Open Tubular Column Lined with a Porous Layer. Anal. Chem.
40:382, 1968.
125. Giddings, J. C. Dynamics of Chromatography: Part I -
Principles and Theory. Marcel Dekker, New York, NY, 1965.
126. Rubey, W. A. Bibliography of Open Tubular Column Gas
Chromatography. Technical Report, UDR-TR-79-11 University
of Dayton Research Institute, Dayton, OH, Jan. 1979.
127. Jennings, W. Gas Chromatography with Glass Capillary
Columns. Academic Press, New York, NY, 1978.
85
-------
128.
129.
130.
131.
132.
133.
134.
135.
136.
Sandra, P., and Verzele, M. Surface Treatment, Deacti-
vation and Coating in (GC)2 (Glass Capillary Gas Chromato-
graphy). Chromatographia 10:419, 1977.
Sandra, P., Verstappe, M., and Verzele, .M.
28, 1978.
HRC and CC 1:
137.
138.
Grob, K. The Glass Capillary Column in Gas Chromatography.
A Tool and a Technique. Chromatographia 8;423, 1975.
Schomburg, G., Dielmann, R., Hussmann, H., and Weeke, F.
Gas Chromatographic Analysis w,ith Glass Capillary Columns.
J. Chromatog. 122:55, 1976.
Novotny, M., Blomberg, L., and Bartle, K. Some Factors
Affecting the Coating of Open Tubular Columns for Gas
Chromatography. J. Chrom. Sci. 8:390, 1970.
Ettre, L., and March, E. Efficiency, Resolution and Speed
of Open Tubular Columns as Compared to Packed Columns. J.
Chromatog. 91:5, 1974.
Rijks, J., van den Berg, J., Diependaal, J. Character-
ization of Hydrocarbons in Complex Mixtures by Two-
Dimensional Precision Gas Chromatography. J. Chromatog.
91:603, 1974.
Blomberg, L. Uniformity of Stationary Phase Films in
Dynamically Coated Glass Capillary Columns for Gas
Chromatography. Chromatographia 8:324, 1975.
Onuska, F., and Comba, M. Comparison of Surface
Modification Techniques Used for the Preparation of Wall
Coated Open Tubular Glass Capillary Columns. In: Pro-
ceedings of Second International Symposium on Glass
Capillary Chromatography, Hindelang, Germany, Institute of
Chromatography, Bad Durkheim, Germany, 1977, p. 283.
Madani, C., Chambaz, E., Rigaud, M., Durand, J., and
Chebroux, P. New Method for the Preparation of Highly
Stable Polysiloxane-Coated Glass Open-Tubular Capillary
Columns and Application to the Analysis of Hormonal
Steroids. J. Chromatog. 126:161, 1976.
Grob, K., Jr., and Grob, K. Are We Using the Full Range of
Film Thickness in Capillary-GLC. Chromatographia 10:250,
1977.
86
-------
139. Madani, C., Chambaz, E. , Rigaud, M., and Bretton, J. Use
of Hydrolyzed Chlorosilanes for the Preparation of High
Resolution Glass Capillary Columns. Second International
Symposium on Glass Capillary Chromatography, Hindelang,
Germany, Institute of Chromatography, Bad Durkheim,
Germany, 1977, p. 249.
140. Blomberg, L. , Buijten, J., Gawdzik, J., and Wannman, T.
Preparation of Thermostable, Phenyl Silicons Coated, Glass
Capillary Columns for Separation of Polyaromatic Hydro-
carbons. Chromatographia 11:521, 1978.
Purcell, J. E., Downs, H. D., and Ettre, L. S. Extra-
neous Peaks in Gas Chromatography: Their Source and
Elimination. Chromatographia 8:605, 1975.
Guiochon, G. Influence of the Apparatus on the Apparent
Efficiency of Columns in Gas Chromatography. J. Gas
Chroma tog. 2:139, 1964. '-
Sellier, N., and Guiochon, G. Influence of GC-MS Coupling
on the Performances of Chromatographic Columns. J.
Chroma tog. Sci. 8:147, 1970. ~~
Eyem, J., The Role of Wall-Coated Capillary Columns in
GC-MS Techniques. Chromatographia 8:456, 1975,
Novotny, M. Coupling of Open Tubular Columns with a Mass
Spectrometer through the Jet-Type Molecular Separator.
Chromatographia 2:350, 1969.
Ducass, A., Gonnord, M. F., Arpino, P., and Guiochon, G.
Simple Techniques for Two-Dimensional Gas Chromatography.
J. Chromatog. 148:321, 1978.
Anderson, E. L., Moskot, M., and Bertsch, W. High
Resolution Columns in Gas Chromatography. Paper presented
at 174th ACS Meeting, Chicago, IL, Aug. 1977.
Cram, S. P. Extended Capabilities in Gas Chromatography
through High Resolution Techniques. Paper presented at
1979 Pittsburgh Conference on Analytical Chemistry and
Applied Spectroscopy.
149. Jennings, W, G. Food and Essential Oil Analysis with Glass
Capillary Columns. Paper presented at 1979 Pittsburgh
Conference on Analytical Chemistry and Applied Spectro-
scopy .
J
150. Fenselau, C. Gas Chromatography Mass Spectrometry: A
Report on State of the Art. Appl. Spectrosc. 28:305,
1974. —
141.
142
143,
144,
145,
146.
147
148.
87
-------
151.
152.
153.
154,
155.
156,
157
Gudzinowicz, B. J., Gudzinowicz, M. J., and Martin, H. F.
Integrated GC-MS Analytical Systems: Part I - Recent
Technological Advances Affecting Integrated GC-MS System
Utilization. Chem. Instr. 8:225, 1977.
Ryhage, R. Use of a Mass Spectrometer as a Detector and
Analyzer for Effluents Emerging from High Temperature Gas-
Liquid Chromatography Columns. Anal. Chem. 36:759, 1964.
Heller, S. R., McGuire, J. M., and Budde, W. L. Trace
Organics by GC/MS. Environ. Sci. Tech. 9:210, 1975.
Milne, G. W. A., and Heller, S. R. The MSDC/EPA/NIH Mass
Spectra Search System. Amer. Lab. Sept. 1976.
Edwards, J. B.
Trace Species.
p. 64.
Combustion: The Formation and Emission of
Ann Arbor Science, Ann Arbor, MI, 1974,
Carnes, R. A., Rubey, W. A., Mescher, J. A., and Duvall,
D. S. The Thermal Decomposition Analytical System (TDAS):
Its Development and Application. In: Proceedings of the
72nd Annual Meeting of Air Pollution Control Association,
Cincinnati, OH, June, 1979.
Hutzinger, O., Sundstrom, G., Karasek, F. W., and Safe, S.
The Chemistry of Some Potential Polyhalogenated Water
Pollutants. In: Identification and Analyses of Organic
Pollutants in Water. Keith, L. H., ed. Ann Arbor Press,
Ann Arbor, MI, 1977, p. 15.
88
-------
BIBLIOGRAPHY
Benedict, R. P. Fundamentals of Temperature, Pressure, and
Flow Measurements. Wiley, New York, N.Y., 1969.
Bird, R. B., Stewart, W. E., and Lightfoot, E. N. Transport
Phenomena. Wiley, New York, N.Y., 1960.
Dal Nogare, S., and Juvet, R. S., Jr. Gas-Liquid
Chromatography. Wiley - Interscience, New York, N.Y.,
1962 .
Danials, T. Thermal Analysis. Wiley, New York, N.Y., 1973.
Deelder, R. S., Kroll, M.G.F., and Beeren, A.J.B. Post-
Column Reactor Systems in,Liquid Chromatography.
J. Chromatog. 149:669, 1978.
Edwards, J. B. Combustion: The Formation and Emission of
Trace Species. Ann Arbor Science, Ann Arbor, Mich., 1974.
Harris, W. E., and Habgood, H. W. Programmed Temperature Gas
Chromatography. Wiley, New York, N.Y., 1966.
Hettmann, E. Chromatography. Van Nostrand Reinhold, New
York, N.Y., 1975.
Jones, C. E. R., and Cramers, C. A. Analytical Pyrolysis.
Elsevier, Amsterdam, Netherlands, 1976.
Kays, W. M. Convective Heat and Mass Transfer. McGraw-Hill
New York, N.Y., 1966. '
Keith, L. H. Identification and Analysis of Organic Pollutants
in Water. Ann Arbor Science, Ann Arbor, Mich., 1976.
Kennedy, M. V., ed. Disposal and Decontamination of Pesticides
ACS Symposium Series No. 73, The American Chemical Society
Washington, D.C., 1978.
Kreith, F. Principles of Heat Transfer. International Textbook
Co., Scranton, Pa., 1958.
Lewis, B., and Von Elbe, G. Combustion, Flames, and Explosions
of Gases. Academic Press, New York, N-.Y., 1951.
89
-------
Littlewood, A. B. Gas Chromatography. Academic Press, New
York, N.Y., 1970.
Mickley, H. S., Sherwood, T. K., and Reed, C. E. Applied
Mathematics in Chemical Engineering. 'McGraw-Hill, New
York, N.Y., 1957.
Miller, J. M. Separation Methods in Chemical Analysis. Wiley-
Interscience, New York, N.Y., 1975.
Purnell, H. Gas Chromatography. Wiley, New York, N.Y., 1962.
Reddick, H. W., and Miller, F. H. Advanced Mathematics
for Engineers. Wiley, New York, N.Y., 1955.
Reid, R. C., and Sherwood, T. K. Properties of Gases and
Liquids. McGraw-Hill, New York, N.Y., 1958.
Sternberg, J. C. Extracolumn Contributions to Chromatographic
Band Broadening. Advances in Chromatography. Vol. 2.
Marcel Dekker, New York, N.Y., 1966.
Wiedemann, H. G. Thermal Analysis. Academic Press, New York,
N.Y., 1969.
Wilkinson, R. R., Kelso, G, L., and Hopkins,. F. C. State-of-
the-Art Report: Pesticide Disposal Research. EPA-600/2-
78-13, U.S. Environmental Protection Agency, Cincinnati.,
Ohio, 1978.
90
-------
APPENDIX A
GAS FLOW ANALYSIS: A DETAILED ANALYSIS OF GAS FLOW
AND DISPERSION IN A NARROW-BORE CIRCULAR
CROSS SECTION TUBE OF FOLDED CONFIGURATION
Residence time is an important parameter in thermal decom-
position studies. To illustrate the importance of residence
time, or more emphatically the residence time distribution,
consider the two dramatically different residence time distri-
butions that are represented in Figure A-l(a). Curve I is
typical of the narrow and symmetric residence time distributions
obtained using laboratory instrumentation. Curve II is typical
of the residence time distributions associated with a large-
scale unit having a high volume throughput and a minimum of
mixing. The thermal decomposition profiles corresponding to
these two different residence time distributions would be as
depicted in Figure A-l(b). Notice that the thermal decomposition
profile corresponding to the wide and asymmetric residence time
distribution shows (1) the onset of thermal decomposition1 at
lower temperatures than the narrow residence time distribution,
and (2) a requirement of higher exposure temperatures for the
same level of efficient thermal decomposition.
Theoretically, the narrower'the residence time distribution,
the more closely the thermal decomposition profile represents the
actual thermal decomposition behavior of a given substance.
Thus in the ideal thermal decomposition experiment, each of the
molecules would be subjected to precisely the same residence
time. Also, the value of this "line-function" 'residence time
would be experimentally adjustable and measurable to a high
degree of accuracy.
In practice, these uncompromising ideal conditions are
unattainable, although accurately determined symmetric and
narrow residence time distributions can be generated in
laboratory-scale experiments. For large-scale work that in-
volves large throughputs, such narrow residence time
distributions are not presently feasible, even though progress
is being.made in this design area. Nevertheless, for purposes
of comparison, interpretation, and extrapolation, it is of vital
importance that the residence time distribution of a substance
be known along with the other thermal decomposition test
parameters.
91
-------
m
3
o
Q. co
o
LU
UJ
cc
TIME
a) RESIDENCE TIME DISTRIBUTIONS
cc
h-
UJ
o
o:
UJ
D_
h-
(3
UJ
100
10
O.I
0.01
T2 .EXPOSURE TEMPERATURE
b) THERMAL DECOMPOSITION PROFILES
Figure A-l. Effect of residence time bandwidth
92
-------
« In this appendix, a detailed transport analysis is present-
ed. The objective of this analysis is to characterize the gas
flow behavior and the residence time variation associated with a
laboratory-scale thermal decomposition analytical system.
Straight Tube of Circular'Cross Section
There are two simultaneous factors, i.e., dynamic processes,
which influence the dispersion of a dilute zone of like gas
molecules as they are being transported by a gaseous carrier
through a straight tube of circular cross section. Funda-
mentally, these two factors are convection and diffusion.
The laminar flow of gases through such a tube exhibits a
paraboloidal velocity profile; therefore, instantaneous axial
velocity depends on radial position. Concurrently, the flowing
gas will experience three dimensional diffusion according to
Pick's second law.
An equation that describes the transport behavior and the
concentration changes associated with these two effects can be
written as
3C
3t
- V7
3C
+ D
(Al)
where C is the concentration of like molecules in the transport
gas, t is time, v represents the gas axial velocity, z is axial
distance, Dg is the intermolecular diffusion coefficient, and
x and y represent the mutually perpendicular Cartesian
coordinates.
Fluid flow through a straight tube of circular cross section
can be examined by equating the applied hydrostatic force to the
opposing viscous force. Then, over an axial fluid increment dz
A dp
m ^
= (n
> dz)
s dr
(A2)
where Am is the cross-sectional area, r\ is the fluid viscosity,
PS is the surface perimeter perpendicular to the z axis, r is
radial distance, and dp/dz is the negative pressure gradient.
Accordingly, for a circular cross^section flow path
irrdp =
2trr dz)
(A3)
93
-------
civ = /r \ d£
dr \2n/ dz
(A'4)
T
•'/-.
r dr
(A5)
where r is the tube inner radius. Then,
v = —
,22'
r ~ r
_
4n / dz
(A6)
and the maximum velocity, V, can be written as
V = v
L • -fe)
(A7)
Therefore, it is seen that the axial gas velocity varies with
r according to ,
V = V 1 - f^rl (A8)
Next, the volumetric flow rate through the tube can be expressed
as
v(2frr) dr .
(A9)
and upon substituting equation A8,
F = 2Vrr
r
|r - ±" 1 dr
r
TrVr
(A10)
Thus the average linear gas velocity, v, can be expressed as
F V
v- =
(All)
m
94
-------
and equation Al can then be rewritten as
It - -^
r2
i _/ r
r2
1 o,
, 32C 32C . 32C
3z I ,..2 ...2 n_2
(A12)
3x
3z'
With the axial velocity term expressed relative to radial
distance, it is likewise advantageous to rewrite the spatial
diffusion term in cylindrical coordinates. Therefore, starting
with the following relationships: , . :.. •
x
= r cosG
(A13)
y = r sinG
(A14)
z = z
(A15)
r = Vx2 + y2
(A16)
the spatial diffusion can eventually be equated according to
1 9 9
'C_ , 3 C 3 C
2 * 2 + ,_2
3C
r* 3
N^ , a
80'
(A17)
However, for the special case involving laminar flow through a
straight tube of circular cross section, it is seen that C is
symmetrical about the z axis. Therefore,
9C_
36,
= 0
(A18)
thus
32C
32C 3 C
2~ 2~
3y 3z
32C
3C
r/ 3r
ifc
3z2
(A19)
95
-------
and finally, the basic equation can be rewritten as
32C
= lg+ 2v
IT (A20)
Prom equation A20, it is seen that
C = f(Dg,r,z,t,v,ro)
(A21)
Therefore, as v represents the axial movement of the
concentration _centroid, and so that the axial point of reference
moves with this centroid, a new coordinate, z,, is introduced
and defined as -
z 1 = z - vt
(A22)
Along with this axial coordinate change, the time coordinate
correspondingly must be revised. Therefore, since
at-j^ = 3t (A23)
and as
3t
= - v
(A24)
then by adopting this new set of coordinates, z and tn, it is
coon -f-Viu-f- J- -L
seen that
3C
3t
3C
3C
F
3t
(A25)
Hence,
3C
3t
3C
- v +
3C
3t,
(A26)
and equation A20 can be rewritten as
96
-------
3r'
32C
3z
- _
~ ^ V J-
r 9C
7T~
9z
l.
3.tn
v 1 -
?r \ ar
£L. \ O\~,
~7}^
(A27)
If now the' stipulation is made (Golay's assumption) that C never
differs greatly from the average radial concentration C~, then
where
C =
C = C + AC
/r
(
C 2-rrr dr = ~
ro
/
«'r\
C r dr
(A28)
(A29)
and
/'
«'<-\
AC r dr
(A'30)
AC « C
(A31)
As equation A28 has been introduced and defined, equation A27
can be rewritten as
D
3r'
2
:1
3C 3AC
3t,
1 -
(A32)
97
-------
and with the following assumed conditions
32C _
3r2 "
(A33)
8 AC
at
-»• 0
1 '
(A34)
32C
3z?
32AC
2
(A35)
3C
3z,
3 AC
3z,
(A36)
then, equation A32 can be rewritten as
°g
3C
9tl
32AC , /1\ 3AC , 32C
9 IT-/ 3v 0
.%„•£ \-L/ or n„^
dr \ / 3z ,
f V
/2r2\
U2 /
3C
8Z1
(A37)
Wow, if each of the terms in equation A27 is multiplied by the
operator
f
r dr
(A38)
and simplified through the use of equations A33 to A36, along
with substitution of the boundary condition
.
3r
= 0
(A39)
r=r
then the following relationship is obtained:
2v
2
f
1 -
2r"
3AC , . 3C
TT—— r dr + TT.—
(A40)
98
-------
Therefore, by subtracting equation A40 from equation A37 and
taking into account the condition set forth in equation A36 ,
then the following important equation is obtained:
Now if
32AC
3r2
+ —
3AC
v
1 -
8C
(A41)
and
3AC
32AC ' _3J_
3r2 = Sr
(A42)
(A43)
then equation A41 can be rewritten as
3r r
v
, \D
1 X -g,
3C / v \ 2r"
3z, ID / 2
1 \ g/ r
(A44)
and a solution is obtained as follows:
dr
(A45)
3C JT_
3z, ID..
3 AC
3r
(A46)
AC =
3C /v \ 1 /". 2
v^~ FT~ —^- I (rr
8Z1 \Dg/ 2r2 7
dr
(A47)
AC = AC +
o 4D
r 3C
(A48)
99
-------
Upon differentiation, the ACQ term vanishes, and
3 AC
3z.
4D
1 -
r2 \ 32C
Ji "g .
Then by substitution into equation A40,
? ; 2
2r / 3zf
o J-
(A49)
o _ 9
afc _ (v)2
"
9 —
3c
(2r4 - 5r2r2
o °
2r4) r3dr
which eventually is rewritten as
Dg"
(vro)2"
48Dg
32C
3z2
3C
-1
(A50)
(A51)
It is now apparent that equation A51 can be expressed in the
same form as Pick's second law, but with an effective diffusion
coefficient defined as
D = D +
e g
(vr )2
o
48D
(A52)
g
If now De is inserted into the one dimensional diffusion
equation, that is
d C dC
dz
dt
(A53)
then consistent with appropriate initial conditions, the
following is obtained as a solution:
C = C.
4Detx
(A54)
This equation is readily recognized as a Gaussian distribution
function. Also, it is seen to have a distribution variance of
(A55)
100
-------
Thus for the defined uniform flow path, the behavior of the
distribution variance, relative to axial distance, can be
equated as
3z
= 2D
2D
v
(A56)
and from equation A52,
2D
v
- 2
24D
(A57)
Thus equation A57 represents the longitudinal (or distance-
based) variation in transport of.like gas phase molecules.
If now the initial condition is
and from equation A57
-z=L
J r\
-»• 0
z=0
(A58)
'2D
v
dz +
L
z=L / - 2
/ vro
I —
O "O
then at distance z=L, it is seen that
\24Dg,
.dz
(A59)
2D L vr L
v
24D
(A.6.0)
Throughout the above examination, D has been assumed
to be independent of concentration and only influenced by
pressure and temperature. (For dilute binary mixtures, this is
indeed a valid assumption.) Therefore, equation A57 also
applies for the situation where like molecules are randomly •
entering and passing through the defined uniform flow path.
However, it would be more convenient if the migration variation
could be represented with a time-based independent variable.
Hence; it is necessary to convert the az value of the Gaussian
distributed population to a time-based sigma value. This is
readily accomplished as
101
-------
atF
VAm ,
(at)2 (v)2
(A61)
Equation A60 can now be rewritten as
~ 2D_L
(at)2 =
r2L
o
24VD
(A62)
and finally it is seen that the time-based variation in
residence time is
(A63)
Folded Tube of Circular Cross Section
What has been stated thus far applies strictly for laminar
gas flow through a nonreactive, smooth-walled, straight tube of
circular cross section. When the axis of the flow path exhibits
significant curvature, then additional considerations must be
included in the descriptive analysis of gas flow.
•At low gas flow velocities, both convection and diffusion
strongly contribute to axial dispersion in a narrow gas flow
path. However, when the average linear gas velocity within a
1.0 mm ID straight tube is large (v > 100 cm sec x),_then the
vastly dominant dispersion mechanism is associated with the
convection. Longitudinal diffusion represents a negligible
contribution to total dispersion at these higher gas velocities.
Thus at high linear gas velocities,
2DgL _^n (A64)
and for the case where v is large in a straight tube, equation
A63 can be written simply as
(A65)
24vD
• 102
-------
However, the velocity profile in a curved tube is not para-
boloidal. It includes an element of flow that is perpendicular
to the tube axis, and this additional element of flow is called
"secondary flow" (see Section 6, and particularly Figure 18).
This additional flow component can be beneficial in that at the
higher gas velocities, increased radial mass transfer occurs,
which thereby produces less axial dispersion.
Equation A65 applies for high velocity paraboloidal flow;
however, the general variance equation for the non-paraboloidal
velocity profile (which contains secondary flow) can be expressed
as '
2 tor L
(at) = (A66)
24vD
g
where the parameter to characterizes the actual flow profile,
which in turn depends on the axial curvature, circular bore
diameter, and surface roughness of the gas flow path, along
with the mean axial velocity and the physical properties of the
flowing gas mixture. Recent curve fitting calculations obtained
from experiments with flowing liquids have empirically expressed
to in terms of the Dean number, NDn, and the Schmidt number, NSc
(see the study by Deelder, R. S., Kroll, M.G.F., and Beeren,
A.J.B. Post-Column Reactor Systems in Liquid Chromatography.
J. Chromatog. 149 : 669 , 1978). Specifically, for values of
N_ J N_ > 12.5
Dn \ Sc
(A67)
it is observed that
= 5.6
-0.67
(A68)
When
N.
Dn
< 12.5
(A69)
then to is equal to unity.
Gas flowing through a narrow7bore tube of folded con-
figuration (such as that shown in text Figure 23) will encounter
dispersion in both the straight and curved sections of the
folded tube assembly. As the circular cross section has been
maintained throughout the entire lengt£ of the tube, the total
103
-------
residence time variance is amenable to calculation. First, the
assumption is made that for this folded tube, there are
instantaneous transitions between the laminar secondary flow and
the concentric laminar flow (i.e., pure paraboloidal velocity
profile) and vice versa. Then by incorporating into equation
A62 the variance contributed by secondary flow as expressed in
equation A66, a general residence time variance equation can be
obtained. Specifically, since independent Gaussian variances
are additive,
(A70)
and as
and
2D
(v)3
•2D
,2
"o
24vDg_,
.2
r
o
24vD
g J
(A71)
(A72)
Then, the general variance equation for this tube will be as
follows:
\ 2
1 tor
^+ —2-
24 v D
r2 L
o s
24 v D
(A73)
where Lc is the total length of curved tubing, and Ls represents
the sum of the straight lengths.
Finally, to complete this analysis of gas flow, the specific
data corresponding to the quartz tube reactor shown in Figure A-2
have been calculated. Table A-l shows the range of various
flow-related parameters that cover the intended mean residence
time extremes of 0.20 to 5.0 seconds. These calculation values
correspond to dilute levels of naphthalene in a flowing air
atmosphere at a reactor internal pressure of 1.0 absolute
atmospheres. With the narrow bore reactor and the relatively
small curvature of the flow'path, secondary flow is only a
104
-------
RACE-TRACK
CONFIGURATION
(3.5 CYCLES)
2.1 cm RADIUS
L = 100.75 cm
l_s = 80.95 cm
Lc = 19.80 cm
Vq = 0.7445 cm
r = 0.0485 cm
X = 43.3
3
*Includes effective volume of connecting tubing
Figure A-2, Quartz tube reactor data.
105
-------
TABLE A-l. CALCULATED FLUID DYNAMIC VALUES
Dimensionless
term
_Lower extreme
v = 20 cm sec
T2 = 1423°K
Upper extreme
v = 500 cm sec
T2 = 473°K
N.
N
N
Re
Dn
Sc
N
Sc
N
Dn
0.71
0.108
4.28
0.050
,0.223
134
20.3
2.97
1224
35.0
106
-------
factor at the very short residence times while at the lower
reactor temperatures.
The effects of temperature, molecular weight, and mean
residence time on the residence time variance were investigated
in a previous study (EPA-600/2-77-288, U.S. Environmental
Protection Agency, Cincinnati, Ohio, 1977, p. 59). Therefore,
in concluding this analysis, Figure A-3 represents the calcu-
lated relative residence time bandwidth as a function of mean
residence time for the quartz tube reactor described in
Figure A-2.
107
-------
40
0.3
t 0.2
0.1
1423°K
NAPHTHALENE IN AIR AT
1.0 ABS. ATMOSPHERES
MEAN RESIDENCE TIME, seconds
Figure A-3. Relative bandwidth as a function of mean residence
time.
108
-------
APPENDIX B
ERROR ANALYSIS OF MEAN RESIDENCE TIME DETERMINATION
The TDAS uses in-line instrumentation for determining the
mean residence of a sample in the high-temperature reactor.
Therefore, as presented in Section 6, equation 7 describes the
mean residence time of a substance that has passed through the
quartz tube reactor. This residence time equation
1 +
Jd
(Bl)
contains many variables, some of which have a pronounced effect
on tr and are. difficult to measure.
Upon investigation, it is found that fused quartz has a
very, _small coefficient of expansion (approximately 0.59 x 10~6)
Therefore, for a given tube, the radius r and the length L can
be assumed temperature invariant. Thus the effective volume
of a specific quartz tube, i.e.
Vg =
(B2)
remains essentially constant,
With the refined instrumentation incorporated into the
T]DAS, there are now five variables that can have an effect on
tr. Therefore, the maximum change in tr, which can be
attributed to an addition of errors in F ,
can be expressed as °
T.
and
3F~
AF
o
Ap
d
at
~3T~
AT
3T.
AT,
(B3)
109
-------
From equations Bl and B2, we find
3t
r
p
' O
V T
9
F~T9
O 2
I 1 +
(B4)
9t
V
F^To
O 2
1 +
(B5)
3t.
3T,
V T.
F T^
O 2
1 +
(B6)
V T
q
(B7)
3t
(B8)
Now the maximum errors typically associated with the selected
instrumentation are
AF0 = 1 x 10~3 cm3 sec"1
AT = 0.2°K
Apd = 2 x 10 atm
• Ap = 1 x 1O atm
o
Next, an example case is presented where
t =1.00 sec
F = 0.3287 cm sec
o
T = 296°K
o
110
-------
T2 = 773° K
Pd = 0.15 atm
PQ = 0.98 atm
V = 0.7445 cm3
By substituting into equation B3 the data for this case
and the typical measurement errors, the worst condition is
found to be
At. =
At
(F>
At
At
At
At
(B9)
At = 0.00304 + 0.00068 + 0.00129 + 0.00177 + 0.00014
Thus
At
0.00692
(BIO)
or, the maximum relative error in t is approximately 0.7%.
With this in-line instrumentation and the use of cali-
bration correction curves, accurate determinations of mean
residence time can be made between the extremes of 0.25 and 5.0
seconds. Even so, most TDAS experiments will be conducted with
the selected mean residence time falling between 0.5 and 2.0 '
seconds. Thus in this region, and with the present instru-
mentation, the measurement of gas flow represents the dominant
source of error.
Ill
-------
APPENDIX C
QUARTZ TUBE REACTOR HEAT TRANSFER ANALYSIS
In designing the fused quartz tube reactor for the TDAS,
one of the specific objectives was to minimize the end effects
(disturbances) produced by the entrance and exit regions of the
reactor. This end effect minimization was accomplished to a
considerable extent by employing the counterflow cooling
principle as shown in Figure C-l. Using this design feature,
an abrupt change in the temperature of the flowing gas _ is
experienced as,it moves from the transport region and into the
high-temperature reactor.
In view of the small bore of the gas flow path, it would be
very difficult to determine experimentally the exact temperature
rise profile as a function of either time or distance. However,
this temperature rise profile is amenable to heat transfer
calculations.
The approach used for obtaining this information was to
determine initially the bulk gas temperature within the fine
bore connecting tube as a function of longitudinal distance.
Then the effects of the longitudinal conduction within the
quartz junction assembly were examined.
First, consider the steady state heat transfer condition
v;here 550°K gas is flowing into a quartz tube that is contained
in a 900°K isothermal chamber (see Figure C-l). Of primary
importance is the time required for the incoming flowing gas to
attain effectively the 900°K equilibrium.
A simplified version of the fine bore connecting tube is
depicted in Figure C-2, and the rate of heat flow by radial
conduction, qk, into the bore of this tube can be expressed as
_
= kA
dr
_
k27rrL —
dr
(Cl)
where k is the average thermal conductivity of fused quartz, A
is radial area, T is temperature, r is radial distance, z is
112
-------
CONNECTING TUBING
70 mm by 0.43 mm ID
2.1 cm RADIUS
70 mm by 0.43 mm ID
980 mm by 0.97 mm ID
HIGH
vp..T TEMPERATURE
^ INSULATION
VENT
CARRIER
+
SAMPLE
L
CARRIER •
" • -K
• PRODUCTS
Figure c-1. Details of quartz tube reactor assembly,
113
-------
Figure C-2. 'Simplified version
of fine bore connecting tube,
114
-------
axial distance, and L is length of the tube. If conduction
the longitudinal direction is neglected, then
in
dr _ k2irL
dT
(C2)
The temperatures corresponding to the boundary conditions
are
at r = r
w w
and
(C3)
T at r = r
3 s
where the subscript s denotes the tube inner surface, and
subscript w represents the outer wall. Thus
(C4)
"w
,T
dr _ 2-rrLk
r q,.
w
dT
T
(C5)
or
27TLk(T -T
w s
r .
r~
(C6)
and therefore for the dz axial element depicted in Figure C-2,
2irk(T -T ]
w s
r
_j
r
dz
(C7)
115
-------
Next, to obtain a comparison of the heat transfer mecha-
nisms, the following steady state energy balance was con-
structed :
dq =
= dqc =
(C8)
where h is the average convective heat transfer coefficient of
the laminar flowing gas, Tfc is the bulk temperature of the gas
in the bore of the dz increment, and qc.is the rate of heat
transfer by convection. (To simplify the analysis, heat
transfer by thermal radiation is neglected.) Upon inserting
equation C7 and incorporating the appropriate ATs to maintain
the balance, the equation may be rewritten as
2frk(AT1J
dz = h 2irr (AT )dz
s *-"
(C9)
If now
1.3 Wm
(C10)
-"quartz at 900°K
and
h = 500 Wm 2 °K
(Cll)
then by evaluating the respective AT's required for maintaining
the balance for the dz element, it is found that
(C12)
Therefore, for the quartz connecting tube of Figure C-2, the
valt?y dominant mechanism of heat transfer's radial conduction
through the tubing wall. Then for the entire connecting tubing
length L, the assumption can be made that
T = T
s w
(CIS)
116
-------
An energy balance can now be written for the heat transfer
case as illustrated in Figure C-2. Specifically,
_
q = h
T -
w
T-+T,
= m c
(ci4)
where To is the bulk temperature of the gas at z=0, m is the
mass_rate of flow, and cp is the specific heat. If the flowing
gas is air, then the properties of the flowing media at 900°K
are as follows: ,
density, p = 0.393 kg m
specific heat, c = 1.12 k J kg"1 °K~1
dynamic viscosity, y = 3.9.0 x 10~5 kg m"1 s
"1 '1
and as
thermal conductivity, k = 0.0628 Wm"1 "^
Prandt number, N =0.696
= 900°K
z=0
= 550°K =
= 0.000215m
o
L = 0.070m
and the average linear gas velocity, v, is
v = 2.56m s
-1
(C17)
117
-------
then the Reynolds number is
pv2r
£. = 11.09
(CIS)
With this low NRe, along with a"constant Tw, the Nusselt
number (from Graetz's classical solution) is
h2re
. = 3.66 = • —^r
Nu .. k
(C19)
and therefore
h =
3.66k
2r_
•535 Wnf 2°K
(C20)
Also
m = pirr2v =1.461 x 10~^ kg s~
s
(C21)
Equation C14 can now be rewritten as
T -
w
T +T,
o b
T,-T
b o
•me ,
P
h2irr z
s
(C22)
which for this particular case reduces to
z =
(T,-550) 0.000453
D
1250-Tb
(C23)
118
-------
Using equation C23, a graph of Tfc, versus :z was prepared and is-.
shown in Figure C-3,. This graph reveals that the flowing bulk
air will experience a 350 °K increase in the very short axial
distance of 0.5 mm. This distance is considerably smaller than
the axial length of the quartz barrier between the two heated
zones (see the enlarged detail drawing of Figure C-l) .
Consequently, as a result of the above calculations, it is
apparent that the major factors that affect this gas phase
temperature increase are: 1) the axial length of the barrier
between the heated zones and 2) the volume flow rate through
this transition region. '..A .simplified sketch of this particular
heat transfer condition is presented in Figure C-4 .
_As was seen from Figure C-3, the axial distance required to
attain thermal equilibrium is approximately 0.5 mm. Therefore,
if this length is added to the barrier length, a worst case
situation is presented. Now if it is assumed that the average
temperature of the air in this transition region is 725 °K, and
as the interior volume, Vt, of this region is ,
zl + A2Z2
— 1.130 mm ( C 2 4 )
and the average air flow rate, F, corresponding to 725°K is
F =
725 * = 300 mm s"1
(C25)
where Vr is reactor volume and tr is'mean residence time. Then
it is seen that the time required- to displace the Vt transition
volume is • .
vt
t = -
t± 0.0038s
(C26)
or 3.8 milliseconds.
From previous measurements and calibrations, the
temperature of the central portion of the quartz tube reactor
has been well established. And now, with the calculation of
119
-------
T,°K
1000
750
500
250
0 . .
0
550
600
700
775
850
890
(mm)
0
0.0348
0.124
0.215
0.340
0.428
1.0 1.5
z , mm
2.0
Figure C-3. Graph of bulk temperature versus distance,
120
-------
550°K
900° K
0.43mm
TEMPERATURE_
TRANSITION.
REGION
Figure C-4 . Sketch of heat transfer condition,
121
-------
the temperature rise time, the gas phase temperature can be
represented relative to mean time as shown in Figure 25, which
is presented in Section 6.
An even more abrupt temperature change occurs at the exit
of the reactor. Specifically, the temperature of the gas emerg-
ing from the reactor changes almost instantaneously to the
axial transport temperature. This very rapid change is
attributed to the combination of induced mixing, greatly reduced
axial velocity, and thermal quenching.
122
-------
APPENDIX D
TDAS TRANSDUCER CALIBRATIONS
To obtain accurate measurements of the various thermal
decomposition parameters, it is necessary to have precise
measurement'devices that have been accurately calibrated.
Therefore, the individual TDAS measurement transducers have
been calibrated using measuring equipment possessing trace-
ability to the National Bureau of Standards. In this appendix,
the callibration data and associated correction curves are pre-
sented for those transducers that were described earlier in
Section 6 under the heading "TDU Instrumentation."
The differential pressure transducer calibration data are
presented in Table D-l. This device was calibrated using a
Merian Instruments 30-PAlO mercury manometer, which was cali-
brated for 22°C and 45° north latitude. The corresponding
voltage readings were made with a Fluke 8300A digital voltmeter.
The calibration data for the mass flow transducer are pre-
sented in Table D-2 and Figure D-l. This particular calibration
was conducted by the manufacturer. Both the pressure and flow
transducer outputs were alternately registered with a Hewlett-
Packard 3465A digital multimeter. Calibration data and
corrective curves for this instrument are shown in Table D-3
and Figure D-2.
Many temperature measurement calibrations were performed.
First, it was necessary to determine the longitudinal dis-
tribution of temperature throughout the reactor region. This
was accomplished using a longitudinally movable unsheathed #24
AWG chromel-alumel thermocouple, and the actual temperature
measurements throughout the reactor region are shown in
Table D-4. Using this same unsheathed thermocouple, the
temperature variations as a function of time at ,a given location
were determined to be less than ±0.5°C over a 10-minute interval,
For the actual measurement of the average temperature of the
reactor, that is T2, a special calibrated sheathed thermocouple
was purchased (Omega Engineering Inc.). This thermocouple was
calibrated by the manufacturer at temperatures commonly used;in
the TDAS-work. These temperature calibrations are shown in
Table D-5. The readout device for the T2 thermocouple was an
123
-------
TABLE D-l. DIFFERENTIAL PRESSURE.TRANSDUCER CALIBRATION*
Transducer output
(millivolts DC)
0.527
1.131
1.743
2.379
2.941
3.601
4.148
5.357
5.80
8.90
12.2
15.1
18.1
21.2
24.4
27.5
30.7
Differential
(in Hg) or
0.20
0.40
0.60
0.80
1.00 .
1.20
1.40
1.80
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
pressure
(mm Hg)
in "t A
J-U . J4
20.68
31.02
41.37
51.71
62.05
72.39 r
QT no ... .
y j . u o
103 4
155.1
206.8
258.5
310.2
361.9
413.6
465.3
517 14
1
correlation =
slope =
intercept =
. t
correlation =
slope =
intercept =
t
0.99991
0.0584
-0.0667
0.99994
0.0599
-0.3764
*Manufacturer: Statham Instruments, Inc
Model #PL-295TC-15-350 Serial #316.
124
-------
TABLE D-2. MASS FLOW TRANSDUCER CALIBRATION*
Transducer output
(volts DC)
Actual air flow
(STP, cm min )
0.005
1.250
2.500
3.750
5.000
2.500
0.006
0
12.479
24.878
37.295
49.909
24.977
0
*Manufacturer: Brooks Instrument Division
Emerson Electric Company
Model #5810-1C214A Serial #7804 H52092
AIR FLOW
CORRECTION
(cm3 min"')
03
0.2
0.1
0
-0.1
-0.2
-0.3
12 3 4
INDICATED AIR FLOW
(cm3 min"1 = VOLTS DCS-IO)
5 VOLTS DC
Figure D-l. Mass flow transducer calibration
125
-------
TABLE D-3.
CALIBRATION OF HP-3465A DIGITAL MULTIMETER
SERIAL 1521AO0452
Standard
(mv)
0.000
5.000
10.000
15.000
20.000 ,
25.000
30.000
35.000
40.000
45.000
50.000
55.000
60.000
65.000
70.000
,75.000
0.06 •
0.04 •
0.02 '
VOLTAGE ^^^^*.- _.
CORRECTION ° >^ ® ®^< ^ xv
(mv) -0.02 - " ^>~^-^-
-0.04 -
-0.06 -....,
O 25
Meter
reading .
(mv)
0.00
5.00
10.00
15.00
20.00
25.01
30.01
35.01
40.01
45.01
50.01
55.01
60.01
65.02 •
70.02
, 75.02 . . .
• i .... j
5O 75
INDICATED VOLTAGE, mv
Figure D-2. Calibration of transducer readout instrument.
126
-------
TABLE D-4. REACTOR' TEMPERATURE MEASUREMENTS
Distance from
reactor centerline
(cm)
Measured temperature (°K)
@ 599* @.903* @ 1198*
-6
-4
-2
0
+ 2
+4
+ 6
597
597
598
599
599
599
599
901
902
903
903
.,,903
903
' 902
1195
1196
1197
1198
1198
1198
1197
*Each of 3 zones of the furnace had the same dial setting
TABLE D-5. OMEGA K-l THERMOCOUPLE CHROMEL-ALUMEL
Actual
temperature
Temperature
deviation*
400
600
800
950
-0.75
-1.79
-2.24 :.; .
-0.90
*When deviation is plus, add to actual reading,
When negative, subtract from actual reading.
127
-------
Omega Model 250 digital temperature indicator. The calibration
of this instrument is presented in Table D-6 and Figure D-3.
Again, there are many temperature measurements made in the
course of a single experiment with the TDAS. With the exception
of the T2 measurement, these temperature measurements are
selectively readout on a single Omega Model 175 digital
temperature indicator. The calibration of this particular
instrument is presented in Table D-7 and Figure D-4.
128
-------
TABLE D-6.
CALIBRATION OF OMEGA DIGITAL TEMPERATURE
INDICATOR-MODEL 250, TYPE K
Standard
(°C)
0
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
(mv)
4.095
8.127
12.207
16.395
20.640
24.902
29.128
33.277
37.325
41.269
45.108
48.828
52.398
54.125
Meter
reading
(°C)
1
101
200
302
399
501
602
702
801
901
. 1002
1103
1202
1300
TEMPERATURE
CORRECTION
0
-3
-6 -
O 3OO 6OO 9OO I2OO °C
INDICATED TEMPERATURE
Figure D-3. Calibration of T2 readout instrument,
129
-------
TABLE D-7.
CALIBRATION OF OMEGA DIGITAL TEMPERATURE
INDICATOR MODEL 175, TYPE K
(°c)
-200
-100
0
100
200
300
400
500
600
700
800
900
1000
Standard
(mv)
-6.891
-4.553
0.000
4.095
8.137
12.207
16.395
20.640
24.902
29.128
33.277
37.325
41.269
Meter
reading
(°C)
-197
-101
0
100
200
298
398
498
597
696
797
898
998
6
4
2
TEMPERATURE _
CORRECTION °
-2
-4
-6
-200 0 200 400 600 800 1000
INDICATED TEMPERATURE
Figure D-4. Calibration of temperature selector readout
instrument.
130
-------
APPENDIX E
PREPARATION OF GAS CHROMATOGRAPHIC OPEN TUBULAR COLUMNS
^h(a 4-^ °f .chemical compounds can be formed upon
the thermal decomposition of an organic material. Therefore
to analyze these complicated mixtures, high-resolution gas '
chromatographic columns are generally needed.
Before a sample is subjected to tests by the TDAS , it is
?Shn?J 3 Vari°US SC'S that Contain high-resolution glass open
tubular columns. And of course, the separation columns that are
used in- the TDAS must be capable of separating complex mix?urS?
There are a number of procedures for preparing high-
resolution OTG's. The preparation procedures that follow are
those currently used in our laboratory for, fabricating OTC ' s .
These preparation procedures have evolved over a long period of
time _ and _ are Continually being updated. Our experience with
anS USe OTC'S dates back to the early 1960 's when
recent '^^^^E sSf a^cleaT
advantage -for 'borosilicate glass (Pyrex) over soda-lime (soft)
glass. Specifically, it has been found that Pyrex is a better
tubing material for high-temperature work, as there is less
stationary --phase bleed. The increased phase bleed with soft
glass has been attributed to the composition of that material
having a eatalytic effect on the thin film of stationary phase.
Consequently, at the present time, Pyrex is used almost
exclusively as the glass tubing stock for, our OTC's.
OnceUithe1 anticipated dimensions of the finished OTC have
SS 2 UP°n' ^ S±Ze °f the glass twbin9 stock ^ be
selected. For example, if the finished bore, size of the OTC
is to be .approximately 0.25 mm ID, then a; 'stock having a 2.5 mm
T«I Ca* b6Viused with a draw ratio of 100 to one. Th6i Ictual
wall thickness of the drawn tubing, is of little importance, as
SS a -i§e variety
131
-------
After selecting the glass stock, it is necessary to clean
the glass tubing before drawing. Some workers resort to cleaning
with acids; however, it has been our experience that a simple
cleaning with distilled water followed by acetone and drying
with nitrogen gas is sufficient. The clean glass stock is next
imSrted in?o" ?he glass drawing machine (Shimadzu Model GDM-1)
anfSSseguently drawn to capillary bore diameters and tubing
lengths of up to 150 meters.
After the capillary tubing has been drawn, it is necessary
to determine She actua/bore size of the finished tubing This
is accomplished by placing small sections of the tubing in a
Nikon optical microscope, Model 43622. With the use of an
optiSa? filar micrometer, Nikon Model 66728, the bore diameters
are measured to within ±0.01 mm ID. The length of the coiled
capillary tubing is determined by merely counting the coils and
multiplying by the circumference of an average coil of the drawn
tubing .
At this point in the column fabrication process, it is
necessary to select the stationary phase with which the tubing
will eventually be coated. The selection of the stationary
ohase will govern much of -the surface modification and de-
activating procedures that will be applied to the tubing surface.
For example, if a dimethylsilicone stationary phase is selected,
then surf ace roughness is a relatively minor aspect, and surface
deactivation is of major concern. These concerns are Precisely
reversed if a relatively polar stationary phase is to be coated
on the tubing wall.
In our laboratory, the non-polar stationary phases are
applied to tubing materials that have been treated with HC1 gas
at elevated temperatures. This is to etch or leach the surface
of the glass. Specifically, this procedure involves f^l^g
the tube with HC1 gas and then sealing the ends with a torch
Thl tubing is then heated in an oven at 380°C for approximately
15 hours. After removal from the oven and cooling to room
temperature, the tube is vented and immediately flushed with
nitrogen gas. Next, the tubing is mounted in the solvent
cleaning apparatus shown in Figure E-l. It is then flushed with
l-mrgSIntities of methanol and methylene chloride, respectively.
After cleaning with the solvents, the tubing is sealed and
stored until ready for the deactivation process, which will be
described later.
For medium polarity and polar stationary phases, the glass
tubing bore is modified by a technique that produces whiskers
of SS Interior surface. This is accomplished by deconjpoaxng in
the tubing a specified amount (approximately 0.3_mm cm .) of
2-chloro-l,l,2-trifluoroethyl methyl ether, or simply, etching
ether. After drawing this quantity of liquid into the central
part of the tubing length, a gentle vacuum is applied to the
132
-------
Figure E-l. Capillary cleaning apparatus'.
133
-------
ends of the tube to remove excess air, and the tubing is quickly
sealed with a torch. The tubing is next subjected'to 375°C for
a period of 15 hours. It is then vented in a hood while still
warm, and a vacuum is quickly applied to remove volatiles. Next,
the tube is flushed with oxygen at 460°C for 15 hours in a
special oven. After this, a whisker-textured surface is observed
on the tubing interior wall. Figure E-2 is a scanning electron
micrograph of these whiskers. Chemically, these whiskers are
considered to be relatively pure SiC>2. After the flushing With
oxygen at high temperature, the ends are sealed until the tube -
is ready for the deactivation step.
At present, there are just two deactivating procedures used
in our laboratory. The first involves the chemical bonding of
polyethylene glycol (Carbowax 20M) to the glass surface. This
is accomplished by passing through.the tube a 5% (w/v) solution ,
of Carbowax 20M dissolved in methylene chloride. With a flow_of
dry nitrogen gas, the solvent is evaporated and the tubing dried.
The tube is then sealed while containing a nitrogen gas atmo-
sphere. Next, the tubing is heating for 15 hours at 280°C.
After this thermal treatment, the tubing is removed front the oven
and flushed sequentially with nitrogen gas and 5 ml quantities
each of methanol and methylene chloride.
The second deactivation procedure involves the chemical
N-cyclohexyl-3-azetidinol (CHAZ). This material is dissolved
in methanol on a 5% (w/v) basis and is passed through the glass
tubing. After removing the solvent and drying with a flow of
nitrogen gas, the ends of the tube are sealed. It is then
heated for 15 hours at 125°C and followed by the same thorough
flushing and clean-up procedures as mentioned earlier with the
Carbowax 20M. These deactivation procedures are applied to
both the HCl-treated Pyrex tubing and whisker wall tubes.
The next step in the OTC preparation procedure is_very
important. It involves mounting the deactivated tube in an
oven with a special "X" tubing connection that permits the
heating of the glass tube interior while in a stream of dry
nitrogen gas (see Figure E-3). This step is performed on each
tube before it is coated with stationary pha,se. Specifically,
the tube is mounted in the oven, and a flow of approximately 1.0
ml min"1 of nitrogen gas is established at the exit of the tub-
ing. Next, the tubing is heated for approximately 3 hours at
150°C. This is to remove excess water from the tubing interior.
After the tube has returned to room temperature, it is ready for
filling with the stationary phase coating solution. It is
important that this filling process be accomplished quickly so
that moist atmospheric air will not back diffuse into the dried
tubing.
A considerable amount of attention should be given to the
selection of the stationary phase. Many workers have found that
134
-------
lOym
Figure E-2. Whisker-textured tubing bore
135
-------
RESTRICTOR
3 m by 0.25 mm
-w—•
METAL CAP
GLASS WOOL
VESPEL FERRULES
GLASS CAPILLARY
MOLECULAR SIEVE
(LINDE 5A)
FILTERED
CARRIER GAS
Figure E-3. Apparatus for drying glass tube interior,
136
-------
for a thermally stable column of high efficiency, it is pre-
ferable to have a gum-type stationary phase, as opposed to the
free flowing liquid phases. This presents some difficulties
with respect to the actual coating of the stationary phase. '
However, it has been observed that not only is there an improve-
ment in efficiency at the elevated temperatures with gum phases,
but also, these columns seem to have somewhat higher thermal
stability. In addition, there seems to be less adsorptivity with
the gum-type stationary phases. The free-flowing liquid phases
do have some advantages when one is trying to separate volatile
substances at the lower chromatographic temperature regions, such
as room temperature or less. The free-flowing coating solutions
can^be applied by the dynamic coating technique, or various
modifications therefore (e.g., the mercury piston coating pro-
cedure). However, the gum-type stationary phases are normally
best applied using the static coating procedures.
In any event, the preparation of the coating solution is an
important step. Clean and filtered solvents are necessary.
Also, the mercury used for the piston should "be free of con-
taminants. The coating solution should be freshly prepared and
placed in a Teflon screw-capped vial. When hot in use, the
coating solution should be stored in the dark.
Immediately after drying and returning to room temperature,
the tubing is ready for application of the coating solution..
For the dynamic coating procedure, approximately 10 percent of•
the tubing length is filled with the coating solution and
immediately followed with about a 4 cm length of mercury, which
acts as a piston. The tubing is filled by applying a partial '
vacuum at the exit and with the use of a rubber suction bulb
connected to the capillary by silicone rubber tubing. The inlet
end of the tubing is immersed in the coating solution, which
readily fills the first 10 percent of the tubing length. Then a
short length of clean mercury is pulled into the tube in the same
manner. The tubing is next placed in the horizontal position.
Silicone rubber tubing is attached to the tubing entrance, and
a nitrogen gas pressure is applied that will push the liquid
coating solution through the tube at an average velocity:of
approximately 0.5 cm sec-1. At the exit end of the tubing, a
short length of the same size capillary glass .tubing is applied
so that the solution coating plug will not be accelerated upon
emerging from the main capillary. After this process, the
solvent is evaporated by flowing nitrogen gas through the-tube
for approximately 1 hour. The .tube is then ready for condition-
ing. This conditioning step will be described later.
For the static coating technique, the entire tube is filled
with a relatively dilute degassed solution consisting of
stationary phase in a volatile solvent (e.g., pentane or
methylene chloride). It is then mounted on a rack, and a seal is
made at one end of the tubing by pulling into the capillary bore
137
-------
Apiezon "N". After the seal has been secured, the tubing
assembly is mounted in a chamber located within a double-door
cupboard in the temperature-controlled laboratory. Flexible
tubing from a mechanical vacuum pump is attached to the open end
of the filled tube, and the solvent is gently removed from the
tubing bore, thus depositing a known and even film thickness of
stationary phase along the tubing length. This static coating
technique is a time-consuming process and can involve many days,
depending on the length and bore diameter of the capillary
tubing. ''
After the coating of the glass tube, it is mounted in the
same oven arrangement as shown in Figure E-3 that was employed
for the drying of the glass capillary tubing. The capillary is
then purged with nitrogen gas as it is .gradually heated to its
elevated conditioning temperature. The conditioning time period
for a newly prepared OTC is usually 15 hours.
After the conditioning of a column, it is mounted in a GC
and evaluated with respect to its chromatographic performance.
If the column meets established criteria of column efficiency,
shape of elution profiles, and thermal stability,.it is removed
from the GC, and immediately the ends are sealed .with a..micro
torch. The column is then stored on a special rack contained
in a darkened cabinet. The column will remain stored in this
manner until ready for use on a routine basis in one- of the
GC's, or in the GC that is an integral part of the ,TDAS .v
138
-------
APPENDIX P - .
APPLICATION OF THERMAL FOCUSING AND PROGRAMMED
TEMPERATURE GAS CHROMATOGRAPHY IN THE TDAS
Appropriate application of temperature programming
techniques is essential to the operation of the TDAS. Programmed
increases of temperature with respect to time are applied
sequentially to the sample insertion chamber, the effluent
collection trap, and eventually the gas chromatographic column.
The transfer of thermal decomposition products from the
collection trap and into the gas chromatographic column is a very
important step and one that is crucial to the success of experi-
ments conducted with the TDAS. This vitally important transfer
is the subject of this appendix.
. Chromatography is basically a dynamic separation process -•
that_uses the concept of differential migration. However, to
obtain chromatographic separations, it is required that the un-
separated mixture be deposited in a small locale at the start
of the chromatographic process. Therefore, in the case of a
mixture of widely different chemical compounds, which are emerg-
ing _in somewhat random order from an effluent collection trap,
it is necessary that the condensable compounds be concentrated
in the inlet region of the gas chromatographic column. If the
multicomponent sample can be placed within a low temperature
region of a gas chromatographic column, it can subsequently be
efficiently eluted, provided it is initially located in the
column as a narrow longitudinal zone.
This thermal focusing process can be explained in funda-
mental migration terms. Specifically, the axial location of the
centroid of like molecules migrating through a gas chromato-
graphic column can be expressed as
/•fc ,
= / Vm CT
Jn
dt
(Fl)
139
-------
where z is the distance along the column axis, t is time, vm is
mobile phase velocity, and k is the partition ratio. The
localized velocity of the migrating solute zone can then be
written as
vs =
dz
dt
= v
m Vl+k
(F2)
Even if k remains constant through the column length, it has
been determined that the solute zone velocity experiences a
very gradual increase as the zone migrates through the column.
This small increase is due to the decompression of the carrier
gas since
v
v
m
= ftp) =
(F3)
where p is pressure, p0 is outlet pressure, ve is_mobile phase
outlet velocity, K is column permeability, and n is the carrier
gas viscosity. However, in usual GC practice, vm does not change
by more than a factor of two or three over the length of the
column. t
Now if k is initially very large and then diminishes with
time, as in PTGC where it can eventually approach zero, it is
seen from equation F2 that vs experiences a many-fold increase
with the passage of time. Therefore, if the given solute can
be deposited in the column inlet while. having a corresponding
initial k or k. such that
k. »
(F4)
where ke is the average partition ratio upon elution, it is _
evident that the initial vs is very small indeed. Thus the time
period over which the solute zone is deposited is of small con-
sequence. If ki is sufficiently large, the sample depositing
time can be tens of minutes.
To capture a solute zone as a narrow distribution in the
inlet portion of the chromatographic column, it is necessary
that there be a pronounced longitudinal temperature gradient
between the heated inlet block and the effective inlet portion
of the chromatographic column. For this in-line thermal
focusing of solute molecules to be efficient, it is important
that the thermal gradient between the heated inlet block_and
the effective region of the column inlet be an even gradient.
That is, there can be no intermediate cold spots or changes in
the sign of the gradient. If this occurs, the chromatographic
performance will definitely be affected, both qualitatively and
quantitatively.
140
-------
The selection of the liquid phase used for thermal focusing
and subsequent PTGC is also important. A good liquid phase
should have a broad liquid range. It should have low viscosity
at the initial temperature of the program, and it should also
have good thermal stability at the elevated final temperature
At no time should the stationary phase be allowed to solidify
because the solute will then be trapped by adsorption and not
liquid partitioning. As a result of such capturing of sample,
the resulting chromatographic peaks will be distorted. Also
the retention characteristics will be unknown, since the com-
ponents experience undetermined extents of both adsorption and
liquid partitioning during their, subsequent migration.
141
-------
r
APPENDIX G
APPROXIMATE COSTS OF TDU COMPONENTS
A number of companies have expressed an interest in the
TDAS and its capabilities. Representatives from the following
companies have visited the Research Institute facilities at the
University of Dayton and observed,the TDAS in various stages of
its development:
Acurex Corporation
Babcock & Wilcox Company
Dow Chemical Company . .
Goodyear Tire & Rubber Company
Tennessee Eastman Company
TRW Incorporated
Shell Development Company
Union Carbide Corporation
There have been many inquiries about the cost of assembling
a TDAS, particularly the costs of the various components that
constitute the TDU. For this reason, this appendix was compiled
showing a breakdown of the approximate costs (estimated 1978
prices) of the TDU components.
Gas purification filters
Pressure regulators, flow regulators, and flow
control valves
Tubing and tube fittings
Temperature programmed power supply
Digital voltmeter and mounting accessories
Pyrolysis probe power supply, probes, and
accessories
Console, instrumentation panels, name plates, etc,
Shielded thermocouple wire, thermocouple selector
switch, and terminals
Digital temperature readout devices
Power switch boxes •
Gas switching valves . . .. •
Variacs, powerstats, and accessories
Pressure transducer and associated electronics ,
Mass flow transducer and associated electronics
Electrical terminals, switches, and wiring
Materials and machining of support brackets
for sample inlet device
395
640
480
970
650
1300
680
390
510
130
360
115
610
645
75
400
142
-------
Fabricated heating blocks and cartridge heaters s
Heating tapes
Quartz stock !.."."!!.!..!!.. *
Quartzware fabrication costs . ] .........'..'.'.. '.
High-temperature, 3-zone furnace and'control "console ".'.'.'.
Rigid table for furnace and reactor
Glass insulating tape and adhesive glass"tape""
Vespel ferrules " "
High temperature insulation ........... "
Trapping media ...]...
Liquid nitrogen reservoir
Glass-lined stainless steel tubing '.....
Valve oven ^ ^ _
Support clamps and associated hardware .'.......'..
Calibrated sheathed thermocouple ,
Total estimated costs* •
*Minus assembly costs.
143
-------
1. REPORT NO.
EPA-600/2-80-098
TECHNICAL REPORT DATA .
(Please read Instructions on the reverse before completing)
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
DESIGN CONSIDERATIONS FOR A THERMAL
DECOMPOSITION ANALYTICAL SYSTEM
RE
August 1980 (Issuing Date]
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Wayne A. Rubey
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
University of Dayton Research Institute
300 College Park
Dayton, OH 45469
SOS#4 Taskll A73D1C
11. CONTRACT/GRANT NO.
R805117-01-0
12. SPONSORING AGENCY NAME AND ADDRESS Gin. , OH
Municipal Environmental Research Laboratory—
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
13. TYPE 01
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Richard A. Carnes (Project Officer) 513/684-4303
16'ABSTRCoTntrolled high-temperature incineration is one of the most pro- _
raising methods for obtaining safe and permanent disposal of highly toxic
organic wastes. A sophisticated laboratory system has been designed and
assembled to provide fundamental thermal decomposition data on a wide
assortment of organic materials. This thermal decomposition analytical
system (TDAS) is a specially designed, closed, continuous system_con-
sisting of a versatile and fully instrumented thermal decomposition unit
Which is connected to dedicated in-line gas chromatography, mass spec-
trometry, and data reduction equipment. The TDAS has been designed to
generate fundamental thermal decomposition data rapidly, economically,
and safely. These laboratory data should be instrumental in
establishing acceptable criteria for controlled high-temperature
incineration of toxic organic wastes.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
Organic Compounds
Degradation
Temperature
Detoxification
Recombination Reactions
Research
Retention Time
Excess Air
13B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisR
UNCLAS SIFIED
158
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
144
•h U.S. GOVERNMENT PRINTING OFFICE: 1980--657-165/0054
-------
-------
r
-------
-------
•T) O
o> -•.
3 =!:
OJ O
r-i ^
CD
m
-o
CTl
O
O
ro
c»
o
i
o
IQ
CO
C O. =5
ill
£?*
HI
m ° ~*
fi) O
ssl
i
li
3-3
o |
5 3
sf
3 01
3 |
C S
s I
5 S.
03
O
> me
3 =!' CD
O O Q.
"* 3 w
D >-«
CD 0)
5 *
0) CO
CD
O
O T O
=• 5, CD
2 O
fj -I *
cn
10
cr>
00
D
<
3
CD
D
5"
en
CD
0)
i-i
m > TI m -n TO!
TJco n ^ 5 pi
> § S. » S|
' 3 CD ™i
CO O O O
GO *< — D
01 o g
3 CD
-OCDl
D) CD |
£
------- |